oEPA
             United States
             Environmental Protection
             Agency
             Office of Drinking
             Water
             Washington DC 20460
Center for Environmental
Research Information
Cincinnati OH 45268
             Technology Transfer
                          CERI-88-23
Workshop on
Emerging
Technologies for
Drinking Water
Treatment

-------
                                         CERI-88-23
      WORKSHOP ON EMERGING TECHNOLOGIES FOR

             DRINKING WATER TREATMENT
                    April 1988
                   Sponsored by:

         U.S. EPA Office of Drinking Water
    U.S. EPA Office of Research and Development

                        and

Association of State Drinking Water Administrators

-------
                        TABLE OF CONTENTS
     Presentation
SDWA Amendments and Regulations 	   I~l
Stephen W. Clark, Chief, Technical Section, Office of
Drinking Water, US Environmental Protection Agency,
Washington, DC


State Implementation of the SDWA	II-l
Frederick A. Marrocco, Vice President, Association of
State Drinking Water Administrators and Chief,
Division of Water Supplies, Pennsylvania Department
of Environmental Resources, Harrisburg


Current and Emerging Treatment Technologies 	 III-l
Robert M. Clark, Director, Drinking Water Research
Division, US Environmental Protection Agency,
Cincinnati, OH


Barriers to Implementing New Technologies	IV-1
G. Wade Miller, Executive Director, Association of
State Drinking Water Administrators and President,
Wade Miller Associates, Inc., Arlington, VA


Technical Session: Filtration  	   V-l
Sigurd P. Hansen, Senior Engineer, CWC-HDR, Inc.,
Cameron Park, CA


Technical Session: Disinfection/By-Products  	  VI-1
Rip G. Rice, President, Rice International
Consulting Enterprises, Ashton, MD


Technical Session: Organics  	 VII-1
John E. Dyksen, Senior Project Manager, Malcolm
Pirnie, Inc., Paramus,  NJ


Technical Session: Inorganics  	   VIII-1
J. Edward Singley, Vice President, James M. Montgomery
Consulting Engineers,  Inc., Gainesville, FL

-------
SDWA Amendments and Regulations
Stephen W.  Clark, Chief,  Technical Section, Office  of Drinking
Water, US Environmental Protection Agency, Washington, DC
                                1-1

-------
EMERGING TECHNOLOGIES -
  REGULA TOR Y IMP A C TS

-------
SOW A  -- MAJOR REQUIREMENTS
       •  83 CONTAMINANTS BY JUNE 19, 1989
       •  TWO TREATMENT TECHNIQUES
             • FILTRATION (SURFACE SOURCES)
             • DISINFECTION (ALL)
       »  25 CONTAMINANTS IN 1991, 1994, . . . etc,

-------
EPA's CURRENT SCHEDULE
    • JUNE 1987 — FLUORIDE & 8 VOCs



    • DECEMBER 1987 ~ FILTRATION



    • JUNE 1988 — 40 COMPOUNDS



    • JUNE 1989 ~ 34 COMPOUNDS



    • JUNE 1991 ~ DISINFECTION & BY-PRODUCTS


-------
BASIS OF MAXIMUM CONTAMINANT
     LEVELS (MCLs)
     9 MCL GOAL (RISKS)

     • MONITORING FEASIBILITY

     • TREATMENT FEASIBILITY

     • COSTS TO METROPOLITAN SYSTEMS

     e NATIONAL IMPACTS

-------
MAJOR CONTAMINANT CATEGORIES
    •  MiCROBIALS



    •  DISINFECTION BY-PRODUCTS



    •  SYNTHETIC ORGANIC CHEMICALS (SOCs)



    *  INORGANIC CHEMICALS (!OCs)

-------
MICROB1ALS
      TREATMENT TECHNIQUES REQUIRED:
          •  MANDATED BY CONGRESS
          •  CANNOT MEASURE
          •  OUTBREAKS (POOR OR NO TREATMENT)
          •  INCREASED OUTBREAKS

-------
SURFACE WATER TREATMENT RULE --
        PERFORMANCE CRITERIA
     •  99.9% REMOVAL / INACTSVATION GIARDIA
     •  99.99% REMOVAL / KNACTIVATION VIRUSES

-------
TECHNOL OGY IMP A C TS -
     SURFACE WATER TREATMENT RULE
       INCREASED USE OF :

          • MULTIMEDIA FILTERS

          o OZONE

          • POLYMERS

-------
DISINFECTANTS / BY-PRODUCTS
    SET MCLs FOR :






     • CHLORINE             *  HALOACIDS




     « CHLORAMINES          *  HALOKETONES




     e CHLORINE DIOXIDE       *  HALOALDEHYDES




     • HALOMETHANES         •  HALOPHENOLS

-------
DISINFECTANTS / BY-PRODUCTS »
       HEALTH CONCERNS
          BLOOD
         • CENTRAL NERVOUS SYSTEM DEPRESSION
          CANCER

-------
CONTROL  APPROACHES
        REMOVE PRECURSORS
      •  ALTERNATE DISINFECTANTS
      •  REMOVE BY-PRODUCTS

-------
IMPACTS
   • AFFECTS >200M PEOPLE
   • HIGH COST

   • BALANCE MICROBIAL SAFETY

   • INCREASED USE OF :

          •  COAGULANTS / POLYMERS
          •  OZONE

-------
ORGANICS --
    HEALTH CONCERNS
     • VOCs :  LIVER, KIDNEY, CNS, CANCER

     e PESTICIDES : CNS, CANCER

     • OTHER SOCs : CNS, LIVER, CANCER

-------
BEST AVAILABLE TECHNOLOGIES (BAT)
         •  GRANULAR ACTIVATED CARBON (GAC)
           PACKED TOWER AERATION

-------
IMPACTS
    • MORE GROUND WATER TREATMENT

    * AERATION & GAG

    • ALSO;  PAG, OZONE, MEMBRANES,
             AND RESINS

-------
KEY INORGANICS (lOCs)
      • LEAD



      • NITRITE / NITRATE



      • RADIONUCLIDES (Rn, U, & Ra)



      • OTHERS (As, Se, Ba, F, etc.)i

-------
LEAD --
   HEALTH CONCERNS
         CNS
         BLOOD FORMATION
       » BLOOD PRESSURE (?)

-------
NITRITE / NITRA TE --
    HEALTH CONCERNS
     • ACUTE TOXICITY IN NEWBORNS

     • BLUE BABY SYNDROME

     • FATAL IF UNTREATED

-------
RADIONUCLIDES -
      HEALTH CONCERNS
   KNOWN HUMAN CARCINOGENS
        • RADON - LUNG
        • RADIUM - BONE

-------
BEST AVAILABLE TECHNOLOGIES
       • LEAD : CORROSION CONTROL
         NITRATE : ION EXCHANGE,        OSMOSIS
         RADON :  AERATION
         RADIUM : ION EXCHANGE,        OSMOSIS

-------
IMPACTS
   LEAD :
    • CORROSION CONTROL - UNIVERSAL
      INCREASED MONITORING
    OTHER IOCS :
     • INCREASED TREATMENT

-------
POINT OF ENTRY CRITERIA
    •  CENTRAL CONTROL



    •  EFFECTSVE MONITORING



    •  EFFECTIVE APPLICATION



    •  MECROBIAL SAFETY



    «  PROTECT ALL CONSUMERS

-------
SMALL SYSTEM TECHNOLOGIES -- e.g.,
         LIMESTONE BED (CORROSION)
          TANK AERATION (Rn & VOCs)
          SLOW SAND (FILTRATION)

-------
SUMMARY"
     REGULATORY IMPACTS
      • 83 CONTAMINANTS  BY 1989

      • DISINFECTION & FILTRATION TREATMENT

      • DISINFECTION BY-PRODUCTS BY 1991

      • FUTURE CONTAMINANTS : SOCs

-------
SUMMARY --
    TECHNOLOGY IMPACTS
        ®  MORE TREATMENT

        *  MORE USE OF OZONE, RESINS,
             MEMBRANES, etc.

        •       GROUND WATER TREATMENT

        •  SIMPLE TECHNOLOGIES (SMALL

-------
                           FACT SHEET                          5 {957
                  Drinking Water Regulations
                            under
                   1986 Amendments to SDWA


Significant directives to EPA's standard-setting program for
drinking water contaminants included in the 1986 Amendments
to the SDWA are provided below:

o EPA is to set MCLGs and National Primary Drinking Water
  Regulations for 83 specific contaminants and for any other
  contaminant in drinking water which may have any adverse
  effect upon the health of persons and which is known or
  anticipated to occur in public water systems.

o Recommended Maximum Contaminant Levels (RMCLs) are now termed
  Maximum Contaminant Level Goals (MCLGs).  No changes were made
  in the basis of an MCLG; i.e.:

    MCLGs are non-enforceable health goals which are to be
    set at the level at which no known or anticipated adverse
    effects on the health persons occur and which allows an
    adequate margin of safety.

o Maximum Contaminant Levels  (MCLs) are to be set as close to
  MCLGs as is feasible.  The definition of "feasible" was
  changed to the following:

    Feasible means with the use of the best technology,
    treatment techniques and other means, which the
    Administrator finds, after examination for efficacy
    under field conditions and not solely under laboratory
    conditions, are generally available  (taking costs into
    consideration).

  Granular Activated Carbon (GAC) is stated in the SDWA as feasible
  for the control of synthetic organic chemicals (SOCs), and any
  technology or other means found to be best available for control
  of SOCs must be at least as effective in controlling SOCs as GAC.

o MCLGs and MCLs are to be proposed at the same time and also
  promulgated simultaneously.

o MCLGs and MCLs/Monitoring requirements are to be set for 83
  contaminants listed in the SDWA.  The best available technology
  (BAT) is also to be specified for each.

  The 83 contaminants are shown in Table 1.  Seven substitutes
  are allowed if regulation of any seven other contaminants would
  be more protective of public health.  The list of substitutes
  must be proposed by June 19, 1987.

-------
                               -2-

o The timetable to produce the MCLGs and MCLs/Monitoring is as
  follows:

  - 9 by June 19, 1987

  - 40 by June 19, 1988

  - 34 by June 19, 1989

o MCLGs and MCLs/Monitoring are also to be set for other contaminants
  in drinking water that may pose a health risk.

  - The 1986 Amendments require the EPA to publish a list
    (Drinking Water Priority List) of drinking water contaminants
    that may require regulation under the SDWA.

  - The list must be published by January 1, 1988, and every 3
    years following.

  - MCLGs and MCLs/Monitoring are to be set for at least 25
    contaminants on the list by January 1, 1991.

  - MCLGs and MCLs/Monitoring are to be set for at least 25
    contaminants every 3 years following January 1, 1991 (i.e.
    1994, 1997,...)-

o Criteria must be established from which States can determine
  which surface water systems must install filtration.  The criteria
  are to be set by December 19, 1987.

o A treatment technique regulation is to be set that will require
  all public water systems to use disinfection.

  - Variances are available.  EPA will specify variance criteria.

  - The disinfection treatment rule must be promulgated by
    June 19, 1989.

o Requirements are to be set for water systems to monitor for
  unregulated contaminants.

  - Minimum monitoring frequency would be five years.

  - State can add/delete contaminants from list.

  - Monitoring regulations are to be promulgated by
    December 19,  1987.

o MCLGS and MCLs/Monitoring requirements are to be reviewed by
  EPA every three years.

-------
                               -3-


o Other requirements/provisions of the 1986 Amendments:

  - Public notification regulations are to be changed to
    provide for different types and frequencies of notice
    depending upon the potential health risk.  Final regulations
    are due September 19, 1987.

  - BAT for issuance of variances is to be set when MCLs are
    set.  BAT may vary depending upon the size of systems and
    other factors, including costs.

  - Exemptions can be extended for systems with 500 connections
    or less.  No limit is placed on the number of extensions but
    certain criteria will have to be met.

o A summary of deadlines pertinent to standard-setting is presented
  in Table 2.
               For additional information, contact:

                   Joseph A. Cotruvo, Director
                   Craig Vogt, Deputy Director
               U.S.  Environmental Protection Agency
                 Criteria and Standards Division
                 Office  of Drinking Water  (WH-550D)
                        401 M Street, S.W.
                     Washington, D.C.  20460
                           202/382-7575

-------
                            -4-
                          TABLE 1
           Contaminants Required to be Regulated
                   under the SDWA of 1986

                 Vp1atile Organic Chemicals
Trichloroethylene
Tetrachloroethylene
Carbon tetrachloride
1,1,1-Trichloroethane
1,2,-Di chloroethane
Vinyl chloride
Methylene chloride
                                    Benzene
                                    Chlorobenzene
                                    Di chlorobenzene
                                    Trichlorobenzene
                                    1,1-Dichloroethylene
                                    trans-1,2,Dichloroethylene
                                    cis-1,2,-Dichloroethylene
                 Microbiology and Turbidity
Total coliforms
Turbidity
G i a r d i a lamb1ia
Arsenic
Barium
Cadmium
Chromi um
Lead
Mercury
Nitrate
Selenium
Silver
Fluoride
Aluminum
Antimony
Endrin
Lindane
Methoxychlor
Toxaphene
2,4,-D
2,4,5-TP
Aldicarb
Chlordane
Dalapon
Diquat
Endothall
Glyphosate
Carbofuran
Alachlor
Epichlorohydrin
Toluene
Adipates
2,3,7,8-TCDD (Dioxin)
                      Inorganics
                       Organics
                      Radionuclides
                                    Viruses
                                    Standard plate count
                                    Legionella
                                    Molybdenum
                                    Asbestos
                                    Sulfate
                                    Copper
                                    Vanadium
                                    Sodium
                                    Nickel
                                    Zinc
                                    Thallium
                                    Beryllium
                                    Cyanide
                                    1,1,2-Trichloroethane
                                    Vydate
                                    Simazine
                                    PAH's
                                    PCB's
                                    Atrazine
                                    Phthalates
                                    Acrylamide
                                    Dibromochloropropane (DBCP)
                                    1,2-dichloropropane
                                    Pentachlorophenol
                                    Pichloraro
                                    Dinoseb
                                    Ethylene dibromide (EDB)
                                    Dibromomethane
                                    Xylene
                                    Hexachlorocyclopentadiene
Radium 226 and 228                  G
*e_a,particle and Ph°ton radioactivity088
                                    Radon
                                          alpha particle activity
Uranium

-------
                                 -5-
                             TABLE 2

                       Summary of Deadlines
                 for Standards under SDWA of 1986
                                                 When
     9 MCLGs and MCLs/Monitoring              June 19,  1987

     Propose Seven Substitutes                June 19,  1987

     Public Notice Revisions                 Sept. 19,  1987

     Filtration Criteria                      Dec. 19,  1987

     Monitoring for Unregulated Contaminants  Dec. 19,  1987

     List of Contaminants                     Jan. 01,  1988

     40 MCLGs and MCLs/Monitoring             June 19,  1988

     34 MCLGs and MCLs/Monitoring             June 19,  1989

     Disinfection Treatment                   June 19,  1989

     25 MCLGs and MCLs/Monitoring             Jan. 01,  1991
Status; National Primary Drinking Water Regulations


Volatile Organic Chemicals (VOCs)

 0  ANPRM March 4, 1982  (47 FR 9350)

 0  Proposed MCLGs June 12, 1984  (49 FR 24330)

    Final MCLGs, proposed MCLs, Monitoring Nov. 13, 1985

 0  November 13, 1985 Federal Register

    -  Extension of public comment period for 45 days on
       tetrachloroethylene MCLG.
    -  NTP Report recently released

 0  Public Briefing:  December 19, 1985  Washington,  D.C.

 0  Public Hearing:  January 13-14, 1986  Washington, D.C,

-------
                                 -6-
Final MCL and Monitoring June 1987;  Summary
             VOCs;  Final MCLGs and MCLs (in mg/1)
   Trichloroethylene
   Carbon Tetrachloride
   Vinyl Chloride
   1,2-Dichloroethane
   Benzene
   para-Dichlorobenzene
   1,1-Dichloroethylene
   1,1,1-Trichloroethane
Final
 MCLG*

zero
zero
zero
zero
zero
0.075
0.007
0.2
                                                        Final
                                                          MCL
0.005
0.005
0.002
0.005
0.005
0.075
0-007
0.2
  *Final MCLGs were published Nov. 13, 1985. The MCLG and MCL
   for p-dichlorobenzene were reproposed at zero and 0.005 jng/1
   on April 17, 1987; comment was requested on levels of 0.075
   mg/1 and 0.075 mg/1, respectively.
BAT for 1412 (MCLs);

 e Packed tower aeration (PTA) and granular activated carbon
   (GAC) for the eight VOCs, except vinyl chloride.

 0 PTA for vinyl chloride.  ,
BAT for 1415 (Variances);

 0 Same as BAT for 1412.
Compliance Monitoring:

 0 Initial Monitoring:   All systems must monitor each sour
   at least once in four years.

    - Surface waters:  4 quarterly samples
    - Ground waters:  4 quarterly samples; state can exempt
      systems from subsequent monitoring if no VOCs detected
      in first sample,
    - Composite samples of up to five sources allowed

-------
                                 -7-
   Phase in by system size (start monitoring:
   January 1, 1988).

      .Size             Complete by    -       • Date
    > 10,000              i year           December 31, 1988
      3300-10,000         2 years          December 31, 1989
    < 3300                4 years          December 31, 1991

   Repeat monitoring:  varies from quarterly to once per
   five years.  The frequency is based on whether VOCs are
   detected in the first round of monitoring and whether
   system is vulnerable to contamination.

Monitoring for Unregulated VOCs;

 0 Initial monitoring: all systems required to sample each
   drinking water source once for unregulated VOCs during
   a four year period.

 * Same phase-in schedules as compliance monitoring.

 0 50 VOCs specified:

    - List 1:  required for all systems (33 VOCs)
    - List 2:  required for vulnerable systems (2 VOCs)
    - List 3:  required at Stat° Discretion (15 VOCs)

 0 Repeat monitoring:  Every five years; EPA will specify
   a new list.

Analytical Methods;  GC or GC/MS

 e Methods 504, 502.1, 503.1, 524.1, 524.2, 502.2

Laboratory Certification Criteria;

 0 Seven VOCs:  + 20%   _>_ 0.004 mg/1

                + 40%   < 0.004 mg/1

 0 Vinyl Chloride:  + 40%  < 0.004 mg/1

Non-transient Non-community Water Systems (NTWS);

 0 Non-community water systems which regularly serve at
   least 25 of the same persons over 6 months per year
   (i.e., NTWS) are required to meet all requirements in
   this rule.

-------
                                 -8-
Point-of-Entry (POE), Point-of-Use (POU),  and Bottled Water;

 0 POE may be used to achieve compliance with MCLs but is
   not BAT.

 0 POU and bottled water cannot be used to meet MCLs.

Variances and Exceptions

 e As a condition of issuing a variance or exemption, states
   have the authority to require the water system to implement
   additional interim control measures.  If an unreasonable risk
   to health exists, the state must require either installation
   of point-of-use devices or distribution of bottled water to
   each customer.
IQCs, SOCs, Microbials

 0 ANPRM  October 5, 1983

 0 Proposed MCLGs, November 13, 1985,  Federal Register
   (50 FR 46936)

 0 Repoprosed MCLGs, proposed MCLs/Monitoring scheduled for
   September 1987.  Final June 1988.
                   Proposed MCLGs for SOCs
                             Existing      Proposed
  SOC                      NIPDWR (mg/1)  MCLG (mg/1)

  Acrylamide                    —           zero
  Alachlor                      —           zero
  Aldicarb,  aldicarb            —           0.009
    sulfoxide and aldicarb
    sulfone
  Carbofuran                    —           0.036
  Chlordane                     --           zero
  cis-1,2-Dichloroethylene      —           0.07
  DBCP                          —           2ero
  1,2-Dichloropropane           —           0.006

-------
                               -9-
Proposed MCLGs for SOCs continued
SOC
  Existing
NIPDWR (mg/1)
 Proposed
MCLG (mg/r)
1, 2-Dichloropropane
o-Dichlorobenzene —
2,4-D 0.1
EDB
Epichlorohydrin —
Ethylbenzene —
Heptachlor —
Heptachlor epoxide —
Lindane 0.004
Methoxychlor 0.1
Monochlorobenzene
PCBs
Pentachlorophenol —
Styrene
Toluene
2,4,5-TP 0.01
Toxaphene 0.005
trans-1 , 2-Dichloroethylene —
Xylene















0.006
0.62
0.07
zero
zero
0-68
zero
zero
0.0002
0.34
0.06
zero
0.22
0.14
2.0
0.052
zero
0-07
0.44

Proposed MCLGs for
Existing
IOC NIPDWR (mg/1)
Arsenic 0.05
Asbestos —
Barium 1«0
Cadmium 0.010
Chromium 0.05
Copper
Lead 0-05
Mercury 0.002
Nitrate 10
Nitrite
Selenium 0.01
MFL = million fibers per liter
IOCS
Proposed
MCLG mg/1
0.050
•7.1 MFL
1.5
0.005
0.12
1.3
0.020
0.003
10
1.0
0.045


-------
                               -10-
        Proposed MCLGs for Microbiological Parameters

                       Existing
  Parameter             NIPDWR           Proposed MCLG

  Total Coliforms      1-4/100 ml          Zero
  Turbidity            1-5 NTU             0-1 NTU**
  Giardia*              —                 Zero
  Viruses*              —                 Zero

  *Analytical methods do exist but they are not considered
   to be technically and economically available for
   Giardia or viruses for use in compliance monitoring.
   Are included in filtration rule.

 **Nephelometric Turbidity Unit.	 _._
Fluoride

 0 ANPRM  October 5, 1983

 0 Proposed MCLG   May 14, 1985

 " Final MCLG, Proposed MCL,  SMCL,  Monitoring  Nov.  14,  1985

 c Final MCL, SMCL,  Monitoring  April 2,  1986

       Final MCLG                  4 mg/1
       Final MCL                   4 mg/1
       Final SMCL                  2 mg/1
       Final Monitoring            1 per  year surface waters
                                   1 per  3 years ground waters
                                   Minimum repeat: 1 per 10
                                    years
Radionuclides
   ANPRM   September 30,  1986
                Draft MCLGs for Radionuclides

     Radionuclide                           Draft MCLG

      Radium 226                               zero
      Radium 228                               zero
      Uranium                                  zero

-------
                               -11-
    Draft MCLGs for Radionuclides (continued)

     Radionuclide                           Draft MCLG
      Uranium
      Radon
      Gross Alpha particle article
      Beta particle and photon radioactivity
                         zero
                         zero
                         zero
                         zero
   Proposed MCLGs/MCLs/Monitoring scheduled for January 1988
   Final December 1988.
Other IQCs and SOCs

 0 Proposed MCLGs/MCLs/Monitoring scheduled for June 1988.  Final
   June 1989.
   IOCS and SOCs:

     Methylene Chloride
     Antimony
     Endrin
     Dalapon
     Diquat
     Endothall
     Glyphosate
     Adipates
     2,3,7,8-TCDD (Dioxin)
     Trichlorobenzene
     Standard plate count
     Legionella
     Sulfate
                   Nickel
                   Thallium
                   Beryllium
                   Cyanide
                   1,1,2-Trichloroethane
                   Vydate
                   Simazine
                   PAH's
                   Atrazine
                   Phthalates
                   Pichloram
                   Dinoseb
                   Hexachlorocyclopentadiene
Substitutes and Drinking Water Priority List

 0 Proposal June 19, 1987.  Final January 1, 1988.

 ° Candidates for removal from SDWA List of 83:
       Zinc
    -  Silver
- Sodium
- Molybdenum
- Vanadium
- Dibromomethane

-------
                                 -12-
   In addition,  other candidates listed for public comment are

    -  Sulfate                   - Phthalates

   Candidates for Substitution into List of 83:
    -  Aldicarb sulfoxide
       Aldicarb sulfone
    -  Ethylbenzene
       Heptachlor
- Heptachlor epoxide
- Styrene
- Nitrite
                  Drinking Water Priority List
    Zinc
    Silver
    Sodium
    Aluminum
    Molybdenum
    Vanadium
    Dibromomethane
    Chlorine
    Hypochlorite ion
    Chlorine dioxide
    Chlorite
    Chi ora mine
    Ammonia
    Trihalomethanes (chloroform,
      dibromochloromethane,  bromo-
      dichloromethane, bromoform)
    Chlorcphenols
    Haloni tri1es
    Selected disinfection related
      chlorinated acids,  alcohols,
      aldehydes, and ketones
    Chloropicrin
    2,4-Dinitrotoluene
    1,3-Dichloropropane
    Bromobenzene
    Chioromethane
    Bromomethane
    1,2,3-Trichloropropane
    1,1,1,2-Tetrachloroethane
    Chloroethane
    2,2-Dichloropropane
    o-chlorotoluene
    p-chlorotoluene
     hexachlorobenzene
    hexachloroethane
    hexachlorobutadiene
    1,1-dichloropropene
    2,4,5-T
    Isophorone
    Ethylene thiourea
    Roron
    Strontium
    Cryptosporidium
Filtration/Col iforms

0 Surface Water Treatment Rule {filtration and Disinfection) to
  be proposed August 1987.  Draft includes following:

  General Requirements and Definitions
     Coverage:  all public water systems using any surface
                water.

-------
                              -13-
     Treatment  technique requirements  in lieu  of  MCLs  for
     Giardia,  viruses,  heterotrophic plate count  bacteria,
     Legionella,  Turbidity

     Treatment  definition is at least  99.9 percent  removal
     or inactivation of Giardia lamblia  cysts  and at  least
     99.99 percent removal or inactivation of  enteric  viruses.

     Technical  definitions are provided  for "surface"  water
     and appropriate treatment processes.

  Criteria under  which filtration is required

     Source water quality conditions under which  disinfection
     only is considered safe (fecal  coliform concentrations
     less than 20 per 100 ml or total  coliforms  less  than
     100 per 100  ml in 90 percent of the samples  and  a
     maximum turbidity level of less than  5 NTU).

     Site-specific conditions to be  met  including disinfectant
     levels, monitoring, watershed management  programs,
     sanitary surveys,  waterborne disease  outbreak  history,
     total coliform and total trihalomethane MCL  compliance
     history.

     Primacy Agency must determine that  all conditions are
     met or filtration is required.

0 Disinfection requirements for unfiltered water  supplies

     All systems  must disinfect.

     Performance  criteria:  achieve  99.9 percent  inactivation
     of Giardia lamblia cysts and 99.99  percent  inactivation
     of enteric viruses (as appropriate).

     Operating criteria:  disinfectant concentration  and
     disinfectant contact times required to obtain  the
     inactivation rates specified in the performance  criteria
     are published in the rule.

     Design criteria:  redundancy of components,  and
     auxiliary power supply.

     Monitoring requirements:  continuously monitor
     disinfectant concentration at the plant and  demonstrate  a
     disinfectant residual concentration in 95 percent of the
     distribution system samples.

* Disinfection requirements for filtered water supplies

     Disinfection with filtration must achieve at least
     99.9 percent and 99.99 percent  removal/inactivation of

-------
                                -14-
    Giardia  and  viruses,  respectively.   Primacy agency defines
    level  of disinfection required, depending on  technology  and
    source water quality.

    Monitoring requirements:   continuously monitor
    disinfectant concentration at  the plant and demonstrate
    a  disinfectant  residual concentration in 95 percent of
    the  distribution  system samples.

  Filtration  requirements

  -  Filtration must be  installed when the system  fails to meet
    source water quality  or site-specific conditions.

  -  Performance  requirements  in terms of turbidity  levels are
    set  for  commonly  used filtration technologies (e.g./
    conventional, direct,  slow sand and  diatomaceous earth
    filtration technologies).

    Other  technologies  (including  emerging and innovative
    technologies) can be  approved  by the Primacy  Agency if they
    achieve  the  performance criteria for Giardia  lamblia and
    virus  removal/inactivation.

  Monitoring  and  Analytical Requirements

    EPA  approved methods  for  total coliforms, fecal coliform,
    turbidity, disinfectant residual, temperature,  and pH
    are  proposed.
     Specified frequency of  monitoring  for  each parameter  and
     numbers  of analyses required,  varies according  to population
     C O T- \T Ct rl .
   Specified
   numbers
   served.

Reporting, Public Notice and Recordkeeping
     Requirements  include monthly  reports  to  the  Primacy Agency
     on  coliform concentrations, turbidity levels,  disinfectant
     concentration,  pH,  water  temperature,  disinfectant contact
     time and disinfectant residuals.

     Unfiltered water  systems  must also  report  annually on
     their watershed control program and sanitary surveys.

     Water-borne outbreaks must be reported within 48  hours.

c  Violations

  -  All systems with  unfiltered surface water  sources must
     meet source water quality and site-specific  conditions
     within 48 months  of promulgation.   If they fail  to
     meet these criteria within  30 months,  filtration  would
     be  required,  but  they would not be  in violation  until
     failing to meet such criteria after 48 months.

-------
                               -15-


      Monitoring requirements for unfiltered, raw water
      quality must be met within 30 months of promulgation.

      Performance criteria and monitoring/reporting requirements
      for the filtered and disinfection treatment techniques
      must be met within 48 months of promulgation.

Total Coliform MCL

 * Maximum contaminant level goal—zero

 e Maximum contaminant levels

      Based on presence or absence of total coliforms in sample,
      rather than density.

   -  Monthly MCL

          No more than 1 coliform-positive sample/month for
          systems which analyze fewer than 40 samples/month.

          No more than 5% of samples can be coliform-
          positive if system analyzes at least 40 samples/
               month.

      Long-term MCL

          No more than 5% of most recent 60 samples coliform-
          positive if system analyzes fewer than 60 samples/
          year.

          No more than 5% of all samples in most recent
          12-month period can be coliform-positive if system
          analyzes at least 60 samples/year.
 0 Monitoring frequency

      For systems serving 3,300 persons or fewer:

          5 samples/month, with less monitoring for systems
          which:

          (a)  filter and disinfect surface water and
               disinfect ground water; and

          (b)  have a sanitary survey at the frequency
               specified in the proposed regulation.

      For systems serving more than 3,300 persons:

          Based on population served

          Similar to current minimum monitoring requirements
          for coliforms, but with smaller number of population
          categories.

-------
                            -16-
Repeat samples

   If system has coll form-positive sample,  system must
   collect five repeat samples all on the same day from
   same location as original sample,  except some may be
   from the next service connection.

   If any repeat sample is coliform-positive,  system
   must:

   —  Analyze positive culture medium to determine if it
       contains fecal coliforms; and

   —  Collect and analyze another set of five repeat
       samples all on the same day,  unless  the MCL has
       been violated and the system has notified the
       State-

Heterotrophic bacteria (HPC)

   No MCLG/MCL is proposed.

   Proposed regulation based on HPC interference with
   total coliform analysis and considers effectiveness
   of disinfection and filtration controlling HPC levels.

   If coliform sample produces a turbid culture in the
   absence of gas production, using the Multiple-Tube
   Fermentation Technique or produces confluent growth or
   a colony number that is "too numerous to count" using
   the membrane filter technique, the system may either
   accept the sample as coliform-positive or declare the
   sample invalid and collect and analyze another water
   sample.  Second sample is analyzed for both total
   coliforms and HPC.  If HPC is greater than 500
   colonies/ml, then sample is considered coliform-
   positive, even if total coliform analysis is negative.

Variances and exemptions—none allowed

Waterborne disease outbreaks—States  must search for and
investigate possible occurrences of such outbreaks.

-------
State Implementation of the SDWA
Frederick  A.  Marrocco,  Vice President,  Association  of  State
Drinking  Water  Administrators  and  Chief,   Division  of  Water
Supplies,  Pennsylvania  Department  of  Environmental  Resources,
Harrisburg

Handout materials not available.
                               II-l

-------
Current and Emerging Treatment Technologies
Robert M.  Clark,  Director,  Drinking Water  Research Division, US
Environmental Protection Agency, Cincinnati, OH
                              III-l

-------
Treatment Techniques   (Robert  M.  Clark,  Director, Drinking Water
Research  Division,  US  Environmental  Protection  Agency,  Cincin-
nati, OH) .

-------
     The Safe Drinking Water Act (SDWA) of 1974 requires  that EPA establish

maximum contaminant level goals (MCLG) for each contaminant in drinking water

which may have an adverse effect on the health of persons.  Each Goal is

required to be set at a level at which no known or anticipated adverse

effects on health occur, allowing an adequate margin of safety.  The SDWA

also requires that National Primary Drinking Water Regulations (NPBWRs)

establishing maximum contaminant levels (MCLs) or treatment techniques, and

secondary drinking water regulations be established.  The purpose of this

paper is to describe the activities of the Drinking Water Research Division

(DWRD) of EPA in evaluating the various technologies that may be available

for achieving these MCLs.
     In addition to conducting research into technological alternatives, the

DWRD provides technical advice and assistance to State agencies, EPA regional

offices, water utilities and professional organizations.  Amendments to the


SDWA have increased regulatory pressure and accelerated the timetable for

establishing MCLs-  DWRD's activities and its role in satisfying the amend-

ments will be discussed in terms of the following SDWA admendment categories:

contaminants to be regulated; surface water filtration; mandatory disinfec-

tion; and prohibition of use of lead materials.  In addition DWRD's work

which is intended to study and characterize the factors that lead to deterio-

ration of water quality in distribution systems will be discussed.
aDirector, Drinking Water Research Division, WERL, U.S.EPA, 26 W. St. Ciair  St
 Cincinnati, Ohio 45268                                                        *'

-------
                            REGULATED CONTAMINANTS
     DWRD is conducting research in three major areas associated with regu-




lated contaminants.  These areas are:  disinfection by-products; organic




contaminants; and, inorganic contaminants.
DISINFECTION BY-PRODUCTS






     Disinfection by-products research is perhaps the most significant and




controversial area of research ever undertaken by EPA in the drinking water




field.  The major and most important disinfection by-products are total




trihalomethanes (TTHM).  In addition over 500 additional disinfection by-




products have been identified.  An objective of this research is to establish




the feasibility of lowering the TTHM standard which of course will have many




ramifications for the water supply industry.  In addition, a limited number




of additional disinfection by-products are being quantified.







     Two major  pathways  of  research are  currently being pursued.  One is  the




identification  and characterization of the by-products of chlorination disin-




fection and the second is the evaluation of various treatment techniques  to




control byproducts using results  from in-house pilot plant studies.  By-




products of all other types of disinfection are being studied as well.




Because of the  complexity of the  research effort a disinfection by-product




workgroup has been formed to bring various groups within ORD together with




the Office of Drinking Water to maintain communication in this research area.

-------
Identification of By-Products

     DWRD has been actively developing methods for identifying the various

by-products associated with chlorine disinfection.  Most of this work has

been conducted at the bench scale.  The Division has also been working closely

with the Technical Services Division of ODW to isolate the various by-products

identified in field studies.

     The major disinfection by-products found to-date include the following:

trihaolmethanes (4 compounds), dihaloacetonitrile (3 compounds), chloroacetic

acid (3 compounds), chlorohydrate, chloropicrin, 1,1,1-trichloropropenone,

Table 1 lists these compounds and the general ranges within which they have

been found.  From 30-60% of TOX in drinking water is associated with these

disinfection by-products.


                  TABLE 1-  COMMON DISINFECTION BY-PRODUCTS
General Compound
 Identification
     Number of
Individual Compounds
   Nominal Concentration
                                             < 10
          in ug/L
          10-100
                                                 > 100
Trihalomethanes

Dihaloacetonitriles

Chloracetic Acids

Chloral Hydrate

Chloropicrin

1,1,1-Trichloropropanone
         4

         3

         3
X

X



X

X

-------
Control of Disinfection  By-Products




     Figure  1  is  a  schematic  of DWRD's  in-house  pilot  plant.   Two parallel




channels are being  evaluated.  The channel at  the  top  of  the  figure is a




control in which  the  pilot-plant  is  being operated as  a conventional filtra-




tion plant with chlorine added at the head works.   In  the bottom channel the




point-of-chlorination has been moved to minimize disinfection by-products,




primarily total trihalomethanes,  and to simultaneously maintain control of




microbiological quality  throughout the  plant.  The influent to both channels




is  spiked with coliform, phage and raw  sewage.   A  side stream from the lower




channel is passed through granular activated carbon (GAC).




     Based on  work  conducted  to-date, microbiological  quality has been main-




tained and TTHMs  minimized when chlorination was moved to a point just before




the filter.  Assimable Organic Carbon (AOC) is also being measured throughout




the plant in order  to characterize the  biological  quality of  the water that




might enter  the distribution  system.  Based on preliminary work AOC is also




reduced when chlorination is moved to a point just before filtration.




     Recent  results indicate  that with  a properly  optimized treatment system




chloramines  can minimize TTHM formation and also maintain microbiological




control.  This area of investigation is continuing






ORGANIC CONTAMINANTS




     Despite the  complexity of setting  MCLs for  health threatening agents,




Congress has been placing great pressure on EPA  to regulate organic contam-




inants more  extensively.   A number of drinking water contaminants have yet to




be regulated.  At this time only  a few  revisions of existing  regulations have




been made although the SDWA requires a  review of regulations  at least every




three years.

-------
 (PHAGE)

(SEWAGE)

   i
STORAGE
                   FUOC
SETHE
 t
CHLOWE

ALUM
             ALUM
               I
              M!X
      FUDC
SETTLE
                         FUER
                       CLEAR
                       WELL
                                      (CHLORIC
                         FUER
                       CLEAR
                       WELL
                                   GAC
                                       FOER
                                         f
                                     CHLORIC
                                    CLEAR
                                    WELL
 FIGURE 1.  PiLOT PLANT SCHEMATIC FOR DISINFECTION BY-PRODUCT CONTROL

-------
      Traditionally EPA's technology program has utilized the approach of

 carefully evaluating process kinetics on the bench, scaling up for engineer-

 ing feasibility at the pilot-scale, and making field-scale evaluations to

 evaluate process economics.  This approach may take several years for comple-

 tion and involve several million dollars for each unit process examined.

 Regulatory pressure will no longer allow for a contaminant-by-contaminant and

 process-by-process evaluation.

      As mentioned previously, recent amendments to the SDWA require a large

 number of Synthetic Organic Chemicals (SOCs) to be considered for regulation

 under stringent time deadlines.^  A key provision for potential regulation

 under the SDWA is that a feasible technology be available to remove each

 contaminant regulated.  For SOCs, granular activated carbon (GAG) by law is

 feasible technology.  Table 2 summarizes the synthetic organic compounds for

 which DWRD is currently developing carbon usage rate data.   In order to

 calculate these data several approaches are being taken.  One approach is to

 develop usage rate data based on microcolumn data and the other is to calcu-

 late usage rates based on field scale studies.   These approaches will be

 described in the following sections.


               TABLE 2.   CARBON USAGE  RATES FOR SELECTED ORGANICS
 2,4-D
 Silvex
 Llndane
 Methoxychlor
 Toxaphene
 Chlordane
 Heptachlor
 Heptachlor Epoxide
 PCS(s)
 Pentachlororophenol
Alachlor
 Carbofuran
Aldicarb
Aldicarb Sulfone
Aldicarb Sulfoxide
Endrin
Phthalate(s)
Adipate(s)
PAH(s)
Hexachlorocyclo pent ad i e ne
Dioxin
Simazine
Atrazine
Diquat
Endothall
Picloram
Dalapon
Dinoseb
Glyphosate
Oxamyl
EDB
DBCP
1,2-Dichloropropane
Cis-1,2-Dichloroethylene
Trans-1,2-dichloroethylene
Chlorobenzene
o-Dichlorobenzene
Toluene
Styrene
Ethyl Benzene
o-Xylene
m—Xylene
p-Xylene
Dibromomethane
1,1,2-Trichloroethane

-------
Microcolumn Data




     Microcolumns are being used to study _GAC adsorption.  The adsorption




protocol  involves developing a set of single solute isotherms in both dis-




tilled  and natural waters.  Using the Freundlich isotherm parameters, the




homogeneous surface diffusion adsorption model (HSDM) is employed to generate




breakthrough curves and predict carbon usage rates for fixed-bed GAC column




operation both at microcolum and at full-scale.  Other inputs to the HSDM are




kinetic parameters estimated from correlations given in the literature and




parameters defining the carbon type and the system hydraulics.




     For  some SOCs, predicted full-scale usage rates are found to be cost




effective, i.e., typically beyond a two-year service life, and these full-scale




usage rates are used as inputs to cost models.  A typical HSDM-generated full-




scale breakthrough curve is given in Figure 2.




     For  poorer adsorbed SOCs, full-scale predications show shorter service




lives, and microcolumn operations are conducted.  Because microcolumn break-




through curves represent empirical kinetic phenomenon rather than correlated




estimates, the microcolumn breakthrough curves are scaled-up to full-scale




conditions.  These scaled-up usage rates are used as inputs to cost models.




     The microcolumns were designed to give constant-pattern breakthrough




curves based on the HSDM.  The apparatus consists of a 4 mm Inside Diameter




(ID) glass column packed with US Standard Seive 100 x 200 mesh Filtraaorb 400




GAC at a depth sufficient to contain the mass transfer zone for the compound.




Other requirements are a prefilter to prevent excessive head loss across the




column,  dampers to smooth out pump delivery and maintain constant pressure on




the column,  and a flexible head space bag to restrict loss of volatile contam-




inants.   The apparatus is described in Figure 3.  For SOCs having full-scale




service  lives  of less than two years, microcolumn breakthrough curves may be

-------
                     Predicted Effluent Concentration Profile for cis-DCE

                           EBCT=15 mln    4 gpm/sqft Loading
    200 n

    180^


5>  160-
C  140-1
O
    120 H


 £  100-


     80-^


     60-^


     40-
 —i

S   20-j
O

O
O
0)
' , , ,  . < |  I

 100    105
                 •  ' •
                       nr i i "1 i i i
                                     T—T


110    115   120    125    130   135

      Throughput Time (days)
                                                          140   145
                                                                       150
           FIGURE 2. TYPICAL BREAKTHROUGH CURVE FOR FULL SCALE SYSTEM

-------
                     PRESSURE GAGE
               PULSE
              DAMPERS
                                   »

                         \   NFUJENT
                     GLASS WOOL
                      PREFUER
                                           ADSORPTON
                                           MICROOOLJJMN
FIGURE 3. ADSORPTION MICROCOLUMN SCHEMATIC

-------
generated in less than 5 days.  This  is  compatible with  the short regulatory

time frame.  Figure A shows a typical microcolumn breakthrough curve.

     Once the full-scale breakthrough curve's are developed for each compound,

field-verified cost equations are used to generate full-scale costs.2  A com-

puter program has been developed that uses breakthrough  curve data to calculate

carbon usage rates, and ultimately calculate costs for a wide number of carbon

systems.  A set of costs generated from  this procedure at various sizes of

application has been developed.  For  the data prepared for this analysis, it

is assumed that the carbon systems were  consistent with  the following condi-

tions: EBCT - 15 minutes; loading rate - A gpm/ft^; carbon type - Filtersorb

AOO.  These were full-scale conditions that were used in the HSDM.

     Several treatment plant sizes were  assumed with emphasis on small systems.

Steel pressure contactors were assumed for GAG adsorption in systems with a

design capacity less than or equal to 1  mgd.  Concrete gravity contactors

were assumed in larger systems.  All  systems were assumed to operate at 50%

plant capacity.  Spent carbon replacement versus reactivation alternatives

were selected based on least cost.  Cost parameters used in the analysis are

presented in Table 3.
                           *
                 TABLE 3.  COST PARAMETERS USED IN ANALYSIS	


               Item                            Value


GAG Price                              $0.70/lb @ AO.OOO Ibs
Capital Recovery Interest Rate         10%
Amortization Period                    20 years
ENR Construction Cost Index            A229
Producers Price Index                  29A
Labor Rate                             $ll/oanhour
Electric Rate                          $0.07/kwh
Fuel Oil Rate                          $0.90/gallon
Process Water Rate                     $0.50/1000 gallons
GAG Bulk Density                       30 lbs/ftj
GAG Loss Rate from Reactivation        HZ

-------
                                         100X200
                                    bed 
-------
     An example of  the SOCs  examined,  their  influent and  effluent concentra-




tions, and  type of  water  studied  in  each case are given in Table 4.  Most of




the  scenarios  are worst-case situations which provide conservative cost esti-




mates for removal of  the  single-solute SOCs.  Table 4 presents a matrix of




GAG  system  costs for  various usage rates and plant capacities.  All estimates




are  determined using  1986-year  prices  and cost indices.   Using the data in




Table 4, two sets of  cost-curves  were  generated: (1) a set of graphs display-




ing  cost as a  function of system  capacity at a fixed usage rate (designated




by specific SOC removal), (2) a set  of graphs displaying  cost as a function




of usage rate  or GAG  bed  life at  a fixed system capacity.  These cost-curves




can  be very useful  in analyzing cost sensitivity with variations in system




size, use rate and  bed life.




     In general, the  most significant  cost impacts appear to be with systems




that have a capacity  of less than 5  mgd and/or a GAG bed  life shorter than 4




months (>0.4 lbs/1000 gal).   At system capacities greater than 40 mgd the




cost curve  has a fairly flat response  showing small changes in cost with




variations  in  system  size.   Likewise,  when the GAG bed life is longer than 1




year, the change in cost  with variation in usage-rate is  small.  For systems




larger than 10 mgd, the fairly  flat  response of the cost  curve extends down




to a GAG bed life of  6 months (<0.23 lbs/1000 gal).  It appears that for




strongly adsorbed compounds  the GAG  bed life could be reduced by as much as




100% and the cost would only increase  by approximately 10 to 20%.  An interest-




ing example is as follows:   if  the GAG usage-rate increases by a factor of 10




(1000Z increase) from 0.0572 lbs/1000  gal (2-year bed life) to 0.6 lbs/1000




gal (69-day bed life), the cost increases by only a factor of 2 (100% cost




increase)•

-------
Table 4.  GAG COSTS FOR VARIOUS USE RATES AND SYSTEM SIZES
Use Rate, lbs/1000 gal .0572 .0626 .0734 .0832 .1176 .178 .2191 .


Service Life, days 730 667 569 502 355 234 191
A B CD E F G
2336
179
H
Design Capacity (MGD) Total System Cost (^/lOOO













A
B
C
D
E
P
G
H
I
J
K
L
M
.25 83.8 84.6 86.2 87.6 92.6 101.1 106.9
.75 79.1 79.8 81.3 82.7 87.4 95.6 101.1
1 67.9 68.6 70.1 71.5 76.1 84.2 89.7
2 40.2 40.9 42.3 43.7 48.2 56.1 61.4
3 35.2 35.9 37.3 38.6 43.1 50.8 56.0
5 30.3 31.0 32.4 33.6 38.0 45.6 48.5
7 27.7 28.4 29.8 31.0 35.4 39.4 40.9
10 25.4 26.1 27.4 28.7 31.1 33.4 34.9
20 21.1 21.3 21.7 22.0 23.3 25.4 26.9
30 18.0 18.2 18.6 19.0 20.2 22.3 23.7
50 15.2 15.4 15.7 16.1 17.3 19.4 20.8
70 13.7 13.9 14.3 14.6 15.8 17.8 18.9
100 12.4 12.6 13.0 13.3 14.5 16.1 18.9
« two-year service life
- chlorobenzene, Cl - 600 ug/L, Ce = 6 ug/L, Mllli-Q distilled water
- chlorobenzene, Ci - 600 ug/L, Ce - 6 ug/L, filtered Ohio River water
- aldicarb, Ci - 500 ug/L, Ce - 1.3 ug/L, filtered Ohio River water
- EDB, Ci - 100 ug/L, Ce - 1, Milli-Q distilled water
- 1,2-dichloropropane, Ci - 100 ug/L, Ce - 2 ug/L Milli-Q distilled water
- trans-DCE, Ci - 100 ug/L, Ce - 70 ug/L filtered Ohio River water
108.9
103.0
91.6
63.2
57.9
49.1
41.5
35.4
27.4
24.2
21.2
19.2
19.4







.2713
154
I
gal)
114.
108.
96.
68.
62.
50.
42.
36.
28.
25.
22.
20.
20.











1
0
5
0
6
5
9
7
7
5
2
2
2







.378
110
J

128.
122.
110.
81.
71.
54.
46.
40.
32.
29.
25.
25.
23.











7
1
4
5
1
4
7
5
4
2
0
0
1







.45
93
K

138.
132.
119.
90.
73.
57.
49.
43.
34.
31.
30.
27.
24.











5
9
6
5
7
0
3
1
9
1
5
5
9







.6046
69
L

159.
151.
139.
109.
79.
62.
54.
48.
39.
35.
34.
3V
30.











3
3
3
7
4
5
7
5
9
2
6
5
8







1.2
35
M

237.8
228.0
213.7
121.0
100.8
83.4
75.4
68.8
55.4
56.8
53.6
51.9
49.5







- 1,2-dichloropropane, Ci - 100 ug/L, Ce - 2 ug/L filtered Ohio River water
« toluene, Ci « 5000 ug/L, Ce *> 100 ug/L, distilled Wausau water
- cis-DCE, Ci - 200 ug/L, Ce - 70 ug/L, filtered Ohio River water
« 93 day service life
» trans-DCE, Ci - 500 ug/L, Ce - 70 ug/L, filtered Ohio River water
- 35 day service life



















































-------
     This cost analysis  suggests  that highly accurate predictions of GAG




adsorption performance for many SOCs may not be necessary because of the




apparent insensitivity of cost with mild changes in usage rate.  From the




cost curves  this appears to be valid for situations where the GAG bed life




is longer than 6 months.  It was  found that capacities and thus costs are




most sensitive to  the parameters  in the Freundlich isotherm equation.  The




costs presented herein were generated using a computer program written for




the IBM-PC/AT.






Other Technologies Studies




     Other technologies  being examined at the bench and pilot scale for control




of organics  include ozone oxidation, reverse osmosis, ultraviolet treatment, and




ultrafiltration.






Ozone Oxidation




     Ozone oxidation is  being studied extensively in DWRD's in-house pilot




plant facilities.^ Controlled pilot plant ozone treatment tests have been




conducted on 29 VOCs in  distilled water and groundwater.  Results showed that




aromatic compounds, alkenes and certain pesticides are well removed by ozone




treatment but that alkanes are poorly removed.  Also, removal efficiency




improved for the alkenes and aromatic compounds with increasing ozone dosage




and for some alkanes with increasing pH.  For most compounds, the efficacy of




ozone was not affected by the background water matrix.  Information from the




literature concerning the ozone treatment of pure materials in the gaseous or




liquid phase  generally predicted  the effectiveness of ozone in treating




aqueous solutions.

-------
Reverse Osmosis




     Revesrse osmosis  (RO) has shown some promise in removing both VOCs and




SOCs from groundwater.^  Most of the efforts by DWRD to date have been on a




pilot- or bench-scale  basis, with some limited application of reverse osmosis




for the removal of organics at one field site.  Primary indications are that




certain RO membranes are very effective in removing a wide range of organic




chemicals.






Ultraviolet Treatment




     Ultraviolet light also shows some promise for removing organic contam-




inants, particularly when combined with ozone.  DWRD is in the process of




funding a project with the Los Angeles Water and Power Company that deals




with the removal of VOCs from groundwater using these two technologies.  If




successful, these chemicals will be oxidized to C0£ and water, and the need




to deal with off-gas control problems will be eliminated.






Ultrafiltration




     Studies have been conducted to evaluate the costs and performance of low




pressure membrane processes (ultrafiltration) for TTHM precursor reduction in




small systems.  A 150 day pilot-plant of two highly organic contaminanted




ground waters (both of which produced more than 400 ug/L THMs when conven-




tionally treated) produced a finished water that easily met the Trihalomethane




Maximum Contaminant Level (100 ug/L).  Costs appears to be reasonable.^




More extensive testing of this technology is underway.






Field-Tested Technologies




Carbon Adsorption




     Extensive field scale studies have been conducted on the use of granular




activated  carbon (GAG) treatment with onsite regeneration.  Past research has

-------
devoted to demonstrating the effectiveness of GAG for  surface water treatment.5




Extensive field testing has been conducted -at Jefferson Parish and Cincinnati,




Ohio.  Other field sites include: Manchester, New Hampshire; Evansville,




Indiana; Miami, Florida; Huntington, West Virginia; Beaver Falls, Pennsylvanis;




and Passaic, New Jersey.




     More recently, DWRD has been conducting studies that incorporate the use




of carbon treatment for removal of VOCs and SOCs from  groundwater at Suffolk




County, New York, California's San Joaquin Valley, and Wausau, Wisconsin.




Each of these projects is designed to examine a different aspect of GAG




application and except for Wausau, are intended to deal with the little-under-




stood  pesticide contamination problem.




     A typical project is in Suffolk County, New York, where the removal of




organics, pesticides, and nitrates is being studied under various flow situa-




tions.  Two parallel treatment systems (one consisting of GAG and ion exchange




and the other consisting of reverse osmosis) are being operated at low flows




similar to home usage.  Costs for these two systems will be established along




with unit operating efficiency so that a large public  water supply system can




be designed and tested.  The results from this study will be applicable to




other  areas, especially in farming communities, where  multiple contamination




of groundwater is identified.




     Two principle objectives of a project being conducted in the San Joaquin




Valley study are to (1) develop cost-effective design  criteria for the removal




of DBCP and other pesticides from water supplies by GAG, (2) to compare treat-




ment methods other than GAG for the removal of these compounds, and (3) to




improve and strengthen existing administrative guidelines and jurisdictional




responsibilities pertaining to both community water systems and private wells

-------
 containing  the  compounds  of interest.   Pilot studies  will  be  conducted with




 GAC,  and  results will  be  compared with those existing operating  systems.




 These comparisons  should  lead to recommendations  for  more  effective design




 and operational criteria  for GAC units used  by water  systems.  The focus  in




 this  study  will be on  point-of-use GAC units that are now  in  use by small




 systems,  individual homeowners,  and farmers  in the area.   Disposal of treat-




 ment  residues is also  being studied.




      The  third  site, in Wausau,  Wisconsin, has multiple contaminants in its




 groundwater source from a nearby superfund site.  Figure 5 shows a typical




 breakthrough curve from the Wausau carbon field study.




      GAC  is the primary technology being studied, but  air-stripping is being




 examined  as a companion technology.  The Wausau project is unique in that




 modeling  techniques are being used to  predict full-scale design  criteria  for




 a GAC plant.  These predictions  will be evaluated against the actual cost and




 performance associated  with building of GAC  contactors onsite.   the results




 from  this study are expected to  provide a useful methodology for extension  to




 other GAC applications.   Modeling techniques will be particularly useful in




 studying  the long  list  of SOCs and VOCs that may be proposed for regulation




 under the current  and  future provisions of the Safe Drinking Water Act.  In




 addition to modeling, rainicolumn technology  is also being studied.  This




methodology allows  investigators to acquire water from a given site and to




study the performance of  a small,  high-pressure column in the presence of a




natural water background.   Results from these minicolumn experiments are




expected to reproduce the  breakthrough curves normally seen in pilot- and




full-scale  facilities.

-------
3,0
   CM-
   T
 c
 o
• *» Influent
D» Effluent; EBCT - .97 mln
0«= Effluent; EBCT •= 2.93 mln
A » Effluent; EBCT « 4.96 mln
o» Effluent; EBCT = 9.98 mln
x m Effluent; EBCT » 20.3 mln
• « Effluent; EBCT - 30.9 mln
                                                        CIa~1,2-Dlchloroethene
                                                           WVQ (12x40)
                                                      Loading Rate « 4.05 m/hr
                                                    Bulk Density « 417 kg/m3
                                                    Ave. Influent Temp. » 13 °C
                                                      Detection Limit » .3 ug/L
           30
                90    120   150   180   210    240    270   300   330    360
                         Elapsed Time of Operation  (days)
               FIGURE  5.  PILOT SCALE DATA FROM WAUSAU, WISCONSIN

-------
Packed Tower Aeration


     Aeration  technology  has  proved  to be especially  effective  for  the  removal


of VOCs.  However,  the  research  at Wausau,  Wisconsin  (which  also incorporates


air  stripping  and off-gas control  technology)  indicates  that aeration may  in


some cases  be  effective for removing compounds that have somewhat lower


Henry's Law Constants than would normally be expected to be  removed by  this


process.  In this project, as with carbon facilities, modeling  is being used


to predict  the performance of packed-tower aeration and  results are promising.


The  off-gas control portion of the air stripping project at  Wausau is being


conducted in cooperation  with the American Water Works Association Research


Foundation.

        *
     DWRD also has  another air-stripping  research project currently underway


at Baldwin  Park, Calfifornia  which is  examinaing the  removal of VOCs from a


groundwater supply.  As in Wausau, the Baldwin Park project  is exmaining the


problems of off-gas control technology.^


     Another field-scale  activity dealing with air-stripping has been con-


cluded at Brewster, New York.  In this project, modeling  techniques and


pilot-scale facilities were used to  determine  the scale-up relationships to


be used for full-scale, air-stripping  facilities.  The purpose of the project


was  to develop a technique that  could  be  used  by consulting engineers to


adequately  predict  the cost and  performance of full-scale facilities using


pilot aeration columns.



Conventional Treatment


     Conventional treatment is unlikely to  be  used for removing organic


contaminants from groundwater.  However,  field studies are being conducted


at Tiffin,  Ohio, where the river source contains high spikes of pesticides

-------
because  of  local  seasonal agricultrual applications.   Powdered activated




carbon,  added  to  the  water normally treated only by conventional  treatment,




appears  to  be  quite effective  for  removing  synthetic  organic  chemicals.






Secondary Discharge Problems




     When evaluating  treatment technology one has to  be aware  of  the possi-




bilities of creating  secondary discharge problems.  For example, while




conducting  GAC research on surface water supplies,  it was discovered that




dioxins  were formed in the reactivation process.   An  extensive evaluation




led  to the  installation of an  afterburner,  which  was  found to  eliminate




those dioxin byproducts when operating at a temperature of 2400CF.'




     At  Baldwin Park,  California,  the  use of high-stack dispersion was inves-




tigated  as  a means of  minimizing the impact of VOC  removal from groundater.




At Baldwin  Park and at Wausau,  Wisconsin, gas-phase carbon adsorption is




being investigated for removal of  both VOCs and SOC from air stripping waste




gases.




     The project  with  the Los  Angeles  Water and Power Company  is intended to




investigate the water-phase oxidation  of VOCs and SOCs to C(>2  and ^0 using




ozone in combination with ultraviolet  light.  Gas-phase oxidation of VOCs and




SOCs by  ozone  and ultraviolet  light  also appears  promising.




     Future work  will  concentrate  on residuals control from various unit




processes such as reverse osmosis  and  ultrafiltration.






Inorganic Contaminants




     Several field scale  studies are being  conducted  for inorganics control.




These technologies include;  ion exchange and reverse  osmosis.  Specific

-------
 contaminants  being removed Include nitrates,  radium and radon.   Each of these

 technologies  is  discussed in the following.paragraph.


 Ion Exchange

      Ion exchange technology has been field-tested  for  nitrate  removal under

 a DWRD-supported cooperative agreement and has  been pilot-tested for uranium

 removal at sites in the western part of the United  States.®»^

      The nitrate removal plant has been operating automatically for  about  3

 years.   The 1-mgd (0.38 ML/d)  demonstration plant is located in McFarland, CA

 and consists  of  three anion exchange vessels  that are designed  to reduce

 nitrate levels to below 10 mg  N03-N/L,  the EPA  MCL  and  also the California

 requirement.  Currently, about 500 gpm (32 L/S) of  water is being treated and
                                                      i
 about 200 gpm (13 L/S)  is bypassed and later  blended with the treated water,

 resulting in  a total  product flow of 700 gpm  (44 L/S).  The blended  water

 adequately meets  the  nitrate MCL and EPA's Secondary Regulation (Fed. Reg.

 1979) for chloride and  sulfate levels.10

      Bench-scale  studies were  initally  conducted at  EPA-Cincinnati for uranium

 removal and later pilot-tested.   The laboratory work showed that when drinking

 water containing  300  ug/L uranium was  passed  through anion exchange  resin,

 more  than 9,000 bed volumes were  treated before breakthrough was observed.

 DWRD then evaluated the  performance  of  twelve 1/4-cubic-foot (0.007 m^) anion-

 exchange  systems  installed in  New Mexico, Colorado,  and Arizona at sites

where uranium levels  in  the raw water exceeded 20 ug/L.   Results confirmed

 the findings shown in the  laboratory.  Because of the high loading capacity

of the anion resins for  uranium, these units are well suited for point-of-use

applications where on-site regeneration is not feasible.  For centralized

treatment,  the resin may be regnerated and recycled  by backwashing it with

sodium chloride solution.

-------
Reverse  Osmosis




      Groundwater  sources  in Illinois,  lowa^  Florida,  Texas, Wisconsin,  and




some  Rocky Mountain states  contain radium in excess of  the 5 pCi/L MCL.   RO




treatment  of  radium-laden ground water was demonstrated in 1977 when DWRD and




Sarasota County,  FL water supply staff undertook a cooperative effort to




study the  operation of  eight RO systems.^  The  systems were located in small




communities serving a population from  39  to  15,000 and  the design capacities




varied from 800 gpd (3  kL/d) to 1 mgd  (3.8 ML/d).  Six  different manufacturers




of RO systems were  represented  and both the  hollow fiber and spiral wound




cellulose  acetate membranes were used. The  study showed from 82-961 Ra-226




removal  for all systems,  resulting in  treated water that contained below  the




EPA MCL  of 5  pCi/L.




      RO  technology  has  also been pilot tested for nitrate removal as part



of DWRD  cooperative  agreements  at three locations.^, 13,14  ^8 mentioned




previously, high  nitrate  levels exist  in  the well waters of Suffolk County,




LI, NY,  along with  several  SOCs.12 This  combination  of  organic and inorganic




contaminants  is the  reason  that RO was selected  as one  of the treatment




technologies  to be  studied  there.   Seven  commercially available membranes




were  evaluated for  their  rejection capabilities.  Nitrate removals ranged




from  75-95%,  and  research is continuing with one of the  polyaraide membranes




that  proved relatively  efficient for both nitrate and SOC rejection.  At




Charlotte Harbor, FL both high  pressure (265-359 psig)  and low pressure




(163-187 psig) RO systems were  studied for the removal  of several spiked




inorganic contaminants, including  nitrate, from  a natural groundwater.^  The




investigation showed that the high pressure  system was  significantly more




effective for removing  all  substances  measured.  The  comparison for nitrate

-------
 removals,  for example,  was 80% vs.   6-24%.   Before ion exchange was  selected




 for  full-scale evaluation at McFarland,  CA^  a 20,000  gpd  (77 m^/d) RO system




 was  examined  for nitrate removal.^  Even though the  system experienced




 frequent  electrical and mechanical  failures,  nitrate  rejection of about 65%




 was  achieved.






 Radium Removal Technologies




      DWRD has sponsored research  to study promising methods for radium




 removal. 15,16  -j^e jon  exchange process,  with both weak acid resin and strong




 acid resin were investigated.   Both resins effectively removed radium from




 water to  well below the 5 pCi/L MCL and,  in most cases, to <0.5 pCi/L, respre-




 senting over  96% removal.   The weak acid  resin in the  hydrogen form also




 removed hardness,  which was not the case  for  the strong acid resin in the




 calcium form.   The maximum capacity of the weak-acid resin was about 2.3




 times that of strong acid resin and much  less spent regenerate per unit




 volume  of water treated was produced from the weak-acid column than from the




 strong-acid column.  Another part of this project was  to determine the feasi-




 bility  of Dow Chemical  Company's Radium-Selective Complexer (RSC) for remov-




 ing  radium from brines  compared with typical  groundwaters.  The RSC is a




 synthetic resin that has  a high affinity  for  radium.   The capacity of RSC was




 observed  to have been about 200 times greater in 450 mg/L TDS water than in




 40,000 mg/L TDS brine (51,000  pCi/dry g vs. 300  pCi/dry g).  However, the




 effect of other parameters,  including calcium, sodium,  and other ions, in




addition to EBCT, needs  to  be  determined.  The application of the RSC for treat-




ment  of radium  from brines  is  presently being studied  at a small community in




Colorado under  a research  cooperative agreement  with DWRD.

-------
Radon Removal Technologies




     In  response  to  the Agency's  recent  emphasis on radon  in  the environment,




DWRD has funded a project to  evaluate  several  treatment  techniques  for  the




removal  of  radon  from  community water  supplies.  Three treatment methods,




packed tower aeration, diffused bubble aeration and granular  activated  carbon




(GAC) will  be evaluated for the removal  of radon from two  community water




supplies in New Hampshire.  The study  will compare the methods for  effective-




ness, costs, operation, maintenance and  other  related factors.  To  accomplish




the objectives, the  three treatment systems will be constructed and operated




at each  of  the two sites.  The  two sites selected are trailer parks, one




whose water supply serves 40  homes with  an average daily flow of 4.6 gpm and




an average  radon  concentration  of 155,000 pCi/L.  The other location has 56




homes, the  average daily water  usage is  6.3 gpm, and the average radon  con-




centration  is 40,000 pCi/L.   After construction of the treatment systems,




they will be operated  for 6-12  months.   At the end of this project, a report




will be  developed as a guideline  to help State Agencies and small communities




select an appropriate  technology  for radon gas treatment in public  water




supplies.






Summary  of  Technologies Studied




     Table  5 summarizes the treatment  technoliges that DWRD is evaluating




for removal of VOCs, synthetic  organic chemicls (SOCs), nitrates, and radio-




nuclides  from water  supplies.   The table indicates carbon adsorption is




effective for removing both VOCs and SOCs.  Packed tower and diffused aera-




tion are best suited for removing VOCs.   Ion exchange has been field-tested




to show effective removal of  nitrates and pilot-tested for uranium  removal




Reverse osmosis (RO) has proven to be effective in the field for radium

-------
removal and pilot-tested for nitrate removals.  Of the technologies that show




promise and are being tested at the bench-.and pilot-scale, conventional treat-




ment with powdered activated carbon (PAC) is effective for removing a few of




the SOCs, ozone oxidation is effective for removing certain classes of VOCs




and SOCs, and certain reverse osmosis membranes and ultraviolet treatment are




also potentially effective against VOCs and SOCs.  Aeration and carbon adsorp-




tion are being examined for their radon removal capabilities.




     Table 6 summarizes the various technologies examined for contaminant




removal, and based on data gathered to date, it attempts to characterize




their relative performances for both Phase I and Phase II organics.






       TABLE 5.  TREATMENT TECHNOLOGIES EVALUATED BY DWRD FOR REMOVING




         VOCs, SOCs, NITRATES, AND RADIONUCLIDES FROM DRINKING WATER
Technology
Status
Field-tested 1.
2.

3.
4.
Pilot-tested 1.
2.
Promising Technologies 1.


2.
3.
4.
5.
6.
7.
8.
Technology Contaminant Class or
Specific Contaminant
Removed
Carbon Adsorption
Packed Tower and
Diff used-Air Aeration
Ion Exchange
Reverse Osmosis
Reverse Osmosis
Ion Exchange
Conventional Treatment
with Powdered Activated
Carbon
Ozone Oxidation
Reverse Osmosis
Ultraviolet Treatment
Ion Exchange
Selective Comp lexer
Aeration
Carbon Adsorption
1.
2.

3.
4.
1.
2.
1.


2.
3.
4.
5.
6.
7.
8.
VOCs, SOCs
VOCs

Nitrates
Radium
Nitrates, Uranium
Uranium
SOCs


VOCs, SOCs
VOCs, SOCs
VOCs, SOCs
Radium
Radium
Radon
Radon

-------
           TABLE 6.  PERFORMANCE SUMMARY FOR ORGANIC TECHNOLOGIES EXAMINED
REMOVAL EFFICIENCyt
Granular
Activated
Carbon
Regulatory Adsorption
Phase Organic Compounds Filtrasorb 400a
VOLATILE ORGANIC CONTAMINANTS
Alkanes
I Carbon Tetrachloride 44
I 1,2-Dichloroethane 44
I 1,1,1-Trichloroethane 44
II 1,2-Dichloropropane 44
II Ethylene Dibromide 44
II Dibromochloropropane 44
Alkenes
I Vinyl Chloride 44
II Styrene NA
I 1,1-Dichloroethylene 44
II cis-l,2-Dichloroethylene 44
II trans-l,2-Dichloroethylene 44
I Trichloroethylene 44
Aromatics
I Benzene 44
II Toluene ++
II Xylenes -H-
II Ethylbenzene ++
II Chlorobenzene 44
II o-Dichlorobenzene 44
I p-Dichlorobenzene 44
PESTICIDES
II Pentachlorophenol -H-
II 2,4-D ++
II Alachlor ++
II Aldicarb NA
II Carbofuran ++
II Lindane ++
II Toxaphene ++
II Heptachlor ++
II Chlordane ++
II 2,4,5-TP -H-
II Methoxychlor ++


Packed
Tower
Aeration


44
44
44
44
44
4

44
NA
44
44
44
44

44
44
44
44
44
44
44

0
0
44
0
0
0
44
44
0
NA
NA
Reverse
Osmosis
Thin
Film
Composite


44
4
44
44
44
NA

NA
NA
NA
0
NA
44

0
NA
NA
0
44
4
NA

NA
NA
•H- .
NA
44
NA
NA
NA
NA
NA
NA


Ozone
Oxidation
(2-6 ppm)


0
0
0
0
0
0

44
44
44
44
44
+

44
44
4+
44
4
4
4

44
4
44
NA
44
0
NA
44
NA
4
NA



Conventional
Treatment


0
0
0
0
0
0

0
0
0
0
0
0

0
0
0
0
0
0
0

NA
0
0
NA
0
0
0
NA
NA
NA
NA
(See footnotes at end of table)
(Continued)

-------
           TABLE 6.  PERFORMANCE SUMMARY FOR ORGANIC TECHNOLOGIES EXAMINED (Contd.)
                                                   REMOVAL EFFICIENCY*
                                   Granular               Reverse
                                  Activated               Osmosis
                                    Carbon       Packed    Thin      Ozone
Regulatory                        Adsorption     Tower     Film    Oxidation  Conventional
  Phase     Organic Compounds   Filtrasorb 400a  Aeration Composite (2-6 ppm)    Treatment


          OTHER

  II  Acrylamide                      NA         0          NA         NA         NA
  II  Epichlorohydrin                 NA         0          NA         0          NA
  II  PCB's                           -H-         -H-         NA         NA         NA
       Excellent            70% - 100%
  + =  Average Removal      30% -  69%
  0 -  Poor                  0% -  29%
 NA ™  Data not available or compound has not been tested by EPA Drinking Water
        Research Division
  a - Excellent removal category for carbon indicates compound has been demonstrated to
      be adsorbable onto GAG, in full- or pilot-scale applications, or in the
      laboratory with characteristics suggesting GAC can be a cost-effective technology.

-------
     Although Table  6 only  provides  a general  guideline  for  removal  of  com-

 pounds,  several  interesting trends are note'd.   Carbon  adsorption appears  to

 provide  removal  for  a wide  range  of  organics whereas conventional treatment

 is  revealed as a poor treatment for  those  compounds listed in  the table.

 Packed tower aeration manifests itself as  an excellent technology for volatile

 organic  compounds and may have application for a limited number of pesticides.

 Ozone oxidation  appears  to  be a good treatment technology for  certain classes

 of  organics such as  simple  alkenes and aromatlcs, as well as certain similar,

 but more complex organic structures.   Although only a  few organics have been

 subjected to long-term testing via reverse osmosis, promising  removals for

 several  low molecular weight organics can  be seen.

     Table 7 shows that  ion exchange  and reverse osmosis each  result in

 excellent contaminant removals.


          TABLE 7.  PERFORMANCE SUMMARY FOR INORGANIC TECHNOLOGIES EXAMINED

Regulatory
Phase
II
III
III
III

Inorganic
Compound
Nitrate
Radium
Uranium
Radon
Removal Efficiency
Reverse Ion Carbon
Osmosis Exchange Aeration Adsorption
++ -H-
++ 4+
++ -H-
* *
-H- - Excellent 70%-lOOZ
 * - Research being conducted by DWRD
     Tables 6 and 7 were generated using a variety of sources, including

EPA-DWRD pilot- and field-scale studies, Henry's Law Constants for predic-

tion of removal of some of the pesticides, as well as the use of oxidative

-------
trends for predicting the removal of complex pesticides.  These tables,

therefore, carry the caveat that the cited-removals should not be used for

design purposes, but that each technology must be tested on compounds, under

field conditions, before the EPA-DWRD will advocate a technology's use.


               SURFACE WATER FILTRATION/MANDATORY DISINFECTION


     DWRD's primary activities in this area are the studies related to Giardia

and viral inactivation.  Current studies on Giardia inactivation are being

conducted at Ohio State.  Hepatitus A and phage inactivation studies are

being conducted at the University of North Carolina.  Field studies related

to inactivation of Giardia are being planned.  A major result of these studies

are the calculation of CT values for various organisms at different levels of

temperature, pH and disinfectant dose.  The following projects are ongoing in

this area:

     0  Reports on concentration and time values for Hepatitis A
        and coliphage inactivation by:

             Chlorine and Chloramines
             Chlorine dioxide and ultraviolet light

     °  Laboratory - scale studies on the inactivation of Giardia Muris
        by chlorine,  chlorine dioxide, and chloramines.

     0  Pilot scale study at Harrisburg,  PA on inactivation 'of virus
        and Giardia by chlorine and chloramines.

     0  Laboratory study on inactivation of Cryptosporidium by chlorine
        and chloramines.

     e  Evaluating treatment plants and distribution systems for
        Legionella.

-------
                        PROHIBITION  OF LEAD MATERIALS






     The amendments to  the SDWA prohibit  the use of lead in any material in




contact with potable water.  DWRD's  major current project dealing with lead




solder is being conducted with the South Huntington Water District on Long




Island, New York.




     Ninety sites were  identified and selected for this study.  Ten homes




were selected in each of nine age groups from zero to over 20 years old.




These sites have been sampled three  times at different pH levels:  less than




6.2, 7.2 and 8.2.  The  homes were randomly sleeted to obtain a geographic




distribution In the 19.4-square-mile service area of the District.  The type




of existing solder was  verified in each home through scraping of an exposed




solder joint and testing by an atomic absorption spectrophotometer.  Of 95




homes tested, only one  had less than 0.5 percent lead In the solder; 67.3




percent of the homes had lead content ranging from 55 to 65 percent In the




solder.



     First-draw samples were tested  for copper and cadmium in addition to




lead.  An additional sample was drawn for determination of various water




quality parameters including pH.  Copper values above the 1 milligram per




liter (mg/L) secondary Maximum Contaminant Level (up to 7.77 mg/L at 6.9 pH)




were found.  These sites were checked for stray electric currents in the




water service pipe to determine whether or not the high copper values were




caused through grounding of electric systems to the water plumbing system.




     During low pH sampling of 64 first-draw tests for cadmium, only one




sample at 42 ug/L was above the drinking water standard.  Since the second




sample at 10 seconds contained only  3 ug/L cadmium, it Is assumed the presence




of cadmium was caused by the faucet.

-------
     On the first-draw lead samples taken, 61.8 percent were above the drink-




ing water standard.  The results ranged from a low of 2.0 ug/L in a 1968 home




at pH - 5.9, to values ranging from 500 to 1,200 ug/L (pH ranges from 5.6 to




6.9) at homes constructed between 1981 and 1983.  However, the highest lead




value (1,300 ug/L) was obtained at pH - 5.9 in a 1968 home.




     There appears to be a correlation between the age of the lead solder and




its ability to leach lead into the drinking water.  Ignoring the first zero-




second flush and allowing the first 300 milliliters (mL) of flush for the




faucet's leaching of lead, the second sample at 10 seconds were compared for




75 homes.  For homes constructed in 1980, 1981, 1982, and 1983, the second




125 mL sample (taken in March and April, 1984) indicated that 23 homes,  or 70




percent, exceeded the 50 mg/L drinking water standard, and 10 homes were less




than this maximum contaminant level.  For homes constructed between 1955 and




1979, only seven, or 17 percent, exceeded the drinking water standard at a




low pH with 35, or 85 percent, less than the maximum contaminant level.




Table 8 summarizes some of the first draw results.






        TABLE 8.   RESULTS FROM HOUSEHOLD LEAD STUDY FIRST DRAW SAMPLES






               pH                   % of Homes Exceeding 50 ug/L






             < 6.2                               58.7




           6.2 -  8.2                             26.6




              78.2                               24.4

-------
     Leaching of metallic solders, particular/ lead solder, appears to be




affected by the following factors:




        pH of water




     0  Plumbing workmanship,




     0  Hardness of water




     0  Time since last use of water




        Percentage of lead in solder.




     Where these conditions are conducive to lead leaching, large numbers of




consumers may be exposed to high-lead, first-draw water.




     Several interesting results have come from this ongoing study that may




provide information on leaching from various kinds of solder from special




pipe loops.  Tables 9 and 10 provide some useful summary data from the project.




As can be seen there is lead in the various alternative solders being tested.




In addition all four pipe loops have experienced copper leaching at very high




levels.



                        TABLE 9.  LEAD SOLDER PROJECT
Solder Type
Tin/ Silver
Tin/ Silver
Tin/ Silver
Highest Lead Value
0.015 mg/L
0.042 mg/L
0.057 mg/L
PH
5.3
5.4
5.3
Standing Time
4 hrs
4 hrs
4 hrs

-------
                TABLE 10.  COPPER LEACHING IN FOUR PIPE LOOPS
pH
5.5
5.5
5.5
5.5
Solder
Tin/ Silver
Tin/ Antimony
Tin/Copper
Tin/Lead
Copper
(fflg/L)
3.80
4.28
4.50
4.28
     Tables 11, 12 and 13 illustrate some of the problems to be encountered




when trying to meet the presently discussed 20 ug/L lead level in households.




These data are from other ongoing research projects.  As can be seen at lower




pHs for short sampling periods in new homes it will be very difficult to meet




the proposed standard.  At higher pH's for longer sampling periods and for




Increasing age of test site the proposed standard can be met although it is




only met consistently in older homes.






        TABLE 11.  PERCENTAGE OF TEST SITES WITH LEAD IN DRINKING WATER




                      GREATER THAN 20 ug/L AT LOW pH (6.4 & LESS)
AGE OF
TEST SITE
(Years)
0-1
1-2
2-3
3-4
4-5
6-7
9-10
15 - 16
20 & Older

FIRST
DRAW
100%
100%
86%
100%
86%
78%
71%
57%
86%

10
SEC
100%
71%
86%
86%
57%
44%
29%
14%
27%

20
SEC
100%
86%
57%
100%
29%
33%
14%
14%
29%

30
SEC
100%
57%
57%
71%
43%
33%
14%
14%
0%

45
SEC
100%
57%
43%
71%
43%
11%
14%
14%
14%

60
SEC
86%
29%
43%
71%
43%
11%
14%
14%
0%

90
SEC
86%
43%
43%
29%
14%
11%
0%
14%
14%

120
SEC
88%
14%
29%
29%
0%
0%
0%
14%
0%

-------
        TABLE 12.  PERCENTAGE  OF  TEST SITES  WITH LEAD  IN DRINKING WATER




                      GREATER  THAN 20 ug/L "AT  MEDIUM pH (7.0 - 7.4)
AGE OF
TEST SITE
(Years)
0-1
1-2
2-3
3-4
4-5
6-7
9-10
15 - 16
20 & Older

FIRST
DRAW
100%
80*
40%
50%
30%
10%
20%
40%
20%

10
SEC
90%
60%
20%
20%
10%
0%
0%
20%
0%

20
SEC
90%
30%
10%
20%
10%
0%
0%
20%
0%

30
SEC
60%
10%
10%
30%
0%
0%
0%
10%
0%

45
SEC
30%
20%
10%
20%
10%
0%
0%
0%
10%

60
SEC
20%
0%
0%
30%
0%
0%
0%
0%
0%

90
SEC
10%
10%
0%
30%
0%
0%
0%
0%
0%

120
SEC
10%
0%
0%
20%
0%
0%
0%
0%
0%
        TABLE 13.  PERCENTAGE OF TEST  SITES WITH LEAD IN DRINKING WATER




                     GREATER THAN 20 ug/L AT HIGH pH (8.0 & GREATER)
AGE OF
TEST SITE
(Years)
0-1
1-2
2-3
3-4
4-5
6-7
9-10
15 - 16
20 & Older

FIRST
DRAW
100%
67%
30%
25%
30%
20%
10%
33%
20%

10
SEC
100%
22%
10%
0%
10%
0%
0%
22%
0%

20
SEC
60%
11%
10%
0%
0%
0%
10%
11%
0%

30
SEC
10%
11%
0%
0%
0%
0%
0%
11%
0%

45
SEC
20%
11%
0%
0%
0%
0%
0%
0%
0%

60
SEC
10%
0%
0%
0%
0%
0%
10%
0%
0%

90
SEC
20%
11%
0%
0%
0%
0%
0%
0%
0%

120
SEC
0%
0%
0%
13%
0%
0%
10%
0%
0%
                        DETERIORATION OF WATER QUALITY






     DWRD is and has been conducting extensive studies into the deterioration




of water quality in distribution systems.  Methodology is being developed for

-------
 calculating the AOC associated  with treated  and distributed water.  Work  is




 being conducted into the identification  of legionella and other organisms




 that may exist in distribution  systems.   The following general areas are




 being studied:




      1.   Microbial Contamination  of Distribtulon Systems




      Research has focused on  corrosion inhibitors and the possibility of




 their acting to promote  microbial growth In  water mains.  Other research




 (2  projects) are examining protection of bacteria by higher organisms as  a



 mechanism for bacteria to be  transported into the distribution system In  a




 viable state.   Still  another  project has been funded to learn about protec-




 tion and transport  of bacteria  by GAG fines.  Present work includes testing




 of  sampling  concepts  to  detect  coliform  bacteria and research on the assimil-




 able  organic  carbon (AOC) in water  and its relationship to bacterial growth




 and proliferation in mains.  Inhouse and  extramural work on AOC started in




 FY'86  and will continue  into FY'88.  Additional work related to biological




 stability of  treated water will be needed, especailly if ozone is to be used




more extensively In water treatment.




     2.  General Corrison Control Studies




     In  addition to the lead study mentioned previously, field studies on




 corrosion control methods for lead, galvanized steel and copper pipes continue




 to take  place at various communities in  the State of Washington.  The project




emphasizes the practicality of providing realistic assistance to utilities to




implement corrosion control measures.  Finally, bench-scale studies on the




chemistry of silicates and phosphates for corrosion control has been completed



at the University of Missouri.

-------
      3.  Modeling  and  Cost Research




      Extramural  and  inhouse research efforts have  focused on the following




 four aspects  of  water  main distribution system deterioration:  (1)  Computer




 models  for mainframe and microcomputers to  predict water  quality and cost




 have been developed.  This model,  for example,  estimates  TTHM levels through-




 out  the distrlbtuion system based  on time-of-travel and mixing of  water from




 wells of varying quality.   Cooperative work dealing with  the efforts of




 hydraulic behavior on  the propagation of contamination in distribution systems




 has  been initiated with  members  of a French-U.S. bilateral team; (2) Case




 studies have  collected main break  data and  applied statistical techniques to




 identify system  characteristics, environmental  factors and installation




 practices that contribute to abnormally high break rates  that can  impair




 water quality; (3) Case  studies  have also analyzed repair, replacement and




 renovation costs and have  to incorparate data base collection and  analysis




 into a  decision  making strategy  for  utilities to minimize  costs and insure




 hydraulic integrity  and  reliability;  and (4) A  study is underway to develop




 standardized  distribution  system costs  and  includes new construction, repair




 and  rehabilitation.








                            SUMMARY AND  CONCLUSIONS






      DWRD is  active  in conducting  treatability and technology studies to




assist In development  of maximum contaminant levels under  the Safe Drinking




Water Act.  The  Division is  also involved in giving technical advice to




states,  EPA Regional Offices, water  utilities and  professional organizations.




An organization  of major interest  to  EPA is the Association  of State Drinking

-------
Water Administrators.  In many ways the State Drinking Water Administrators




are the first line of implementation of the SDWA.  It is the purpose of this




paper to help provide information to the State Adminstrators and to provide




a point of contact for future information concerning the evaluation of various




kinds of technologies for meeting SDWA maximum contaminant levels.

-------
                                  REFERENCES
1.    Safe Drinking Water Act 1986 Amendments, United States Environmental
      Protection Agency, Office of Drinking Water, EPA 570/9-86-002, August
      1986, Washington, DC  20460.

2.    Clark, Robert M. and  Dorsey, Paul, "A Model of Costs for Treating
      Drinking Water", Journal of the American Water Works Association,
      Vol. 74, No. 12, pp.  618-d627, December 1982.

3.    Fronk, Carol Ann, "Removal of Volatile Organic Chemicals in Drinking
      Water by Ozone Treatment".  Submitted for publication to the Journal
      of  the American Water Works Association.

4.    Sorg, Thomas J., and  Love, 0. Thomas Jr., "Reverse Osmosis to Control
      Inorganic and Volatile Organic Contamination" Submitted for publication
      to  the Journal of the American Water Works Association.

5.    Lykins, Benjamin W.,  Geldreich, Edwin E., Adams, Jeffrey Q., Ireland,
      John C., and Clark, Robert M., "Granular Activated Carbon for Removing
      Nitrohalomethane Organics from Drinking Water (Project Summary)".
      EPA-600/S2-84-165, WERL, Cincinnati, OH, 1984.

6.    U.S. EPA Research Cooperative Agreement CR809974, "Evaluation of Packed
      Tower Aeration for Removal of Volatile Organics from Drinking Water",
      Project Officer:  R.  J. Miltner, Drinking Water Research Division,
      Cincinnati, OH, 1982-1985.

7.    Miller, S. E., DeRoose, F. L., Howes, J. E., Tabor, J. E., Hatchel, J. A.,
      Sueper, C. V., Kohler, D. F. and Degner, K. B., "Determining the Effective-
      ness of an Afterburner to Reduce Dioxins and Furana".  EPA/600-2-86-039,
      WERL, Cincinnati, OH, March 1986.

8.    Lauch, Richard P. and Guter, Gerald A-, 1986.  "Ion Exchange for the
      Removal of Nitrate from Well Water", Journal American Water Works Associa-
      tion, Vol. 78, No. 5, pp. 83-88.

9.    Reid, George W., Lassovszky, Peter, and Hathaway, Steven, 1985.  "Treat-
      ment, Waste Management and Cost for Removal of Radioactivity from Drink-
      ing Water", Health Physics, Vol. 48, No. 5, pp. 671-694.

10.   Federal Register, 1979.  National Drinking Water Regulations, 40CFR Part
      143, Vol. 44, No. 140, Thursday.

11.   Sorg, Thomas J., Forbes, Robert W. and Chambers, David S., 1980.  "Removal
      of Ra-226 from Sarasota County, Florida Drinking Water by Reverse Osmosis",
     Journal American Water Works Association, Vol. 72, No. 4, pp. 230-237.

12.  Lykins, Benjamin W.,  Jr. and Baler, Joseph A., 1985.  "Removal.of
     Agricultural Contaminants from Groundwater".  In Proceedings:  American
     Water Works Association Annual Conference, June 23-27, 1985, pp. 1151-1164.

-------
13.  Huxstep, Martin R., 1981.  "Inorganic Contaminant Removal from Drinking
     Water by Reverse Osmosis", (Project Summary,  EPA-600/S2-81-115, WERL,
     Cincinnati, OH.

14.  Guter, Gerald A., 1982.   "Removal of Nitrate  for Contaminated Water
     Supplies for Public Use:  Final Report".   (Project Summary),  EPA-600/S2-
     82-042, Cincinnati, OH.

15.  Myers, Anthony G., Snoeyink, Vernon L. and Snyder,  David  W.,  1985.
     "Removing Barium and Radium Through Calcium Cation Exchange",  Journal
     American Water Works Association, Vol. 77, No.  5, pp.  60-66.

16.  Syner, David W., Snoeyink, Vernon L. and Pfeffer, Julie L.,  1986.
     "Weak-Acid Ion Exchange  for Removing Barium,  Radium and Hardness,
     Journal American Water Works Association,  Vol.  78,  No. 9, pp.  98-104.

-------
Barriers to Implementing New Technologies
G. Wade Miller, Executive Director, Association of State Drinking
Water Administrators and President, Wade Miller Associates, Inc.,
Arlington, VA
                               IV-1

-------
  AN INFORMATION  SERVICE  OF THE  AWWA  RESEARCH  FOUNDATION
                        Research
                       Quarterly
PROJECT REPORT
Overcoming Barriers to the Introduction of
New Technologies

G. Wade Miller
President
Wade Miller Associates, Inc.
Arlington,  Virginia

 In June 1984, Wade Miller Associates, Inc., and the
Utah Bureau of Public Water Supplies initiated a
project to identify barriers to the introduction of new
technologies into the drinking water field and to
determine how some or all of these barriers can be
eliminated.  Sponsored by the AWWA Research
Foundation, this project focused on the roles and
interactions of equipment suppliers, state plan review
engineers, consulting engineers, and  utilities in the
process of designing and constructing water treatment
facilities and appurtenances. The project was
designed to identify solutions since it was known that
barriers to new technologies exist.
 If the results of the study could be distilled into one
statement, it would be that there are no heroes or
villains responsible for introducing  or blocking new
technology in the potable water supply industry. Each
of the involved parties operates from  a different
perspective and also has a different set of concerns.
These concerns often are either not appreciated or
fully understood by the other parties. These
perceptions and concerns impact specific actions
taken or not taken. All parties are sincerely interested
in seeing that water systems have well-designed, cost-
effective technologies. The failure to achieve this goal
is rooted in the basic decision structure in which all
the major participants operate.
 The rate of introducing  new technologies into water
supply, and in fact into all facets of local government,
has historically been slower than it has been in many
other sectors of the economy. There are several
reasons why this is true.
   « Several types of organizational entities are
     engaged in the process, and their interactions
     often lead to miscommunications and
     breakdowns in the system.
   • The water industry itself is a mature industry; the
     emphasis is on reliability and  use of proven
     treatment techniques and products. Although
     there is innovation and change in some aspects
     of water supply, the rate of change is
     considerably slower than in the field of
     microcomputers or telecommunications, for
     example.
   • Because of public health concerns and the
     emphasis on proven techniques, the number of
     new products being introduced is  low relative to
     other industries. Some products that are
     considered new, such as ozone and  reverse
     osmosis, have actually been used  in  water
     treatment plants since 1906 and 1967,
     respectively.

Organizational Entities

 The five major organizational entities that interact
continuously in the course of water plant  design,
construction, and operation, are: state drinking water
officials who review and approve plans  and
specifications, consulting engineers who design  the
facility, equipment suppliers, the general contractor,
and the utility.
 State regulatory agencies. State drinking  water
administrators are public health officials charged with
the primary responsibility for protecting the public
health by ensuring that utilities deliver a potable, safe
supply of water to the consumer. They  view their role
as one of primary oversight of designers,  equipment
suppliers, and water utility owners and  operators. In
this role, they often must evaluate complex
technologies and rule on the  use of substances whose
chemistry is not always well understood. Long after
the engineer and contractor have completed  their
work and the warranty on the equipment  has expired,
state drinking water officials must monitor drinking
water quality and work with problem systems.
 As a group, state regulators perceive themselves as
understaffed and less than well  equipped in terms of
resources and access to knowledge needed for the
review process. Consequently, there is  a strong
predisposition to favor "tried and true"  technologies.
The principal concern is system operation; cost is

-------
      Water
 Research
Quarterly
                                AN INFORMATION SERVICE OF  THE  AWWA  RESEARCH  FOUNDATION
 relegated to a secondary consideration. Thus, state
 regulators tend to favor old-line technologies with
 conservative design parameters. By requiring unit
 processes to be lightly loaded and/or to have very
 long detention times,  it is believed that the operator
 will have  more time to respond to problems. Because
 of their lack of resources and their commitment to
 protection of the public health, state regulators often
 appear to be arbitrary and conservative in their
 judgments.
  Consulting engineers.  Design engineers are concerned
 with designing a water treatment plant that will
 function efficiently. This generally means that
 equipment selected is tried and proven. The engineer
 has little incentive to specify new, often unproven
 technology. In fact, the engineer has a disincentive;  if
 the plant  does not function properly, liability claims
 and/or damage to the firm's reputation can be
 significant.
  The consulting engineer is the lead professional in
 most water supply system projects. He is responsible
 for determining the client's needs and designing and
 overseeing the installation of a water treatment system
 that will meet those needs within the regulatory and
 public health framework. In effect, he is the primary
 procurer of products and technologies.
  The survey of 20 consulting engineers conducted in
 this study served to confirm that designers are highly
 conservative in their philosophy and not inclined to
 take a leading role in  advocating new technologies
 and products. The key factor that drives consulting
 engineers to rely on well-established older
 technologies is liability. The engineering firm generally
 holds virtually all of the  liability associated with the
 design of a treatment  plant. State regulatory personnel
 have no liability. Equipment supplier liability is limited
 to product warranties. So long as the plant is
 constructed properly,  the contractor is not liable for its
 performance. As a result of this situation, engineering
 firms  must carry expensive liability insurance;
 insurance carriers are quick to cancel or raise fees to
 exorbitant levels if a claim is filed. Hence, the engineer
 has a substantial amount to lose if a new technology
 fails to perform.  Since many of the smaller water
 systems are designed by small engineering firms, a
 major problem can force the firm to close down.
  Equipment suppliers and general contractors. The most
 elegant plans and specifications will not  treat a single
 drop of water until they are converted into the
 physical structure of the treatment plant. The
 contractor and the equipment  supplier, working with
 the engineer, create the actual system from the paper
 abstraction. General contractors are the favorite target
 of criticism by those who conveniently ignore the
 highly competitive, low-bid environment  in which
 general contractors must work.
 The equipment supplier is the primary source of
innovation and forward movement in water treatment
technology. The supplier's environment stimulates
change and innovation since the company's survival
and prosperity depend to a large part on its ability to
improve on the state of the art and thereby create a
salable advantage over competitors in a highly
competitive and price-driven environment. The
supplier performs many valuable services, including
research and  development (R&D), pilot testing,
training and supporting design engineers, and
providing supplies and services to operators. Much of
the funding for professional activities, publications,
and exhibitions comes directly or indirectly from
suppliers seeking to advertise their products.
 Equipment suppliers  are concerned primarily with
selling a product that satisfies the demands of the
marketplace. The supplier is profit motivated.
However, his  profits are often  invested in research or
product development that ultimately benefits the
industry.
 Utilities. Utilities are often the forgotten organizational
entity. Water utilities are usually as conservative as
their state counterparts. Like the consulting engineers,
they have  little incentive to try new products or
process equipment. When they do opt for installation
of an innovative process or treatment technique, the
incentive is the promise of a significant reduction in
costs—capital costs, operation and maintenance costs,
or both.
 Others. The role of the U.S. Environmenta! Protection
Agency (USEPA) in product introduction should be
mentioned briefly. Although the USEPA is the primary
developer  of drinking water regulations, it specifically
does not get involved  in product testing or
endorsement  and is not directly involved in the
"decision loop" of states, consulting engineers,
utilities, and equipment suppliers. It does get involved
in the approval-acceptance of direct and indirect
additives to drinking water,  but only on an advisory
basis. The USEPA's advisory opinions on treatment
chemicals and paints and coatings have never been
mandatory. The state regulatory agencies have
primary authority and  responsibility in this area.  Many
choose to rely primarily on  USEPA's advisory rulings,
making them  mandatory for approval. Since the
USEPA limits its work to issuing advisory opinions on
new  products that are essentially identical in chemical
formulation to previous products, states that limit their
own  approvals to products that have been "approved"
by the USEPA basically bar the introduction of any
new  products of this type in their states.

Specific Barriers

 Certainly one of the principal barriers to introduction
of new technologies is the methodology by which
                                  (continued on page 8)

-------
8
    AN  INFORMATION  SERVICE  OF THE AWWA  RESEARCH  FOUNDATION
     (continued from page 7)

     water treatment systems are designed, purchased, and
     operated. It has been called the "great American low-
     bid process," the "eternal triangle" of the consulting
     engineer who designs the plant, the general
     contractor who builds it (using equipment from the
     supplier), and the owner who operates it. While having
     no direct fiduciary interest, the state drinking water
     administrator, as the principal guardian of the public
     health,  is also very much involved.
      When  something goes wrong, an all too frequent
     occurrence, the situation quickly dissolves into a
     triangle of "finger pointing," each party trying to shift
     the responsibility to the other. The entire system is full
     of loopholes: the contractor may substitute cheaper,
     inferior equipment that local or federal procurement
     regulations force the engineer or owner  to accept; the
     engineer fails to "do his homework," resulting in less-
     than-efficient processes and hardware. The owner
     may take the best designed and constructed plant in
     the world and produce poor quality water by using
     unqualified, untrained operators.
      This process does not  establish a clear, single source
     of responsibility that can efficiently design, construct,
     and operate a plant. The net effect of the system is to
     drive the technology and equipment quality to the
     lowest common denominator, and drive  the costs up,
     to the benefit of no one.
      Another barrier is the lack of a uniform set of
     guidelines  for reviewing new or old technologies.
     Consequently,  the supplier finds that he must meet
     widely divergent standards for treating waters of
     similar  composition that simply happen to be located
     in different geographical and/or political areas. This
     serves to drive up the cost of equipment and to
     complicate system design.
      State regulatory agencies that do not periodically
     review and update their standards may appear to have
     little technical basis for existing standards. In some
     extreme cases, an entire policy having major
     economic  impact on the state's water utilities may  be
     rooted in the personal prejudices of a long-term
     senior employee who has not kept abreast of technical
     evolution. Policies in states operating on this basis
     may be characterized by abrupt changes that
     correspond to the retirement or reassignment of key
     personnel.
      About  one-half of the states rely on the "Ten States
     Standards" (TSS) for all or a major part of their basic
     criteria.  Those states that are not a part of the TSS
     committee  that generates these guidelines tend  to
     interpret the standards in a much  more rigid manner
     than was intended by the authors. This has led to the
     phenomenon of new technologies being approved by
     TSS committee states, while the same are rejected by
     an outside  state citing the TSS as the reason. Use  of
the TSS by so many states reflects a strong desire
among state officials for greater commonality of basic
standards. The problem with the TSS is that they have
institutionalized some  basic criteria that are not based
on best available technology. For example, even
though modern filters  provide effective treatment at
rates far in excess of the two-gallons-per-minute-per-
square-foot guideline listed in the TSS, this
requirement remains in the standards long after it has
been rendered obsolete.
 The combination of the low-bid environment and the
lack of serious prequalification of suppliers has placed
severe profit constraints on the equipment supply
industry. Suppliers have responded by cutting back on
services, reducing warranties, substituting cheaper
designs and materials  in products, and reducing
product R&D efforts. The first cost of the product may
be lower; however, the service life cost and
maintenance headaches associated with cheap
equipment may dwarf  any up-front savings.
 The great lack of communication between equipment
suppliers and state regulatory officials contributes to
an atmosphere of distrust and misunderstanding
between them. Survey responses and conversations
with members of both groups during this study clearly
established that each party suffers from considerable
misconceptions about the other and these misconcep-
tions are allowed to interfere in the working
relationship to the detriment of both parties.

Conclusions and Recommendations

 Even though there are clearly several barriers to the
introduction of new products, the system  does work.
As noted earlier, most water treatment professionals-
state administrators, consulting engineers, equipment
suppliers,  and utility managers—have a common goal
of providing a high  quality product to the consumer.
 The  procurement system, as currently structured, is
inefficient and unwieldy, however. Virtually no
incentives exist for risk taking, either by the consulting
engineers or the state plan review engineers. Utilities
sometimes can benefit from risk taking  in the form of
lower costs of an innovative technology. For the most
part, however, the water treatment equipment supplier
is  selling to a conservative constituency that places a
high premium on reliability and, under most  state
laws,  low-bid prices.
 Of all the key actors  surveyed'during this study, only
the consulting engineers appear to accept the current
system used for procurement, design, and
construction of water  treatment facilities.  The state
drinking water administrators recognize that they are
perceived as "barriers to  innovation" and  would like to
take actions to ensure that they are not barriers. The
more progressive utilities may be frustrated by
conservative state guidelines and would like to see

-------
      water
 Research
Quarterly
                                AN INFORMATION  SERVICE  OF THE AWWA  RESEARCH  FOUNDATION
 changes. Most equipment suppliers are frustrated by
 the present way of doing business, but are relatively
 powerless to change the status quo.
  In addition to the inefficiency of the procurement
 process, the industry is plagued  by a clear lack of
 communication. This lack of communication cuts
 across all groups. The  states, until the recent
 formation of the Association of State Drinking Water
 Administrators (ASDWAi, have not had a forum for
 sharing information with each other. Equipment
 suppliers often appear to develop new products
 without consideration for the concern of state
 administrators. They market their products to the
 consulting engineer and typically consider the
 engineer, and not the utility, to be the client.
 Consulting engineers are criticized by states and
 equipment suppliers alike for not being willing to "take
 more chances."  Engineers are especially criticized for
 not sticking by a specification once written.
  Principal recommendations for actions that can break
 down some of the existing barriers are as follows:
   1. Consideration should be given to the
 development of a uniform set of  national guidelines
 that can be used by all states in the plan review
 process. The feasibility of such a set of standards is a
 controversial topic; however, the states appear to want
 and need the  consistency that would result from this
 action. It is exceedingly important that any such
 guidelines or standards be kept as current as possible.
 Continued use of outdated design parameters based
 on outmoded  technology can be a  major barrier to the
 introduction of new technologies if such parameters
 are incorporated into national guidelines or standards.
   2. There is  a clear consensus among the majority of
 state agencies and most of the other participants for a
 national "clearinghouse" that would support direct
 communication between state regulators, equipment
 suppliers, consulting engineers, and water system
 operators. The clearinghouse could serve as a vehicle
 for disseminating new product information and test
 data, and generally foster the spread of design,
 product, and operational information within the
 industry.
   3. The current practice of  low-bid purchasing
 inhibits the introduction of new technologies by
 causing credible suppliers to expend unnecessary
 resources while watching low-bid suppliers win
 procurements on the basis of the first cost, not service
 life costs. The process  makes it more difficult for state
 officials to approve new technologies, particularly
 packaged treatment systems, for fear of inferior
 substitutions to approved products  at the time of
 bidding. Numerous examples of successful equipment
 and service procurement on a negotiated or
 prequalified bid basis exist in the municipal sector
overseas (i.e.,  as in France where two large, vertically
integrated companies dominate the water industry),
and in the municipal privatization and industrial
sectors in the United States.
 Feasible alternatives to the current practice should be
defined, and steps should be taken to educate
municipal buyers and consulting engineers in a more
effective  procurement approach. The lead role would
have to be taken by USEPA, the American Water
Works Association (AWWA), and ASDWA since
consulting engineers seem to accept the current
procurement process.
  4.  In the current environment, exhorting the
consulting engineer to "take more  risks" on new
technology will  have no significant impact unless
some mechanism for reducing his  liability in the event
of failure is also developed.  A definitive analysis of the
current liability  situation and development of
alternative models that would share liability among the
principal  participants (i.e., designers, equipment
suppliers, contractors, system owners, and states)
should be undertaken by AWWA, ASDWA, or other
interested parties.
  5.  State regulatory agencies need to look closely at
their decision making to assure that it is consensus-
based and not dominated by one, or several, strong-
willed staff members who have not kept  pace with the
development of water treatment technology. States
should emphasize, even require, participation in
professional educational activities by those staff
members who are reviewing plans  and specifications
and who  are responsible for design criteria. Resources
should be made available for site visits to treatment
plants that use new  technologies, for participation in
pilot studies, for sharing information and exchanging
ideas with plan  review engineers in other states, and
for travel to regional or  national conferences or
seminars dealing with new technologies.
  6.  Equipment suppliers need to recognize the
importance of state drinking water officials and make
a concerted effort to work with and understand their
constraints and  perspectives. In particular,
manufacturers need to understand the public health
implications that drive the regulatory process and
address those issues in  their product submittals and
research  efforts. Also, new product development
should include input from key state officials as early
as possible, so that health and operational concerns
can be recognized and  incorporated.
  7.  One of the most important and immediate actions
that should be taken is  sponsorship of a workshop (or
a series of workshops) that would  bring  equipment
suppliers, states, and consulting engineers together in
a forum conducive to constructive  dialogue.  USEPA's
Office of  Drinking Water in conjunction with the
Drinking  Water  Research Division should provide
funding for such workshops, which should be
conducted by ASDWA alone or with the AWWA
Research Foundation.

-------
Technical Session: Filtration
Sigurd P. Hansen,  Senior Engineer, CWC-HDR,  Inc.,  Cameron Park,
CA

Material not enclosed — see separate handout.
                               V-l

-------
Technical Session: Disinfection/By-Products
Rip G.  Rice,  President, Rice  International  Consulting Enterpri-
ses, Ashton, MD
                               VI-1

-------
                           CONTENTS

I.   INTRODUCTION   	   1

II.  DEFINITIONS	   2

III.  EPA DISINFECTION REGULATIONS	   4

    A.  STATUTORY REQUIREMENTS	   4

        1.    Variances and Exemptions	   5
        2.    Anticipated Timing of Disinfection Regulations	   6

    B.  THE PROPOSED SURFACE WATER TREATMENT REGULATION  7

        1.    Applicability	   7
        2.    Microorganisms to be Regulated  	   7
        3.    Appropriate Treatment Technologies   	   9
        4.    Turbidity Limitations   	  10
        5.    Minimum Disinfection Requirements   	  10

             a.   For Systems Which  Do Not Filter	  10
             b.   For Systems Which  Filter	  13

IV.  EPA DISINFECTION BY-PRODUCTS REGULATIONS	'15

    A.  STATUTORY REQUIREMENTS	15
    B.  CANDIDATE DISINFECTANT/OXIDANT BY-PRODUCTS TO BE
        REGULATED	  16
    C.  ANTICIPATED TIMING OF REGULATIONS	  18
                 e
V.  BEST AVAILABLE TECHNOLOGIES AND TREATMENT TECHNIQUES 20

    A.  STATUTORY REQUIREMENTS	  20
    B.  DEVELOPMENT   OF  CRITERIA  FOR  BEST  AVAILABLE
        TECHNOLOGIES	  21
    C.  BAT FOR DISINFECTION	  21
    D.  BAT TECHNOLOGIES FOR DISINFECTION BY-PRODUCTS  .  22
    E.  STRATEGIES FOR CONTROLLING DISINFECTION BY-PRODUCTS
         	  23

VI.  DISCUSSION OF  DISINFECTANTS AND OXIDANTS	24

    A.  BASIC ISSUES  	  24

        1.    First  Approach	  25
        2.    Second  Approach    	  25
        3.    Third Approach   	  27

-------
     B.   TECHNICAL ISSUES » DISINFECTION AND DISINFECTANTS   28

         1.    For Systems Treating Surface Waters	28

              a.   The "CT" Value   .	   28
              b.   Suggested No Adverse Response Levels (SNARLs) ...   29
              c.   Possible Regulatory Consequences	   30

                  1.   EPA'S PERSPECTIVE ON RECOMMENDED NAS
                       SNARLS	31

         2.    For Systems Treating Groundwater	   33

              a.   "CT" Values  and SNARLs	33
              b.   Possible Regulatory Consequences	   34

     C.   TECHNICAL    ISSUES    --   DISINFECTION/OXIDATION
         BY-PRODUCTS	   34

         1.    Non-Halogenated DOBs	   35

              a.   From Ozonation	   35
              b.   From Chlorination   	   37

     D.   COMPARISON OF DISINFECTANTS - OXIDANTS	37

         1.    General Considerations    	   37
         2.    Chlorine   	   39
         3.    Chlorine Dioxide  	40
          	   41
         4.    Monochloramine   	   42

              a.   Chloramine Summation	43

         5.    Ozone    	   44

              a.   General Considerations	   44
              b.   Disinfection With Ozone   	   45
              c.   Microflocculation	   46
              d.   Promotion  of Biodegradability	   46
              e.   By-Products of Ozonation	   47
              f.   Catalytic Ozonation	   47
              g.   Ozone  Oxidation of Bromide Ion   	48
              h.   Summation for Ozone	48

         6.    Ultraviolet Radiation for Groundwaters   	49

VII.      DISCUSSION    OF   TREATMENT   TECHNOLOGIES   FOR
         DISINFECTION AND TO MINIMIZE  PRODUCTION OF DOBs    51

     A.   THE PROBLEM   	   51
     B.   THE STRATEGIES  	   51

                                    ii

-------
C.   DISINFECTION TECHNOLOGIES   	   51

     1.    Primary Disinfectants   	   52

           a.    Chlorine	52

                i.    Chemistry of Chlorination   	   53
                ii.   Establishing A Chlorine Residual	   55
                iii.   Factors    Affecting    Disinfection   Efficiency    of
                     Chlorine   	   57
                iv.   Disinfection With Chlorine Gas	57
                v.   Disinfection With Sodium Hypochlorite  Solution   .   58
                vi.   Disinfection With Solid Calcium Hypochlorite   .  .   60
                vii.   Chlorination System Design    	   60
                viii. Chlorination Systems Costs   	   67

           b.    Ozone	86

                i.    Characteristics  and Properties of Ozone	86
                ii.   Generation of  Ozone   	90
                iii.   Contacting of Ozone With Water	91
                iv.   Destruction of Contactor Exhaust Gas    	93
                v.   Chemistry of Ozone in Water   	   93
                vi.   Establishing An Ozone Residual   	94
                vii.   Factors   Affecting  the  Disinfecting  Efficiency   of
                     Ozone	.97
                viii. Ozonation System Design	   97
                ix.   Costs of Ozonation Systems   	107

           c.    Ultraviolet Radiation	 120

                i.    General  Description of the UV Process	120
                ii.   UV  Disinfection  System Design	122
                iii.   Estimating the Average  Intensity  in  a  UV  Reactor
                     	124
                iv.   Water Quality  Considerations  in  the  Design  of  a
                     UV  Disinfection  System	124
                v.   System  Design and  O&M Considerations  for  the
                     UV  Process    	125
                vi.   System    Design    Considerations   for   Effective
                     Maintenance   	127
                vii.   Costs for Disinfection  With UV Radiation .... 128

     2.    Secondary Disinfectants   	130

           a.    Chlorine Dioxide	133

                i.    Generation of  Chlorine Dioxide    	136
                ii.   Oxidation-Reduction Reactions  of Chlorine  Dioxide
                     •   "-.	139
                iii.   Establishing a  Chlorine Dioxide Residual   .... 141
                                     111

-------
                    iv.   Factors   Affecting  the   Efficiency   of   Secondary
                         Disinfection  With  Chlorine Dioxide    	141
                    v.    Chlorine Dioxide Systems Design	142
                    vii.   Costs of Chlorine Dioxide Generating Systems   .  .  146

               b.   Chloramination	152

                    i.    Chemistry of Chloramination	153
                    ii.    Establishing  a Chloramine Residual	153
                    iii.   Chloramination System Design  	154
                    iv.   Costs for Chloramination   	155

          3.    Oxidants	157

               a.   Potassium Permanganate	157

VIII. CASE EXAMPLES OF EMERGING TECHNOLOGIES	  .  162

     A.   OZONE CASE HISTORIES	162

          1.    North Andover, Massachusetts

               Ozone Disinfection  for  Giardia lamblia   	162

               a.   The Problem   	162
               b.   The Interim Solution	163
               c.   The Results	.....' 164
               d.   For The Future	  164
               e.   Note - Sturgeon Bay, WI	165

          2.    Kennewick, Washington  (Cryer, 1986) -

               Preozonation  For THM Control	  165

               a.   The Problem	165
               b.   Pilot Plant Study Results	166
               c.   Plant Design	166
               d.   Operational Experiences	  166

                    i.    General   	166
                    ii.    Water Quality   	168
                    iii.   Applied  Ozonation   Dosages  -  Dissolved  Ozone
                         Residuals    	168
                    iv.   Ozone Equipment Operational Experience   ....  168
                    v.    Costs Of Ozonation	169

               e.   Implications of Proposed Surface Water Treatment  Rule
                    Disinfection Conditions	169
                                        iv

-------
B.   UV RADIATION  CASE HISTORY	17°

     1.    Ft.  Benton, Montana

          UV Radiation for Primary Disinfection   	170

          a.   The Problem   	170
          b.   The Solution   	170
          c.   UV Radiation Conditions	171
          d.   Costs	171
          e.   Operating Experience	171

C.   CHLORINE DIOXIDE CASE HISTORIES    	172

     1.    Evansville, Indiana
          Predisinfection for THM Control	172

          a.   The Problem   	172
          b.   Pilot Plant Study	172

               i.    Optimization Phase    	172
               ii.    Long-Term Evaluation	172

          c.   The Full-Scale Plant   	173
          d.   Operating Experience	173
          e.   Implications of the Proposed Surface Water Treatment
               Rule    	174

     2.    Hamilton, Ohio

          Primary/Secondary Disinfection  With Chlorine
          Dioxide	175

          a.   The Problem   	175
          b.   The Treatment Process	175
          c.    Generation of  Chlorine Dioxide	175
          d.   Effects of Installation of Chlorine Dioxide    	176
          e.   Costs for Chlorine Dioxide	  176
          f.    Implications of the Proposed SWTR CT Values    ...  176

     3.    Galveston, Texas

          Preoxidation With C1O2
          Post-Disinfection With C1O2  + Chloramine   	177

          a.   The Problem   	177
          b.   The Original Treatment Process	177
          c.    Study of Alternatives	178

               i.    Study Results	178

          d.   The Results	180

-------
     C.    CHLORAMINE CASE HISTORIES	181

          1.    Bloomington, Indiana

               Prechlorination, Post-Chloramination	181

               a.    The Problem	181
               b.    The Solution   	181
               c.    Performance	  181

          2.    Philadelphia, Pennsylvania

               Pre-Chlorine Dioxide + 1-Hour Prechlorination
               Post-Chloramination	183

               a.    The Problems	183
               b.    Process Modifications	183

                    i.     Chloramination  of Finished  Water  (Reduction  of
                         Free  Chlorine  Contact  Time from  96-hours to 24-
                         hours)  	184
                    ii.    Reduction  in  Chlorine  Treatment  at  Raw Water
                         Basin   	184
                    iii.    Utilization  of  Chlorine  Dioxide  at  Raw Water
                         Basin Inlet (5 Hours Free Chlorine Contact Time)
                         	  184
                    iv.    Installation of a  New   Chlorine  Application Point
                         (One Hour Free Chlorine Contact Time)   ....  184
                    v.    Ten Minutes Free Chlorine Contact  Time  . .  .  . • 185

               c.    Economics	185
               d.    Operational Improvements    	185
               e.    For the Future   	186

IX.  SUMMARY RECOMMENDATIONS FOR DISINFECTION STRATEGIES
     AND  FOR  THE  CONTROL  OF  DISINFECTION/OXIDATION  BY-
     PRODUCTS   	186

     A.    For Disinfection   	186
     B.    For Controlling Disinfectant/Oxidant By-Products   	188

X.   REFERENCES	189
                                       VI

-------
         EMERGING TECHNOLOGIES FOR COMPLIANCE WITH

   DISINFECTION AND DISINFECTION BY-PRODUCTS REGULATIONS


                             Rip G. Rice. Ph.D.

                   Rice International Consulting Enterprises
                             1331 Patuxent Drive
                             Ashton, MD  20861
I.  INTRODUCTION

The  Safe Drinking  Water Act  Amendments of 1986  as  related to  Disinfection
and  Disinfection  By-Products  will  require  all  community and non-community
water  supplies to disinfect their  drinking water  supplies.   However, this must
be accomplished in  such a  manner  as to minimize or eliminate the formation
of by-products  of the  disinfection process.    Since  some of  the  disinfectants
employed  in  drinking  water  treatment  (e.g.,  chlorine,  chlorine  dioxide,  and
ozone)  also  are  used for oxidative purposes  other  than  disinfection, the  same
by-products  can be produced during such non-disinfection applications.

Historically,  chlorine  has  been  the  primary  disinfectant  utilized by American
water  utilities.   Its  use  has  produced (and  continues  to  produce) excellent
results  in  terms  of  killing  and/or  inactivating pathogenic  microorganisms,  and
also  has  provided  many  additional  water  treatment  benefits  (e.g., iron,  and
manganese removal,  color removal, sometimes  taste and odor  control, etc.).

On  the   other   hand,   chlorine  also  is   an   excellent   chlorinating   agent,
producing  trihalomethanes  (THMs)  and  other  halogenated  organic  materials
during  water treatment.  Consequently,  other disinfectants  are being  studied
and  utilized  (chlorine dioxide,  monochloramine,  ozone)   as  well  as  combina-
tions   of  oxidants/disinfectants   (i.e.,  ozone   followed  by  chlorine,  chlorine
dioxide,  or  monochloramine)  to "replace"  chlorination,  or  at  least  to lower
the amounts  of  chlorine  applied  rather  cavalierly in  the  past without  regard
to the chemical by-products produced  by such  practice.

Analogously,  each  of  these  alternate  disinfectants  or   combinations  thereof
can be expected  to produce  its own  set  of  oxidation/disinfection  by-products.
Therefore,  EPA's  major problem with  respect to this subject  area  is to  study
the  efficacy   of   each   potentially   promising   alternative   water  treatment
scheme,  identify  the   by-products  which  can  be  or   are   produced,  and
determine their health effects.

Disinfection  By-Products  currently are  discussed  by  some  as   a  topic  related
solely  to the  "disinfection1',, step, without  considering  that  the  identical  by-
products can be  produced- during  non-disinfection  applications of the  same
oxidant/disinfectant.   When  breakpoint chlorination is employed for ammonia
removal  early  in  the treatment  process, the same  types  of  halogenated  by-

                                       1

-------
products  can be produced  as  when chlorine is added for  disinfection  after
filtration,   given   the   presence   of  the  same   types  and  concentrations  of
organic materials, and the same  water parameters.

Ozone and potassium permanganate  (KMnO4) are  prime candidates for coping
with  many  Synthetic  Organic  Chemicals  (SOCs) to  be  regulated  under  the
SDWA by oxidation  early  in  the water  treatment process.   As  a  result of
adding  ozone  to  raw  water supplies  (for turbidity,  color,  and/or  taste  and
odor  control,  microflocculation,  oxidation  of THM precursors,  or  iron  and
manganese   oxidation,  for  example),  oxidation  by-products will  be   produced
which  are of the same  general  types as  when  ozone is employed late in the
water treatment process as  the primary disinfectant.

Similarly,   when   potassium  permanganate,  rarely  used,  if  ever,  as   a
disinfectant,   is   applied   for   oxidative  purposes   (iron,   manganese,  color
removal,  taste and odor control), many  of the same  oxidation products  are
produced  as when  ozone,  chlorine dioxide,  or even chlorine are  employed as
disinfectants  or  oxidants (see  Rice  &  Gomez-Taylor,   1986,  and   references
cited  therein).


II.      DEFINITIONS
   A.    DISINFECTION   ~   The   killing   or   inactivation   of  (pathogenic)
         microorganisms.

Attainment   of  the   desired   degree  of  disinfection  is  a  function  of  the
concentration of the  disinfecting  agent  (in  mg/L)  times the contact  time (in
minutes)  (the  "CT1  value,  in  mg/Lrmin).    For example,  with chlorine at  a
concentration of 2  mg/L and  a  contact  time  of 30 minutes,  the product  of
[concentration  x  contact  time]  is  60.  On the  other hand, with ozone applied
under  viral  inactivation conditions (0.4  mg/L maintained over  4 minutes;,  the
"CT" product is 1.6.


   B.    PRIMARY  DISINFECTION -  The major  disinfection step  normally
         practiced  after  filtration   in   a   surface   water  treatment  plant
         employing   conventional   or   direct   filtration.     In  groundwater
         treatment  plants,   primary  disinfection  also  can  be  applied  post-
         filtration  (if  filtration  is  used).   If filtration  is not  incorporated  in
         groundwater   treatment,   primary   disinfection  is  practiced   in   the
         treatment  plant, prior to the  water entering the distribution system.

Oxidants/disinfectants   which  are   used  currently  for  primary  disinfection  are:
chlorine, ozone, ultraviolet radiation, chlorine  dioxide, and monochloramine.


   C.    SECONDARY DIS|NFECTANT  (RESIDUAL  DISINFECTANT)  -  A
         disinfectant  addecp to   finished  water  to  maintain   a  disinfecting
         residual  concentration throughout the water distribution system.

-------
Current  secondary  (residual)  disinfectants  are:     chlorine,  chlorine  dioxide,
and  monochloramine.


   D.    POINT  OF  DISINFECTION  -  The  point  in  the  water  treatment
         process  at  which  disinfectant is added for the  purpose  of  killing  or
         mactivation of microorganisms.


   E.    DISINFECTION/OXIDATION BY-PRODUCTS  -- Materials produced  or
         formed whenever a disinfectant or oxidant  is added  to water.

These  by-products can be materials formed by:

(a)      Reduction  or   disproportionation  of  the  disinfectant   (or  oxidant)
         itself,   e.g.,  chlorite   and   chlorate  ions   from   chlorine   dioxide,
         dissolved oxygen from ozone, chloride ion from chlorine,

(b)      Oxidation  of organic materials present in  the water,  i.e., aldehydes,
         ketones,  alcohols,  and  carboxylic acids from   the  reactions  of humic
         acids with ozone, chlorine, chlorine dioxide, KMnC>4, or

(c)      Halogenation   of  organic  materials  upon  the  addition  of  halogen-
         containing  disinfectants,  e.g.,  trihalomethanes   and  other  halogenated
         organics   during  chlorination,   organic    chloramines   produced   by
         reaction of monochloramine with organonitrogen compounds, etc.

If bromide  ion is  present in the  raw water,  it can be  oxidized  by  ozone  or
chlorine  (not by  chlorine  dioxide  or chloramine)  to  form  hypobromous  acid,
which,  in  turn,  can  brominate  many  organic materials.    Bromine-containing
trihalomethanes,  for  example,  are  known   to  be  formed  by  this  reaction
mechanism.

Until recent years,  it had been  considered  that chlorine is added to drinking
water  supplies primarily  to disinfect  or to remove  ammonia.    When chlorine
is added to  raw water to  remove  ammonia  by  breakpoint  chlorination,  the
oxidative  function  of  chlorine  is  utilized.    Although  disinfection   also  is
obtained  when  the  chlorine  is  added,  nevertheless  the  primary  purpose  of
breakpoint  chlorination   early  in  the  water  treatment  process  is   oxidative
removal  of nitrogen (ammonia).     On the  other  hand,  since  the chlorine is
added  at  the point  of greatest  concentration of  organics capable  of  being
chlorinated,   the   maximum  concentration of  halogenated   oxidation  products
will be produced simultaneously under these circumstances.

Therefore, and  in  light  of current, more  complete  understanding  of the  roles
of various  disinfectants   as  also  being strong chemical oxidants,  it should  be
recognized that anywhere  in  the  water treatment process  a strong oxidant is
added  for  the   purpose  of  oxidation,  disinfection  may  occur  simultaneously.
The  degree  of  disinfection  gbtained under  these  circumstances  will  depend
upon a   number  of  factors,  "Including  the   types   of  organisms  present,  the
specific  oxidant/disinfectant added, the  water temperature,  and  the  time  the

-------
disinfectant  is  in  contact  with  the  microorganisms  present  (i.e.,  the  "CT"
value).

Conseouently,  the  point(s)  of  disinfection  today  should  be  redefined  as  any
point(s)   in   the   water  treatment  process  at  which   a   specific   oxidant/-
disinfectant  is added for  any  purpose.   The  same  "disinfectant  by-products"
can  and will be produced, given the presence of  the  same types and levels of
"disinfectant   by-product  precursors",  and   the  same  water   conditions   (i.e.,
temperature, pH, ionic strength,  alkalinity, hardness, etc.).
From  a  nomenclature  point of  view,  therefore,  it  is  more appropriate  to
refer  to  "Disinfectant/Oxidant  By-Products"  (DOBs),  rather  than  merely  as
Disinfection By-Products (DBFs).


III.      EPA DISINFECTION REGULATIONS

   A.    STATUTORY REQUIREMENTS

Section  1412(b)(8)  of  the  Safe  Drinking  Water  Act  Amendments  of  1986
(enacted  June  19,   1986)  requires  the  EPA  Administrator  to  propose  and
promulgate  national primary drinking  water  regulations requiring disinfection
as a treatment technique  for all public water  systems  within 36  months  (by
June 19,  1989).   Simultaneously,  the  Administrator is  required  to promulgate
a  rule  specifying  criteria  that  will  be  used  by  EPA or  delegated  State
authorities  to  grant  variances  from  this  requirement,   according   to  . the
provisions   of  Section  1451(a)(l)(B)   and  14l5(a)(3).     This  disinfection
requirement   applies  to  all  waters  used  for  drinking purposes,  including
groundwaters.

Section  1445(a)   requires  the EPA  Administrator  to  promulgate  regulations,
not  later than  18  months  after  enactment of  the Safe Drinking Water Act
Amendments  of  1986  (by December  19, 1987),  requiring  every  public water
system  to   conduct  a  monitoring  program  for  unregulated   contaminants.
These  regulations shall require  monitoring  of  drinking  water  supplied  by  the
system,   and   shall   vary    the   frequency  and  schedule   of   monitoring
requirements  for  systems  based  on the  number  of persons served  by  the
system, the source  of  supply, and  the  contaminants likely to be  found.   Each
system  shall be  required  to   monitor once  every five  years  after  the  effective
date  of  the  Administrator's  regulations,   unless  the   Administrator  requires
more frequent monitoring.

This  monitoring  requirement will apply  to  all  disinfection  and  disinfection
by-product  regulations  to  be  discussed  in  this  presentation,  as  well  as  to
compounds  listed as Volatile and  Synthetic  Organic  Chemicals  (VOCs  and
SOCs).

-------
         1.    Variances  and Exemptions

A water utility  will be required  to disinfect unless it successfully applies  to
the  state primacy  agent for a  variance.   At this time, criteria for obtaining
variances from  disinfection  requirements  are  not  yet  available.    However,
proof  of microbiological safety during  the  variance period  can be  assumed  to
be   a   determining  factor,  and  the  State  must  find  that  in  granting  the
variance, an  unreasonable risk to the public health  will not result.

The   respective   roles   of  variances   and  exemptions  under   the  SDWA
Amendments  differ, but  the  logic  is  clear.   Variances can  be granted only  on
the  basis that because  of  poor raw water  quality, the  MCLs  cannot be  met,
even after application  of Best Available Technology.   There  does  not appear
to be  any endpoint  for  compliance  under  the variance  provision, although the
states  must  impose a  compliance  schedule as  a  condition  of the  variance
[Section 1415].

In  EPA's  proposed  Surface  Water  Treatment  Rule  (U.S.  EPA, 1987a),  EPA
states:

   "...  due  to the acute and high risk associated with  poor disinfection  of
   surface waters, EPA  is proposing  that  no variances be allowed."

EPA  also states that  no  variances  will be allowed  in the  requirements for
filtration  of  surface  water supplies,  in the  same  proposed  Rule  (U.S.  EPA,
1987a).   This position  is taken under  Section  1412(b)(7)(C)(ii) of  the  SDWA,
which  states  that  in  lieu  of the variance  provisions,  the  EPA is to  specify
criteria  by  which  States  will determine which  public  water  supplies  will  be
required to  filter.

Exemptions from any requirement respecting a maximum contaminant level  or
any  treatment technique requirement,  or from both, are allowed  based on the
inability  to  comply due to  "compelling  factors",  which  may include economic
factors.   As  with  variances, the  state must  find  that  the  granting  of  an
exemption shall  not  result  in an  unreasonable  risk to  health  [Section 1416(a)-
(1)].   However,  the  maximum length of time for  an exemption is  three  years,
and  at  the  time  the  exemption  is   granted,   the  State  must   prescribe   a
schedule  for  compliance  [Section  1416(b)(2)(B)].    For  smaller  connection
systems  (under  500),  the  exemption  can   be  extended for  several  two-year
periods  [Section  1416(b)(2)(C)],  if  the  system  establishes  that  it is  taking  all
practicable steps  to  meet the  requirements of Section 1416(b)(2)(B).

In the proposed Surface Water  Treatment Rule  (U.S. EPA,  1987a),  EPA  is
proposing that no  exemptions  be allowed  from  the  requirement  of providing
disinfection,  but  that  exemptions  be  allowed  for the  degree  of  disinfection
required  and  for  meeting the  filtration requirements.    For  example,  under
certain   conditions,  an  unfiltered  system  might  obtain an  exemption   if  it
achieved  a   99%  inactivation  of   Giardia.  but not  the  required  99.9%
inactivation.   Guidance  for  determining  conditions under  which  an  exemption
might  be granted is discussed in the Surface Water Treatment  Rule Guidance
Manual (US. EPA,  1987e).

-------
Additional  consequences  of  requiring  disinfection   also  can   be  anticipated.
For example,  many  groundwaters are  not  disinfected  today.    Those  which
contain   iron   and    manganese  will   form   insoluble   oxides    when    a
disinfectant/oxidant  such as  chlorine,  chlorine  dioxide, or  ozone  is  added.    In
turn,  this  will  require  the  addition  of  a  filter for  their removal.    Thus  the
affected groundwater  systems  may  also  be  forced  to  install  filtration as  a
result of being  required  to disinfect.

Having  installed  filtration   to  cope  with  insoluble  materials  produced upon
oxidation,  the  question  now  becomes,  should  primary  disinfection  gjso  be
practiced  after  filtration?     If  so,   this  will  open the  door  to  two-stage
oxidation/disinfection.     This   is  a   prevalent  treatment  practice   in  many
modern groundwater treatment  plants.

EPA   also   may   require   redundancy   of  the  necessary  disinfectant  feed
equipment,  along   with an   auxiliary  power  source,  automatic  start-up  (and
alarm)  to ensure  that  continuous disinfection is provided.   Many  utilities  fall
short of having such facilities today.


         2.   Anticipated Timing of Disinfection  Regulations

EPA anticipates promulgating  the Surface Water  Treatment Rule  (SWTR)  in
mid-late  1988,  which  will  include  requirements for  filtration and  disinfection.
Treatment  processes  (filtration  and  disinfection) will be  discussed,  individual-
ly  and  in combination,  for  removal  of specific contaminants.   Such  processes
will include   the use  of powerful oxidants (e.g., ozone,  chlorine, and chlorine
dioxide),  which also  can   provide  primary  disinfection   at  their   point(s)   of
addition.

In  addition  to  the  Surface Water Treatment Rule, EPA  intends  to  promulgate
additional  regulations   specifying  disinfection  requirements  for  systems  using
groundwater  sources,  and  possibly more  specific  disinfection  requirements  for
surface  water  systems.     This  Groundwater  Treatment  Rule currently  is
planned  to   be  proposed   early  in   1991.   It  will  include   regulations   for
Disinfection   By-Products (when  health  risks  associated   with  disinfection   by-
products   are  adequately   evaluated),  revision  of  THM   regulations,   and
proposed  MCLs  for  the  disinfectants  themselves  (e.g.,   chlorine,  chlorine
dioxide, chlorite/chlorate ions, and monochloramine).

In  early  1988,  the  Environmental   Protection  Agency  (U.S.  EPA,  1988a)
promulgated  the   First  Drinking  Water   Priority   List,  which  contains   53
substances or classes of substances that  are candidates  for  regulation.   These
include  the  above  disinfectants, ammonia,  the  four  THMs  currently  regulated,
ozone   by-products,  cryptosporidium.   and  many  additional halogenated organic
materials, some of which are by-products of chlorine disinfection.

At  the  same time  (U.S. EPA,  1988a), EPA finalized the list of 83 substances
mandated  by the  Safe  Drjijking Water Act Amendments to be regulated by
June  1989.    Many of these  substances   are  at  least  partially  reactive with
chlorine dioxide, ozone  and potassium  permanganate.   Consequently,  they  may
be  destroyed, or  at  least   converted  in chemical  form  by  pretreatment with

-------
strong  oxidants,  giving  rise  to  byproducts.    Such  preoxidation  will  change
the  way  these   oxidized  derivatives  are  removed  in  subsequent  processing
steps.


    B.    THE PROPOSED SURFACE WATER TREATMENT REGULATION (U.S.
         EPA, 1987a)

         1.    Applicability

This  regulation  will be  applicable  to all  water  treatment  systems which use
surface water, because all  of these are at risk, at least  to  some degree, from
contamination  by Giardia  lamblia.  enteric  viruses,  and pathogenic  bacteria.
Surface  water is  defined  as all  water open  to  the  atmosphere  (e.g.,  rivers,
lakes,   streams,    reservoirs,  impoundments),   and   all   springs,  infiltration
galleries,  wells,  or  other   collectors  that  are  directly  influenced  by  surface
water.

"Directly  influenced  by  surface  water"  means  that  the  source  is subject to
pathogen  contamination  from  surface waters.    This  determination is   to  be
made  on a  case-by-case basis.   The  State  will be responsible for  determining
which   systems   use  surface  water  and,   therefore,   will  be  subject   to  the
requirements of this  rule.

EPA  estimates  that more  than  90%  of the  systems  which  do  not  filter
currently  serve  less  than  10,000 people.   Therefore,  the  impact  of  these
proposed regulations will be maximum on small treatment systems.


         2.    Microorganisms to be Regulated

EPA  plans  to  regulate  the microorganisms listed in  Table  I,  plus  turbidity,
which   is  included  because it   is  an  indicator  of  disinfectant  performance.
MCLGs for  Giardia cysts  and  enteric  viruses  are  given  as "zero",  not with
respect to  a specific water  volume,  but rather conceptually.   EPA's  intent  is
to  approach  the  MCLG  by requiring treatment technologies  (e.g.,  filtration
and disinfection).

Under  the  Surface  Water  Treatment Rule,  all community  and  noncommunity
public water  systems using  any  public water  source  will  be  required to treat
their  surface  water  source(s) so  as  to  achieve at least 99.9 percent  removal
and/or  inactivation  of  Giardia   cysts,  and  at  least  99.99  percent   removal
and/or  inactivation  of  enteric  viruses.     This  can  be   accomplished  by
disinfection  alone,  or  by  a  combination  of filtration  and  disinfection.   No
variances  or exemptions will be allowed for disinfection requirements.

At  a  minimum,  treatment  for any  surface  water  will  include disinfection.   In
addition,  unless  the system  meets  certain  site-specific  criteria,  treatment  also
will  include  filtration.    No variances  from  filtration  will  be  allowed;  but
exemptions will be  permitted^

-------
     TABLE I.  MICROORGANISMS AND DISINFECTION GOALS
Parameter
Total Coliforms
Turbiditv
Giardiad
Proposed11
MCLG
0/100 mL
zero
Proposed11
MCL
b
0 S NTTJC
1,000 fold removal
     Pathogenic
        Viruses"           zero

     Heterotrophic       	
  Plate Count (HPC)e
     Legionella             zero
     and/or inactivation

      10,000 fold removal
      and/or inactivation

  compliance with filtration and
disinfection criteria will ensure
     HPC control.

  compliance with filtration and
     fection requirements will
     remove and/or  inactivate.
a)  U.S. EPA, 1987a.

b)  Proposed MCL is based  on presence or absence  of coliforms in a  sample,
    rather than  on density.   No more than 5  percent of the  most recent 60
    consecutive  samples can  be  coliform-positive  for  systems  analyzing fewer
    than  60 samples/year  (3,300  persons  or   fewer),  and  no  more   than  5
    percent  of the total  number of  samples analyzed  in  the  most  recent  12-
    month  period can be  coliform-positive  for  systems analyzing  at least 60
    samples/year  (3,300  persons or  more).   In  addition,  no  more than  one
    sample/month  if less  than 40  samples are  collected per month,  or  5% of
    the  samples/month if 40 or  more  samples/month  are  collected,   can  be
    coliform-positive.

c)  NTU  =  Nephelometric Turbidity  Unit;  filtered water turbidity.   States  may
    allow  less stringent levels (<,  I NTU) for  systems using ozonation at CT
    values  achieving  99.9  percent  inactivation   of  Giardia  cysts.   All  systems
    are  expected to optimize treatment  so as  to achieve the lowest turbidities
    feasible  at all times of the year.

d)  Existing   analytical  methods  are   not  considered   to  be  technically  and
    economically  available  for use  in  compliance monitoring.    Consequently, a
    treatment technique regulation is proposed.

e)  Former  terminology - Total Plate Count


Systems  with very clean and i protected  source waters  (i.e.,  low total  coliform
or  fecal  coliform  levels  and  low turbidity levels as  specified in  the Rule)
would  only  be  required to use  disinfection  to  achieve  the  required  degrees of
inactivation  of  Giardia cysts and viruses.   If  such  systems  are  continually
                                        8

-------
able to  meet  certain disinfectant "CT"  requirements specified in  the  Rule, the
water  system can be assumed to be  in compliance with  the  above inactivation
requirements  for  Giardia  and  viruses  without  having  to monitor  for  these
organisms.

Systems   with   no  potential  sources  of  human   enteric  viruses  within  the
watershed  (at  the  discretion  of  the  State)  would  not  be  subject  to  the
requirement for  99.99  percent  removal and/or  inactivation  of enteric  viruses.
It  should  be   noted,  however,  that  this  provision will  only  benefit  systems
using  chloramination for  primary  disinfection,  since  other  disinfectants  (e.g.,
ozone, chlorine, chlorine dioxide), which  achieve  99.9  percent inactivation of
Giardia  cysts  also  achieve  much greater  than 99.99  percent inactivation of
enteric viruses at the same time.

Systems  required  to filter  could  use  a variety of technologies  to  meet  the
minimum   99.9  and  99.99   percent  performance  levels.     To  satisfy  this
requirement, systems  would be required to meet certain  turbidity removal  and
disinfection  performance  criteria,   and  comply   with  design  and  operating
criteria specified by the State.


         3.    Appropriate Treatment  Technologies

Conventional    treatment   (which   includes    coagulation,    flocculation,
sedimentation,    rapid   granular   filtration,   and  disinfection)   has   been
demonstrated to  achieve at   least  99.9 percent  removal  and/or  inactivation, of
Giardia   cysts   and   99.99  percent   removal  and/or   inactivation  of   enteric
viruses,  under  appropriate  design and operating  conditions  (U.S.  EPA,  1987a).
Filtration  systems  without  disinfection,  but  with  proper  pretreatment  - (to
produce  water having turbidity  levels  <  0.5 NTU), can be  assumed to  provide
99  to 99.9 percent  inactivation of   Giardia  lamblia  and 90  to  99.9  percent
removal  of  viruses.    Disinfection  is needed to  supplement  filtration  so  that
the  overall treatment   achieves  greater  than   99.9  percent  removal  and/or
inactivation of  Giardia  cysts  and 99.99  percent  removal  and/or  inactivation of
viruses.

EPA   considers   conventional  treatment  with   disinfection  to   be  the  Best
Available  Technology  (BAT)   for  most source  waters  in the   United  States
because  of the multiple barriers of protection that it provides.

Direct   filtration  (which  includes   coagulation),   slow   sand  filtration,   and
diatomaceous earth  filtration, each with disinfection, have been demonstrated
to  achieve at  least  99.9 percent  removal and/or  inactivation  of Giardia  cysts
and 99.99  percent  removal  and/or   inactivation  of viruses  under  appropriate
design conditions (U.S.  EPA,  1987b).   The Surface Water Treatment Rule will
allow  their  use under   certain  source  water  quality  conditions  as determined
by  the State.

The same  comments about  the effectiveness of the  filtration step  to  remove
Giardia  cysts  and viruses,  arid  the   need for supplemental disinfection,  apply
equally  to  direct  filtration,   slow  sand  filtration,  and diatomaceous  earth
filtration.

-------
Filtration  technologies  other than  those  specified  above can be  used  if it is
first  demonstrated  through pilot  plant  challenge  studies,  using  Giardia cysts
and   viruses,   or   equivalent   indicators,   that   the   filtration  technology,   in
combination  with  disinfection,  can achieve  at  least 99.9  percent  and  99.99
percent   removal   and/or  inactivation   of   Giardia   cysts   and   viruses,
respectively.


        4.    Turbidity Limitations

For  systems  using  conventional  treatment or  direct  filtration,  the  proposed
Surface Water Treatment  Rule requires  that  filtered  water turbidity be  £  0.5
NTU in 95%  of the measurements taken  every  month.   Other  levels could be
specified    by   the   State   after    on-site    demonstration   that   effective
removal/inactivation of  Giardia  lamblia cysts,  or effective  removal of  Giardia
lamblia cyst-like  particles  can  be  obtained  at  other  filtered  water  turbidity
levels.     For   example,   the   State   could   allow  less    stringent   turbidity
performance  criteria  (up  to  1.0  NTU)  for  systems  demonstrating  effective
Giardia removal  by  pilot plant  studies.  However,  in such  cases,  the proposed
rule  will require  that the  maximum filtered water  turbidity level  be  <,  1  NTU
in 95% of  the  measurements  taken   each month,  and  at no  time  exceed 5
NTU.

All systems will  be  expected  to  optimize their  treatment so as to  achieve  the
lowest  turbidities  feasible   at  all  times  of  the year.   This would  promote
optimal removal  of Giardia cysts  and other  pathogens, and provide optimum
conditions for disinfection.

For  systems using slow  sand  filtration, the filtered water turbidity must be  <
1  NTU in 95%  of the  measurements taken each month,  and at  no time exceed
5  NTU.    However,  the  State could  allow a  turbidity value  greater  than 1
NTU in 95%  of  the measurements if the filter effluent at  the  plant  prior  to
disinfection  consistently  meets  the  long-term  coliform  MCL as  required  for
the disinfection system (see footnote b,  Table I).

For  systems  using diatomaceous   earth  filtration,  the  filtered  water  turbidity
must  be <  1  NTU in 95% of the measurements taken  each month, and at no
time exceed 5  NTU.

For   systems   using  other  filtration   technologies,  the  performance  criteria
would be  the same  as  for conventional  treatment and direct filtration.   The
State  could  allow  a filtered  water  turbidity  value  greater  than  0.5  NTU  in
95%  of the measurements  taken  (at no  time exceeding  5  NTU)  if the system
demonstrates effective performance at such  levels.


        5.    Minimum  Disinfection Requirements

              a.     For System,? Which  Do  Not Filter

At a  minimum,  a surface  water treatment system,  or a groundwater  treatment
system  "directly   influenced  by  surface  water"  (to be  determined   by  the

                                        10

-------
State),  and  which  does  not  filter  will  be  required  to  provide  disinfection
operational  conditions  which  inactivate   99.9%  inactivation   of Giardia  cysts
and  99.99%  inactivation   of  enteric  viruses.    Both  of  these  levels   of
disinfection  can  be  attained  by application  of chlorine, chlorine dioxide, and
ozone, according to the "CT values  listed  in Table  IIA.

  Table  IIA. CT VALUES FOR ACHIEVING 99.9% INACTIVATION OF
              Giardia lambliaa (U.S. EPA, Nov. 3, 1987)	
                                       Temperaturr
     Disinfectant     pH  0.5°C    5°C  10°C   15°C   20°C  25°C
Free Chlorine'3
(2mg/L)


6
7
8
9
171
261
377
521
122
186
269
371
91
140
201
279
61
93
134
186
46
70
101
139
30
47
67
93
   Ozone             6-9   4.5      3     2.5     2      1.5   1.0

   Chlorine Dioxide   6-9    81     54     40    27     21    14

   Chloraminesc      6-9  3,800  2,200  1,850  1,500  1,100  750
    (preformed)

   a     These  CT  values  for  free  chlorine,  chlorine  dioxide,   and
         ozone  simultaneously  will   guarantee  greater   than  99.99%
         inactivation of enteric viruses.

   b     CT   values  will  vary   depending  on  concentration  of   free
         chlorine.   Values  indicated  are  for  2.0  mg/L  of free chlorine.
         CT   values   for  different  free  chlorine  concentrations   are
         specified  in  tables  in  the   Guidance  Document (U.S.  EPA,
         1987d).

   c      To   obtain   99.99%   inactivation   of  enteric   viruses   with
         preformed chloramines  requires CT values >  5,000 at tempera-
         tures of 0.5, 5,  10, and 154C.


To avoid the  filtration  requirement,  EPA's proposed Surface  Water Treatment
Rule   will  require   that a   system   practice  disinfection and  have  backup
disinfection  capability provided  by backup  components,  including  an  auxiliary
power supply, with  automatic  startup  (and alarm), to  ensure that  continuous
disinfection is provided.

The  unfiltered  system  also  will  be required  to  demonstrate  by  monitoring
certain  disinfection   parameters  that it   is  achieving  disinfection  operational
conditions  which  inactivate  99.9  percent of  Giardia  cysts  and  99.99  percent
of  enteric viruses on  a daily  basis.    To  make  this  demonstration  without
resorting  to  isolating   and   counting cysts   and  viruses,  the  system  must
monitor  and  report  the disinfectants)  used, disinfectant residual  concentra-
tion^),  disinfection  contact  time(s), pH,  and  water temperature.    The system

                                       11

-------
must apply  these data  to  determine  if their "CP  value exceeds  that  required
by  the  Rule  for  the   specific  disinfectant  or  combination  of  disinfectants  to
determine if the minimum percent  inactivations required  are  being  achieved.

A  subset of these "CT" values  for  the various disinfectants  to  provide  99.9
percent  inactivation of  Giardia lamblia  cysts  is  shown  in Table  IIA.   These
values   are   based  on  laboratory  studies,  and  include  appropriate  safety
factors.    Since   Giardia  cysts  are  much  more  resistant  to free   chlorine,
ozone,  and  chlorine  dioxide than  are enteric viruses, it can  be  assumed  that
if  a system achieves  99.9 percent  inactivation  of Giardia  cysts  using  these
disinfectants, it  will  achieve much greater than 99.99  percent inactivation  of
enteric viruses simultaneously.

On   the  other   hand,   enteric  viruses   are  significantly  more  resistant   to
chloramine  residuals than  are Giardia  cysts.    Systems  using chloramination
for  primary disinfection,  therefore,  will  have  to  demonstrate that  they are
achieving  99.99  percent  of inactivation  of  enteric  viruses   as well  as  99.9
percent   inactivation of  Giardia  cysts.   On-site  pilot plant  challenge  studies,
using seeded organisms,  will be needed to  determine whether lower CT values
for  chloramines   achieve  the  required  levels  of  inactivation.    Such  studies
require  a high   level  of expertise  to carry  out,  and  specialized  independent
(commercial)  laboratories   or   university   research   personnel  to   make  such
determinations.

For  the  purpose  of   calculating  CT values,  disinfection  contact  time,   in
minutes   (for  chlorine,   chlorine   dioxide,   or  monochloramine)  is  the  time
required  for the  water to  move  between  the point  of  disinfection  application
and  the  first customer.   For  ozone,  because  of  its short  half-life in  water,
disinfection  contact  time  must  be defined  as the  time  the  water  is  exposed
to   a continuous  ozone  residual  concentration  during  the  water  treatment
process.

Residual  disinfectant  concentration  (in  mg/L)  for  chlorine,  chlorine  dioxide,
or   chloramine   is  the   concentration   of  the  disinfectant at  a  point  in  the
distribution  system  before  or  at  the  first  customer.     Contact  time   in
pipelines  must  be  calculated    based  on  "plug flow"  (i.e.,   all  water  moves
homogeneously  in  time  between  two  points)  by dividing the  internal  volume
of  the  pipeline  by the  peak  hourly  flow through that pipeline.   Contact time
within  mixing  basins  and  storage  reservoirs  must be   determined  by  tracer
studies or an equivalent demonstration.

For ozone,  the   residual of disinfectant concentration probably will  be  defined
as  that  ozone residual  measured  between  two specific points  in the treatment
process.    These  two  points  of ozone residual  measurement  will also  define
the contact  time  for purposes of calculation  of CT-values.

If  disinfectants   are   applied   at   more   than  one  point,  only  the   percent
inactivation  of  all disinfectant  sequences  prior to  the   first  customer  may  be
considered in  the determination  of total  percent inactivation.    In  making this
determination,  the  disinfectant  residual  of  each   disinfection  sequence  and
corresponding  contact  time must  be   measured  before  subsequent  disinfection
application point(s).

                                         12

-------
If  multiple   disinfectants   are   used  (e.g.,  primary  disinfection  with   ozone,
followed  by  later  secondary  disinfection  with  chloramines  or  chlorine),  the
percent  inactivation  achieved  by  each  of  the  disinfectants   is  additive  and
would apply in determining the overall disinfection performance provided.


              b.    For Systems  Which Filter

For systems  which filter,  and  which produce  a filtered water turbidity  <  0.5
NTU, it may be  assumed  that  99 to 99.9% removal of Giardia cysts and 90 to
99%   removal   of  viruses  is   attained   by   the   filtration   system.     EPA
recommends  that  systems  credit   filtration  systems  with  2-logs   of  Giardia
inactivation,   but  only  one-log  of  virus  inactivation.    Therefore,  disinfection
will be  required  to provide  an  additional  90%  inactivation (1-log  inactivation)
of  Giardia cysts and 99.9% inactivation  (3-logs inactivation)  of viruses.

Table  IIB  lists  CT  values  necessary   to   achieve  90  percent  inactivation  of
Giardia  lamblia cysts.   With  the  exception of  chloramines, where  higher  CT
values  than  those  indicated  in  Table  IIB  might  be needed,   these conditions
will achieve greater than 99.99%  inactivation of viruses simultaneously.

If  multiple disinfectants  are used,  the percent inactivation achieved by each
of  the  disinfectants  is additive  and  will apply   in  determining   the  overall
disinfection provided.    The clarification/filtration  processes used,   the  degree
of  fecal  contamination  in the  water  at  the  point(s)  of   primary   disinfection,
and  the  formation  of disinfection  by-products  (currently  trihalomethanes), will
influence the  type(s) of disinfectant used and the level of CT value(s).

Under   the   proposed  SWTR,   the  entire  treatment  train  must  achieve   a
minimum  overall removal  and/or inactivation of 99.9% for Giardia cysts  and
99.99% removal of  enteric viruses, respectively.

If  a   water  treatment  system   has  demonstrated  that filtration  is  achieving
greater  than   a  2-log  removal  of  Giardia  cysts,   the Primacy  Agency  may
permit  the disinfection requirement to  be  reduced  to  the  Ct  requirements  for
0.5  log  reduction  of  Giardia  cysts  --   but  not  for systems  using  direct
filtration!.    CT  values which   provide  0.5-log  reduction  of levels  of  Giardia
cysts   simultaneously   provide   greater   than   4-logs   reduction  in   levels   of
viruses,  but  only with  ozone,  chlorine  and  chlorine dioxide.   Table IIC  lists
CT values for attaining 0.5-log inactivation of Giardia cysts.

Based upon  raw water total coliform levels at  the point  of  disinfection, EPA
recommends  that  for  systems  which filter, sufficient disinfection  be  provided,
in  addition   to  the  2-logs  of  inactivation of   Giardia cysts   and  one-log  of
inactivation of viruses  credited  to the  filtration  system,   the following levels
of additional  disinfection be provided :
                                         13

-------
Table IIB.  CT VALUES FOR ACHIEVING 90%  INACTIVATION OF
           Giardia lambliaa  (U.S. EPA, 1987a,d)

Temoerature
Disinfectant
Free Chlorine0
(2 mg/L)
pH
6
7
8
9
0.5°C
60
90
130
170
5°C
40
60
90
120
10°C
30
40
60
90
15°C
20
30
50
60
20°C
15
23
34
46
25°C
10
15
22
31
Ozone
Chlorine Dioxide
Chloraminesc
(preformed)
6-9
6-9
6-9
1.5
27
1,270
1
17
730
0.8
13
620
0.7
9
500
0.5
7
366
0.3
5
260
a
      These  CT  values  for  free  chlorine,  chlorine  dioxide,  and
      ozone  simultaneously  will  guarantee  greater   than  99.99%
      inactivation of enteric viruses.

      CT   values  will  vary   depending  on  concentration  of  free
      chlorine.   Values indicated  are for 2.0 mg/L of free chlorine.
      CT   values  for  different   free  chlorine   concentrations   are
      specified  in tables  in  the  Guidance Document (U.S.  EPA,
      1987d).

      To   obtain  99.99%   inactivation  of  enteric   viruses   with
      preformed   chloramines  requires   CT  values   >   5,000   at
      temperatures of 0.5, 5, 10, and 15°C.
Raw Water Total
Coliforms *
(#/100 mL)
< 100
< 500
< 1,000
< 5,000
< 10,000
Giardia Cyst
Inactivation
(logs)
1.0
1.5
2.0
2.5
3.0
Enteric Virus
Inactivation
(logs)
3.0
3.5
4.0
4.5
5.0
               at the point of disinfection
                                     14

-------
  Table IIC.  CT VALUES FOR ACHIEVING 0.5-LOG  INACTIVATION
              OF Giardia lambliaa  (U.S. EPA, 1987d)

Temoerature
Disinfectant pH 0.5°C
Free Chlorineb
(2 mg/L)


Ozone
Chlorine Dioxide
Chloraminesc
(preformed)
6
7
8
9
6-9
6-9
6-9

28
43
63
87
0.8
13
690

5°C 10°C 15°C 20°C 25°C
20
31
45
62
0.5
9
363

15
23
34
46
0.4
7
337

10
15
22
31
0.4
4.5
250

8
12
17
23
0.3
3
181

5
8
11
15
0.2
2
130

  a      These  CT  values  for  free  chlorine,  chlorine   dioxide,  and
         ozone  simultaneously  will  guarantee  greater  than  99.99%
         inactivation of enteric viruses.

  b      CT  values  will  vary   depending  on  concentration  of  free
         chlorine.   Values indicated are for 2.0 mg/L of free chlorine.
         CT  values  for  different   free  chlorine   concentrations   are
         specified  in tables  in  the  Guidance Document (U.S.  EPA,
         1987e).

  c      To  obtain  99.99%  inactivation  of  enteric  viruses   with
         preformed chloramines requires  CT  values > 5,000 at tempera-
         tures of 0.5, 5, 10, and 15*C.
IV.     EPA DISINFECTION BY-PRODUCTS REGULATIONS

   A.   STATUTORY  REQUIREMENTS

Section 1412(b)(3)(A)  requires  the  EPA Administrator to  publish  Maximum
Contaminant  I^evel Goals  (MCLGs)  and promulgate  national primary drinking
water  regulations  for  each  contaminant  which,  in  the  opinion  of  the
Administrator,  may  have  any  adverse effect  on  the  health of  persons,  and
which  is  known  or  anticipated  to occur in  public water  systems.   Not  later
than  January   1,  1988,  and at  3-year  intervals  thereafter,  the  Administrator
must  publish a list  of contaminants  which  are known  or  anticipated to occur
in public  water systems  and which may require regulation under  this Act.  It
is   this   Section  under  which   disinfection/oxidation  by-products   will   be
identified  and   regulated  by  the  EPA, although  the  specific  wording   also
includes  synthetic organic  chemicals   (SOCs) which  are not produced  during
the water  disinfection process.
                                      15

-------
These regulations  will  be  subject to  the monitoring  requirements  of Section
1445(a)  of  the  Safe  Drinking  Water  Act  Amendments,  which  have  been
discussed earlier.
   B.    CANDIDATE DISINFECTANT/OXTOANT BY-PRODUCTS TO BE
         REGULATED

At present,  EPA  is regulating  only the four  trihalomethanes  as  DOBs.   The
total  concentration  of these  four  compounds currently  is regulated  at  100
jug/L.   Other  compounds will  be listed for regulation  by EPA as their health
effects become known.

In  early  1988,   the  Environmental  Protection  Agency  (U.S.  EPA,   1988a)
promulgated  the  First   Drinking  Water   Priority  List,   which   contains  53
substances or  classes  of  substances that are  candidates for regulation.   These
include  the  above disinfectants,  ammonia,  the  four  THMs  currently regulated,
ozone by-products, cryptosporidium.  and  many additional  halogenated  organic
materials,  some  of which  are  by-products  of  chlorine  disinfection.    EPA
currently is  conducting  research on  health  effects,  analytical chemistry,  and
treatability for  most of  these  compounds.   Within  three years  of publication
of  this  first  Drinking Water  Priority  List  (by  January  22,  1991),  EPA  is
required to promulgate regulations for 25 of the substances listed.

Ten  operating  water utilities  were   surveyed  recently  by  EPA  scientists
(Stevens  et  al.,  1987a,b,c).    Some  200  discrete DOBs  present  in  drinking
water  and   produced  by  chlorination  were  isolated,   and  many  have  been
identified.     Although many   of   these   DOBs  are   chlorinated,  many  are
oxygenated with no chlorine atoms in their  molecular  structures.

Table  IV lists  the  frequency  of  occurrence  in  the   10-utility  survey  of 22
chlorinated   by-products   of  greatest   current  interest to   EPA's  Office  of
Drinking Water  (Stevens  et  al., 1987a,b,c).    Other  disinfectants/oxidants  can
be  expected  to   produce corresponding  lists  of compounds  which may  be
proposed for  regulation   as  DOBs  in  the  future.   Many  of the  compounds
listed in Table IV have  been  proposed for possible  regulation (see Table  III).

Table   V  summarizes  the  currently   known   health  effects  associated  with
several   types  of  chlorination by-products   (Akin et  al., 1987).   Much  of the
information  learned   about  disinfection  by-products  in  drinking  water  has
been   inferred  from   studies  of chlorination   of naturally  occurring   organic
materials present in source waters.

Studies   on   the  toxicity  of the  readily identifiable  compounds are  continuing
with  the recognition  that  there is  no  reliable way  to determine  whether all
substances  found  to  possess   some  biological  activity are   being  identified.
Because  the  concentrations  of disinfection by-products in  drinking  water are
low,  it   may not  be  possible  to  detect any  biological activity directly.   Thus
it  is  not  possible  to determine   if  efficient  recovery of  biologically  active
substances  is  being   achieved  by  the  standard  or  advanced  isolation  and
identification methods that have been investigated for  this purpose.


                                        16

-------
  Table III.   DISINFECTANTS AND DISINFECTANT BYPRODUCTS
              LISTED  IN THE FIRST DWPL (U.S. EPA,  1988a)


  Disinfectants:

   Chlorine, Hypochlorite  Ion, Chlorine Dioxide, Chlorite Ion,
   Chlorate Ion, Chloramine, and Ammonia

  Four Trihalomethanes

   Chloroform,  Bromoform,  Bromodichloromethane, Dichlorobromo-
   methane,

  Haloacetonitriles:

   Bromochloroacetonitrile, Dichloroacetonitrile, Dibromoacetonitrile,
   Trichloroacetonitrile

  Halogenated Acids. Alcohols.  Aldehydes,  Ketones. and Other Nitriles

  Others:

   Chloropicrin  (trichloronitromethane), Cyanogen Chloride,  Ozone
   Byproducts
However, after chlorination,  mutagenic  activity  can be  determined  directly in
solutions  containing about  2 g/L of  humic acid.  Thus recovery of mutagenic
activity   during  subsequent  extraction   and   fractionation   can  be  followed
unambiguously.    From  these  types  of  studies,  it  has been  found  that  the
total  mutagenic  activity  of the  compounds  in  those  samples  that  could  be
identified by  GC/MS  and  for which  sufficient  quantities  were available  for
Ames   testing,  accounted   for  only  a   few  percent  of  the  total  activity
observed  in  the solution.   Obviously some biologically  active  substances were
produced that were not being extracted  and characterized.

Extensive  studies   of  the  by-products   of   ozonation,  chloramination,   and
treatment with  chlorine  dioxide  have  not yet  been  conducted  (Akin,  1987).
However, it  is well-known that the by-products  of  ozonation, for  example,
are more  highly  oxygenated and  polar in nature  than the  starting materials.
However, even  fewer of  these by-products of ozonation have  been identified
than  in  the  case  of  chlorination.     Currently,   studies  are  underway  to
determine  if  ozonized,  concentrated humic  acid  solutions  are mutagenic,  or
produce any  effects  in subchronic bioassays in rodents.   These same  solutions
will  have  to  be  tested   for  biological   activity  after  subsequent  treatments
with either chlorine,  monpchloramine,  or  chlorine  dioxide,  because in actual
practice,  ozonation   applied   as   the   primary  disinfectant   after  filtration
normally  is  followed  by  one  of these  secondary  disinfectants  to  provide  a
bacteriostatic residual for  the distribution system.
                                       17

-------
  Table IV.  OCCURRENCE OF CHLORINATED DISINFECTION BY-
             PRODUCTS AT 10 WATER  UTILITIES (Stevens et al,
             1987a,b,c)

  ~                          Number of       Range of Values      ~
   Compound                Locations             (ug/L)

   High Confidence

  Chloroform                     10 of 10           2.6 to 594
  Bromodichloromethane           10 of 10           2.6 to  77
  Chlorodibromomethane           10 of 10           0.1 to  31
  Bromoform                      6 of 10           0.1 to   2.7
  Dichloroacetonitrile              10 of 10           0.2 to   9.5
  Dibromoacetonitrile               3 of  7           0.4 to   1.2
  Bromochloroacetonitrile            7 of  7           0.2 to   4.0
  Chloropicrin                      8 of 10           0.2 to   5.6

   Low Confidence

  Chloroacetic Acid                 6 of 10                < 10
  Dichloroacetic Acid              10 of 10           < 10 to >  100
  Trichloroacetic Acid               6 of 10             10  to 100
  Trichloroacetaldehyde
   (as Chloral Hydrate)            10 of 10             10  to 100
  1,1,1-Trichloropropanone          10 of 10             10  to 100
  2-Chlorophenol                   0 of 10
  2,4-Dichlorophenol                0 of 10
  2,4,6-Trichlorophenol              0 of 10

   Qualitative Only

  1,1-Dichloropropanone             0 of  8
  l,l-Dichloro-2-butanone            0 of  8
  3,3-Dichloro-2-butanone            1 of  8
  l,l,l-Trichloro-2-butanone          0 of  8
  Cyanogen Chloride               1 of  7
  Dichloroacetaldehyde              0 of 10
   C.   ANTICIPATED TIMING OF REGULATIONS

Before DOBs can be proposed  for regulation, they must be  identified as being
present in drinking  water supplies  and  produced during  the  disinfection (or
oxidation) process,  isolated, and their health effects  determined.   At  present,
only the  four  THMs are being  regulated  as  disinfection  by-products.   The
current MCL for TTHMs is  100 /ng/L.   However, the National Academy  of
Sciences  (1987)  has  recommended that this level  be  lowered.   EPA regulators
concur in this  thinking, and  plan  to propose a new THM  standard when the
Groundwater  Treatment  Rule  is  proposed in late  1990  or early  1991  (U.S.
EPA, 1987a;  1987d).
                                      18

-------
  Table V.   SUMMARY OF HEALTH EFFECTS ASSOCIATED  WITH
             CHLORINATION BY-PRODUCTS (Akin et al, 1987)
   Chemical
    Class
Example
Toxicological
  Effects*
  Trihalomethanes
  Haloacetonitriles
  Haloacid
   Derivatives

  Chlorophenols
Chloroform
Dichlorobromomethane
Dibromochloromethane
Bromoform

Chloroacetonitrile
Dichloroacetonitrile
Trichloroacetonitrile
Bromochloroacetpnitrile
Dibromoacetonitrile

Dichloroacetic Acid
Trichloroacetic Acid

2-Chlorophenol
2,4-Dichlorophenol
2,4,6-Trichlorophenol
    C, H, RT
    H, RT
    H, RT
    H, RT

    G, D
    M, G, D
    G, D
    M, G, D
    G, D

    MD, C,  N, OL, A
    HPP

    F, TP
    F, TP
    C
Chlorinated
Ketones
Chlorinated
Furanones
Chlorinated
Aldehydes
1,1-Dichloropropanone
1 , 1 , 1-Trichloropropanone
1 , 1,3,3-Tetrachloropropanone
MX
2-Chloroacetaldehyde
M
M
M
M, Cl
G
        C = Carcinogenic;  H = Hepatotoxic; RT =  Renal Toxic;  G =
        Genotoxic;  D = Developmental;   M  =  Mutagenic;   MD =
        Metabolic  Disturbance;    N  =  Neurotoxic;    OL =  Ocular
        Lesions;   A = Aspermatogenesis;   HPP = Hepatic Peroxisome
        Proliferation;   F  = Fetotoxic;  TP = Tumor Promoter;   Cl =
        Clastogenic
From the  number  of  halogenated  compounds  listed  for  possible regulation in
the  First  Drinking Water Priority List  (Table III),  from  the  22  chlorinated
compounds  listed  in  Table  IV,  and from the  nearly  200  chlorination  by-
products  identified  in   the  10-utility  survey (Stevens  et al.,  1987a,b,c),  it is
apparent  that   most  of  EPA's  current   concerns  are  for  by-products  of
chlorination.
                                      19

-------
V.       BEST AVAILABLE TECHNOLOGIES AND TREATMENT TECHNIQUES

   A.    STATUTORY REQUIREMENTS

The SDWA Amendments require the  EPA Administrator to specify  MCLs  for
each  contaminant  to be regulated at a  level which  is as close to the  MCLG
as   is   feasible   with   the   use  of   best   available   technology,  treatment
techniques,  and  other  means which  the Administrator finds, after examination
for   efficacy  under    field   conditions,   are   available   (taking   cost   into
consideration)  [Section  1412(b)(4)].   The  Administrator  also  is  required  to  list
the  Best Available Technologies  (BAT) which are  capable  of meeting MCL
regulations.

Granular Activated Carbon  (GAC) technology  is defined by  the Safe Drinking
Water Act  Amendments of  1986 as  being feasible for the control of synthetic
organic  chemicals,  and  any  technology,  treatment technique, or  other   means
found  to  be  best available for  the   control   of synthetic  organic  chemicals
must  be at  least  as   effective  in   controlling  synthetic  organic  chemicals  as
GAC [Section  1412(b)(5)].

At  first  glance,  this last paragraph  may  not appear to be  applicable  to  this
discussion  of  disinfectant/oxidant  by-products.   However,  primary disinfection
usually  is  practiced  after all filtration  steps, including  after GAC  adsorption.
Many  of the organic  precursors  of  DOBs  will  be  adsorbed  by  the GAC; thus
when  disinfection  now  is  practiced,   lower   quantities  of  DOBs will   be
produced.

Conversely,  the  practice of preoxidatipn, using  ozone, KMnO4,  C1O2, or even
chlorine,  for  the  purposes  of oxidation of  color, iron  and manganese, , taste
and  odor  materials, microflocculation,  or  of synthetic  organic  chemicals,  will
provide  at  least  partial disinfection,  according  to  the  CT  values cited  in
Tables  HA and  HB,  and depending on the total  coliform levels  at  the point
of  preoxidation/predisinfection.     In  some  cases,  particularly  those  involving
ozone,  chlorine  dioxide, and chlorine  at  the  lower  pH  value,  it  is  possible
that the  primary  disinfection  requirement may  be  satisfied at this point in
the water treatment  process  (prior  to filtration).    This  is  especially  true in
the  case of ozone,  for which  the  CT  values  necessary  to  provide   99.9%
inactivation  of  Giardia  cysts and  greater  than  99.99%  enteric viruses  range
from  1  to 4.5 mg/L-min (see  Table IIA).

If the water treatment  process  is modified  to include  the  addition  of  strong
oxidants/disinfectants   ahead  of   filtration  (including   GAC   filtration/adsorp-
tion), the  chemical nature  of  many of  the  organic  contaminants  (VOCs  and
SOCs) may  be  altered,  thus promoting their biodegradation  in  the  filters  and
changing  their   GAC   adsorption  characteristics.    At  the  very  least,  even
partial oxidation  of  those  easily  oxidized  VOCs  and SOCs  will  decrease  the
adsorptive loading on the GAC filters advantageously.

The  Administrator also is .authorized [Section  1412(b)(7)(A)]  to promulgate a
national   primary  drinking--* water  regulation  that  requires   the use   of  a
treatment  technique in  lieu  of  establishing an   MCL,  if  the  Administrator
makes  a finding  that  it  is  not  economically or  technologically feasible  to

                                        20

-------
ascertain  the  level of the  contaminant.   In such case, the Administrator shall
identify  those  treatment techniques  which,  in  the  Administrator's  judgement,
would prevent known or  anticipated adverse effects on the health  of persons
to  the   extent  feasible.     Such   regulations  shall  specify  each  treatment
technique  known  to the Administrator which meets the requirements of this
paragraph,  but the  Administrator  may grant  a  variance  from any specified
treatment technique.


    B.   DEVELOPMENT    OF   CRITERIA    FOR    BEST   AVAILABLE
         TECHNOLOGIES

EPA's overall  approach to the  development of BAT criteria  currently  includes
the following five  major steps (U.S. EPA,  1987c):

1.  Identification  of  all  candidate  treatment  technologies, or  combinations of
    treatment technologies;

2.  Initial screening of technologies;

3.  Identification   of  available   treatment   technologies  (or  combinations  of
    treatment   technologies)   for   major   contaminant   categories    and
    subcategories;

4.  Evaluation   of  available   technologies   (or  combinations   of   treatment
    technologies) in each category;

5.  Determination   of   BAT   for   each   major   contaminant   category  or
    subcategory.


    C.   BAT FOR DISINFECTION

The   following   processes   for   removing  or   inactivating  microbiological
contaminants   already   have   been   identified  by   EPA  during   the   initial
screening of technologies:

    1.    Coagulation Sedimentation/Filtration

    2.    Direct Filtration

    3.    Pressure Diatomaceous Earth Filtration

    4.    Slow Sand Filtration
    5.    Cartridge Filtration

    6.    Disinfection/Oxidation

    7.    Ultraviolet Radiation
Evaluation of  these  technologies by  EPA is  proceeding  with  the objective of
developing BAT Criteria Documents  for each.   UV radiation has been  found
to  be ineffective  for  providing 99.9% inactivation  of Giardia lamblia  cysts
(Rice  &  Hoff,  1981).   Therefore  its  use  as a  primary disinfectant probably


                                       21

-------
will  be limited  to  groundwaters which  are not "directly influenced  by surface
waters".

Of the seven  technologies listed,  #1 (Conventional Treatment),  #2, and #6
can  be considered  as well-known and in  widespread practice.   The  other four
can  be considered  as "Emerging Technologies", even  though slow sand filters,
for example,  are  utilized  widely in  other parts  of the  world.   For  example,
the  United Kingdom  relies  extensively  on  this technology in  both  small and
large municipalities  (Rachwal,  et  al., 1987).


   D.   BAT  TECHNOLOGIES  FOR  DISINFECTION  BY-PRODUCTS

At the present time, a  BAT  document  exists only  for  THM removal  (U.S.
EPA,  1981).   Available or Potentially Available  Treatment  Methods described
in this 1981 document are  grouped into the following three categories:

1. Best  Generally  Available  Treatment Methods  for  Reducing  (concentrations
   of) TTHMs;

2. Additional  Treatment Methods for  Reducing (Concentrations of)  TTHMs;

3. Granular Activated Carbon (GAC) and Biological Activated Carbon (BAG).

Under  category 1   are  the  following  Best  Generally  Available  Treatment
Methods:

   a.   Use  of chloramines  as  an   alternate  or  supplemental  disinfectant or
        oxidant,

   b.   Use  of chlorine dioxide as  an  alternate or  supplemental  disinfectant
        or oxidant,

   c.   Improving existing clarification for THM precursor removal,

   d.   Moving  the  point of  chlorination  to reduce  the amount  of TTHM
        formation and,  where  necessary,  substituting  for the use  of chlorine
        as  a  preoxidant,  chloramines, chlorine  dioxide, hydrogen  peroxide, or
        potassium permanganate,

   e.   Use  of powdered activated  carbon for reduction of concentrations of
        THM  precursors  or  TTHMs seasonally  or  intermittently  at  dosages
        not to exceed 10 mg/L  on an annual average basis.

Under  Category  2  are  the  following  Additional   Treatment  Methods  for
Reducing  (concentrations of) Trihalomethanes:

   a.   Off-Line Water Storage,

   b.   Aeration (after production of  THMs),

   c.    Introduction of clarification where  not  currently practiced,

                                      22

-------
   d.    Alternate source of raw water,

   e.    Ozone as an alternate  or  supplemental disinfectant or oxidant.

At least items  a,  c,  and  d  are technologies  which  are  well-known.   Only
aeration and  the use of ozone can be  considered as  "Emerging Technologies ,
even  though   the  use  of  ozone  in Europe  has been  extensive,  particularly
since World War II.

Finally,  GAC  and  BAG  technologies, because  of their higher capital costs and
lack  of  widespread  adoption   in the  USA,  both  must   be  considered   as
"Emerging Technologies".


   E.    STRATEGIES FOR CONTROLLING DISINFECTION BY-PRODUCTS

Currently,   EPA  has   identified   three   distinct  strategies   for   controlling
disinfection by-products:

1. Application of alternative disinfectants,

2. Removal of precursors prior to  disinfection,

3. Removal of disinfection by-products once they are formed.

Initial   screening  of  technologies   has   identified   the   following   available
treatment  processes   or  unit  process   steps  for controlling  disinfection  by-
products, specifically chlorination by-products (THMs):

   1.    Oxidation

   2.    Coagulation,  Sedimentation, Filtration

   3.    Direct  Filtration

   4.    Lime Softening

   5.    GAC Adsorption (Precursor Removal)

   6.    PAC  Adsorption

   7.    GAC Adsorption (By-Product Removal)

   8.    Packed Column Aeration
   9.    Diffused Aeration

Items  #5,  #7,  and  #8  are   "Emerging  Technologies",  whereas  the  others
generally are well-known.

However, it  is logical to  expect  that   EPA may expand  this list  to  include
combinations    of  treatment   technologies   for   removing   synthetic   organic
contaminants,  including disinfectant/oxidant by-products,  such as:

   10.    Oxidation,    Coagulation,   Sedimentation,   Filtration   (Conventional
         Treatment with  Preoxidation)


                                        23

-------
   11.   Oxidation, Direct Filtration (Direct Filtration with  Preoxidation)

   12.   Oxidation, Slow Sand Filtration

   13.   Oxidation, GAC Adsorption

In  at least  half of  the  more than  2,000 water  treatment plants throughout
the  world using ozonation,  for  example  (including more than  some  35 of  the
40   ozone  plants  currently  operating  in  the   U.S.A.),   ozone   oxidation  is
practiced prior  to  conventional treatment  or  direct  filtration.    Preozonation
is practiced  early  in  the  treatment  process  for  microflocculation,  turbidity
removal,  taste   and   odor   control,  color  removal,  iron   and  manganese
oxidation,  oxidation  of THM  precursors,  etc.   In  this manner,  combinations
of  oxidation  followed  by  conventional  treatment  or  direct  filtration  reduces
the  concentrations  of  chlorine-demanding  organic  materials   at  the  terminal
chlorination (currently the primary disinfection) step.

Although these  process  combinations are  well-known throughout  Europe, they
must be considered  as "emerging" in the U.S. today.

In  process  combinations  #12  and  #13,  the  oxidation  step  is  designed   to
convert   normally    biorefractory   organic   materials    into   more   readily
biodegradable  materials.    During  the  slow sand  filtration  or  GAC  adsorption
steps which  follow  oxidation,  considerable biochemical  conversion  of dissolved
organic materials  to CC>2 and water occurs.

In subsequent studies, EPA  plans  to evaluate the sensitivity of process design
criteria as functions of treatment performance and  cost.


VL      DISCUSSION OF DISINFECTANTS AND OXIDANTS

   A.    BASIC ISSUES

The    current   interest  in  regulating   disinfectants   and   disinfectant/oxidant
by-products  has  been  instigated  by  the  recognition  that  when  chlorine  is
added   for   disinfection   purposes,  it  also   produces  halogenated   organic
materials.    Their  formation   is  a  function   of  a  number  of  factors,   in
particular  the  concentration  and  types  of organic  materials  present  in  the
water  when  chlorine  is  added,  and  the  chlorinating  conditions  (amount   of
chlorine  dosed,  pH,  temperature, reaction time,  etc).

Therefore,  the   salient  objective  of water treatment  now  is  to reduce  the
amounts  of  halogenated  materials  formed  anywhere during the   treatment
process.   In turn, this has prompted the three  strategies identified by EPA:

1. Remove the undesired compounds after  they have  been produced;

2. Change  disinfectants  tp  those  which  do  not   produce  the   undesired
   by-products;
                                        24

-------
3. Reduce   the  concentrations  of  organics  in  the  water  before  chlorine  is
   added.

Modifications  of  these  same  three   strategies  also  can  be   considered  as
applicable  to  coping  with  SOCs  and  their  disinfection/oxidation  by-products.

Of the  three  approaches,  the  first two can  be criticized as  disregarding  the
problem  and  treating  the  symptom ...  keep  the  water  treatment  process  the
same,  but  either  incorporate additional processing  to  remove  the  undesired
materials  produced  by  water  treatment,  or  switch  to  another  disinfecting
material which will not  produce the undesired materials.

Only the third approach recognizes and treats the  problem  itself,  that is,  the
concentration  of   organics  in   the   water,  and  how   to   remove   these
"precursors" of  undesired  by-products  or  lower  their  concentrations before
adding  chlorine.   If  one  assumes  that technological  logic  will win  the day
over  partial solutions which only treat the symptoms, it can be expected that
most water utilities  will opt to  apply  the third  strategy, provided  that   it  is
affordable.
         1.    First Approach

Once  the  halogenated organics  have  been  produced,  they  are  very difficult
and/or  costly  to  remove  from  water.   Oxidation  is  ineffective,  even  with
ozone,  for  removal of THMs.   Adsorption by  GAC  is  effective  only  for srjort
periods of  time,  and  GAC reactivation  costs  are  quite high.    Adsorption by
powdered  activated  carbon  (PAC)  is  costly  and  increases  the  quantities  of
sludge  for  disposal.   Air  stripping  of  volatile  organic chemicals  (VOCs)  is
efficient,  but GAC  may be  required  to adsorb the  compounds  (to prevent
transfer  of  the  pollutant  from  water  to  air),  and the  same  problems  of
reactivation  are  present.    Besides,  after  removal from the  water,  the  VOCs
now must be disposed of, and most of them can be expected to be subject  to
RCRA  compliance  under the  appropriate hazardous  waste  disposal  regulations.


        2.    Second Approach

Switching   to an   "alternative  disinfectant"  to  chlorine  (used  as the primary
and   secondary   disinfectant)   is   a   logical   approach,   provided    that  the
alternative disinfectant  will:

(a)     not produce halogenated  organics,

(b)     not produce other undesired disinfection by-products,

(c)     provide  guarantees  of  microbial  disinfection or  inactivation  at  least
        equal to  those provided by chlorine,  and

(d)     provide a stable residual  disinfectant in the  distribution system.
                                        25

-------
For   economic   reasons,   most   utilities   also  would   like  the   alternative
disinfectant to  be no more  expensive  than chlorine.   All  of this, particularly
the  "same  cost"  requirement, is  asking a  great deal  of any single  candidate
alternative  disinfectant.    Given  the fact  that  chlorine  has  been  applied  as
the  disinfectant of  choice throughout  the world since  the  early  1900s,  if  a
cheaper  and   better  disinfectant  were  available,  it  should  have  been  in
commercial use by this time.

Candidate  "alternative  disinfectants"   currently  being   considered  by   many
water  treatment  specialists are:   chlorine  dioxide,  monochloramine, ultraviolet
radiation, and ozone.  However,  both  ozone and UV  radiation can  be rejected
as  candidates  because  neither  provides a  stable residual level  for distribution
systems.    Miller  et  al.  (1978),  after  studying   European  and  Canadian
applications  of ozone for  disinfection,  concluded that ozone can be  used  as
the  secondary,  residual   disinfectant,   but  only  if  five conditions  are  met
simultaneously:

1.  Water temperatures must be cool (to slow biological regrowths),

2.  Water  must  be   free  of  iron  (to  eliminate  growth   of  iron-consuming
    bacteria),

3.  Water  must  be  free  of  ammonia  (to   eliminate   growth  of nitrobacter
    microorganisms),

4.  TOC  values must be  less  than 1  mg/L  (to minimize  its accumulation  as
    food for microorganisms),

5.  Residence  time  in the  distribution  system  must  be less  than  12 hours  (so
    that microbial regrowths will be  minimal).

Even   in  France,   where  post-filtration  ozonation   for primary  and   some
secondary  disinfection  has   been  practiced  since  the  early 1900s,  all  five
conditions were found by Miller et  al. (1978) to  coexist  only  rarely.

Chlorine  dioxide can be  used  as the  secondary  disinfectant  in  distribution
systems, and   in  the water  treatment plant for the primary disinfection step.
However,  it must be realized  that C1O2  is  reduced  when  it disinfects  and/or
oxidizes,  producing  some  level  of chlorite  ion,  C1O2",  from  which  chlorine
dioxide is generated.

Chlorine   dioxide  produces   hematological   effects   in  both    humans  and
laboratory animals  (NAS,  1987,  pp.  89-90).   The  mechanism of these  effects
is not  known;  however  it  is believed  to be  related to  the oxidant properties
of  chlorine  dioxide  and  its  aqueous  reaction  products,   chlorite  and  chlorate
ions.    In  addition,  thyroid  and  developmental  neurological  effects have been
observed  in  laboratory  animals.    These  thyroid  effects  are  thought  to   be
caused  by  its  oxidation  of  dietary iodide  ion  in  the gastrointestinal  tract.
The oxidized  iodine  then  binds  to either food or  tissue,  and  is  unavailable
for absorption.   The  mechanisms  of the  neurological  anomalies  are  unknown.
                                        26

-------
Because  of  this  behavior,  EPA  currently recommends  that  the total oxidant
levels  C1O2  usage (chlorine  dioxide  +  chlorite  ion  +  chlorate  ion)  should not
exceed 1  mg/L.   This  means  that  usage of chlorine  dioxide  as the  primary
disinfectant  must  be  limited  to  rather  clean  waters,  requiring   very  low
applied dosages, i.e., 1.2-1.4 mg/L.

In  Germany,  for  example, CIO?  is  employed  for  distribution  system  residual
at  dosage levels not  to  exceed t).3 mg/L (Miller et  al,  1978).   In Switzerland,
restrictions  on  the use  of CIO?  are  even  more rigid,  0.15  mg/L  being the
maximum allowable dosage (Schalekamp, 1986).

In  laboratory  studies, Werdehoff  and  Singer (1986)  have  shown  that the  1
mg/L  total  oxidants  from  chlorine dioxide will  not be  exceeded  if  the  dosage
of  CIO?  does not exceed  1.2  to  1.4  mg/L.   This level  has  been confirmed in
pilot  plant  and full-scale  plant studies at Evansville,  Indiana  by  Lykins  and
Griese (1986) (see Case  History of Evansville,  IN ~ Section VIII.C.1).

By  elimination,  this   appears  to  leave  monochloramine  as  the only  apparent
candidate  "ideal  alternative  disinfectant" to  chlorine.    On  the  other  hand,
monochloramine is a  very much  weaker  disinfectant than chlorine (see  Tables
IIA,  IIB,  IIC,  and  Section  VLB  on  Technical  Issues  -  Disinfection   and
Disinfectants).    Because of  the very  high  CT  values  of monochloramine for
inactivating  99.9%  of Giardia  lamblia  cysts and 99.99%  inactivation  of  enteric
viruses  (see  Table IIA),  it is impractical  to  consider  its  use  as  a  primary
disinfectant   for  surface  waters  or  groundwaters   "directly   influenced  by
surface water".   Therefore, monochloramine  should be considered only  as the
secondary disinfectant  for these types of systems  (U.S. EPA,  1987a,d).


         3.    Third Approach

Without  much  argument,  except   perhaps cost, the  concept  of reducing the
concentration(s)  of organics in water  before chlorine  is added will produce
fewer  halogenated  organic  DOBs.   This approach addresses the basic problem
of  removal  of  organics.    A  more  restrictive  subset  of  this  approach  is to
remove  organics  or  lower their  concentrations before  any  strong  oxidant is
added.    This   is  because  the  organic  oxidation  products  of  disinfectants/-
oxidants  other  than  chlorine  will  not  be well  identified  for  some  time,  since
more  sensitive  analytical procedures  and  health effects  data  continue  to  be
developed.

A  compromise  approach  is  to employ  non-chlorinous  oxidants  in   the  early
stages  of water  treatment  (e.g.,  ozone, potassium   permanganate,  hydrogen
peroxide,  before or  during rapid  mix,  and/or  prior to  filtration) to  assist in
the  removal  of organic materials by  partial  oxidation  of  the  organics,  more
efficient   flocculation   and  filtration.     In   such  cases,  the   non-chlorinous
oxidants   do   not  produce  halogenated  oxidation  products,  except  for  those
waters which may contain  substantial  amounts  of bromide  ion (which  can  be
oxidized  by  ozone ~  as well  as  chlorine  --  to  hypobromous acid,  which  then
may produce  brominated  organics, such  as   bromoform  —  see  Section  VI.C,
Technical Issues - Disinfection By-Products).


                                        27

-------
This  third  approach,  although  installed  in  many water treatment  plants  for
the   purpose  of  lowering  the   concentrations  of  disinfection/oxidation  by-
products,   also  is  applicable  to   the   simultaneous  removal  or  reduction  in
concentration of many SOCs and VOCs.


   B.  TECHNICAL ISSUES « DISINFECTION AND DISINFECTANTS

It is paramount that the  disinfectant,  applied  for the  purpose  of disinfection
(as  opposed to  being  applied  for  oxidation)  be  capable  of  providing  the
required  levels  of   microorganism  kills   or  inactivations.     In  addition   to
reducing  levels  of total  cohforms,  heterotrophic  plate  count  organisms, and
legionella  bacteria,   disinfection  currently  is  defined  by  EPA  (U.S.   EPA,
1987a)  to  mean a  99.9% reduction in Giardia lamblia cyst levels, and  a 99.99%
reduction in enteric virus concentrations.
         1.    For Systems Treating  Surface Waters

              a.   The "CT Value

Disinfection   should   be   recognized   as   a   [concentration  x   contact   time]
phenomenon  (C  x  T  product).    That  is,  the  weaker  the  disinfectant,  the
longer  it must  contact  the water  being disinfected.    Therefore,  one  of  the
consequences of changing disinfectants  will be  the appropriate  lengthening  or
shortening  of  the   contact times,   thus  changing  the  sizes  of  disinfectant
contact chambers.

In  EPA's proposed  Surface  Water  Treatment Rule (U.S.  EPA,  1987a),  several
tables of CT values  are  provided as guidance  for  water  treatment officials in
coping  with  the  microorganisms  to be  regulated  (see Table  I).   Table  IIA
compares the CT  values  for the four  major disinfectants over  the  pH  range
of  6 to  9  at  0.5°C  to  25°C with  respect to  their  abilities  to  achieve  99.9%
inactivation   of   Giardia  lamblia  cysts:    free  chlorine  (at  2  mg/L),  ozone,
chlorine dioxide, and preformed  chloramines.

Several important conclusions can  be drawn from Table IIA:

1.  Ozone is  by far  the  most  efficient  disinfectant  of the four  (CT  values  of
    4.5  at 0.5°C to   1  at  25°C).   This  means that  a 0.5  mg/L  residual  ozone
    concentration would  have   to  be   maintained   over  9   minutes  at   0.5°C
    ranging  down  to  2   minutes  at  25°C  in  order   to  guarantee   99.9%
    inactivation  of Giardia  lamblia cysts and  more than 99.99%  inactivation  of
    virus inactivation.

2.  Chlorine  dioxide  over  the pH range  of  6-9 is at  least  twice as effective
    as free chlorine  at pH 6 only.  At  pH 7, 8,  and 9, free chlorine becomes
    progressively less  effective than chlorine dioxide.

3.  A residual chlorine dioxide  level of  1  mg/L would  have to  be maintained
    for 81 minutes at  0.5°C ranging down to 14  minutes at 25°C in order to
    guarantee  99.9%  inactivation  of Giardia lamblia cysts.

                                        28

-------
4. A  residual  of  2  mg/L  of  free  chlorine  at  pH  7  would  have  to  be
   maintained  for  130 minutes at 0.5°C ranging down to  23.5 minutes at  25°C
   in order  to guarantee 99.9% inactivation of Giardia  lamblia  cysts.  If the
   water   being  disinfected   also  contains  high   trihalomethane   formation
   potentials,  using  chlorine   as   the   primary  disinfectant   under  these
   conditions  will   produce   high  THM  and  other  halogenated  organic   (see
   Table  IV) concentrations.

5. When   CT  values  are  attained  with  free  chlorine,  chlorine  dioxide,  and
   ozone   to  inactivate  99.9%  Giardia  lamblia  cysts,  simultaneous  attainment
   of 99.99% inactivation of enteric viruses  is assured.

6. Preformed  chloramines  are  by  far  the weakest  disinfectant  of  all.   To
   achieve 99.9% inactivation of Giardia lambiia  cysts with a 2  mg/L residual
   would   require contact times  of 1,650  minutes  at  0.5°C ranging  down to
   375 minutes at 25°C.

7. Much   longer  contact  times  are  required  with  preformed  chloramines to
   attain  99.99% inactivation  of  enteric viruses   (CT values  >  5,000 at  0.5-
   15°C)  than  to   attain  99.9% inactivation of  Giardia  lamblia  cysts.   This
   means  that  attainment   of  the  required  99.9%   inactivation  of  Giardia
   lamblia  cysts  with  preformed  monochloramines  does  not  assure 99.99%
   inactivation of enteric viruses.

This has  led EPA  to the recommendation that  surface  water  systems wishing
to use monochloramine  as  their primary disinfectant will  have to demonstrate
efficacy  by  conducting   pilot  scale  tests  with   challenge  organisms  (Giardia
cysts  and  viruses)  ~ a  costly  and time-consuming  process.    Thus,  from  a
practical  and cost  standpoint,  monochloramine  should not  be considered  as  a
primary disinfectant  (U.S. EPA, 1987a;d).

              b.    Suggested No Adverse Response Levels (SNARLs)

For  some substances not  regulated as  known  or  suspected  carcinogens  and
for which  there  are  adequate toxicity  data available  from prolonged  ingestion
studies  in  humans  or  animals,   the   Subcommittee  on  Disinfectants   and
Disinfectant  By-Products  of  the  Safe  Drinking  Water  Committee  of  the
National  Academy  of Sciences  has  calculated Suggested  No-Adverse-Response-
Levels (SNARLs).   In   a  recent publication  (National  Academy of  Sciences,
1987),  SNARLs  for  adults   and children  are  presented  for  chlorine  dioxide,
chlorite ion,  chlorate  ion,  and  monochloramine.    These specific  values  are
based  on  the  assumption that 20% of  the daily  intake of  these  substances  is
by ingestion  of drinking  water, and are summarized in Table VI.


Also  included  in   Table  VI  are   recommended  SNARLs  for   some of  the
halogenated  disinfection   by-products listed  by the EPA  in the  first  Drinking
Water  Priotiry  List for possible regulation (U.S. EPA 1988a).
                                       29

-------
  Table VI.   NAS RECOMMENDED SNARLs FOR  DISINFECTANTS
              AND  DISINFECTANT BY-PRODUCTS (NAS,  1987)
    Disinfectant/By-Product         for 70-kg adult   for 10-kg child

    Chlorine Dioxide                  0.21 mg/L        0.06 mg/L

    Chlorite &  Chlorate Ions          0.024 mg/L       0.007 mg/L

    Monochloramine                  0.581 mg/L        0.166 mg/L

    Dichloroacetonitrile                0.056 mg/L       	

    Dibromoacetonitrile                0.161 mg/L        0.046 mg/L

        The Subcommittee does not recommend  this SNARL because
        of concern  that DCAN may be carcinogenic.


              c.    Possible Regulatory Consequences

Values  for  SNARLs  recommended  in  Table  VI  are  quite   significant,  by
themselves, but  also when  considered  in  light  of  the  CT values  for  the
disinfectants   listed   in  Table  IIA.    Considered by  themselves,  they  are
indications of  possible MCLs (or  at  least  MCLGs)  which  EPA  may  propose
and promulgate  in  the  future  (i.e., when  the  Groundwater Treatment  Rule  is
proposed   in  1991).   If these SNARL  values do become  the  basis  for  the
corresponding MCLs,  it is  also probable  that  EPA will set the  MCLG and/or
MCL to  protect the  70-kg  adult, because exposure times  would  be calculated
over a human  lifetime.

In  this  event,   a  projected  MCL  of 0.21  mg/L for  chlorine  dioxide  may
eliminate   this  material  from consideration  as a  primary  disinfectant, except
in  very  clean waters, and  will  reduce  its  utility   as  a secondary  disinfectant
(except in  very clean waters).

At  the same  time,  a projected MCL of 0.581  mg/L  for monochloramine  will
virtually  eliminate  this  material  from consideration as a  primary  disinfectant
because  of the  excessive  contact  times   which would be required to  attain
99.9% and 99.99%  inactivations  of Giardia lamblia  cysts  and  enteric viruses,
respectively, at  this concentration.   For example,  at  0.5°C, a contact  time of
5,690 minutes  (94  hours  =  3.95  days)  would  be required  to  assure 99.9%
inactivation of  Giardia lamblia  cysts.   At  25°C, a  contact  time  of  1,293
minutes  (21.6  hours) would  be  required.   To  attain  99.99%  inactivation  of
enteric  viruses  at   0.5-15°C  would  require  a  contact time  with  monochlor-
amine of  > 8,620 minutes  (> 143 hours = > 2.4 days).

The hypothesized 0.581 mg/L  MCL for  monochloramine would not  eliminate
this  material  from  consideration   as a  secondary   disinfectant,  however, since
EPA  is   likely   to  require   only  a  "detectable  residual"  in  the  distribution
system when  the Surface  Water  Treatment Regulation is promulgated during
1988.    Under these  projected  regulatory restrictions,  the  concept of  a  single

                                       30

-------
"alternative  disinfectant"  disappears,  to  be  superseded  by  the  concept  of
"alternative  disinfecting  systems".      Such  alternative  disinfecting  systems
would involve  the use  of combinations of  a primary  disinfectant (e.g., ozone,
ultraviolet  radiation)  to  be  followed   by a  secondary  (residual)  disinfectant
(e.g., chlorine,  chlorine dioxide, or chloramine) for the majority of cases.

For systems whose  raw  waters  have very  low THM  formation potentials,  or
in  which  THMFP  can  be  lowered  appropriately  prior  to  disinfection,  then
chlorine  can  continue  to  be  both the  primary  and secondary  disinfectant.
However,   the   luxury   of   continued  application   of   chlorine    as   the
predisinfectant   and   as  the  primary  and  secondary   disinfectant   will  be
dictated  by  the  future  THM  regulations,  as well  as future  regulations for
other halogenated organic by-products of chlorination (see next  section).


                   1.    EPA'S   PERSPECTIVE  ON  RECOMMENDED   NAS
                         SNARLS

In  the  proposed  SWTR (U.S. EPA, 1987a), EPA  rejects  the NAS  assumption
that only  20% of the  daily exposure to  chlorine dioxide  and  monochloramine
comes   from  drinking   water.     EPA   expects   that   exposure   to   these
disinfectants  from   sources   other   than   drinking   water  will   be  minimal.
Therefore,  assuming  that  nearly  100%  of  exposure comes  from  drinking water,
the  relative  SNARLs  might  be estimated  as  1   mg/L  for  chlorine dioxide
(including  its  oxidation/reduction products,   chlorite and   chlorate  ions)  and
2.5  mg/L   for  chloramine.     This   is  the   current  recommendation  that
concentrations  of these  two  disinfectants do  not  exceed  1.0  and   2.5  mg/L,
respectively.  These  levels were proposed by  EPA earlier (U.S.  EPA,  1979).

It  is  also  possible  that  EPA  might  apply  a  different   uncertainty  factor
(1,000)  than used  by the  NAS  (100)  in  calculating SNARLs (actually MCLs)
for  chlorine  dioxide,  chlorite and  chlorate   ions,  and  monochloramine.   The
significance of this  possibility is  illustrated  in the  following  example  SNARL
calculation  for  monochloramine:

         8.3 mg/kg bw/day x 70 kg x 0.2  =   0.581 mg/L, or
         	     581 ug/L
                100 x 2 liters

where:

   8.3 mg/kg bw/day  =  dosage in mg/kg of body weight per day

   70 kg            =  average  adult weight

   0.2               =  20% of total daily intake

   100               =  uncertainty factor

     2 liters          =  daily adult water consumption
                                       31

-------
It  can  be  seen  that  changing the  estimate  of percent of  total intake from
0.2 to 1.0 multiplies the  calculated  SNARL by a factor  of  five.   On  the other
hand,  changing  the  uncertainty  factor  from  100  to  1,000,  decreases  the
calculated SNARL by an  order of magnitude (a factor of 10).

For  example,  using  an uncertainty factor  of 1,000,  the  NAS  SNARL  for C1O2
for adults  becomes  0.021.  When this figure  is multiplied by 5  (to correct  for
100%  daily exposure  rather  than 20% exposure),  the  SNARL  becomes 0.105
mg/L.    Similar  calculations  for  monochloramine   gives  a  revised  projected
SNARL of  0.29 mg/L.
To  put   these   potential   regulatory  limits  into  perspective,  the   three
possibilities  discussed   above  are  summarized  in  Table  VII  for  chlorine
dioxide, chlorite and chlorate  ions, and monochloramine.

From this table,  the range of highest and lowest potential  MCL levels can be
projected, Variations A and B.   In  the proposed SWTR, EPA has rejected  the
NAS-proposed  values  and is proposing  values  close to  the  higher  (Variation
A) levels.   However,  further health  effects studies  are  in  progress,  and these
issues  are  to  be  revisited by  EPA  in  about  a  year.   At  that  time, it  is
possible  that  the current recommended  maximum  levels  for chlorine  dioxide
and  monochloramine may be  lowered.


   TABLE  VII.    PROJECTED MCLs FOR  CHLORINE DIOXIDE AND
                   CHLORAMINE
     Disinfectant     Projected MCLs Based on Variations in Adult
                                     SNARLs
     or By-Product   June 87 NASa  Variation  Ab   Variation Bc
chlorine dioxide
chlorite/chlorate
ions
0.21 mg/L
0.024 mg/L
1.05 mg/L
0.12 mg/L
0.105 mg/L
0.012 mg/L
    monochloramine       0.581  mg/L    2.9 mg/L         0.29 mg/L
  a     20% daily exposure; uncertainty factor =  100.  The 20% daily
        exposure  level  number has been rejected  by  the EPA in favor
        of 100%  (U.S. EPA,  1987a).

  b     100%  daily exposure; uncertainty factor = 100

  c     100%  daily exposure; uncertainty factor = 1,000


Therefore,  the  following  possibilities  appear with  respect  to chlorine  dioxide
and monochloramine:       •••"••
                                       32

-------
 1.  If an  MCL for  chlorine dioxide is set at 0.10 mg/L (Variation  B),  chlorine
    dioxide  would  be  eliminated  as  a  primary and probably  as  a  secondary
    disinfectant, except  in very  clean waters.

 2.  An  MCL  for  chlorite/chlorate  ions  at either  the  highest  or lowest value
    given  may  eliminate  chlorine  dioxide  from  consideration  as  a  primary
    disinfectant.

 3.  For monochloramine,  an MCL of  2.9 mg/L (Variation  A) will  allow  this
    material  to be  applied  both  as a  primary and  secondary  disinfectant.   In
    fact, EPA's current proposal (U.S.  EPA,  1987a;d)  is to allow a maximum of
    2.5 mg/L.

    On  the  other hand,  an MCL of 0.29 mg/L (Variation B)  would eliminate
    monochloramine  from  consideration  as  a  primary  disinfectant,  but  would
    not eliminate monochloramine as a secondary disinfectant.

 Under  these  assumptions  of  total  human   exposure  from  drinking   water,
 chlorine   dioxide  currently  can   be   considered for  primary   and  secondary
 disinfection,  within  the  constraints   of  the  appropriate  "CT values",  and
 monochloramine can be  considered  as  a  secondary  disinfectant.    However,
 there is a  strong  probability  that  EPA will propose  MCLs  by  1991 (Ground
 Water  Treatment   Rule)  which  will  be  lower  than  currently  recommended
 levels,  and  which  will   reduce   the   practicality   of  both   disinfectants  as
 primary, and probably as secondary disinfectants as well.


        2.    For Systems Treating Groundwater

 As  pointed  out  earlier,  many   groundwater  systems  which  do  not  now
 disinfect  will be required  so to do, but the  Groundwater Treatment Rule  will
 not  be  proposed  by EPA until 1991.   Since cysts normally  are not present in
 groundwaters   (except  in  those  groundwaters  "directly  influenced  by   surface
 waters),  requirements  to  inactivate  these  types  of  microorganisms  are  not
 expected.     However   the  other  microorganisms   listed   in   Table  I  (total
 coliforms,    pathogenic   viruses,   heterotrophic   plate   counts,   and  legionella
 organisms can  be  expected to be  present, and therefore  to be regulated.  If
viruses  are  regulated,  the  same  99.99%   inactivation   requirement can  be
expected.


             a.    "CT" Values and SNARLs

It  can  be  anticipated that CT values  similar  to those  given  in  the  Surface
Water Treatment Rule will apply  to  the  various disinfectants,  as  well  as the
SNARLs  and/or the  resulting  MCLs/MCLGs.   Therefore, the  same reasoning
discussed  above  with  respect  to  chlorine  dioxide,   chlorite  and  chlorate ions,
and  monochloramine  should  apply  to  eliminate or  reduce  the  viability  of
these materials from consideration  as primary  and secondary disinfectants.
                                       33

-------
              b.   Possible Regulatory Consequences

This  reasoning   leads   to   the  same   conclusions   regarding   disinfection   of
groundwater  systems  as  for   surface  water  systems:    in  the  majority   of
systems,  the  application of  a  primary  disinfectant  (e.g.,  ozone,  UV radiation)
will   be  necessary,   followed   by   a  secondary  disinfectant   (e.g.,  chlorine,
chlorine  dioxide,  or  chloramine)  for  the  "detectable  residual"  to  be  carried
throughout  the distribution system.

As before,  the  ability  to  use  chlorine  as both the  primary and  secondary
disinfectant  will  rest  on  the  future  MCLs  for  THMs  and  for  other  by-
products of chlorination to be proposed  by EPA in 1990 or 1991.

Ground  water  systems  may  face  an  additional  complication,  however,  when
they  are   required   to  disinfect.      Those  ground  waters   which   contain
significant   levels  of  iron  and  manganese  will  produce  insoluble  iron  and
manganese  oxides  when treated with ozone or chlorine.   This  will necessitate
the  addition  of  filters  to  remove  oxidized insoluble  materials.    In  addition,
both  ozonation   and  chlorination  also  may  cause  flocculation  of  dissolved
organics, and precipitate turbidity-causing particulates.

Formation  of  these   types  of   insoluble  materials   will  require  secondary
disinfection after filtration.

One  major  advantage  of  UV  radiation  as  the  primary   disinfectant  for
groundwaters is that  at appropriate applied  energjy  levels,  bacterial disinfec-
tion  occurs  readily,  without extensive  oxidation.    Therefore,   for  groundwater
systems  with only traces of iron  and  manganese  in their raw waters,  use  of
UV  radiation  followed  by  secondary  chlorination  might  be  the  least-cost
alternative disinfecting system.


    C.   TECHNICAL ISSUES - DISINFECTION/OXIDATION BY-PRODUCTS

Although to  date, EPA is regulating only the four  THMs  as disinfection  by-
products, several  additional  halogenated organics  have  been  listed  in the  first
Drinking  Water  Priority List  (U.S.  EPA,  1988a)  for  possible regulation.   Also
included on  the  first  DWPL  are the  three  disinfectants  themselves (chlorine,
chlorine   dioxide,  chloramine),   chlorite   and   chlorate  ions,  ammonia,   and
"ozone   byproducts".     This  latter  expression  is   intended  as   a  catchall
category.    In the  event that  specific  byproducts of  ozonation  are identified
and  shown  to  be   detrimental  to  public  health, they will  be  proposed  for
regulation individually.

Akin  et  al.  (1987) describe the  current  EPA Health  Effects Research Program
dealing   with  compounds   which  have  been  identified  as   disinfection   by-
products  and  for which  health  effects  data   currently  are  being developed.
All of  those  under  current investigation  are halogenated compounds  which
are listed in Tables  IV  and V.

The  major  technological  issue  in  controlling  current DOBs is how  to  reduce
the   concentrations   of   these    halogenated   organic    materials   without

                                        34

-------
compromising   microbiological   safety.     Some   will   add,   "...and   without
producing  other  by-products  which  are  as  potentially  toxic  (or  of  public
health  concern)  as those  produced  from  chlorine".   Still others will  add,  "...
and at  no increase in costs .
         1.    Non-Halogenated DOBs

As  has  been  concluded  in  an  earlier  discussion,  only  chlorine  and  ozone
(possibly  UV radiation  for  groundwater)  appear  to  be  the  major primary
disinfectants  of  the future for  public  water  supplies.   Ozone  is  known  to
produce  oxygenated  materials   as   oxidation  products.    High  energy  UV
radiation   also  can produce  oxidation   products,  provided  that  appropriate
dissolved   oxygen  and  peroxide  levels  are  present.    What  types  of  non-
halogenated oxidation  products  are  produced  by ozone  and/or  UV radiation,
and what is  known of their  toxicities?   Are  similar types of non-halogenated
oxidized organic  materials produced from  the use  of  chlorine?


              a.     From Ozonation  (Rice & Gomez-Taylor,  1986)

Ozone   generally   is   not   reactive   with   saturated   aliphatic   hydrocarbon
compounds.     However,   other  types   of  aliphatic  compounds,   particularly
olefins,   are   readily   oxidized   to    unstable   ozonides,   peroxides,   and
hydroperoxide intermediates  to  form primarily aldehydes,  ketones,  and acids.
A  by-product of   nearly  all  (if not  all)  organic  oxidations  with  ozone,  is
hydrogen  peroxide.   In the  presence of  additional  ozone, however,  the  H2O2
is decomposed  into hydroxyl  free  radicals, which are  more powerful  oxidizing
agents  than   the   ozone   molecule   itself.    Hydroxyl  radicals  are  rapidly
destroyed  by  bicarbonate   and  carbonate  ions,   which  comprise  the  natural
alkalinity  of raw  waters.

Although  ozone  is capable  of oxidizing  many organic  compounds  completely
to CO2 and  water, this  conversion usually requires  large  doses  of ozone (> 3
moles   O3/mole  of  organic  compound)   and  long  reaction times  (sometimes
hours).    Under  drinking  water  treatment plant  ozonation  conditions  (1  to 5
mg/L   applied  ozone   dosage;  5   to  20   minutes  contact   time),   organic
compounds   usually  are   only   partially  oxidized.      The  oxidized   organic
materials  are more polar,  of  lower  molecular  weight, more biodegradable,  and
usually  more readily removed from solution by flocculation and filtration.

Oxalic  acid (HOOC-COOH) is  found commonly  as  a  "final" oxidation product
of  many  aliphatic, aromatic,  and  heterocyclic organic  materials,  because  of
its  very  slow  rate  of  oxidation with  ozone.   Acetic  acid is  another  "final"
oxidation  product   which   also  is  resistant  to  further  oxidation,   even  with
ozone, under  drinking water treatment plant conditions.

Formic  acid  and  formaldehyde  also  are  formed during  the  later stages  of
ozone  oxidation  of many  organic materials.   However,  these  two  by-products
are  readily  oxidized to CO2  and  water,  provided  sufficient  ozone  is  present
in solution.   Ozone oxidation  rates of intermediate  organic oxidation  products
usually  are slower than the oxidation rates of the  original compounds.

                                       35

-------
In  a few instances,  ozone  oxidation  products have been isolated  which  are
more  toxic  than  the original  compounds.    For  example,  ozonation of  the
pesticide   heptachlor  produces   heptachlor   epoxide,   almost   quantitatively.
Heptachlor   epoxide  is   quite  stable   to   continued  ozonation  (Hoffman  &
Eichelsdorfer, 1971).

Parathion  and  malathion produce  paraoxon  and  malaoxon,  respectively,  as
initial  oxidation products  upon  ozonation.   Both  of the oxons are at  least as
toxic  than   are  the  original  thion  pesticides.    However,  upon  continued
ozonation, the  intermediate  oxons  continue   to  oxidize,  producing  innocuous
final oxidation products (Laplanche et al., 1983).

Ozonation  of  phenol  can   produce   resorcinol,  which   is  a  known  THM
precursor.  This  may  explain the  observations made  by several  groups that in
following  ozonation  of some surface  waters   directly with chlormation, higher
levels  of  THMs  have been  produced  than by chlorination alone (Rice, 1980).
In   most   cases,   however,  ozonation  followed  immediately  by   chlorination
produces  lower  concentrations of THMs.

Infrequent  finding   of  higher  THM  levels   after  ozonation  and  immediate
chlorination  has led to moving  the  initial  ozonation  point(s)  to the rapid  mix
or  before,  and just sometimes just prior to filtration,  to  allow  flocculation
and even biodegradation   of  the  more  polar  initial  ozone  oxidation  products
during  filtration.     Chlorination   then   follows  at  a  later   treatment  stage.
Under these conditions, lower THM levels always are found.

The National Academy of Sciences  (1987)  report  (p.  67)  notes that oxidation
products  of  ozone  (as  well  as  non-chlorinated  organic oxidation  products
from chlorine  and  chlorine  dioxide)  are  similar  to  the  organic  compounds
formed by   natural  oxidation   processes.    In  other words,  a  surface  water
source  such  as   a  lake  will   be  experiencing  prebiological  and   chemical
oxidative  processes   for  months,  perhaps  longer,  by  processes  which  are
similar in their  chemistries to  oxidation  processes used in water treatment.

It is  concluded by  the National Academy  of  Sciences  (1987,  p.  195)  that little
is known  about  the types  of  by-products produced by  ozonation  of natural
organics,  and   that  well-conceived studies   need  to  be  conducted  which  will
focus on  the stable compounds expected from ozonation  reactions  with humic
materials.    Since  saturated  aldehydes  (non-toxic)  are  well-known  oxidation
products  of  organic materials,  it  is  suggested  (NAS,  1987)   that  particular
attention  should  be  paid to the  search  for  unsaturated  aldehydes  (some  of
which are known  toxicants) and  for hydroxyhydroperoxides.

However,   unsaturated aldehydes,  if  formed   during ozone  oxidation, should
oxidize rapidly at  the double  bond  with ozone,  thus destroying  them.

Finally,  the  National Academy of Sciences (1987, p. 196) states:

   "Notwithstanding  the   fact that these  studies  need  to  be  carried  out,
   drinking   water  suppliers  should   not  dismiss  the  possibility  of  using
   ozone   as   an   alternative   to  chlorine   and   chloramines  in  water
   treatment.   Ozone  is an  excellent  disinfectant  (although  it  must be

                                        36

-------
    used  in  combination  with  a   secondary  disinfectant   to  maintain  a
    residual in  the distribution system); ozone  is  an excellent  oxidant  for
    the  various  needs  of  water treatment;  it  does not form  chlorinated  by-
    products;  and  the  admittedly  inadequate studies now available  point to
    lower toxicities of  ozonated water than of chlorinated water."
              b.    From Chlorination

 From  the  aqueous  chlorination  of  various  humic  materials,  Stevens  et  al.
 (1985)   identified   98   specific   organic   compounds,   including   47  discrete
 compounds  containing   chlorine  and  51   discrete  compounds   that   did  not
 contain  chlorine.    Christman  et  al.  (1980)  showed  that   the  products  of
 chlorination   of  aquatic  humic  materials  generally   fall   into  three  broad
 structural   categories:     non-chlorinated   substituted   aromatics,   chlorinated
 straight chain acids, and  non-chlorinated straight  chain aliphatic  acids.

 Seeger  et al.  (1984)  showed that at low  chlorine  doses, which  are  typical of
 those  conditions currently found  in  drinking  water  treatment  plants  and their
 distribution  systems,  many  ring-chlorinated  aromatic   acids  are  obtained,  in
 addition  to  the  numerous  nonchlorinated  aromatic  and  aliphatic  compounds
 identified by earlier investigators.

 In  reviewing  the available  literature  on products  isolated  and  identified from
 the  treatment  of  humic  and  fulvic  acids  in  aqueous solution  with  ozone,
 chlorine,  chlorine dioxide,  and  potassium  permanganate, Rice  & Gomez-Taylor
 (1986)  concluded  that   the  non-chlorinated  oxidation  products  formed  by  all
 four oxidizing agents are similar.

 More   recently,  Stevens  et   al.  (1987a,b,c)   isolated  nearly  200  discrete
 compounds from  10 operating  water  utilities  practicing  chlorine  disinfection.
 Of those  compounds which were positively identified,  31  are non-chlorinated,
 and  are  similar   to  those  identified   in  other  studies   of  ozonation  of
 humic/fulvic acids.

 Therefore, consideration  of  ozonation  as a water  treatment  agent  should not
 be  postponed  simply  on the basis  of the  myriad  of  oxidation  products  which
 are formed.  Many  of the same compounds  are  formed  during  chlorination.


   D.    COMPARISON  OF  DISINFECTANTS - OXIDANTS

         1.    General Considerations

The  capability   of  one  substance   to   oxidize  another is  measured  by  its
Oxidation   Potential,   normally   expressed   in  volts  of   electrical    energy
(referenced to  the  hydrogen electrode).   The  oxidation  potential is a  measure
of the  relative  ease  by which  a substance is  able to  lose  electrons, thereby
being converted to  a higher state  of  oxidation.   If  the oxidation  potential of
substance  A  is  higher   than that of  substance B,  then substance B  can  be
oxidized    by   substance    A.       Oxidation    potentials   of   representative


                                        37

-------
oxidants/disinfectants commonly  used  or  encountered in  water  treatment are
listed in Table VIII.
Table VIII. OXIDATION POTENTIALS OF WATER TREAT]
OXIDANTSa
Species
hydroxyl free radical
* ozone
hydrogen peroxide
permanganate ion
* hypochlorous acid
* chlorine
* hypobromous acid
* bromine
hypoiodous acid
* chlorine dioxide
* iodine
oxygen
Oxidation
(OH)-
(03)
(H202)
(MnO4')
(HOC1)
(C12)
(HOBr)
(Br2)
(HOI)
(C102)(aq)
(12)
(02)
Potential. Volts
2.80
2.07
1.76
1.68
1.49
1.36
1.33
1.07
0.99
0.95
0.54
0.40
  a   Source:  Handbook of Chemistry  & Physics, CRC Press, Inc.

  *   excellent disinfecting agents
Although the relative  position of an oxidant in Table  VIII is  indicative  of its
ability  to  oxidize  other  materials,  its  oxidation  potential  does  not  indicate
how  fast  one  material  will  be  oxidized  by  another,  nor  how  far  toward
completion   the  oxidation   reaction  will   proceed.    One  cannot   tell   from
oxidation  potentials  alone  whether  a  specific   organic  compound  will  be
oxidized completely  (to  CO2  and water)  or  only  to  the   first  of  several
intermediate  stages.

One  significant  fact can  be learned  from  Table  VIII, however,  at  this  point.
As  has been  discussed  in  earlier sections of  this   report,   it  is  rare  that
organic compounds treated  with  an  oxidant  even  as powerful  as  ozone will be
converted  totally  to  CO2  and  water,  under  conditions  normally  encountered
in water treatment plants.   Therefore,  no  other  commonly employed and less
                                        38

-------
powerful   water  treatment   oxidant   (such   as  chlorine,   bromine,  chlorine
dioxide,  etc.),  all  of which  have  lower  oxidation  potentials than  ozone,  will
oxidize  an  organic  material  completely  to CO2 and water  if ozone will  not.

All  oxidants  weaker   than   ozone   will  be   less   effective  than  ozone  in
converting  organic  compounds  to carbon dioxide  and  water,  and thus  may
produce  higher  quantities  of partially oxidized organic  materials  under  water
treatment   plant  conditions.     It  is  important,  therefore,  for  the   water
treatment   professional  to   understand   the  chemistry  of  the  organic  com-
ponents   of the  particular   water  supply  when  considering  the  use   of  any
oxidant in the  processing.

When  oxidizing  agents  are  added  to  water  supplies  containing   specific
organic  compounds  as impurities,  it  is  not  sufficient  simply  to  follow  the
reduction  in  concentration  of that compound.  Other  parameters,  specifically
the  total  organic carbon  (TOC), provide  needed  information.   With  chlorine,
the  TOX  (total  organic halogen) also is  a necessary parameter.   In terms of
the  newer  halogenated compounds  to be  regulated  by EPA,  TOX Formation
Potential   (TOXFP)  will   become  important,   particularly  the  non-purgeable
TOXFP  (the NPTOXFP).   This is  a  measure of the  non-volatile TOXFP in the
water.

It  is  also  desirable  to   know   the   products of   oxidation  of  the   organic
impurities  which  are being  treated by  the oxidizing  agent.   When a  discrete
organic compound  is  oxidized,  it  may  be  totally destroyed by  oxidation,  but
without  a  decrease  in TOC content, In  this  instance,  the  concentrations  of
oxidation  products may be  as significant as the  concentration  of  the  original
impurity.

It is also important not  to  attempt  to  relate  the  disinfection  capability of a
specific  oxidant   to  its  oxidation  potential.    Ozone  is  the  most  powerful
oxidant listed  in  Table VIII, and  it  is also the  best  disinfectant  (i.e., its  CT
values  are  less  than  for  any  other  disinfectant  for  a  given  species  of
microorganism).   However  hydrogen  peroxide and  potassium  permanganate,
which  follow  ozone  in  Table VIII,  are known to be  poor  disinfectants,  while
iodine and  chlorine dioxide, which   have oxidation potentials  less than  half
that of ozone,  are very good  disinfectants.

Table  VIII  is  useful  to  the water  treatment  specialist  in  understanding  the
degree  of  chemical  transformation   which  can  be  expected  when   various
oxidizing agents are  used at various stages of the water  treatment  process.

Table  IX  lists  data for the  various  disinfectants,  comparing the  dosages  and
contact  times  required for  99% inactivation   of  Escherichia Coli.   Table X
lists  similar data  for  99% inactivation  of Poliovirus  Type I.


        2.    Chlorine

This  material  is  an excellent disinfectant,  an  excellent  chemical  oxidant,  but
unfortunately  also  is  an  excellent   chlorinating  agent.     Its  production  of
THMs  is   well-known,  but  it   also  produces  a  wide   variety   of additional

                                        39

-------
halogenated compounds  (TOX).   In addition,  chlorine is known  to  produce
significant  quantities  of  non-halogenated oxidation  products.   Some of these,
from  humic and fulvic acids  for  example,  are identical to those produced by
potassium  permanganate,  ozone,  and   chlorine  dioxide  (Christman,   et   al.,
1980).

  TABLE  IX. COMPARISON OF DISINFECTION PERFORMANCE TO £,
              COLI a
Disinfecting
Agent
HOC1
(OC1)-
NH2C1
ozone
ozone
C102
C102
iodine
iodine
H2O2
KMnO4
KMnO4
a From
Concen- Contact
tration Time,
mg/L min
0.1
1.0
1.0
0.04
0.0125
0.30
0.80
1.3
0.30
90
1
16
0.4
0.92
175.0
0.50
0.33
1.8
0.35
1
2
-360
45
25
CxTb
0.04
0.92
175.0
0.02
0.004
0.54
0.28
1.3
0.60
32,400
45
400
Drinkine Water and Health. Vol.
National Academy
Press, 1981
0), Chapter
Temperature
pH *C
6.0 5
10.0 5
9.0 5
7.2 1
7.0 12
7.0 5
7.0 25
6.5 2-5
9.1 20-25
6.5 ambient
5.9 0
9.2 20
2 (Washington, DC:
2.
  b     Concentration of disinfectant times contact time.
On  the  positive  side,  chlorine  provides  a  stable  residual  for  the  water
distribution  system,  provided  that  the  water  is free  of chlorine-demanding
ammonia and  organic materials.
        3.   Chlorine Dioxide

This  chemical   is  a  powerful  oxidant   and  disinfectant.     Because  of  its
instability,  it   must  be  generated  on-site,   which  can  be  considered   a
disadvantage.   In its  pure  state,  C1O2  does  not  produce trihalomethanes  in
the presence of organic materials which do produce THMs with chlorine.

Some  procedures  for  synthesizing  chlorine  dioxide  from sodium chlorite and
elemental   chlorine  involve  the  use  of  excess  chlorine.     Thus,   in  these

                                      40

-------
instances,  some   free  chlorine  is  present  in  the  water,  resulting  in  the
production of some THMs.   The  more that  C1O2  can be  synthesized without
the  need  for  excess  chlorine,  the lower  the concentrations  of THMs  which
will  be produced  by the excess  chlorine.   One procedure for generating C1O2
free of  excess chlorine  is by  addition  of  mineral  acid  to  solutions of sodium
chlorite.

TABLE  X.   COMPARISON OF DISINFECTION PERFORMANCE TO
              POLIOVIRUS TYPE I a
Disinfecting
Agent
HOC1
HOC1
(OC1)-
(OC1)-
NH2C1
NH2C1
ozone
ozone
C1O2
C1O2
iodine
iodine
Concen-
tration
rng/L
0.5
1.0
0.5
1.0
10
10
0.042
<0.3
0.3
0.8
1.25
20
Contact
Time,
min
2.1
2.1
21
3.5
90
32
10
0.13
16.6
1.5
39
1.5
CxTb
1.05
2.1
10.5
3.5
900
320
0.42
<0.04
5.0
1.2
49
30
Tempera-
pH ture, °C
6.0
6.0
10.0
10.0
9.0
9.0
7.0
7.0
7.0
7.0
6.0
7.0
5
5
5
15
15
25
25
5
5
25
25-27
25-27
a  From Drinking Water  and Health. Vol. 2 (Washington, DC:
   National Academy Press,  1980),  Chapter 2.

b  Concentration of disinfectant times contact time.
Although   chlorine  dioxide  does   not   produce  the  variety   of   chlorinated
organic materials  as  does chlorine,  nevertheless, some organic  compounds do
form  chlorinated oxidation products, although in much smaller  quantities  than
from  chlorine (Rice & Gomez-Taylor, 1986).  Mostly, however,  C1O2 oxidation
products of organic materials are non-chlorinated.

A major  disadvantage of  chlorine  dioxide  is  that as  it performs  its oxidation
or  disinfection  work,  about  half  of  it  reverts  back  to  the chlorite  ion.
Although  not all  of  the  lexicological parameters  of chlorite  ion  are  as yet
known,  it  does  produce  hematological  effects (Condie, 1986).   Consequently,
EPA  currently  recommends  that the total residual  concentration  of chlorine
dioxide,  chlorite  ion  and  chlorate  ion  not exceed  1  mg/L  in finished water
                                       41

-------
because   of  concerns   about   hematopoietic   and   possibly  thyroid  effects
(Cotruvo and Vogt, 1985).

On the  other hand, chlorine dioxide is  not known to  oxidize bromide ion  in
water  to  produce  hypobromous  acid (as does  chlorine  and ozone), which can
form bromine-containing organic materials.


         4.    Monochloramine

This material  is synthesized at  the  water  treatment  plant  by the  reaction  of
elemental  chlorine  with  ammonia  in  equimolar  quantities  by  one  of  three
procedures:

1. Ammonia can be added  to water  already containing chlorine,

2. Chlorine  can be added to water already  containing ammonia,

3. A  preformed  solution  of monochloramine  can  be  added to  the  process
   water.


Procedure  #1:

If chlorine  is first  added  to  the  water to be treated  to  attain  disinfection
over  a  specific  period  of  time, a  CT contribution  toward  disinfection  from
free  chlorine  will  be  attained.    However,   the  longer  the  free   chlorine
residual  is  sustained,  the  more halogenated  organic  materials  which  are >not
desired  will be  produced  during  the  disinfection  contact time.   Thus  there
will  not  be  major  benefit  to   the  use of  chloramine,  in that  instantaneous
halogenated  organic  levels will  be the  same.   However,  addition of  ammonia
will  insure  that THMs  will  not continue  to  be  generated  after  monochlor-
amine  has  been formed.   In this  manner, TTHM levels  can  be  held  to  their
initial  levels.
Procedure #2:

If ammonia  is  first  added  to  the  water, then  chlorine, provided  that very
good mixing  is  available, none of the  water  should be  exposed  to  significant
concentrations   of   free   chlorine.     This   will   lower  concentrations  of
halogenated   materials,  but   not   necessarily   guarantee  disinfection   since
monochloramine  now is the disinfectant.

In certain cases, however, this approach  may be even  less  effective.   When
chlorine  is  added  to  waters  which  contain  organic  nitrogen materials, such as
proteins   and  aminoacids,  the  chlorine  reacts  much  more   rapidly  with the
organo-nitrogen  compounds  to  form  organic N-chloramines   than  with free
ammonia to form  the  inorganic monochloramine  (Weil  &  Morris,  1949;  Morris,
1967).    In   turn,  the  organic  N-chloramines are   even weaker  disinfectants
than is monochloramine.
                                        42

-------
Even  worse,  standard  methods  of  analysis  for   monochloramine  based  on
iodometry  (i.e.,  DPD)  do  not  distinguish  between  inorganic  mpnpchloramine
and  organic  chloramines,  because  both oxidize iodide  ion to  iodine  in  the
determination of "combined residual" chlorine (NAS,  1987, pp. 62-63).

Thus,  a water  utility with  raw  water  containing  organic  nitrogenous materials
which  adds  ammonia,  then chlorine to produce "monochloramine" for primary
disinfection  may  be  seriously  overestimating  the  ability  of  its   treatment
system to provide the desired degree of disinfection.


Procedure  #3:

To be  assured  of  producing the  minimum  quantities  of  chlorinated organics,
the  chlorine  and   ammonia  can  be  mixed  in  organic-free  water,  and  the
monochloramine  solution  can  be  added  to  the  water  being  treated.   This
approach  "preforms"  the  chloramine.    However,  this   approach  produces
minimal  disinfection  and  also is subject to  the same competition with organic
nitrogen compounds as  discussed above.  The  organo-nitrogen compounds still
present  can  steal  chlorine from the inorganic  monochloramine,  producing  the
much  less  effective organic chloramines,  which are  not  distinguishable  from
inorganic chloramines by field iodometric analytical  procedures.


              a.    Chloramine Summation

Because chloramine  is  a  much weaker disinfectant than  is  free  chlorine,  or
any   of  the  other  disinfectants,   the  contact   time   required  to  assure
disinfection  of   any particular  organism  will  be  longer  than  with chlorine,
chlorine  dioxide,  or  with ozone.

In  some cases,  such as  with   patients who are  on  kidney dialysis  machines,
the  ingestion of chloramine-containing  water  can be fatal.   In  municipalities
using C1NH2  as the  terminal  disinfectant,  the utility  generally  advises  local
hospitals and health maintenance centers of  the  presence  of monochloramine
in their tap  waters, and advises  these institutions  to  employ  distilled  water
for their dialysis  patients.

As with chlorine dioxide, the  use  of monochloramine  as a  bacteriostat can
follow post-filtration disinfection  by means  of  ozone, for  example,  to provide
minimum production of halogenated organics.   Monochloramine is a very  poor
chemical oxidizing  agent.   However, it is known  to dissociate slowly in water
to  produce  small  quantities  of hypochlorous  acid,  which  in  turn  produce
traces of halogenated organic materials  (Rice & Gomez-Taylor, 1986).

In   assessing   all   currently    known  facts   about    monochloramine,    the
Subcommittee  on Disinfectants  and Disinfectant  By-Products  of the National
Academy of  Sciences'  Safe  Drinking   Water  Committee  has  stated recently
(NAS, 1987,  pp.  194-195):                                                      '

   "Because  it   is   a much  weaker disinfectant than  chlorine,  chloramine
   must  be  used  at   higher   concentrations  and  for   longer  periods   of

                                        43

-------
   contact  to achieve  sufficient  disinfection.   Even with  extended  contact
   time   and   higher   concentrations,   however,   chloramination  is   not
   recommended  as  a  primary  disinfectant,  especially  where  virus   or
   parasitic cyst contamination is  potentially present."

   "Preformed monochloramine is undesirable as a primary disinfectant."

   "The  use  of  marginal   chlorination  as   a  method   of  introducing
   chloramines  into a  water supply  system  is  specifically not  recommended
   because,   along  with   depletion   of  chlorine  to   produce  inorganic
   monochloramine, organic  chloramines that  have even lower  efficacy  as
   disinfectants   are   formed.      Organic  chloramines   have   also   been
   implicated  as  major  contributors  to the   mutagenicity  of  chlorinated
   drinking and natural waters."

   There  are   currently  no  suitable   methods  for  fully  quantifying   the
   organic    chloramine   fraction   in   the   presence   of    inorganic
   monochloramine.    Until   such  methods  are  developed,  utilities  that
   handle  water  supplies  containing  high   concentrations   of  organic
   nitrogen  run  the  risk  of overestimating  the  ability  of their  systems  to
   maintain  adequate disinfection."

   "When free  chlorine  is  used  as  the primary  disinfectant,  an amount
   should be used  that  is sufficient  to produce  a slight residual of  free
   chlorine  above  that  required  to  oxidize  nitrogen,  followed by  addition
   of ammonia to form monochloramine and limit THM formation.
         5.    Ozone

              a.    General Considerations

This  gas  also   must  be  generated  on  site,  at  the  water  treatment  plant,
because  it  is  too  unstable  to  be  stored  for  significant  periods  of  time  in
cylinders.   Additionally, it is only partially  soluble in water (about  13 times
the  solubility of  oxygen).   Therefore, one key  to  its successful  use is proper
gas/liquid contacting.    If  all of  the  ozone generated  is  not  solubilized  and
reacted  with  water  constituents,  the  excess  ozone  present  after  contacting
must be reused  or  destroyed,  in order  to prevent  unnecessary  exposure  of
plant operating personnel to this material.

Because  the  generation,  application,  and handling  of  ozone  is  so foreign  to
the  classical  water  treatment  processor  in the  USA,  and because  of  the
relatively high  capital  cost  of  ozone  generation  and  application  equipment,
the  acceptance of  ozone  in North  America has been  slow  to  develop, even
though  it  is  used  abundantly in  other  countries of the  world  (Miller  et  al.,
1978).

On  the  positive  side,  ozone is  the most  powerful disinfectant  and  oxidizing
material  which  is available  to  the water  processor.   It  kills  or  inactivates  all
organisms   tested  in   shorter   periods   of  time   than   does   any  other
disinfectant/oxidant available  (see Tables IIA, IIB, IIC, VIII - X).

                                        44

-------
Ozone   cannot   produce   any   halogenated   organic   materials,   other  than
indirectly,  e.g.,  by oxidation  of  bromide  ion  (if  that material  is  present  in
raw waters being treated) followed by bromination of organic  precursors.

Being a  powerful oxidizing  agent,  ozone  used in  very low  applied dosages  is
able  to   neutralize   charges  on   colloidal   particles,  thus  causing  them   to
precipitate.    In higher  applied  dosages,  ozone  is quick  to  oxidize   organic
materials,  but  usually  only  partially.    In  rare  cases,  organic  matter  can  be
converted  quickly  to  CC>2 and water by ozone.  However, under normal water
treatment  conditions  (i.e., dosages  of 1-5  mg/L  of ozone and  contact/reaction
times  of 5-15 minutes), most of the  organic materials contained in water are
only partially oxidized.


               b.   Disinfection With Ozone

For purposes of  disinfection, it  is necessary  to generate ozone, on-site,  and
apply  the gas (normally air  +  1-5%  ozone)  to  the  water to be  disinfected
(using  an   appropriate  gas/liquid  contacting  device)  for  a  length  of  time
appropriate   to   provide  the  CT  values  given  in  Table IIA  for  99.9%
inactivation  of  Giardia  lamblia  cysts.   In  practice,  dissolved  ozone levels  of
approximately  0.4  to   0.5  mg/L  are  attained   and  maintained  during  the
disinfection  process.     Ozone  contact  times   in  currently operating  water
treatment  plants  using  ozone  for  bacterial  disinfection   and viral  inactivation
are a minimum  of 10 minutes, and  range up  to 15-20 minutes.

EPA's proposed  Surface Water Treatment Rule  requires attainment  of 99.9%
inactivation  of Giardia  lamblia cysts.   From Table IIA,  the highest CT value
for ozone is  4.5  (for  water temperatures  of  0.5°C).   This  means  that  for  a
dissolved  ozone  concentration  of  0.5  mg/L, a  maximum  contact  time of  9
minutes  would  be required  at  0.5°C  to  attain  the  CT value  of  4.5.   This
length  of   contact  time   is  easily   within  the  design  capabilities   of  the
technology,  and would   not  produce  unique  consequential problems of design
and installation of ozonation systems.

As  will   be   discussed   in  the  following   subsections,  ozone  has   many
applications    in   drinking   water   treatment    which    utilize   its   oxidation
capability,  and  which  are  entirely  unrelated  to  its  disinfection  capabilities.
On  the   other  hand,  whenever  ozone  is  used  for  oxidative  purposes,  it  is
possible   to   attain  primary  disinfection   simultaneously,  for  most   oxidative
applications   of   ozone.    The  simplest  method  of  attaining   simultaneous
primary  disinfection  is  to extend  the  contact  time  appropriate  to  attain  the
necessary CT value at the  particular water temperature.

However,  knowledge  of the  dissolved  ozone  concentration during oxidation
also  is  critical  to  attainment  of  the  appropriate  CT  value.     In   certain
oxidative  applications,   specifically  ozone  oxidation  of  waters  containing  high
levels   of   iron  and   manganese  (groundwaters),   and   preozonation   for
microflocculation  and   turbidity  control  (surface  waters), measurement  of  a
dissolved ozone residual is  inappropriate.
                                        45

-------
In  the case  or iron  and  manganese,  the  dark,  precipitating  oxides produced
upon  ozonation will  interfere with  measurement  of dissolved ozone.    In  the
case  of  microflocculation  or turbidity  control,  the  objective  is  to  apply  a
very  low dosage of  ozone,  which means  that  a  measurable  concentration of
dissolved  ozone may  never  be  attained.   In  these  cases,  ozone  disinfection
would have to  be provided  as  a separate treatment  step,  normally practiced
just  before  or after filtration.

However,  in  all  other  oxidative  applications,  it  is  not  only  possible,  but
actually  feasible  to  design  the  ozone  dosage  and  contact  time  so  as  to
produce the  CT values  required  in Tables IIA,  IIB, and IIC.  In  taking credit
for  ozone  disinfection  prior  to  filtration,  attention  must be paid  to the total
coliform levels at  the  point  of primary  disinfection.  Successful attainment of
both  oxidation  and   primary  disinfection  before  filtration  means  that  only
secondary disinfection will be  required after filtration.


               c.    Microflocculation

Ozone  oxidation  of   humic/fulvic materials  can   proceed  to  a   variety   of
endpoints.    With   low  applied  ozone  dosages, oxidation  occurs primarily  on
the    pendant   groupings  without   cleaving   the   high    molecular   weight
humic/fulvic  polymer chains.

Partially  oxidized   organic  materials  contain  a   plurality  of  polar,   oxygen-
containing  groupings,  which  now   allow  ready   combination  with   cationic
flocculating  agents.    Consequently,  ozone can  be effective  in  aiding   in  the
removal of organics when used  in or before the rapid mix step, provided that
ozone  oxidation  is   followed   by  coagulation   then   conventional  or   direct
filtration.

On  the  other hand, when larger amounts of ozone are  applied to solutions of
humic/fulvic  acids,  not  only  does the oxidation  of pendant  groupings occur,
but  also  oxidative  scission  of the high molecular  weight  humic/fulvic  polymer
chains.  This  produces low molecular weight,  polar compounds,  which  are  not
as easily flocculated with cationic  flocculating agents.

Thus,  for  optimum   application  as   a  flocculation aid,  low  applied  ozone
dosages  should  be employed.   German   experience  in  the  use of  ozone  for
microflocculation  shows  that  the  ratio  of  ozone  applied  to  DOC (dissolved
organic  carbon) should  be in  the range  of 0.1  (mg/L  of applied  ozone  per
mg/L of DOC) (Sontheimer, 1985).


              d.   Promotion of Biodegradability

These same  partially  oxidized  organic  materials  also  are  more  biodegradable,
and  advantage is  taken  of  this  behavior  in  many European  water treatment
plants.   For example,  slow  sand filters  remove organics  from drinking  water
by  operating  in a biological mode.    Preozonation of  the  water  fed   to  the
slow  sand   filters   increases  the  biodegradability  of  organic  materials,   and


                                        46

-------
makes  the combined  oxidation/filtration  step  more effective  (Rachwal,  et al.,
1987).

In  a similar  manner,  if GAC  adsorption  is required  for  removal  of refractory
organic  materials,  and  if  some  or  all  of  the  organic  materials  can  be
partially  oxidized  by  ozonation,  then  preozonation  prior  to  GAC  filtration
can promote  biological  removal  of the  partially  oxidized  organic  materials in
the  GAC  medium.    The  adsorptive  capacity  of  such  Biological  Activated
Carbon  filters  thus  can  be  restricted to  the ozone-refractory  organics,  while
the   ozone-sensitive   organics   (partially  oxidized)   are   converted  to  carbon
dioxide and water  biologically.

The   consequences   of   not   providing   a  biological  treatment   step   after
ozonation  in   the  water   treatment  plant are   biological   regrowths  in  the
distribution  system,   unless   a  sufficient   level  of  secondary  disinfectant  is
present.   If  the  secondary disinfectant is chlorine,  then it is  important that
sufficient  removal  of  organics  has  occurred  before  the  chlorine  is  added.
Otherwise,  chlorinated  organics  still  can  be   produced,  sometimes  in  higher
quantities  than  without ozonation  (Rice, 1980).


              e.   By-Products of Ozonation

However,  the   partial   oxidation   of  organic  compounds   has  raised   some
concerns  as   to the  possible  toxicities of  the intermediate  oxidation products
formed  upon  ozonation.    In  the  majority  of cases,  these  oxidation products
are  oxygen-containing  derivatives  of  the  original  organic materials,  mostly
aldehydes, ketones, alcohols,  and carboxylic acids.   Ozone  is quick  to  rupture
many  unsaturated  linkages   in  organic  molecules,  producing  aldehydes  - and
ketones.

There  are, however,  a  few  organic compounds  from which ozone  has  been
shown  to  produce   toxic   oxidation  products.    For  example, the pesticide
heptachlor,  although   rapidly  oxidized  to  "destruction"  by  ozone,  produces
heptachlorepoxide  nearly  quantitatively.   This  example  illustrates  the  absolute
necessity  of  knowing  what  specific  compounds are present  in the  water  being
oxidized  and/or disinfected,  in  order  to  determine  appropriate  pre- and/or
post-treatment  procedures  to  cope  with these  undesired oxidation  products.


              f.    Catalytic Ozonation

For  many organic compounds  refractory  even  to  so strong  an  oxidizing  agent
as  ozone, the  simultaneous  application  of  ultraviolet  radiation   or hydrogen
peroxide  along  with  the  ozone   can  accelerate  otherwise  sluggish  reaction
rates significantly.    Acceleration  is  brought  about  by  catalytic  formation  of
hydroxyl free  radicals,  which  are  stronger  oxidizing  agents than  ozone   itself
(see Table VIII).
                                        47

-------
              g.    Ozone Oxidation of Bromide Ion

Ozone  does not  form  halogenated organic compounds,  unless  bromide  ion is
present in  the  raw water.   If  bromide  ion is  present,  hypobromous acid  will
be produced, which can form brominated organic materials:

              Br"  +   03   	>  (OBr)-

           (OBr)'  +   H+  	>  HOBr

If bromide ion  is present  in sufficient quantity to result  in  the formation  of
bromine-containing  organics  in  quantities  which  exceed  current  or  projected
MCLs,  either the  bromide  ion  should be  removed, or chlorine  dioxide  should
be  considered  as  the  primary  disinfectant,  followed  by  chloramination  for
secondary   disinfection,   or  an   alternate  source  of  raw  water  should  be
utilized.
              h.   Summation for Ozone

Because it  is  such a  powerful oxidizing  agent and disinfectant (the strongest
available  for  water  treatment)  and  does  not  form  halogenated  by-products
(except  when   bromide   ion  is  present),   ozone   is   the   most   versatile
oxidant/disinfectant for  water treatment.

As an oxidizing  agent,  ozone is  useful  added  in very  small quantities .for
turbidity  control  and  microflocculation.    In  higher  applied   dosages,  ozone
oxidizes   many  troublesome   inorganic  materials   (iron,  manganese,   sulfide,
nitrite,  arsenic),  and  destroys  or  alters  the  chemical  structures of many
organics responsible  for  tastes  and  odors,  colors,  THM  formation,  and other
precursors  of  halogenated  organics.   Such  alteration  of  chemical structures
can  make  the  oxidized  organic  materials  readily removable from  the.  treated
water  by  flocculation/filtration,  and/or  by  biological  means.    Many   of  the
compounds  listed as Synthetic Organic Chemicals  by  the  EPA in  the list  of
83  compounds to  be  regulated   by  1989  (U.S.  EPA,  1988a)  are at  least
partially oxidizable by ozone, although at varying reaction rates.

The  more  refractory,  halogenated, organics which may  be present  in the  raw
water,  which  may not be  oxidized with  ozone  at a  practical rate,  can  be
oxidized at  greatly  increased  rates  by combining  ozone  with UV  radiation  or
with  hydrogen  peroxide.

The  recent  National Academy  of Sciences  Report  (NAS,  1987,  pp. 195-196)
summarizes   the  latest  thinking   of the  Subcommittee  on  Disinfectants  and
Disinfectant  By-Products  of the  NAS  Safe  Drinking  Water  Committee with
respect to water treatment and ozonation in particular as follows:

   "When  possible, organic precursors  should be   removed prior  to  the
   disinfection  process.   Thi$  can  be  achieved  by  changing the  order of
   the  procedures  of  conventional   treatment.     A  better  approach,
   however,  is to  improve  specific  conventional  treatment  processes to
   remove  organic compounds   and  to  add  processes  such  as  carbon

                                        48

-------
   adsorption  and  preoxidation.    Initial  removal  of  organic  by-product
   precursors  precludes  the  need  for  reducing  (disinfectant) contact  tune,
   thus   improving   the   efficacy   of   the   disinfection    processes   and
   minimizing  formation  of organic chlorine  by-products."

   The  use of  alternative oxidants,  especially  ozone and  chlorine dioxide,
   will increase  in  the  United  States  in  the  coming decades.   Little  is
   known   about  the  types  of  by-products  produced  by  ozonation   of
   natural  organics.   Well-conceived studies need  to be   conducted  that
   will focus on  the stable  compounds  expected from  ozone reactions with
   humic  material.  ... Particular attention  should  be  given  to  the  search
   for unsaturated aldehydes and the  hydroxy-hydroperondes."

   "Following  these studies,  further  health  effects  studies   are  needed  to
   determine whether  ozone by-products are mutagenic or  carcinogenic  or
   produce  other  adverse  effects.  These  studies should  take  into account
   variations  that  are  likely  to  occur  when  the   oxidation   process  is
   carried out in  different  matrices (pH,  O^/TOC ratio,  alkalinity)/

   "Notwithstanding  the  fact  that  these studies need  to be  carried  out,
   drinking  water  suppliers  should  not  dismiss  the   possibility  of  using
   ozone   as   an   alternative   to   chlorine   and   chloramines   in   water
   treatment.   Ozone  is   an  excellent  (primary)  disinfectant  (although  it
   must be used  in combination with a secondary  disinfectant  to  maintain
   a  residual  in   the  distribution  system);    ozone  is  an  excellent  oxidant
   for  the various  needs  of  water   treatment;      it   does  not  form
   chlorinated  by-products;    and  the   admittedly  inadequate  studies  now
   available  point  to   lower   toxicities   of   ozonated   water   than   of
   chlorinated  water."
         6.    Ultraviolet Radiation for Groundwaters

The  effectiveness  of  ultraviolet  radiation  as  a  bactericide  and  virucide has
been  well  established  (U.S.  EPA,  1986), but   is  not  appropriate  for the
inactivation of  Giardia lamblia cysts (Rice  &  Hoff,  1981).    It  is  a physical
disinfecting agent  compared  to   the  other  disinfecting   agents   being  used,
whose actions  are chemical.   Radiation  at  a wavelength  of 254 nm  penetrates
the  microorganism  cell  wall and  is  absorbed  by  the  cellular  nucleic  acids.
This  can  prevent  replication  (reproduction)  and  cause  death  of  the  cell.
Since  UV radiation  is not  a chemical  agent, no potentially toxic  residuals are
produced.   Although  the structures  of  certain  chemical  compounds  may be
altered by  the  UV  radiation, the  energy  levels  used  for  disinfection are too
low for this possibility to be  a significant cause for concern.

UV  radiation at  a wavelength  of 254  nm  is  readily  available commercially  in
standard equipment  from a  number  of suppliers.   It  was  reported  (Angehrn,
1984)  that in 1984,  some  2,000 communities in Western Europe  now disinfect
their drinking  water supplies  .by  means of UV radiation.

Major advantages  of  UV disinfection  are  its simplicity, lack of impact on the
environment,  and  minimal   space   requirements.      There   is  a   negligible

                                        49

-------
likelihood  of producing  harmful  chemicals  in  the  water.    Required  contact
times  are  very  short, on  the  order  of seconds  rather  than  minutes.   The
equipment   is  simple  to  operate  and  maintain,  but  fouling  of  the  quartz
sleeves  or  Teflon tubes  (housing the  UV  bulbs)  must  be  dealt  with  on  a
regular basis.   Fouling  of these items  normally is controlled  by  mechanical,
sonic, or chemical cleaning.

On  the  other  hand,  waters  containing  high  suspended  solids  concentrations,
color,  turbidity,  and  soluble organic  matter  can  react  with  or absorb  the UV
radiation,  thus  reducing  the  disinfection  performance.    In addition,  if  the
amount  of radiation  received  by  the organism  is   not  a  lethal dose,  photo-
enzymatic  repair  can  occur, and the  organism can  become  viable  again.   The
phenomenon  is  termed   "photoreactivation", and  the  enzymatic  mechanism
generally  requires subsequent  exposure  to  light  at wavelengths  between 300
and  500   nm   (available  in   sunlight,   incandescent,  and  fluorescent   light
sources).

The major operating  costs  are  power consumption  and annual  replacement  of
the  UV  lamps.   Increased  popularity  and  lowered  costs have occurred due  to
improvements in  modern lamp  and  system designs,  increased competition, and
improved reliability and simplicity of operation.

The biocidal  properties  of UV radiation  make  it   a  candidate  for use  as  a
primary  disinfectant,  which must  be  followed   by  a  secondary  disinfectant
since UV  radiation provides no disinfecting  residual.  In addition,  at biocidal
wavelengths  (254  nm)  and energies,  UV  radiation  does  not  produce  much
oxidation   of  organic   materials.    However,  at  higher energy  intensities  and
low UV wavelengths  (i.e., 184.9 nm),  oxidation of organic  materials can  occur
readily to produce CC>2 and water.

The UV  light  spectrum  is  broad  enough  to  include  wavelengths  of  unequal
effects  upon  different  organisms.     Even  for  efficacious  wavelengths,   the
influence  of  the  UV radiation may  differ.   Also,  the devices  designed  to
discharge  UV  light  are  not all alike  in  the  radiant  energy   they  generate.
Moreover,  the  UV emanations  are  susceptible to absorption  by  the molecules
of  the  organismsuspending medium.    Their   germicidal  influence  may  be
reduced by the  time  they reach  the microbe.

That UV   radiation   is  of practical  utility  in   controlling  most  organisms  is
beyond  challenge.    The   devising   of  effective  UV  systems  requires   being
engineered, however.    More than  an ultraviolet  radiation  source  is  needed.
Finally,  where  the  organisms  are   killed  by the   UV  radiation,  subsequent
removal  of  the dead  microbes  may  be required,  as by  filtration, when  the
presence  of  particles is  undesired,  or  when  their  catabolic   products  may
manifest  total  organic carbon   (TOC) levels, or  pyrogenic  lipopolysaccharides.
                                        50

-------
VII.          DISCUSSION   OF   TREATMENT   TECHNOLOGIES    FOR
              DISINFECTION AND TO MINIMIZE PRODUCTION OF DOBs

    A,    THE PROBLEM

In  providing  disinfection  under  the  projected  EPA drinking water regulations,
water  treatment  systems  will be  faced  with  the challenge  of how to disinfect
surface and groundwaters to:

1.  assure 99.9% and  99.99%  inactivation of Gjardia lamblia cysts and  enteric
    viruses, respectively,

2.  assure control of other microorganisms,

3.  not   impart   toxicity    to   the    disinfected   water   (mutagenicity,
    carcinogenicity, etc.),

4.  minimize   formation  of  chlorination  and  other undesired  disinfection  by-
    products,  and

5.  do  all  this without exceeding the projected MCL levels  for  the  candidate
    disinfectants.
   B.    THE STRATEGIES

Before   primary  disinfection   is  practiced,  the   concentrations   of  organic
materials should be  as  low as  practicable, so  as  to minimize  the  production
of  oxidation  and/or  halogenation  by-products.    For  secondary  disinfection,
the turbidity  should be  < 0.5 NTU.

If  the  raw  groundwater  has   the   appropriate   levels  of  turbidity  and
concentrations   of   dissolved   organics,   pretreatment   can   be   eliminated.
Otherwise,  at least filtration must be  applied.

Concentrations of dissolved  organics can  be reduced by  applying conventional
treatments   of  flocculation,  perhaps  sedimentation,  then filtration.    Newer
techniques  of coping with  organics  concentrations  include  oxidation  (before
filtration, with ozone, potassium  permanganate)  and adsorption (powdered  or
granular activated  carbon).

An emerging oxidation technology is  the  use of ozone coupled simultaneously
with  UV  radiation   or  with  hydrogen  peroxide.    This  technique  generates
hydroxyl free radicals,  which are  more  powerful oxidizing agents  than ozone
alone.
   C   DISINFECTION  TECHNOLOGIES

According  to  the  reasoning  given  earlier  in  this  document,  the  use  of
chlorine for  pretreatment  and  primary disinfection  should be  minimized.   The
current  100  ^g/L MCL  for THMs  is expected  to  be lowered.   In addition,

                                      51

-------
regulation  of  some  of  the  halogenated by-products  of chlorination  listed in
Table  III may  place  even greater restrictions on the use of  chlorine for some
utilities.

Projected MCLs  for  chlorine  dioxide  and  its decomposition  products (chlorite
and   chlorate   ions)   effectively   eliminate   this   material   as   a   primary
disinfectant,  except  for  very  clean waters and/or  short  distribution  systems.

The  known  limited  capability  of  monochloramine  to  achieve  the  mandated
percent  inactivations   of  Giardia  lamblia  and  enteric  viruses  eliminate  this
material from the role of primary disinfectant.

Therefore,   for   surface   water  treatment,  only   two   primary   disinfectants
remain:   chlorine  and ozone.   For  groundwater  (unaffected by  surface  water),
the   same   two  primary   disinfectants   remain,  plus   UV  radiation.     For
secondary  (residual)  disinfection,  chlorine  remains  as  the  most  viable,  with
occasional  use  of chlorine  dioxide  and  monochloramine.    The conditions of
chloramination  should be  those  recommended  by  the  National  Academy of
Sciences (NAS,  1987):   chlorination to  the  nitrogen breakpoint,  followed  by
addition of  ammonia.

Each disinfection  approach  will  be discussed in this section.   Much  of  this
discussion is  adapted  from  an  earlier  EPA publication  (U.S.  EPA,  1983)  and
supplemented  by  a  more   recent  publication  (U.S. EPA,  1986)  which is  a
design  manual  for   disinfection  of  wastewaters  by means  of  chlorination,
ozonation  and  UV   radiation.    Nearly  all  of  the  factors  appropriate  to
disinfection   of  wastewaters  is  directly  applicable to drinking water  disinfec-
tion  as well.   Since  most of  the  utilities  which  will  be  affected  by  the
upcoming Surface Water  and  Groundwater Treatment  Rules serve  less . than
10,000  persons,  the   emphasis  of  this  discussion   will be   on  the  smaller
utilities.
         1.    Primary Disinfectants

              a.   Chlorine

Chlorine, symbolized  chemically  as Cl2,  is  the  disinfectant  most  commonly
used by U.S.  water utilities.  It  is available in  three forms:

         Form           Formula             Name

         gas               Cl2            chlorine gas

         solid            Ca(OCl)2        calcium hypochlorite

         aqueous          NaOCl          sodium hypochlorite
          solution

The  gaseous   form  is   used   most  frequently,  especially   by  larger   water
utilities, because it is  the lowest cost form of chlorine.


                                        52

-------
                   i.    Chemistry of Chlorination

In  the  gaseous form, chlorine will react with  water to form hydrochloric acid
and hypochlorous  acid:

    C\2     +    H2O   <-—>         HC1       +       HOC1

  chlorine    water                hydrochloric          hypochlorous
                                        acid                 acid

The  hypochlorous  acid  then  will  react with  the water  by  dissociation  to an
extent determined by the pH of the solution:

         HOC1  <--—>   (OC1)-     +    H+

      hypochlorous     hypochlorite     hydrogen
         acid              ion               ion

pH  is  a measure of the  concentration  of hydrogen  ion in  the water.    The
more hydrogen ion present,  the lower  is the  pH.   Conversely,  the lower  the
hydrogen ion  concentration, the higher will be the pH value.

At  neutral pH  (pH  =  7.0),  almost 80% of the chlorine  is  present  in  its most
effective  disinfecting  form,   hypochlorous   acid;  the  remainder  exists  in  the
less  effective  hypochlorite  ion  form.   As the  pH  increases,  however,  an
increasing  amount  of HOC1  will  react  with water  to  form  more  hypochlprite
ion.  At pH  8.0, for  example,  nearly  80% of  the chlorine present exists as
the  hypochlorite ion,  almost  a  complete reversal  of the  situation  which  exists
at pH 7.

From the  CT data given in  Tables  IIA, IIB, and  IIC for chlorine  inactivation
of  Giardia  lamblia  cysts,  it  can  be  seen  that the CT  values for a  2  mg/L
concentration  of  chlorine increase  rapidly  for each  temperature  as  the  pH
rises  above  7.0.   Consequently,  effective  pH  control  is  essential  in  order to
guarantee  the   amount  of  disinfection   designed   in   chlorination   systems.
Figure   1  shows   the   relationship   between  pH   and  the  concentration  of
hypochlorous  acid,  the  desired  chlorine  species  in aqueous  solution at 0°C
and 20°C.

When the chlorination  step  is conducted by adding either sodium  hypochlorite
or  calcium  hypochlorite,  the  chemical  reactions  which  occur  result  in  an
alkaline  (basic) product as  compared  to   the  acidic  product  obtained   when
using the gas:


         NaOCl    +    H2O  	>HOC1     +     NaOH

       sodium       water     hypochlorous   sodium
    hypochlorite                    acid         hydroxide
                                       53

-------
  Ca(OCl)2

  calcium
hypochlonte
  H20 -->   2HOC1

water     hypochlorous
              acid
                                                   Ca(OH)2

                                                  calcium
                                                  hydroxide
           IOC
                                                   10      11
Figure 1.      Distribution   of  hypochlorous   acid  and  hypochlonte  ions  in
              water at different pH values  and temperatures of 0°C and  20°C
              (AWWA, 1973).

The resulting  hydroxides  increase  the pH  values  of  the  aqueous solutions
Since  an  increase  in  pH  results  in  lower  concentrations   of  HOC1  and
therefore,   poorer   disinfection  (requiring  much  longer   contact   times  to

                                       54

-------
provide  the  appropriate CT value  to  guarantee 99.9%  inactivation  of Giardia
lamblia  cysts  and  99.99%  inactivation  of enteric viruses),  the ability to adjust
and   control  pH  becomes  critically  important  when  using  the  hypochlonte
forms  of chlorine.
                   ii.    Establishing A Chlorine Residual

Hypochlorous acid is one  of the most  powerful oxidizing agents  known.   This
means  that  it  will  react  with  many  substances  present  in the  water  being
disinfected,  in addition  to the  target organisms.   This fact is  clear from  the
known   production   of   trihalomethanes,   and   many  of  the   halogenated
compounds  listed  in  Tables III and IV.

In  order to  achieve a  concentration  of  chlorine  sufficient to  guarantee  the
required level of disinfection,  it is  necessary  to add  enough chlorine  to react
with  all of  the  reactive  substances  which are  likely  to be present,  and then
provide  the  excess  "residual"  chlorine.   These reactions consume  chlorine and
are   collectively   called   the   "chlorine  demand"  of  the  water.    Thus  the
chlorine demand   of a  water  must be   satisfied  before   a residual  of  free
chlorine can be provided and an adequate job of disinfection can be  attained.

The  concentration of chlorine determined  by  an  analytical  procedure  is called
the  "available  chlorine  residual",  and it means only  that   amount  of chlorine
originally added  which  remains  available   for   disinfection.    This residual  may
be   either   a  free   available  residual,  a  combined  available  residual,   or   a
combination  of the  two.   Free  available  chlorine  is  essentially  the sum  of
concentrations   of  hypochlorous  acid  and  hypochlorite   ions.     Combined
available  chlorine  is  the  sum  of  the  concentrations  of mono-  and.  di-
chloramines,   plus   nitrogen   trichloride   and   organic   nitrogen   chlorine-
containing   compounds   (see  later   discussion   of  Chloramines   —   Section
VILClb).

Intuitively,  one would  expect  that  each  mg/L  of chlorine added  to   water
would be  measurable  as hypochlorous acid or hypochlorite ion.    This  is  not
the  case, because chlorine  reacts with  many  substances present in the  water
in complex manners.  To understand some of these  complex reactions  better,
Figure  2 shows what is  called  a  "breakpoint  curve".   The  amount of chlorine
added is shown  on  the  horizontal scale  and  the  amount of available chlorine
determined by an  analytical procedure is shown on the vertical scale.

Assume  that  the  chlorine  is  added  slowly and that  the water  contains  small
amount  of   reduced  substances   such   as   sulfides,  ferrous   iron,   organic
materials,  organic  nitrogen  materials  (aminoacids  and proteins),  and  some
ammonia,  all  of  which   exert  a chlorine  demand.    The  initial  amount  of
chlorine  added  will  be  taken  up  by reactions with  the   reduced  substances,
and  the analysis  for free  available   chlorine [HOC1 +  (OC1)']  will show that
none is present.

After  the  chlorine   demands  of  the reduced  substances have   been  satisfied,
then  the HOC1 will  react  with  ammonia,  organic  nitrogen  materials, and some


                                        55

-------
of   the  organics  present   to   yield   chloramines  and  chlorinated   organic
compounds.
       -I  6
       V
       i  4
       in
       \tt
       K
       U
       i  2
       3
       z
       u
             ImrrndioU
            •-Dtmond  —
            H, S,Ft",ttc
                            Fret Chlorine
                             4        6        8        10
                                  CHLORINE  APPLIED, rog/L
                                    12
                    14
  Chlorine  and
   Ammonia
or similar compounds
Free  Rtsiduals
Figure 2.     Graphical  representation  of the  breakpoint chlorination  reaction
              (U.S. EPA,  1983,  p.  111-10).   The  straight  line at the  left shows
              that chlorine  residual is  proportional to  dosage  in  pure  water.
              When  impurities are present, they exert a chlorine  demand.


When all  of  the ammonia and  other chlorine-demanding organics  have  reacted
with  chlorine, the addition  of  more  chlorine  results in  the  hypochlorous  acid
oxidizing  some  of the  same materials  it just helped  to create.   The  strange
phenomenon  observed  is  that  the  addition  of  more  chlorine  results  in  a
decrease  in  the  amount  of  residual   (at  this   point   a  combined  residual)
indicated   by   the analytical  procedure.    When  this  oxidation   is  complete
(called  the breakpoint),  then the  addition  of  still  more  chlorine  results in an
increase   in   the   amount  of available  chlorine   measured.    Note  that the
breakpoint  must  be   surpassed  before  a   free   residual  of  chlorine  can
accumulate  and persist.

It  is  important  to be  aware  that  the  above illustration is  considerably  more
complex   than   described,  because  the  reactions  taking   place  are   time-
dependent.    For  this  reason, a breakpoint curve  is  difficult to  recreate and
predict.   Individual  tests must  be run, seasonally,  and the data plotted to
define the breakpoint for  each water.
                                        56

-------
                   iii.    Factors Affecting Disinfection Efficiency of Chlorine

As  indicated  by  the data  of Tables  IIA,  IIB,  and  IIC,  chlorine in  the free
state  [HOC1  +  (OC1)'] is quite an effective disinfectant.   Rapid inactivation
of most microorganisms can  be  obtained in a  matter  of minutes.   However,
effective disinfection  with chlorine requires careful attention to:

1. Concentration  of  free  available  chlorine  -  high  enough in  the  plant  so
   that  it  is  always   detectable  at  the  farthest  points in  the  distribution
   system  (time basis),

2. Maintaining the  pH  as  close  to  7.0  as  is  practical  or  consistent  with
   other  water quality  aspects, so  as to  maintain  as much  of  the chlorine
   residual in the HOC1 (hypochlorous acid) form,

3. Time  of contact long  enough  to  achieve  the desired  degree  of microbial
   inactivation  (i.e.,  to   attain  the   CT  value   commensurate  with   the
   concentration of chlorine at the appropriate temperature).

4. Mixing  --  baffle  the chlorine contactor well to  eliminate  the  possibility  of
   short-circuiting of flow or provide an external  mixing device.

Other  factors also  influence  the  chlorine  disinfection  process.    Temperature
has some  effect on  the ratio  of HOC1 to  hypochlorite  ion  (see Figure 1) and
has more  effect on  the disinfection  rate  (see  Tables IIA, IIB, IIC, and Tables
XI  through XVI  -  taken from  the  October 10,  1987 Guidance  Manual for the
Proposed Surface Water Treatment  Rule, U.S.  EPA,  1987d),  disinfection  being
faster  at  the  higher  temperatures.    However,  usually  there  is  no  means
available  to  the  operator  to control temperature.   Other tables  are  given  in
the Guidance Document  (U.S.  EPA, 1987d)  relating residual  chlorine  levels  to
pH,  concentration, and logs of Giardia inactivation.


                   iv.    Disinfection With Chlorine Gas

Chlorine  is  a toxic,  yellow-green  gas at ordinary  temperatures  and  pressures.
It is  supplied in  high  strength steel cylinders,  under  sufficient pressure  to
liquefy the  chlorine.    When  chlorine  is  required,  simply  opening  the  gas
valve  allows  rapid  vaporization of  the liquid.   As the  liquid  evaporates,  its
temperature  falls.   This  normally  results  in   a  slower  rate  of evaporation,
thus requiring manifolding  of containers or use of a vaporizer.

There  are  two basic  types  of gas  chlorinators:   (1}  pressure  operated,  direct
gas feed  and (2) vacuum  operated,  solution  feed.   The former  allows chlorine
gas,   under  pressure,  to  be  fed directly  into  the  water  to  be  disinfected.
Solution  feed  units  mix  the  gas  with  a  side stream  of  water  to  form  a
solution of hypochlorous  acid and  hypochlorite  ion,  which  then is mixed with
the main stream.
                                         57

-------
  TABLE XI. CT VALUES  FOR 99.9% INACTIVATION OF Giardia
              CYSTS  BY  FREE CHLORINE AT 0.5°C *
Free
Residual
(mg/L)
< 0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0

6.0
129
138
145
151
156
160
164
168
171
174
176
179
181
183

6.5
160
172
181
188
194
200
205
209
213
216
220
223
226
228

7.0
196
211
222
231
238
245
251
256
261
265
269
273
277
280
pH
7.5
238
255
268
279
288
296
303
310
315
321
326
330
335
339

8.0
284
305
321
333
344
354
362
370
377
383
389
395
400
405

8.5 9.0
335
360
379
394
407
418
428
437
445
453
460
466
472
478


392
421
443
461
476
489
501
511
521
530
538
546
553
559
         These CT values  achieve greater than a 99.99% inactivation
         of  enteric viruses.
                   v.   Disinfection With Sodium Hypochlorite Solution

Liquid  chlorinators  meter  a previously  prepared  hypochlorite solution  directly
into the  water to be disinfected.   If the water supply system  cannot afford
the  capital  costs  (see  later  sub-section) and  requirements associated  with
storing  and  handling  chlorine gas, solutions  of  sodium  hypochlorite  can be
purchased.   It must  be  remembered that sodium hypochlorite  solutions  are
more  costly   per  pound   of  available   chlorine  and  do  not  contain   the
concentrations  of  chlorine   available   in   cylinders  of  chlorine  gas.    Also,
hypochlorite  solutions  decompose  if   stored  for  prolonged periods.   Thus,
small systems  using sodium  hypochlorite  should  plan  to store no more than  a
one-month supply.

In  recent  years,  methods  for  on-site   electrolytic  generation  of  aqueous
solutions  of  hypochlorite  ion  have been  developed.    In  a  two-cell  unit,  a
brine  solution  (salt   in  water)   is  electrolyzed,  producing  a  solution  of
hypochlorous acid in  one cell  and a  solution  of caustic (sodium hydroxide) in
the other:
                                       58

-------
              Na+  +   Cl-   +    2H2O +  e'  —>

         sodium chloride           water     electron

              HOC1 +   NaOH+   H2

         hypochlorous    sodium    hydrogen
              acid       hydroxide

   TABLE XII.CT  VALUES FOR 99.9% INACTIVATION OF Giardia
              CYSTS BY FREE CHLORINE AT 5°C *
Free
Residual
(rng/L)
< 0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0
nH
6.0
92
98
104
108
111
114
117
119
122
124
126
127
129
131
6.5
114
123
129
134
139
142
146
149
152
154
157
159
161
163
7.0 7.5
140
150
158
165
170
175
179
183
186
189
192
195
197
200
* These CT values achieve greater
of enteric viruses.
169
182
191
199
206
211
216
221
225
229
232
235
239
242
than
8.0 8.5 9.0
202
217
229
238
245
252
258
264
269
273
277
281
285
288
a 99.99%
239
257
270
281
290
298
305
311
317
323
328
332
337
341
280
300
316
329
339
349
357
365
371
378
383
389'
394
399
inactivation
The  advantages of this  procedure  are that  purchasing  and storing  of gaseous
chlorine  and   hypochlorite  solutions  are  avoided.     The  primary  technical
disadvantages  are  the  generation of  hydrogen (which poses fire  and  explosion
hazards), and  the  need to  dispose  of  the  caustic  generated.     In  addition,
the cost per  pound,  on a  chlorine basis,  typically is  more  than  double for
on-site  electrolytic  generation  of hypochlorite  ($0.30  to  $0.35/lb)  versus  the
cost   of  gaseous   chlorine  ($0.08   to  $0.15/lb).      However,   site-specific
considerations   may  make  on-site   hypochlorite   generation  the  process  of
choice.
                                      59

-------
TABLE XIII.  CT VALUES FOR 99.9% INACTIVATION OF Giardia
              CYSTS BY FREE CHLORINE AT  10°C *
Free
Residual
(mg/L)
< 0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0
* These
enteric
oH
6.0
69
74
78
81
83
86
88
90
91
93
94
96
97
98
6.5
86
92
97
101
104
107
109
112
114
116
117
119
121
122
7.0 7
105
113
119
123
128
131
134
137
140
142
144
146
148
150
CT values achieve greater
viruses.
.5
127
136
144
149
154
158
162
166
169
172
174
177
179
181
than
8.0
152
163
171
178
184
189
194
198
201
205
208
211
214
216
a 99.99%
8.5
179
193
203
211
218
224
229
234
238
242
246
249
253
256
9.0
210
225
237
247
255
262
268
273
279
283
288
292
296
299
inactivation of
              vi.   Disinfection With Solid  Calcium Hypochlorite

Solid calcium hypochlorite  is  stable when properly packaged and  sealed.   A
water  supply  system  can  purchase   its  annual   requirements  in  a   single
procurement.   Simply mixing  the proper  amounts  of  solid and water  to  allow
metering without clogging of pumps or metering valves is all that is  required
for  use.   Normally,  an entire drum  of  calcium  hypo-chlorite  is made into
solution.     This  avoids  the   partial  use  of  a  container,  with  attendant
uncertainties  of  proper resealing and loss of strength.


                   vii.  Chlorination System  Design

Choice  of  the   form  of chlorination  system  to  be used, whether  gaseous
chlorine,   dry calcium hypochlorite,  sodium  hypochlorite  solution or on-site
generation, depends  upon a number of factors which include the following:

   o    availability of chlorine  source chemical,
   o    capital  cost of the facility,
   o    operation and maintenance  costs for  the equipment,

                                       60

-------
    o    chemical costs,
    o    location of the  facility,
    o    operator skills available,
    o    safety.

   TABLE XIVCT VALUES  FOR 99.9%  INACTIVATION OF Giardia
CYSTS BY
Tree
Residual
(mg/L)
<, 0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0
FREE CHLORINE
AT 15°C
*

DH
6.0 6.5
46
49
52
54
56
57
59
60
61
62
63
64
65
65
* These CT values
of enteric viruses.
57
61
65
67
69
71
73
74
76
77
78
79
80
81
7.0 7.5
70
75
79
82
85
87
89
91
93
95
96
97
99
100
achieve greater
85
91
96
100
103
106
108
110
112
114
116
118
119
121
than
8.0
101
109
114
119
123
126
129
132
134
137
139
141
142
144
a 99.99%
8.5
120
128
135
140
' 145
149
153
156
159
161
164
166
168
170
9.0
140
150
158
164
170
174
179
182
186
189
192
194
197
199 .
inactivation
Each of the  methods of  chlorination will  provide  the same disinfecting power
on  a pound  for pound basis of available chlorine when utilized at the same
pH.  However, each  of the systems must be approached  differently in terms
of basic design and safety.

Sufficient  chlorine  must  be  provided  to  satisfy the  chlorine  demand  of  the
water  at  the  point  of  chlorine   addition,  plus  an   additional  amount  to
maintain the  required  residual  after  a specified contact  time  (see CT values
in Tables XI to XVI).   The relative  dosages of  the various chemical  sources
of  hypochlorite  ion [(OC1)-] in solution  can be  determined;  these  frequently
will depend  upon the point of chlorine application in the  process.

The chlorine  demand  of raw water  usually is far  higher than  that of finished
water.    In  any  case,  a  minimum  contact time,  commensurate  with  the  CT
                                       61

-------
value,  the pH  of  the  water,  and  the temperature,  must be maintained  to
assure  the appropriate inactivation of Giardia lamblia and viruses.

  TABLE XV.CT  VALUES FOR 99.9% INACTIVATION OF Giardia
              CYSTS BY FREE CHLORINE  AT 20°C *
Free
Residual
(rng/L)
< 0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0

6.0
34
37
39
40
42
43
44
45
46
46
47
48
48
49

6.5
43
46
48
50
52
53
55
56
57
58
59
60
60
61

7.0
53
56
59
62
64
66
67
68
70
71
72
73
74
75
pH
7.5
64
68
72
75
77
79
81
83
84
86
87
88
90
91

8.0
76
82
86
89
92
95
97
99
101
102
104
106
107
108

8,5
90
96
101
105
109
112
114
117
119
121
123
125
126
128

9.0
105
113
119
123
127
131
134
137
139
142
144
146
148
150
        These CT values achieve greater than a 99.99%  inactivation
        of enteric viruses.
Chlorination With Gaseous  Chlorine:

Chlorine  is  supplied  in  high strength  steel  cylinders  with minimum chlorine
capacities of  100 and 150  pounds, up  to one  ton,  and  in  tank cars  under
sufficient  pressure  to liquefy  the  chlorine.   Major  manufacturers of gaseous
chlorine  are listed  in Table XVII.   However,  the  quantity consumed by  small
water  systems  normally would be  purchased  from  local  suppliers  which  are
listed  in   the   local  telephone   directory  yellow  pages  under   "Chemical
Suppliers" or "Swimming Pool Suppliers".

Direct feed chlorinators add  gas  under pressure directly into  the water to be
disinfected.   This type of unit normally is  used only when electrical power is
unavailable  or  insufficient  water  pressure  differential  to  operate  a solution
feed  unit.   This  is  a site-specific  application  which  will  not  be discussed
further.
                                       62

-------
   TABLE XVICT VALUES  FOR 99.9%  INACTIVATION OF Giardia
              CYSTS BY FREE CHLORINE AT 25°C *
Free
nH
Residua]
(mg/L) 6.0 6.5 7.0 7.5
< 0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0
*
23
25
26
27
28
29
29
30
30
31
31
32
32
33
These CT values
of enteric viruses.
29
31
32
34
35
36
36
37
38
39
39
40
40
41
achieve
35
38
40
41
43
44
45
46
47
47
48
49
49
50
greater
42
46
48
50
51
53
54
55
56
57
58
59
60
60
than
8.0
51
54
57
59
61
63
65
66
67
68
69
70
71
72
a 99.99%
8.5 9.0
60
64
68
70
73
75
76
78
79
81
82
83
84
85
70
75
79
82
85
87
89
91
93
94
96
97
99
100
inactivation
Solution feed units mix  chlorine gas with  a side  stream of water to form  a
hypochlorous  acid  solution,  which  then  is  injected  into  the  main stream.
Solution  feed  chlorinators   operate  on  a  vacuum  controlled   basis,  auto-
matically  shutting  off if the  side  stream flow  is  interrupted.   This  type  of
unit,  shown  in  Figure  3,  is preferable for safety reasons over  direct  feed
units.

The basic solution  feed gas chlorinator includes the  following components:

   o    gas shut-off valve to  interrupt gas flow;
   o    vacuum regulator, or
   o    gas flow indicator;
   o    adjustable  gas flow controller;
   o    check valve;
   o    venturi  type gas injector.

The  market  for  supply  of gas  chlorinators  is quite competitive,  as  illustrated
by Table  XVIII,  which  provides a  partial listing  of suppliers  of these types
of units.

                                       63

-------
  TABLE XVII.
MANUFACTURERS OR PACKAGERS OF GASEOUS
CHLORINE
                       Name

  Ashland Chemical Company
    Petrochemical Division

  Dow Chemical U.S.A.
  FMC Corp.,
   Industrial Chemical Group

  Georgia-Pacific
  Kaiser Aluminum & Chemical Corp.
  Industrial Chemical Division

  Kaiser Chemicals
  Cleveland, OH  44124

  Kuehne Chemical Co., Inc.
  Kearny,

  Occidental Chemical Corp.
   Industrial  &  Specialty  Chemicals
  Olin Corporation
  Pennwalt Corporation
   Inorganic Chemicals Division

  Stauffer Chemical Co.

  Vulcan Materials Co.,
   Chemicals Division
                                  Address

                    P.O.  Box 2219,  Columbus, OH
                         43216

                    2020  Dow Center, Midland, MI
                    48640

                    2000  Market Street, Philadel-
                    phia,  PA 19103

                    P.O.  Box 105605, Atlanta, GA
                    30348

                    300  Lakeside Drive, Oakland,
                    CA  94643

                    30100   Chagrin   Boulevard,
                    86   Hackensack   Avenue,
                    NJ  07032

                    360 Rainbow Blvd. South, P.O.
                    Box  728,  Niagara  Falls,  NY
                    14302

                    120   Long   Ridge   Road,
                    Stamford, CT   06904

                    3 Parkway, Philadelphia, PA
                    19102

                    Westport, CT   06881

                    P.O. Box 7689,  Birmingham, AL
                    35253
Well  established  standards  for  design  of gas  chlorination  systems  exist in
standard  waterworks  industry  literature  (Great  Lakes  -  Upper  Mississippi
River Board  of State  Sanitary Engineers,  1985;   Am. Water Works Assoc.,
1985).    The  following   points  are  provided  to   highlight  possible  design
questions:
                                     64

-------
       O'
       c
       •-I
       ft
                                                    TO VENT
O\
s.°
       Q.
     i—i
     a
     OOP
c
3.
i
^-»
§
      i
       o>
                            LEAD
                            GASKET
                 CHLORINE
                 CYLINDER
                 VALVE
              YOKE
              CLAMP
                                                         RATE
                                                         VALVE
                                   VACUUM SEAL
                                   "0" RING
                                                             OUTLET CONNECTION
                                                      ' VENT VALVE
                                                                RATE INDICATOR
                                                            REGULATING
                                                            DIAPHRAGM
                                                            ASSEMBLY
                                                                           VACUUM LINE
                                                                            EJECTOR AND
                                                                            CHECK VALVE
                                                                            ASSEMBLY

                                                                       WATER SUPPLY  —
                                                                                                               CHLORINE
                                                                                                               SOLUTION
                                                     CHLORINE CYLINDER
       a
      8

-------
   TABLE  XVIII.  SOME GAS CHLORINATOR MANUFACTURERS
         Name                          Address

  Capital Controls  Co., Inc.              Box  211, Colmar,  PA   18915

  Chlorinators  Inc.                       733 NE Dixie Highway, Jensen
                                        Beach, FL  33457

  Fischer & Porter Co.                  County Line Rd.,  Warminster,
                                        PA  18974

  Hydro Instruments  Inc.                492 Richlandtown  Pike,  Box
                                        615,  Quakertown,  PA   18951

  Modern Process  Systems Inc.          14834 Highland  Road, Baton
                                        Rouge, LA   70810

  Wallace & Tiernan,                   25 Main Street,  Belleville, NJ
   Division  of Pennwalt Corp.           07109
Chlorination With  Sodium Hvpochlorite

Sodium  hypochlorite,  usually   supplied   in  concentrations   of  5  and   15%
available chlorine,  is  available  commercially only in solution.   In  this form,  it
is  easier  to handle  than gaseous chlorine or calcium  hypochlorite.   However,
sodium  hypochlorite  solutions  will  lose   their   disinfecting   (oxidizing)  power
during storage,  and should be  stored  in  a cool,  dark,  dry area.   The material
is  supplied  in  glass  or  plastic  bottles,  carboys,  or  lined drums ranging in  size
from  0.5  to 55 gal.   Bulk shipment by tank truck also is  a common form of
transport.   No  more than  a  one-month supply  of  the  chemical  should  be
purchased, to  prevent loss of available chlorine.


Chlorination With  Calcium Hvpochlorite

This  material is supplied as a  white  solid which is  quite corrosive,  and gives
off  a  strong   chlonnous  odor.    It  contains   approximately  65%  available
chlorine, is readily soluble  in  water,  and is  available in granular, powdered,
or  tablet form.   It  is provided in  2,  5,  8,  and 35  Ib cans and  in 100, 300,
and 800 Ib  drums.   The  containers generally are  resealable.

Calcium  hypochlorite  is  hygroscopic   (readily   absorbs  moisture),  and   reacts
slowly with atmospheric  moisture to  form  chlorine  gas.   Therefore, shipping
containers   must  be  emptied   completely  or carefully resealed.    It is  not
feasible to handle this material in bulk  handling systems.

The contents  of  a calcium  hypochlorite   container  are emptied into  a  mixing
tank  where  it  is  readily and  completely dissolved  in water.    The  resulting
corrosive   solution  is   stored   in   and   fed  from  a  stock  solution   vessel
constructed  of  corrosion-resistant  materials  such  as  plastic,  ceramic,  glass,

                                       66

-------
or  rubber-lined steel.  Dosage of the  solution at 1% or 2%  available chlorine
content is  by a diaphragm type, chemical feed/metering pump.


                   viii.        Chlorination Systems Costs (U.S. EPA, 1983)

In  the following discussion, cost data are  presented  for  the  three types  of
chlorination  systems  discussed above,  and  as  related  to  their use at  small
water  treatment systems,  sized to  treat  water volumes  up to  at  least  1  mgd.
The  larger  systems  will  have  to  change  chlorine  cylinders  more  frequently
than  will  the  smaller systems.   For example,  a 1  mgd water  treatment  plant
using  an  average  chlorine  dosage  of  5  mg/L will   use   nearly  42  Ibs  of
chlorine per  day.   Thus,  a  150-lb  cylinder  of chlorine  will  last between  three
and four days at this size  plant.

Operating  and  maintenance  costs  presented   do not   include  chemical  costs.
These will vary depending  upon the volume  of water   treated and  the  dosage
required.

Cost  data were  obtained  primarily  from  two  sources  (Hansen  et  al.,  1979;
U.S.  EPA,   1983),  and  updated   by  calling  various  vendors in  mid-1987.
However,  a   more   recent  publication   by   Gumerman   et  al.  (1986)  contains
additional  cost  data developed  specifically  for  small  water  systems.    Cost
curve figures  from this later publication are included, as appropriate.


Solution-Feed Chlorination With Gaseous Chlorine

                              Equipment Costs

Table  XIX shows a detailed cost  breakdown  obtained  during  May  1980  from
three vendors of chlorination equipment.   Data are presented in terms  of a
basic gas  chlorination  system, as  well as  costs for five increasingly complex
systems.   The  basic system  includes equipment  to  handle  two 150-lb  chlorine
cylinders,   two      cylinder-mounted   chlorine   gas   regulators,   automatic
changeover valve,  and chlorine gas  flow and rate  valve ejector  (with  system
backup).    Alternate  #1   adds  two   scales,  a  gas   mask,  and   a   diffuser
corporation cock (to allow  connection  under  water  line pressure).    Alternate
#2 adds  a  flow-pacing chlorine  addition system.   Alternate  #3  adds  a  flow
meter.  Alternate  #4 adds  a booster pump and piping. Alternate  #5  adds a
chlorine leak  detector.

The  cost  comparisons  in   Table   XIX  present a basic-to-most-sophisticated
comparison  between the   various   system   configurations   in   which   gaseous
chlorination  systems  can  be  purchased.   Costs are comprised of  equipment,
installation,  safety   enclosure,  contractor's  overhead  and   profit,   plus   10%
engineering fee for the basic system  estimates.

The basic  (lowest  cost) gaseous  chlorination  system  costs about $9,350;   with
all  options  added,   the  most  sophisticated  gaseous chlorination  system  costs
$16,050, in May  1980 dollars.


                                        67

-------
TABLE XIX.
CAPITAL COSTS FOR GAS CHLORINATION (1980
Dollars*)
EQUIPMENT COSTS FOR A SYSTEM OF 100 Ibs/day (2 kg/h) OR
LESS
Basic System**
Alternate #1 - add scales, mask
diffuser, corporation cock
Alternate #2 - add flow pacing -
existing signal
Alternate #3 - add flow meter
& signal, 8 in. or less
Alternate #4 - add booster
pump & piping
Alternate #5 - add Cl2 gas
detector
INSTALLATION
SAFETY ENCLOSURE
Average $1.873
High 2,300
Low 1,320
Average
Average
Average
Average
Average
Average
High
Low
Average
High
Low
770
1,694
2,068
792
1,382
1167
1,500
1,000
3.500
6,000
2,000
CONTRACTOR'S OVERHEAD AND PROFIT (20%)  1,869
ENGINEERING FEES (10%)
TOTAL CAPITAL COST
               Basic System
               Most Sophisticated
               (with Alternate  #5)
*     May 1980 quotes (three vendors)
                                   934
                            $9,343
                           $16,049
**
      Basic  system  includes  two  150-lb  chlorine  cylinders,   two
      cylinder-mounted   regulators,   automatic   changeover   valve,
      chlorine  gas  flow  rate  valve,  and    ejector  (system  with
      backup)
                                 68

-------
                            Construction Costs

Figure 4  shows construction  cost  curves for gas feed chlorination  systems  up
to a maximum  chlorine feed rate of 80 Ibs/day (Gumerman et al.,  1986).


                     Operation and Maintenance Costs

Hansen et al.  (1979)  state that  in general, operation  and maintenance  costs
for  chlorination systems  treating  2,500  gpd  to 1  mgd  are  independent  of
flow.   Process  energy requirements  are  for  the booster  pump only,  and are
about  1,630  kWh/yr.   Building energy  requirements for a  25 ftz  building to
house  the  system would  be 2,560  kWh/yr.   Maintenance material requirements
would  be  only for miscellaneous  repair  of  valving,  electrical  switches,  and
other equipment, and  would total  about $40/yr.   Labor  requirements are for
periodic checking of equipment, with an average requirement  of 0.5  h/day, or
183 h/yr.

O&M  costs of  $2,45 7/yr  are summarized  in  Table XX.  Note  that power  costs
were  taken at  $0.07/kWh and  labor at $10.00/h.  These  were prevalent  rates
in 1982 and  were used  (U.S.  EPA, 1983)  to  update the  corresponding energy
and  labor costs made  earlier by Hansen et al. (1979).

  TABLE XX.O&M SUMMARY FOR SOLUTION-FEED GAS
             CHLORINATION
   Item                Requirements*                 Costs

  ELECTRICAL ENERGY:

   Process             1,630 kWh/yr x $0.07  =    $  114.10
   Building            2,560 kWh/yr x $0.07  =       179.20
             Total     4,190 kWh/yr x $0.07  =     $  293.30

  MAINTENANCE MATERIAL:               =    $   40/yr

  LABOR:                  183 h/yr x $10/h =     $ 1,830

   TOTAL ANNUAL O&M COST                $2.457

*  Estimates  of energy,  maintenance, and  labor  made  by Hansen et
   al. (1979).
O&M  cost curves are depicted in Figure  5 (building energy,  process  energy,
and  maintenance  material) and  Figure  6 (labor  and total  O&M costs), both
taken from Gumerman et al.,  1986).
                                     69

-------
         to,OOP.OOP
\
6
5
4
3
2
1.
ft
* «
* 3
H
W „
0 2
0
Z
0 1<
CONSTRUCT
> w * atat^toMi

-------
1C
3
a
6
S
4
3
2
1
9
I
*:
•»
i 3
j
< 2
oc
Ul
H-
< 1(
-1
Ii
2 4
5 »
i *
11
]
5
4
3
2
1
,000 1,
: &
8
a
s
4
3
2
300 1
5
4
3
2
>.
»0 | 1(
i~ I i i i i i i
ENERGY - 1
•H 	 •* «* * o»«^o»«
E f
a
9
4
3
2
1<
1
000.00




00.000
ag-M«ae^i*if •;.?-.


^


).000






100^
fiiyiiiiii.uiniiini 'iii
HM«TH*itfflnaiiaa^a




)0
0


















mmnm
marrm



























asm
mm



























SOS




























BHBtm















































BEBM







X
/"















H

aoooi







^
ma)
















V






(
r
ta
















•






^

•















h
• •





?
r

B
• •















(AINTEN/



ROCESS
>



UILDING















iNCE



ENE




ENE















: M



RQ




RG




















A"
••M


Y




Y








re
••
























R
••
























IA
•
























(,
• •

















2 3 4 S 678i 2 3439 789 2 3 4 S 3 789
.0 10 100 1000
AVERAGE PEED RATE • It/day
                                      10                  100
                             AVERAGE FEED RATE - kg/day
Figure 5.     Operation   and   maintenance   requirements   for   gas   feed
             chlorination   systems  --   building   energy,  process  energy,  and
             maintenance material (Gumerman et al., 1986).
                                      71

-------
  100.000     100.000
I
3
4
3
2
1C
S
CO
O
° 1C
rr
O 7
O 4
"3
2
1<
8
S
4
3
2
11
lows*- » « •» « w S IQ
8
1 1 1 i i i i i 3i
E »
6
S
4
3
i
£
»00 » 1C
•i OC 9
0 1
07
< 8
4
3
2
4
K) 1C
f 8
5
4
3
2
) K
•
[•HI 	 Illl Illl 	 i




1.000




p_»
00






Mmmaum
»0







3
«^ *
*»aaj










•si^ai
™-^--
l—











Effil?

—f-

j
i
J

B»Oi









jass


^-i












tat

SiESi.
=«



'
J
—

n-gan-J




,-










-°





s»







•5S-—
~~.










JHE





ra












.
















a






























8






























Jeaswassss*





























H^H


















BOTSSBffil
"—







B^l

















fcr
—









ntt«
1st

















F
"










s





























m





























B8







f






(


2 349 8789 2 3 4 S 6 78S
18 1
AVERAGE FEED RATE - It)/








































































TOTAL O&M COST
'(EXCLUDING
CHEMICALS)






LABOF





































































































B B
















t 2 3 4 S@7SS
00 1060
day
                                       10                  100

                               AVERAGE FEED RATE - kg/day
Figure 6.     Operation   and   maintenance   requirements   for   gas   feed
             chlorination  systems --  labor and  total O&M cost  (Gumerman  et
             ah,  1986j.
                                      72

-------
                                Chemical Costs

In  150-lb  cylinders,  chlorine  cost  $0.47/Ib  in  the  Washington,  DC/Baltimore
area  in  January  1983.    In  1987, the  cost  for  gaseous  chlorine  in  150-lb
cylinders  is  $0.37/lb,  when  purchased  in  ton quantities.   For purposes of  this
discussion applicable  to  small  water  systems,  the  $0.47/lb  will  be  considered
as current.

Assuming a chlorine  dosage of 5 mg/L,  gaseous chlorine  chemical  costs would
be  about $18/yr to  treat 2,500  gpd,  and about $7,150/yr  to  treat 1 mgd.

To  calculate  costs  for  gaseous  chlorine  at  lower  or  higher  dosages,   the
following formula can be  used:

  dosage (mg/L) x  no. of L dosed/day  x C\2  cost/lb
                                     ————	  = C\2 cost/day
         1000 (mg/g)  x   454 g/lb)


Sodium  Hvpochlorite  Solution Feed

                              Equipment Costs

Table XXI displays  estimates obtained during May  1980 from two vendors  of
sodium  hypochlorite  chlorination  equipment.    Data  are   presented  for  the
basic  liquid  hypochlorination system which includes  two  metering  pumps (one
serves  as  standby),  solution  tank,   diffuser,  and   appropriate   quantities   of
tubing.    However,   two  types  of basic  system  are  costed,  one  activated
electrically, the   other   activated  hydraulically.    The  basic  systems  can  be
supplemented, and  costs  of  four  increasingly  sophisticated   alternatives  also
are  presented in  this table.   Alternate  #1  adds  a diffuser  corporation cock
and  anti-siphon   backflow  preventer.    Alternate  #2  adds  a safety- housing
enclosure.   Alternate #3 adds  a flow-pacing  system,  and Alternate #4  adds  a
flow  meter and signal.

The total capital  costs for the basic and most  sophisticated systems are:

                                         most sophisticated
                         basic system	system	
 Electrically Activated         $3,643           $10,738

 Hydraulically Activated    $5,023              $15,121


                              Construction Costs

Figure  7  is  a  cost  curve  (Gumerman  et  al.,  1986)  which  shosw  that  the
construction  cost  for   hypochlorite  solution  chlorination  systems  is   constant
for systems feeding up  to 100 Ibs/day.
                                        73

-------
     TABLE XXI.  CAPITAL COSTS  --  LIQUID CHLORINATORS*
EQUIPMENT COST
(basic system**)



INSTALLATION
SITE WORK
Electrically
Activated
Avg. $ 1,800
High 2,300
Low 1,300
500
250
Hydraulically
Activated
$ 2.266
2,782
1,750
1,000
250
CONTRACTOR'S OVERHEAD & PROFIT (20%) 729   1,004
ENGINEERING FEES (10%)          364

 Alternate #1:  add diffuser           165
 corporation cock & anti-siphon
 backflow preventer

 Alternate #2:  add safety           6,930
 enclosure (housing)

 Alternate #3:  add flow pacing
 existing signal

 Alternate #4:  add flow meter
 signal, 8 in. or less

TOTAL CAPITAL  COST:

 Basic System (Equipment +        $3,643
 Installation  + Site Work +
 Overhead & Profit  +
 Engineering Fees

 Most Sophisticated               $10,738
 (with Alternate #2)

 (with Alternate #4)

*     May  1980 quotes  (two vendors)

**
  503

  231



 6,930


 1,485


 1,452




$5,023
$15,121
      Basic  System  includes  two  metering  pumps  (one  standby),
      tubing, solution tank, and diffuser
                                  74

-------
1,
8
S
4
3
2
D
8
S
» 4
1 3
CO -
0 *
0
i t(
i-TT
o i
JE «
co 8
i ^
0 3
2
1C
9
3
4
3
2
1C
1
000.00
^



M^MHHHM*
30,000






1,000
—





100







>0
0




aE=E£T'TE^




























, L
T
L
pz
i














U
;'








I







m
1!
i!







—- -1




















i





















lUt
1












JL
i


























^^r^fsoii



F

























|
























z



























I


























FT



























2 3 4 S 87i$ 2 3 4 S i 7i
10
MAXIMUM PEED RATE - 1

-— — — -
























































UUKl
























L






5Bfl^















H
r

•



sea
•a















t
JHI
^




^a«
^
M
V
aw
s




|
EK6















B















MOB
(B 09
91 B
K D
HI •
• •
m v
E as





m as






•








9 2 3 4 36799
100 1900
b/day
1 10 100
                            MAXIMUM FEED RATE ' kg/day
Figure 7.     Construction  cost  for  hypochlorite solution  chlorination  systems
             (Gumerman et al.,  1986).

                                      75

-------
                     Operating and Maintenance Costs

As  with  solution-feed gas  chlorinators,  O&M requirements  are independent of
flow for  plants  treating  2,500  gpd to 1  mgd  (Hansen et al., 1979).  Process
energy requirements  are  for the diaphragm metering  pump  and amount to 570
kWh/yr.   Building energy  requirements for  a  25 ft^ building would be  2,560
kWhyyr.   Maintenance materials would be required only  for  minor component
repair --  costs which  are  estimated at $20/yr.

Labor  is  required for periodic  mixing of  the  sodium hypochlorite solution, as
well as for checking  of  the  equipment.   Based  on a requirement of  1  h/day,
the annual labor requirement would be 365 h/yr.

Annual O&M costs of $4,108  are  summarized in Table XXII.  Note  again that
power costs are  based on $0.07/kWh and labor  costs of $10.00/h.

TABLE XXII. O&M COST SUMMARY FOR SODIUM HYPOCHLORITE
             SOLUTION FEED (U.S. EPA, 1983)
        Item                 Requirements*              Cost
ELECTRICAL ENERGY:

   Process                 570 kWh/yr  x  $0.07  =  $   39.90

   Building               2,560 kWh/yr  x  $0.07  =  $  179.20

             Subtotal    3,130 kWh/yr  x  $0.07  =   $ 219.10

MAINTENANCE MATERIAL                       $  20/yr

LABOR                    365 h/yr   x  $10/h   =   $3,650

        Total Annual O&M Cost                     $4.108

*  amounts estimated by Hansen et al. (1979)


Figure  8   shows  constant   O&M  requirements   for   hypochlorite  solution
chlorination  systems   (building  energy,   process  energy,   and   maintenance
material)  for systems feeding  up  to  100  Ibs/day of chlorine.   Figure  9  shows
similar  constant  O&M  maintenance  requirements  for  hypochlorite  solution
chlorination systems   (labor and total  O&M  costs).    Both  curves  are  from
Gumerman et al. (1986).


                              Chemical Costs

Sodium  hypochlorite   customarily is  sold as  a  15%  (by  weight)  solution.   In
July  1987,  the  cost of a 15%  solution in 1,500 gal  tanks  in the Washington,
DC/Baltimore  area  was   $0.50/gal.    A  small water   utility treating   2,500
gal/day  with  5   mg/L  of  chlorine  as  sodium  hypochlorite  solution  would
require about 64 gal/yr.    At  $0.50/gal,  this utility would spend  about $32/yr.


                                      76

-------
1(
6
9
4
3
2
1(
* 9
*» 4
: *
E 2
UJ
t-
<
a K
ui "T
0 8
Z 7
< 0
z a
Ul a
5 4
< 3
a
2
K
1
3
2
1
3,000 1
- 9
: *
a
9
4
3
2
)00 1
- 9
: i
a
9
4
i
3
2
>»
io i K
(£. 9
UJ
^ Z 4
3
2
<
i
1 K
8
» S
4
- 3
- 2
1G
1.
000. OC
M§^^BC"EI»TK-
30.000
.— «x

).000






B»Bs«3Ki
fclUHimillKH
)00







0
0
sasrssa
t-
F
i
|
EKSffia'.I^
jg^SSSGESL





WLS^T
f™1'^
C3!:
i' '






L
I



!=•
r
K:
r~






_1T~ j
i! i
5H3S&&S3S
^,i:.-,'i:.-
^^^^








rn






















-44
r:t







fl



rcT~~-"
t i^i !
S j
!
"^IR3 "i*Y*as&~li—
"TT



















rr"




















iEjraesiinaJi^JDsWsKSJii-
t )







"""^IH^"1
assaa






























»»|n>o
|
[ T





















BKfl




















«








6%





























IKpS!




















U J^**


r





•J MAJNl
" ' RJATI









- i









I
jLzE'JZJSTi.

_



PEN.
fl9A!








ESS
INQ








Jst.i3W
tfywm.
JU
rT1

4_



%N(
D








^
$£&£'-





^


t- — B
KB H E





















ENERGY
i i i i
ENERGY
























2 3 4 S 6 7 Si 2 3 4 9 8 789 2 3 4 i€7i
0 10 100 tO
AVERAGE FEED RATE - Ib/tiay
^"I
T^
n wi
n n
H •




!



















IS
00
                    1.0
        10

AVERAGE FEED RATE
                                                             100
Figure 8.     Operation   and   maintenance   requirements   for   hypochlorite
             solution   chlorination   systems     -   building   energy,   process
             energy, and maintenance material (Gumerman et al.,  1986).
                                      77

-------
  100.000    100,000
i
0
5
4
3
2
J<
9
4
£ 3
*»
1 2
t-
CO
O
0 1(
II
<:
2 »
2
1C
I
4
3
2
1C
= I
"- 7
a
5
4
3
2
I.OOO 1C
S
4
3
2
h»
>
>.
IOO f 1(
id i I I l I I 1 1
LABOR
!;• MO*, »»-*tant
: 1
0
5
4
3
2
> 1C
1.







1,000







100
















1 "f<^--~
































H














-"


-








;
_IIIII^ ' •' !
	 H


10







I
JL
^^LTTTT



L_._L.IJ_































































LJ
SSB

























0^





























2 3 4 S 6789
0 11
AVEI


«^-—















































































































































































































































. -(EXC
CHgl






























kL 0
LUC
MGt







)R





















&M
n»^
^L£





























c
3
1}





























;c





























>s





























T
















2 3430 789 2 3 4 58788
3 100 1000
*AGi PEED RATE • Ib/day
i 	 1 	
1.0 10 100
                               AVERAGE FEED RATE - kg/day
Figure 9.     Operation   and   maintenance   requirements   for  hypochlorite
             solution  chlorination  systems    -  labor  and  total  O&M  cost
             (Gumerman et al.,  1986).
                                       78

-------
A  1 mgd  facility  using  5  mg/L  of chlorine would require  400  times  more
hypochlorite solution,  and would spend $12,800/yr.

However, it should be realized  that the  cost for  sodium  hypochlorite  solution
is  about three  times  the  cost of  gaseous  chlorine  in  equivalent quantities,
and that 0.9 Ib  of 15%  NaOCl solution is equivalent  in  oxidation  potential to
one  pound  of  gaseous  chlorine.    These  ratios  between  the  two  forms  of
chlorine will assist in  obtaining more  accurate  cost calculations.


Calcium Hypochlorite Solution Feed

Equipment   costs  and  operating  and  maintenance  costs  for  this  method  of
disinfection   should  be  very  close  to  those  for  sodium  hypochlorite  feed
systems.   Solutions  of  calcium  hypochlorite  are  prepared in  a mixing  tank,
then transferred  to  a  day  tank (a  tank which  holds enough  solution to  last
for  one day),   then   injected  into  the  water   stream   using  a   diaphragm
metering pump.


                               Chemical  Costs

Solid   calcium   hypochlorite,  Ca(OCl)2,  contains  65%   available   chlorine.
Therefore,  one  pound  contains  65  Ib  of  available chlorine.    Since  a  2,500
gal/day  treatment  plant  requires 0.104 Ib  of chlorine  per day (at  an  average
dosage  of  5 mg/L), 0.104/0.65  = 0.16 Ib of  Ca(OCl)2 per  day is required with
which to prepare a solution for metering  into  the water to be treated.

During   July  1987,   calcium  hypochlorite  was   selling   for   $0.95/lb   in,  the
Washington, DC/Baltimore area.   Over  a one year  period, the 2,500 gal/day
facility  will  require 0.16 Ib x  365  days  =  58.4 Ibs of Ca(OCl)2, x $0.95/lb =
$55.48/yr.   A 1  mgd  facility, which  uses  400 times the  amount  of  chlorine,
therefore will spend $22,192/yr for Ca(OCl)2.


Pellet Feed  Chlorinators

Figure  10   shows  that  construction  costs  for   pellet  feed  chlorinators  are
constant for  systems  feeding  up to  15 Ibs/day.    Figure  11  shows  operation
and  maintenance  requirements  for  pellet  feed   chlorinators  (process  energy
and maintenance  material);    Figure   12  shows  operation  and  maintenance
requirements  for  labor  and  total  O&M  cost.   These  curves  are  taken from
Gumerman  et al.  (1986).


Erosion Feed Chlorinators

Figure  13  shows constant  construction  costs  for  erosion  feed  chlorinators
feeding   up  to  six  Ibs/day   of  chlorine.     Figure  14  shows  operation  and
maintenance requirements for  erosion  feed  chlorinators  (process  energy  and
maintenance material);    Figure  15  shows  O&M  requirements  for labor  and
total O&M  cost.   All  three figures  are taken  from  Gumerman et al., 1986.

                                        79

-------
         100,000
6
9
4
3
2
1(
6
9
4
3
2
1
i\
e
9
4
3
2
1(
^
6
9
4
3
2
1'
0.







LSpJL.








JOO







Vt?


—




•MB9
iimmtMfflg











































«






=

)

3








_™




™



«
EBC3
-











^SSE
_
""•

"
?se
23SU
ESS
_




















E

"


















SJL.
r~


i
:(

,:
"•"""-f
i rf
-
^
-


























-vTB&l
4























)




















































^±
=«j

I
!































































































































































































"—













































ISKLlT^W
Eaasaai










i s
T I ' _ ~^1
r ~*i



















-




_


I





































:







i
rift
i_LX
h


















1
4

















1
J^
1 "
jj
1
^SB
















2 3 4 9 8789 2 3498 789 2 3498 71®
1 1.0 10 101
MAXIMUM FEED RATE • Ib/day
      *»
       I
      t-
      co
      O
      o

      z
      o
      H
      u

      oc
      t-
      w
      z
      o
      o
               0.1
         1.0                   10

MAXIMUM FEED RATE • kg/day
Figure 10.     Construction cost for pellet  feed  chlorinators  (Gumerman et al.,
              1986).                                        v
                                       80

-------
10(
1
a
9
4
3
2
10
9
MAINTENANCE MATERIAL - $/yr
p M ** *• W»-*aHt»|-. M (•» * (HOt^OMCJO M U *> UK_-4fe
)0 10
•» ^
a
9
4
3
2
9
4
3
r jl
* in
IO 1 1 1 1 1 1 1
ENERGY *
JO M U *. MtOt-^OH
a
4
3
1 lo
0.
000





12_




— J
1)












^
y
6:











®m














>R<
•2 »
1















^^^.-.^=5— .,--•.- .•
fc:"™^^
" „, 	 i 	 	



iNTENA
E^S



3CESS i
p-












ICE



•Nil










_„„„.„_,[_.,„__!
--^-^^IL™,
"~~™"™[™"










M



IQ1











.; .^HS
IBM




AT



'



UOf
BBB
•«•
m




E








E
•a




3






BS
•B
IB
H




^fl








iTl,
i






tunJ—



















j
j
i











T* f
HOT
• M
• •




L












i
-hn
i »



H



2 34387§a 2 3498 789 2 3 4 3S7§§
T 1.0 10 100
AVERAGE CHLORINE FEED RATE • ib/day
                   0.1                  1.0
                       AVERAGE CHLORINE FEED RATE  -  kg/timy
Figure  11.    Operation   and   maintenance   requirements  for   pellet   feed
             chlorinators   --  process   energy  and   maintenance   material
             (Gumerman et  al., 1986).
                                     81

-------
  100.000   100.000
1
8
S
4
3
2
12
9
4
5 3
4»
* 2
CO
O
o io<
< s
i;
2
1
S
4
3
2
IP.
8
9
4
3
2
.goo 10
8
8
4
3
x 2
JO ^ 10
tD 7
: 5 :
4
3
2
I m
i i
4
3
2
10
0,
noM^noan^.
(MMBVH^^DW*
•^••••••iii


GPP






DO


MWH
^^•MKE£^X5
B^1BMK&££3?3
5


•••in imt'n imimm
BSggg




I-— •—•
I — -
L ™












msst

_



••«••••




r""L±C"~*c'n'*'
teHSEuBafeti-sa^Kt. h_i.
™~L!_
ffi«^«








ncrfa*;*

=_
fM»J!





~"
L
GK.TJ

-.
-





~
bd
h







_
_
__,L,i._L«

i





^i









»sa












!*•












a





1
I











36







^



















^^K—'






.„ ...IIM-iJ
B^e£!^



















X
.^^





riid
™
•MHMMI
•i^H^HH



















X*
^












pi












>
h-






^








TO
MEffi










^







f








T"
i
1
1






T
(

f







*


















01
EX
t*








**









\








fAL 0«D
CLUOIN
®














4 C<
3 C







LABOR
X



























3S'
HE





















I







r
M









C











































At


















BKBSBSffi;









.S)


















2 3498789 2 3498 789 2 3 4 8 8 7SS
1 1.0 10 100
AVERAGE CHLORINE FEED RATE • Ib/day
1 ' ' ' ' 1 ' ''''''I ' . . i i i i • , 11
0.1 1.0 10
                          AVERAGE CHLORINE PEED RATE - kg/day
Figure  12.    Operation   and   maintenance   requirements   for   pellet  feed
             chlorinators - labor and  total  O&M cost  (Gumerman et al,
             1986).
                                     82

-------
1.
4
•f
6
9
4
3
2
1
I
6
S
4
r '
§ «
o
i H
1 S
O 4
° 3
2

000.00
mmrmnBX^
00,000
"""""" (at
~ c
c
3.000








1000
8
9
4
3
2
1C
0.







(0
0
fe=^
taHEWia^-
I


URVi
URVI
URVi









^E^EEJSS







pHBSBm
l"™"

: A
• B
: C

















t
i^mK


a
a
•
















h
P



•
•
j
<
f

















i , i
h-
rf



PRE
ro
PRE
^FT
iftA1
40UI























SSURE W,
WELL B
SSURE M
ER AN Ai
IflTY OPI
NTED AB(























^Ef
EFOf
^TEF
30VI
•RAT
3VE






















t S
IE
I S
: G
ID
A






















YS
A
YS
RO
F
SI

















toast





Ti
S!
11
UE
E!
Qi

















i mini™





EM
JB^i
EM '
€; i
•DEI
lASi





























- CHLORINE
ERSIBLE W
. CHLORINE
'UHP
1-CHLORINA7
1 TANK
i





AD
IP
AD
'OR













































DI
DI





















T
T;





















[ON
[ON















2 349 6739 2 3 14 9 8 r88| 2 3 4 5i7SS
1 1.0 10
MAXIMUM FiiO RATg • Ib/day






.















i
100
                0.1
        1.0                   10
MAXIMUM FiEO RATE • kg/day
Figure 13.    Construction  cost for  erosion  feed chlorinators  (Gumerman  et
                                      83

-------
   1000
100.000
                               AVERAGE FEED RATE • kg/day
Figure 14.    Operation  and  maintenance  requirements  for  erosion   feed
             chlorinators   --  process   energy   and   maintenance  material
             (Gumerman et  al., 1986).

                                      84

-------
1C
6
5
4
3
2
JC
5
4
» 3
<0
ft
O «
2
1C
T
6
4
3
2
10
10,000 1
6
$
4
3
2
>,000 11
a
6
S
4
3
2
5*
JOO f 1(
tnof~9 W « » M ^j
HO8V1
II 1 1 1 1 1 I 5i
^ 3
S
4
3
- 2
1C
C
30,000
-»S_B«=4=— .
~ '^""^^


5 000





)00
5i^*1





0










a£LU,
















^

















+






a*4&
T.1"™:1 	 T-fr1"1 ""
.... _.j«^ 	
«*8B*fjp
OB*35a!5a^»^sai3»


1



















J»—
f-










^






f








M










'






0



















*'






>*


















/
{





_jt
X

















J*






4
r
















mA
(/






.
/-










.39eLC£
«,




__J
/"




•BH
A
*











UC5BC
•»Pui-





W





••
rjj
r













""










F-


















1 TOTAL











M«=











IBB





f"?





. O&M COST
UQING
CAL5)



• LABOR




































































































Kupn





• m
• •
• •


















2 34SS789 2 3489 789 2 3 4 i i 789
.1 1.0 10 100
AVERAQg FEED RATE - Ib/day
                   0.1
1.0
10
                              AViBAQE FEED RATE • kg/day
Figure 15.    Operation  and  maintenance   requirements  for  erosion  feed
             chlorinators  -  labor and  total  O&M  cost  (Gumerman  et  al.,
                                      85

-------
On-Site Electrolytic Hypochlorite Ion Generation

Figure  16   shows   construction  costs   for  on-site  hydrochlorite   generation
systems producing  up  to  100  Ibs/day of chlorine.   Figure  17  shows  operation
and  maintenance  requirements  for  on-site hypochlorite  generation  systems-
building  energy,  process  energy,  and  maintenance  material;  Figure  18 shows
operation and  maintenance  requirements  for  labor and  O&M  cost.   These
curves are taken from  Gumerman et al.  (1986).


              b.    Ozone

                   i.    Characteristics and Properties of Ozone

Ozone  (03)  is  a  very  powerful  oxidizing agent,  second only  to  elemental
fluorine  among  readily available chemicals.   Because  of  its  strong oxidizing
ability,  ozone  also  is  a  powerful  disinfectant.    It  is   an  unstable  gas  at
ambient  temperatures  and pressures,  and  decomposes  rapidly to  oxygen (from
which  it is  made)  at temperatures  above 35°C.  For  this  reason, it cannot be
manufactured and  packaged  at a central manufacturing plant,  as can chlorine.
Therefore, ozone must  be generated on-site  and used at  once.

Ozone  has  a  characteristic  odor which  can be detected  by  most  humans  at
low  concentrations  (0.01  to  0.05  ppm  by volume),  far   below  the  levels  of
acute  toxicity.   However,  olfactory  fatigue  has been  noted in some  instances.
This   means   that   as   the  length   of   exposure   to   an   ozone-containing
atmosphere  increases,  the odor  of  ozone  may become  less noticeable  to the
individual being exposed.
                                                                         >
When  added to  water,  ozone dissolves only  partially.   That  is  to say  that
ozone  is  only  slightly  soluble in  water,  about  2-10  times  the solubility  of
oxygen,  depending  on  the  water  temperature.   The  solubility  of  ozone  in
water  is governed  by  Henry's Law,  which  states  that  the  mass  of  ozone  that
will  dissolve in a  given volume of water,  at  constant  temperature, is  directly
proportional to  the  partial  pressure  of   the  ozone  gas above  the   water.
Consequently,  proper   design   of   the   ozone  contacting  system  is   very
important to  the proper  application  of  ozone  to  water  (see  later  discussion-
 iii).

The  stability of  ozone  is  greater in  air than  in  water, but is not  excessively
long in  either case.   The  half-life  of  ozone  in  water  has been reported  to
range  from  8 minutes  to  14 hours,  depending  on  the  phosphate  and carbonate
concentrations  of the water  (Grunwell et  al.,  1983).   With no  phosphates  or
carbonates  present  and the  water adjusted to pH  7.0  with sodium hydroxide,
the half-life  was 8 minutes.
                                        86

-------
     0)
     O
     o

     z
     o
     H
     o

     (C
     h-
     co
     2
     O
     O
1(
7
6
5
4
3
2
t
8
9
4
3
2
1C
i
a
9
4
3
i
2
1(
6
5
3
2
1C
1
), 000,0
.~-™-~-
-?nrg^ ^aiggtaa^ijCE






)0,000






^•B^SSS!ES!S
J.OOO


00



i















2ESQSH



I




)00











•S3SB2L













ft^H



_

























^c_


















"™"
„«


















««S2ffl











1 1









=>-»




















































--































#*>




















P

«=™*l=r~'
.a.,


















*
T
onJ

^-



1




























p-
57^1







^5^a^:JGa^^.^^raa»J5»?iJMigJ^a^
|

















£!
-.








L ~ I

















->

1











-n^sea^
t^uatiffitf
jagwran-gr









^ai-JM"







"




EBDBCTt^
m^ra^njr
jnv
















L
=
sar-
I







i
_









:
6
n
r"













m»t— 1






























[i














•i
*m
M
U




















1
1



i
B M
P M
• Pi















B



2 3436789 2 3436 789 2 34517 89
.0 10 100 1000
MAXIMUM FEED RATE - Ib/day of chiorin* equivalent
1.0 10 100
                MAXIMUM  FEED RATE - kg/day  of chlorine equivalent
Figure 16.    Construction  cost  for  on-site  hypochlorite  generation  systems
             (Gumerman et al.,  1986).
                                       87

-------
    10.000
1,000.000
                        MAINTENANCE MATERIAL
                                               PROCESS ENERGY
                                                    BUILOINQ ENERGY
                                                            2  3  4 SS788

                                                                       1000
                     AVERAGE FEED RATE • Ib/day o* ehlorlne- equivalent
                     i I      i   I   I f i i i '[       i   I   <  I i i  i rr      *
                     1.0                  10                  100

                      AVERAGE FEED RATE « kg/day oJ chlorine equivalent
Figure  17.    Operation   and   maintenance   requirements    for   on-site
             hypochlorite   generation  systems   --  building  energy,   process
             energy, and maintenance material (Gumerman et al.,  1986).

-------
CO
O


2
•a
O

_J
<
t-
o
1C
6
S
4
3
2
JJ
6
3
4
3
2
1(
T
6
9
4
3
2
1C
T
s
4
3
2
1C
10,000 K
7
8
5
4
3
2
XOOO K
e
s
4
3
„ 2
w
)00 . 1C
IQ 1 1 1 1 1 1 I 1
LABOR
>|-^ M U * OK»-^OHC
e
s
4
•» 3i
2
» K
1
30,000





3,900
__m™»_

"cr
—




(




)00
^





!0




mim^maammgam
***•£
>




-*-**1
Kn_










z&>


I




^



r




c:
h





-

i

|

^

'



/











*«s












^*












grf
















I*
T-S









!»•

















K1










j?





rc^=ZL~ ' »— — i-^—






i
, t





--— H--=— •
™(



y
"«^™"J
""^^








1









^

rf




















1


/











/







i

























1
Cj— :f—
II



,/







n


,
. TOTAI
• (EXCL










. Ot
.UDI
ICA










M







|






r









•»


(










?BB^E5


K*










•1

1
>




LABOR








i











I




kM
NG
LS]












C






































35
r
























iT
™





















EV M H


•MBMM



















2 3 4 5 6789 2 3458 739 2 3 4 58789
.0 10 100 1000
AVERAGE FEED RATE - tb/day of chlorine equivalent
                    1.0
                     	I

                          10
	I

      100
                    AVERAGE FEED RATE • kg/day of chlorine equivalent
 Figure 18.
Operation    and   maintenance   requirements   for   on-site
nypochlorite  generation   systems   --  labor  and  O&M   cost
(Gumerman et al.,  1986).


                        89

-------
                   ii.   Generation  of Ozone

For  the  on-site  generation  of ozone   by  means  of  corona  discharge   (the
procedure  used  most  for  water  treatment),  an  oxygen-containing  gas   (air,
oxygen-enriched  air,  or  pure  oxygen)   is  dried   and  cooled,  then  passed
between  two  electrodes   separated by  a  discharge   gap  and   a  dielectric
material across which  high  voltage  potentials are  passed.   In  recent  years,
the more  modern  ozone  generating equipment has been  designed  to operate
at high frequencies rather  than high  voltages.

Figure  19 is  a  schematic  diagram  of  the essential components  of a  corona
discharge ozone  generator.    For small  water  plants, ambient  air  dried  to  a
dew point  of minus 65°C  (-65°C)  will   be the  source  gas fed  to the ozone
generator.   It is imperative  that the  ambient  air be  dried rigorously in order
to  maximize   the  yield  of  ozone   produced,  as   well  as  to  minimize  the
formation of corrosive nitrogen oxides.
                                i HEAT  1

                                T1    ?
                         %
                 — ELECTRODE
                 — DIELECTRIC
                    02
DISCHARGE
   GAP
Figure 19.    Typical  corona discharge cell ozone generating configuration.

With properly dried air, the  output  from  a corona  discharge ozone generator
will  be dried, cooled air,  containing \%  to 3.5%  of ozone, which  is  partially
soluble  in water  (about 2-10 times  the  solubility of  oxygen).   This produced
mixture of ozone in air then must be mixed  with  the water to be  treated,  by
a process known as contacting (see later discussion -  iii).

Ozone also  can be generated  by passing UV  radiation  through  ambient  air.
However,  in   this  procedure, only small quantities of  ozone can  be generated,
and  at  much lower  concentrations (0.1  to 0.001%)  than  can  be generated  by
corona  discharge.    This  concentration  is  considered  insufficient   to  provide
enough   ozone  in  water   to   cause  a   significant   amount  of oxidative  or
disinfective treatment.
                                        90

-------
When  oxygen  is  used  as   the   feed   gas  to  the  ozone   generator,   the
concentration of  ozone  produced  is  effectively doubled  over  that  produced
when properly dried air is  employed.  This  means that  oxygen containing  2%
to  7%  (by  weight)  of  ozone  is generated  in  the  same  size  corona  discharge
ozone generator  and for the  same expenditure of  electrical  energy  as when
dried air  is used.

In  the  United States today, there are at least  35  operational  water  treatment
plants using  ozone.   Only  the  new  540 mgd  plant  in  Los  Angeles  is using
oxygen  as the feed gas.   In the rest of  the world, where  ozone is used more
extensively  than   currently  in  the  U.S.,  fewer  than  a dozen water  treatment
plants are  known to be using oxygen  as the feed gas, although more  interest
is being shown in its use.

When the  properly  dried  air  or oxygen passes  through the  ozone generator,
part  of  the oxygen  dissociates  as  a result of being  exposed  to  the  high
energy electrical field of the corona  discharge:

        O2  +   e'   	>    2[O]    "fragments"

      oxygen

These oxygen "fragments"  are highly reactive,  and  they  combine  rapidly .with
molecular oxygen, forming the  triatomic molecule,  ozone:

        2[O]     +   2 O2  	>   2  03

      oxygen        oxygen           ozone
      fragments

The overall reaction to produce ozone is  the sum of the above reactions:

        3 O2   +   e'   <— ->   2 03

      oxygen     energy          ozone

The  reaction  to   produce  ozone  is  reversible,  meaning  that  once  formed,
ozone decomposes   back  to   oxygen.     This  reverse  reaction occurs quite
rapidly  above  35°C.     Therefore,  because  reactions  involving  high  energy
electrical  discharges also are  accompanied by generation of considerable heat,
ozone generators  are designed to  include  a high  degree of cooling,  in order
to minimize  ozone losses by decomposition.


                   iii.   Contacting of Ozone  With Water

Because  ozone  is only  partially  soluble  in water (2-10 times  that   of  oxygen),
the  manner with which it  is  contacted   with  the  water  to  be treated is of
primary importance.    In turn,  the manner of contacting ozone   with water
depends upon the particular ozonation job or  jobs to be accomplished.
                                       91

-------
For   purposes  of  illustration,  there  are  basically  two   types  of   reactions
involving   ozone:     fast  and  slow.     Fast   ozonatipn  reactions  include  the
destruction  or  inactivation  of  microorganisms  (disinfection),  oxidations  of
iron,   manganese,   sulfide  ions,  nitrite   ions,  some  organics,  and   lowering
turbidity  levels.    On   the   other   hand,  other  ozone   oxidations,   such  as
oxidation   of  many organic  materials  (specifically  many pesticides  and  VOCs),
are kinetically rather slow.

As a  result,  for  the  fast  ozone reactions,  the contacting  objective  is to add
the  requisite   amount  of ozone  to  the water as  rapidly  as  possible,  so that
the  solute being   oxidized  or  inactivated will  react  as  rapidly  as  possible.
The  rate  of  this  type   of ozonation  reaction  is  dependent  on  the  rate  of
transfer  of  ozone  into  solution.    This   type  of  reaction  is  termed  "mass
transfer controlled", and  contact times, therefore, are relatively short.

For example, during  disinfection  with  ozone, ozone is  added first  to satisfy
the  ozone demand of  the  water  and to  create a residual  dissolved ozone
level,  then to  maintain  the requisite  residual of  ozone for a given  period  of
time.

Slower ozone  oxidations  are  dependent upon  maintaining  a  low  residual
concentration  of ozone  in  the water, but over  a much  longer  length of time,
because  the  chemical  reaction rates  are  slow.   These types of  reactions are
termed "reaction rate controlled".

Ozone  can  be   generated  under   positive   or  negative   air  pressure.  .  If
generated   under   positive  pressure,  the  ozone-containing  air  normally  exits
the  ozone  generator  at  approximately  15  psig.   This is  sufficient  pressure for
the  gas  to pass  through  porous diffusers installed  at the  base of a  16 foot
column or tank of water.   Fine  bubbles  containing ozone  and  air (or  oxygen)
rise  slowly through  this  water column.    During the  time  of bubble   rise, the
ozone contained  in the  bubbles diffuses  from  the gas  phase  into  the  water
phase, where  oxidation  and/or disinfection takes  place.   The  16  ft  height  of
water  column  maximizes  the  amount of  ozone  transferred  from   the  tiny
bubbles as they rise.

Other types  of  positive  pressure  ozone  contactors  include  packed   columns,
static  mixers,   and  high speed agitators.   The  water to be treated  also can be
sprayed  through   small   orifices   into   an  ozone-containing  atmosphere  (the
atomizer,  or spray-drying principle).

When ozone  is generated under  negative  pressure, the air  or oxygen   is drawn
through  the  feed  gas  preparation  system, then through  the ozone  generator
and  into   the  contactor.   The  sub-atmospheric pressure  can  be  created by
employing  a  submerged  turbine as the  contactor.   Thus, when  the  turbine  is
turned  on, its whirling   action  creates  a  slight  vacuum,  drawing  air  through
the entire  ozonation system.

Other  methods  of  creating  sub-atmospheric  pressure  which  are  in   common
practice  are   by   use  of  injectors  or  Venturi-type   nozzles.     Water  to be
treated is  pumped  past  a  small  orifice  (injector)  or  through a  Venturi  nozzle.
In  either  case, a  slight  vacuum is  created,  which  can  be used  to  draw the

                                         92

-------
air/oxygen/ozone  output  of  the  ozone  generating  system.   Contacting  takes
place during the mixing of gas with  flowing water.

It can  be  appreciated  that  the  diffuser  and  packed  tower contactors require
no energy  to  operate,  whereas  high  speed  agitators,  static  mixers,  and all of
the negative pressure  contactors require energy  for their operation.

As excellent  discussion  of  the  various  types  of  ozone  contactors  has  been
assembled by Masschelein (1982).


                   iv.    Destruction of Contactor Exhaust Gas

No  matter how well  designed the  ozone  contactor,  it is  rare, if ever,  that
100%   transfer  of  ozone  to  the  water  occurs,  unless  insufficient  ozone is
provided.   Therefore, there  will  always  be some  excess  ozone in  the  gases
which   exit the  ozone   contacting  system.    Particularly  for disinfection  with
ozone,   during   which   a   dissolved  residual   of   ozone   is  present  for   a
predetermined  length  of  time,  there will  always  be ozone present  in  the
contactor  exhaust   gases  (off-gases).     This  excess  ozone either  must  be
reused  or  destroyed before discharge of the gases to the plant atmosphere.

Reuse  of  ozone  can   be  effected  if   two  stages   of  ozonation  (one  for
oxidation,   one  for  disinfection)   are   designed  into  the  water  treatment
process.    In  this  event,  contactor  exhaust  gases  from  the later  ozonation
step are  drawn  into  the earlier ozonation stage, for example, by means of  a
submerged  turbine.   However,  even well-designed  reuse of ozone in a second
stage still will  leave  some excess ozone in the final  exhaust gases, and  this
must  be  destroyed   before   discharge.     Destruction   of  ozone (to  produce
oxygen  from which it  is generated) is accomplished by  several methods:

   o     thermal  destruction  --  at   300-350°C,   ozone  decomposition .requires
         only a few seconds,

   o     catalytic destruction  - at  ambient temperatures  with the  appropriate
         catalyst (metal or metal oxide),

   o     adsorption and  decomposition on moist granular activated carbon,

   o     exposure to UV radiation (254 nm wavelength).

This  latter method  is  in use only  in  very  large  treatment plants, because
even at 254 nm wavelength, there  is a small amount of ozone generated.


                   v.     Chemistry of Ozone  in Water

Unlike  chlorine, ozone  does  not  react  with  water   to  produce   disinfecting
species.    However, above pH about  6,  ozone  does  decompose  in water  to
produce the more reactive hydroxyl free radical:

             03   +  H20  —>    O2   +  2(OH)-

                                       93

-------
This reaction is accelerated  at the higher pH  values  (above 8).   In  addition,
hydroxyl  free   radicals  are   produced  whenever  ozone  is  exposed   to   UV
radiation,  or when  ozone  is  applied in  the  presence  of hydrogen  peroxide.
Ozone   oxidation  rates   of   some  refractory  organic  materials   are  greatly
accelerated by the simultaneous  addition of UV radiation  or  H2®2 because of
the formation of hydroxyl free  radicals.

Thus,  when  ozone is added  to water it can react as  the  03 molecuJe, as the
hydroxyl free radical, or  as a mixture of both.

Bicarbonate  and  carbonate   ions  are  excellent  scavengers  of  free   radicals;
therefore  waters  containing   high  alkalinity  levels  will  quickly   destroy  the
hydroxyl free radicals  produced  by the decomposition  of  ozone,  and  eliminate
the benefits  of the free radical oxidative mechanisms.
                   vi.    Establishing An Ozone Residual

In  order  to attain  the appropriate CT value  to  achieve 99.9% inactivation of
Giardia lamblia  cysts  (Table  IIA) a  measurable  residual level of ozone must
be  developed in water and maintained  for  the appropriate  period  of time.   In
operating    drinking  water   treatment  plants  using   ozone  for   disinfection
(primarily  in France,  but also  in  many  other  countries of the world), it is
customary  to attain  a  dissolved ozone  residual  concentration  of  approximately
0.4  mg/L,   then  to  maintain  this concentration  for  a  period of time of at
least four   minutes  (laboratory)  and  6-8 minutes  in operating water treatment
plants.

It  can  be  appreciated  that   the  water  entering  the  ozone-for-disinfection
contacting  chamber  has the highest  ozone  demand, whereas water  exiting  the
contacting   chamber  has  the  lowest ozone demand.    Therefore,  part of  the
function of the  ozone  contacting system  is  to  satisfy  the  short term,   or
immediate,  ozone demand of the  water.   Since disinfection  is  a  mass  transfer
controlled   reaction   (fast), a  certain  percentage  of ozone  is  utilized  initially
to  satisfy   the   short  term  ozone  demand.    After   this   ozone  demand is
satisfied,  the contacting  task  becomes  one  of  maintaining  a  specific ozone
concentration for a specific time  period.

These  two  tasks   normally   are  conducted  in  two  different  contacting
chambers,   with  the  larger  amount   of  ozone   being applied  to  the   first
chamber.    This  procedure,  which is   employed  to minimize  the   amount  of
ozone added, is  shown  in  Figure 20.   Approximately two-thirds  of the total
ozone required  is  added  to  the first  chamber, in which the  ozone  demand of
the  water  is  satisfied,  and  a  residual  ozone  level   of  about  0.4  mg/L is
attained.    The  balance  of  the  ozone  (one-third) is  added  to the second
chamber, in which  the  0.4 mg/L ozone residual is maintained.

To  attain  the necessary  contact  time  for  the ozone  with  a 0.4  mg/L ozone
residual, it  is simply a matter  of dividing  0.4 into  the CT  value from Table
IIA  taken  at the   appropriate  temperature.    The product is the minimum
contact time  (in  minutes) required.   Table  XXIII  shows the  CT  values   and


                                        94

-------
   contact  times  required  for  an  ozone  residual  of  0.4  mg/L  at  various
   temperatures.
UNOZONATED
  WATER
                                                 CONTACT
                                                 CHAMBER
                                                 OFF-GAS
         U'AvB^
           «.*  it >'".!«• ..vv
     OZONE-RICH
        AIR
FLOW METER (TYPICAL)

VALVE  (TYPICAL)
   Figure 20.    Two-compartment ozone contactor with porous diffusers.
    TABLE XXIII.  CT VALUES AND CONTACT TIMES FOR OZONE
                  TO ACHIEVE  99.9% INACTIVATION OF Giardia
                  lamblia AND > 99.99% INACTIVATION OF ENTERIC
VIRUSES FOR 0.4 mg/L OZONE
Temperature (°Q
Disinfectant pH
Ozone (0.4 mg/L) 6-9
CT value
Contact Time (min)
0.5

4.5
11.25
5

3
7.5
10

2.5
6.25
15

2.0
5
20

1.5
3.75
25

1.0
2.5
                                   95

-------
If the dissolved  ozone  residual can  be  raised to  0.5  mg/L, the  contact  times
will  be somewhat  lower.   If the ozone residual  is lower  (i.e., 0.3 mg/L),  the
contact time  will be  somewhat longer.   If only  a  1-log or  0.5-log inactivation
of  Giardia  is required, as  when  preozonation  disinfection  is  practiced  prior
to  filtration),  the  CT value  (and  therefore, the  ozone contact time)  required
will be even lower (see Table IIB and IIC, respectively).

Many  current ozonation systems  in  the  U.S.  today are  designed to  optimize
the  oxidative  capabilities   of   ozone,   not   necessarily   to   optimize  ozone
disinfection.    In these  instances  it  is   customary  to  provide   a  single  ozone
contact chamber.   If single  contact  chambers  are  used  for  ozone disinfection,
and the  residual   ozone  concentration  is  measured   at  the  outlet  of  the
contact chamber,  how  should  the  ozone concentration  and contact  time  be
determined?

The initial  response  is to  assume  that the  ozone residual  concentration  at
the  outlet is  present  throughout the contact chamber.    Were  this to  be true,
one would   simply  multiply the  detention  time  in   the  contactor  by  the
concentration  of  ozone  in  the water  exiting  the  contactor  to determine  the
CT value.

However,   this   assumption   is   incorrect.      Considering   only   the   first
compartment  of the  2-compartment  ozone  contactor in Figure  20, it  must be
realized  that  the water entering the  upper  part  of the compartment  contains
no  ozone  and  constantly  exerts  an ozone  demand.    Therefore  it  can  be
concluded  that  the  upper   portion  of   this  compartment  contains  much  .less
ozone  than the  lower  portion.   However,  the ozonized water exits from  the
lower  section  with  a measurable dissolved ozone residual.

It  is  clear  from this reasoning that the  true CT value  of a single  contact
chamber is  lower  than  that  obtained  by  multiplying the  detention  time by  the
ozone   concentration   in  the  water   exiting   the  chamber.     As  a   first
approximation only,  one  should   assume  that the average  concentration   of
ozone  throughout  the contact  chamber   is,  say, 50% of that measured at  the
outlet.  But even this  approximation may not be true.

If  single  ozone  contact chambers  are  to be   supplied,  it  is  advisable   to
measure  the  dissolved  ozone concentrations  at various points  in  the  chamber
at  the various  water  flow rates and ozone  dosages with  the  actual  water  to
be  treated.    In  this  manner, an ozone  concentration profile  can be  developed
for  the  particular  contactor,  and  CT  values can be  calculated  which  are
closer  to the  true value.

It must be  recognized  that if  ozone is  employed early in  the  water treatment
process  to   oxidize   iron  and  manganese,  and  if the   levels of  iron  and
manganese  are  sufficiently  high,  copious  amounts  of brown/black precipitate
will  be   produced.    These  insoluble   materials  will   interfere  with   the
measurement  of a  dissolved  ozone residual, even  though  a  significant residual
may be present.
                                        96

-------
              vii.  Factors Affecting the  Disinfecting Efficiency of Ozone

It  should  be  appreciated  that  because ozone is  such  a powerful  oxidizing
agent, it  is  not  particularly  selective.   In other words,  if  ozone is  used  early
in  the  water treatment  process,  for  example,  to  oxidize  iron  and  manganese,
or  color,  or  tastes  and  odors,  a  high  degree  of disinfection  also  will  be
obtained.     Conversely,  if  ozone  is  used  after  filtration  for  disinfection,  it
will also oxidize  any  easily oxidizable materials still  present.

Ozone's   ability   to  disinfect   is   affected   little  by   changes   in  water
temperatures or  pH, and  it  does  not  react with  water, except as  noted
earlier.    Also,   ozone does  not  react with  ammonia  at  any  significant  rate,
except  above pH  9.   However,  ozone  does  have  a short  half-life in water,
decomposing to  oxygen  at  a  rate  dependent upon the  water  temperature.
This  means that  ozone  will  not have  a  lasting  residual  in  the   distribution,
and  therefore  normally  cannot  be  employed   as  the  secondary   disinfectant.
Although   there  are  a number  of  European,  Canadian,  and  even two  U.S.
water treatment  plants  which do  use  ozone   as  the  secondary   disinfectant,
these are  the  exception  rather than the rule.   In those cases  in which  ozone
is  the   last treatment   step,  a   combination   of  five  factors  must  occur
simultaneously (Miller et al.,  1978):

    o   cool water temperatures,
    o   clean and short  distribution system,
    o   short residence time (<  12 hours)
    o   low levels of organics,
    o   no  ammonia present.

In  all  other cases,  a secondary  disinfectant which  provides a  stable  residual
is  added   after  ozone  has  been  utilized  as  the   primary  disinfectant.   One
advantage  of employing  ozone  as the primary  disinfectant is  that  the oxidant
demand  of  the  water  is  satisfied  simultaneously,  which  allows  much  lower
dosages  of  chlorine,  chlorine  dioxide,   or  monochloramine   to  be  utilized,
normally less than  0.5  mg/L,  to  provide a  stable, detectable, residual for the
distribution system.


              viii.        Ozonation System Design

Figure  21  illustrates  the  five basic components of an ozonation system  which
employs ambient air as  the  generator  feed gas.    The  essential   components
include  air  preparation,   electrical  power  supply,   ozone  generation,  ozone
contacting,   and  ozone  contactor  exhaust  gas   destruction.    The   amount  of
ozone  produced  by  an   ozone   generator   is   affected  by  the  physical
characteristics of the  equipment,  the power supply to  the ozone  generator,
the  moisture content  and  dust  content  of  the feed-gas,  the  temperature  of
the  ozonized gas,  and  the  oxygen  content  of  the  feed gas.    Each of these
basic system components  will be discussed below.
                                        97

-------
THE FIVE  BASIC
 COMPONENTS
     OF  THE
   OZONATION
    PROCESS
     Electrical
       Power
      Supply
               2
       Gas
  Preparation
              1
dry
gas
  Ozone
Generator
         3
                                  A
                     HoO
o,
                   Ozonated  Water
                        Exhaust Gas
                         Destruction
                                   5
        Contactor
4
 Figure 21.   The five basic components of an air-fed ozonation system.


 Air Preparation

 Ambient  air must  be  dried  to  a maximum dew  point  (a  unit of moisture
 content  of  air)  of minus 60°C.   Even  drier gas  is  preferable  (lower than-
 60°C).    This  unit  process   must  be  designed  conservatively,  especially  for
 warm, humid climates.   Use of air having a  dew point higher than -60°C  will
 result  in  lower   ozone  production,  slowly  foul   the  ozone   production
 (dielectric)  tubes  or plates,   and increase  corrosion  problems in  the  ozone
 generator unit and downstream equipment as well.

 Air  feed systems  typically are  classified  by their operating  pressure.   The
 most common is a  low  pressure system, which  operates  at  pressures  ranging
 from 69  to 103 kilopascals (kPa =  10 to  15 psig), although  pressures  up to
 275 kPa  (40 psig) have  been reported when the  gas pressure  is  reduced prior
 to the  ozone  generator.   High pressure systems operate  at  pressures  ranging
 from 480 to 690 kPa (70  to 100 psig), reduce the pressure prior to the ozone
 generator, and  typically  are  used  in  small  to medium sized applications.
 Either system may be  used  in  conjunction  with most ozone generators  and
 with  all of the contacting systems  described.

                                    98

-------
The  decision to  use a  high or  low  pressure  air  preparation system  often is
based on a qualitative  evaluation  of potential  maintenance requirements,  in
addition  to  the  quantitative  capital  cost  evaluation.   Some  of the  issues  to
consider are listed below:

1.  High   pressure   air   pretreatment   equipment   generally   has    higher
    maintenance requirements for the air compressors;

2.  High  pressure  air pretreatment equipment generally  has lower maintenance
    requirements for  the  desiccant  dryers.

3.  High  pressure air pretreatment equipment  generally  has a lower  capital
    cost.    At small  to  medium  sized installations  this lower  capital cost  may
    offset  the   additional  maintenance  required  for  the  air  compressors  and
    associated equipment,  such  as  filters  for the  oil-type compressors.    The
    design  engineer  should  investigate  the  potential  maintenance  associated
    with  the  high  and low pressure systems  rather  than  evaluating  the  design
    on the basis of capital cost alone.

Another  type  of  air  feed-gas   treatment system   is  the  "nominal  pressure"
system,  which  typically  operates   at  a  negative  or  in  some cases a  slightly
positive   pressure.    This   type   of air  pretreatment  system  is  designed  to
operate  in  conjunction  with aspirating  turbine,  injector,  and/or Venturi  type
contactors, all  of  which  create the  partial vacuum  necessary  to  draw  ambient
air  through  the  air  pretreatment  and   ozonation  systems.    Since  ozone
disinfection  of  drinking  water  involves   maintenance  of   a  specific  ozone
contacting time  to attain  the  required  CT value,  this  type  of  contacting' will
require  higher  energy expenditures than the positive  pressure contactors (and
positive  pressure  air pretreatment  systems).   Therefore,  nominal pressure  air
pretreatment systems  are  not  considered  appropriate  for  drinking  water
disinfection.

Figures  22  and  23  are schematic  diagrams  showing  low  pressure and  high
pressure   air   feed   gas   pretreatment   systems,   respectively.      Figure  22
illustrates  a dual   component  process,   showing  desired  flexibility   for   the
equipment provided.   The  precompressor  filters are  provided  to  protect  the
air compressors from damage  due to  large  particles.    The air  compressors
typically   are   positive   displacement,   oil-less  units.    Positive  displacement
compressors   are   used   in  order  to  obtain  constant  air  flow  at  variable
operating  pressures.     Variable   pressures  often   are  encountered   due   to
variable   pressure  losses   in  downstream   equipment  and processes  such  as
filters  and ozone  contact  basins.    Oil-less compressors are  used to  eliminate
oil  contamination  of  the  downstream  desiccant  dryer  medium   and  ozone
generator  dielectrics.    Liquid-seal and  rotary  lobe  compressors   have  been
used most frequently.

The compressors  may be  followed  by  an after-cooler  or a  refrigerant  dryer.
These  components  are  depicted  by dotted  lines in  Figure  22,  which  indicate
that they are optional.   Typically,  either one or the  other  option  is  provided.
These  cooling mechanisms  are used to remove moisture in  the  air  at  minimal
operating expense.


                                        99

-------
         Pre-Compressor
             Filter
          Compressor
          — — —>— — —
           After-Cooler
         __ _ ___.
                            ._„]- Optional
  Pre-Compressor
       Filter
   Compressor
  ,___L__.
 •               »
 1  After-Cooler   !
 I	__ .„.. j
           Refrigerant
          . - Dryer -—.
                I	
 ___J__...
 •   Refrigerant
 L-- Dryer	
	I
          Pre-Desiccant
             Filter
HH
Towe
rA
To

/ver B
          Post-Desiccant
              Filter
                           Heat-Reactivated
                           Desiccant Dryers
   Pre-Desiccant
       Filter
  Tower A
                                                     Tower 8
   Post-Desiccant
       Filter
        To Ozone Generator
To Ozone Generator
Figure 22.    Example  low  pressure  air  feed-gas  treatment  schematic  for
            ozone generation (U.S. EPA, 1986, p. 125).
                                 100

-------
          Pre-Compressor
               Filter
            Compressor
            After-Cooler
           Pre-Oesiccant
               Filter
         Tower A
                  Tower B
          Post-Desiccant
               Fitter
   Heat-Less
Desiccant Dryers
                    Pra-Compressor
                         Filter
                      Compressor
                       After-Cooler


                     Pre- Desiccant
                         Filter
    1
Tower A
                            Tower B
                              I	
                                                   t_	_r
                     Post-Desiccant
                         Filter
                                   PRV - Pressure Relief Valve
         To Ozone Generator
                   To Ozone Generator
Figure 23.    Example  high  pressure  air  feed-gas  treatment  schematic  for
             ozone generation (U.S.  EPA, 1986, p. 126).
                                    101

-------
The compressed,  cooled air is directed  to a  predesiccant  filter,  which  is used
to  remove  dust  and dirt  particles greater than 3  to 5  microns  in  diameter.
Particulate  removal  prior  to   the desiccant  dryers   reduces  plugging  in  the
desiccant medium.

Probably the most  important  component  of the  air  treatment process  is  the
desiccant dryer,  which  consists  of  two  towers  containing  moisture-adsorbing
media.   One tower  operates in the  adsorption  mode  while the other  tower is
being  regenerated.    The  low  pressure  system  desiccant  dryer uses  heat  for
reactivation of the desiccant.

Post-desiccant  filters  are installed  to  remove  particulates  smaller  than  0.3  to
0.5  micron  in  diameter.   Two-stage  filtration  is  preferred.   The first  stage
filter  removes  particulates  greater  than  1  micron   and  the  second  stage
removes particulates  less than 0.3 to 0.4  micron in  diameter.

In  the  high  pressure  air pretreatment  system  (Figure 8), the  precompressor
filters    are   used   to   remove   larger   particulates   and   protect  the   air
compressors.   These compressors  typically are  oil-less  units;  however,  oil-seal
compressors can be used if followed by extensive oil removal equipment.

Following  the  high  pressure   compressors,  it  is  essential  to  provide  after-
coolers,  to  remove  the  heat   of  compression.    The   filter(s)  before,  the
desiccant dryer are  used  to  remove  particulates less  than  3  to 5 microns  in
diameter when  oil-less  compressors are  used.    When  oil-seal  compressors  are
used, filtration to remove oil droplets  less than 0.03 micron is provided.

The  high  pressure  system   desiccant  dryer   consists  of two  towers  with
moisture adsorbing  media.   One tower  operates in  the adsorption  mode while
the  other  is regenerating.   Regeneration is  accomplished without  additional
heat.    Thus,  the  high  pressure   desiccant  dryers  are called  heat-less units.
The  post-desiccant  filters  remove  particulate  matter  less  than  0.3.  to   0.4
micron  in diameter.   The high  pressure  system also  has  a pressure reducing
valve to regulate  operating pressures in the ozone generator.

Air  compressor,  refrigerant  dryer, and  desiccant  dryer  design  considerations
are discussed in detail in U.S.  EPA, 1986.
Electrical Power Supply

Supply  line  voltage  (220/440  V),  or frequency  in some  cases,  is  varied to
control  the  amount  of  ozone  being  generated  and  its  rate  of  generation.
Because these  two parameters are  varied  in many ozone  generation  systems,
the  electrical   power  subunit  can  represent  a  proprietary  product  of  the
ozonation  system supplier.   As  a result, the  electrical power system must be
specified  as  an  integral  power   supply  system  specifically  designed  for  the
ozone  generator to be  supplied.   In other  words, the power system  should be
designed for and purchased from the ozone  generating system supplier.

Because the  ozone  generator uses  high voltage  (>  10,000 V)  and in some
cases also high  frequency electrical  current  (up  to  a  maximum  of 2,000 Hz),

                                        102

-------
special  electrical  design  considerations  must  be implemented.   For  example,
special   insulation   must  be   provided  for  the  electrical   wire.    A  cool
environment for  the  high  voltage  transformers  should be  provided,  and the
electrical  transformers  should  be  protected from ozone  contamination  due to
minute  ozone leaks  which might occur on a periodic  basis.

The  electrical  considerations  for  an  ozone   system  should receive  special
attention.   For example, a  number of problems have  been reported with dry-
type   potential   transformers.      Oil-cooled   transformers    apparently   have
performed  more reliably.   In view  of  the  dependence  of ozone  generation  on
high  frequency   or  high  voltage  electrical   energy,  the   ozone  generator
supplier  should   be  responsible  for  designing  and   supplying  the  electrical
subsystems.   However,  the specifications  should  require  that  the  frequency
and voltage  transformers  be  high  quality  units designed  for ozone   service.
The ozone  generator  supplier  should be  requested  to  provide  a record  of
successful electrical  equipment performance.

Another item  to  consider in  the  design  of an  ozone generating  system is the
power  factor.   This is  the  ratio  between the  apparent power (kw) measured
by  a watt-hour meter  and  actual  power (kVA)  measured in  terms of  voltage
and amperage.  The relationship is shown by the  equation:

         Power Factor  =  Apparent Power/Actual Power

                    pf  =  kW/kVA

The power  factor  is  UNITY when  the  voltage  and current  of  an alternating
current  power supply are "in-phase"  with each other,  for  example  in  a purely
resistive circuit like a heating  element.    In  a  purely capacitive circuit,  such
as  an   ozone generator which  has not reached the  ionization  potential  of the
dielectric,  the  voltage  and  current  are  90  degrees out  of  phase.    In  this
case, the  power  factor  is  ZERO.    For  a  generator producing ozone,  the
voltage  and amperage will  be  somewhere  between  0  and  90 degrees  out  of
phase;  thus, the power factor  will be  less  than  1.0.   The  actual  power factor
will  vary  depending upon  the power  supply  to  the ozone  generator  and the
amount of electrical resistance developed within the electrical  circuit.

The power  factor  may  be corrected  by installing  inductors  in  the  electrical
circuit   or   by  using the  inductance created  by the  operation  of motors  in
other areas of  the treatment  plant.    However,  caution   must   be exercised
when using other  plant  equipment  for power  factor correction,  because  of
the  variable operating conditions   of  the  equipment from  hour  to hour and
from day to day.    The inductance of other equipment  should be  used only to
control   the  low   power  factor   of   the  ozone  generator  when  consistent
equipment operation can  be  assured.

An operating ozone generator  can decrease  the power  factor to 0.3  to  0.5,
depending   on  the   generator  setting.    Corrections   normally will  be  cost-
effective,   since   utilities   that   supply   electrical   power    typically    impose
penalties for a low power factor.
                                       103

-------
Ozone Generators

The most common  commercially available  ozone  generators  can  be classified
as follows:

o  horizontal tube; one electrode water-cooled,

o  vertical tube; one electrode water-cooled,

o  vertical tube; both electrodes cooled,

o  plate;  water- or air-cooled.

The  operating   conditions  of  these  ozone  generators  can  be  subdivided  as
follows:

o  low frequency (50/60 Hz), high voltage (> 20,000 V),

o  medium frequency (600 Hz), medium voltage (< 20,000V),

o  high frequency (> 1,000 Hz), low voltage  (< 10,000 V).

Currently, low  frequency   - high  voltage  units  are most common, but  recent
developments  in electronic  circuitry  are  resulting in  higher  frequency  units
(and lower voltages)  being  used.

To  determine  the  amount  of ozone  required  to  provide  disinfection,  the
following  steps  are recommended:

1. Estimate  the initial  ozone  demand of the water at  the  point  of entry  into
   the  first  chamber  of  the  ozone  contactor.   The  initial  ozone demand  is
   that   quantity of   ozone  necessary  to   develop   a   stable   residual  ozone
   concentration at  the  outlet of the first  contact chamber.   In most  ozone
   disinfection  processes,  the   ozone residual attained  is 0.4  to 0.5 mg/L,  and
   the  initial ozone demand   is between  1  and 6  mg/L, depending upon the
   water  quality at this point.

2. If  the initial  ozone  demand  is  estimated,  assume  a 90% ozone  transfer
   efficiency for the  2-chamber bubble diffuser contactor.

3. An additional amount  of ozone  will be  required in  the  second chamber  of
   the  ozone contactor to  maintain the  residual of ozone for the  appropriate
   period of time to  attain the necessary CT value.   Assume this amount  of
   additional ozone will be 50% of  that applied  in the first chamber.

4.  Determine the  peak water  flow  through  the plant, and from this, calculate
   the peak  ozone requirement.

In operating ozone  treatment  systems, a  single  generator  designed for  peak
use  is  never  installed, in  order to  minimize  energy  costs  during  operation,
provide  backup capability, and  to  facilitate  maintenance   operations.    The
minimum  production rate  from an  ozone generator is 10 to 25  percent of  its

                                       104

-------
maximum  production  capability.   On  the  other hand,  the most cost-effective
production  rate  of  an  ozone  generator  occurs  at 60-70%  of  its maximum
production rate.

Therefore, if the  treatment plant requires,  say,  100 Ibs/day  of  ozone  during
normal  production,  and  150  Ibs/day  for  peak  operation, it is cost-wise  to
purchase three  60  Ibs/day generators,  and  operate  all  three at about  65%  of
their   capacity for  normal  production.    This approach  allows   satisfying the
occasional  peak   ozone  demands,   provides  one  standby  generator   during
normal flows, and allows the  down  generator  to  be maintained  on a  periodic
schedule.
Qgone Contactors

The  different  types  of contactors have  been  discussed  earlier.    Selection of
the  specific type  of  contactor  should  be  based  upon  the  specific  type of
ozonation  task  contemplated.     For disinfection,  the  most  commonly  used
contactor is the two-chamber,  porous plate  or porous  tube diffuser  contactor
shown  in   Figure  5.    Housings  for  such  contactor  systems  normally  are
constructed  of  concrete.
Destruction of Excess Ozone

Ozone  destruction removes excess  ozone in  a contact basin  off-gas prior to
venting or  prior  to  recycle or reuse of  the  exhaust gas.   Safety is  the  major
consideration.     The  maximum  ambient  air  ozone   concentration  currently
allowed by the OSHA for exposure of human beings over an  8-h  working day
is 0.0002  g/m3 (0.1  ppm  by  volume).   This concentration  is  significantly less
that  the   ozone  concentrations  in  contactor exhaust  gases,  normally greater
than  1 g/rn3 (500 ppm by volume).

The  primary  methods  for treating  excess   ozone  in  the exhaust gases  are:
thermal  destruction  (300-350°C  for  3  sec),  thermal/catalyst   destruction,  and
catalyst  destruction   (metal   catalysts  or   metal   oxides).     Moist  granular
activated  carbon  is  used quite extensively in  small  scale European ozonation
systems  generating  g/h   quantities  of  ozone  (<  2  mgd  water supplies;  large
public swimming  pools,  etc.),  but is  not  recommended  (U.S.  EPA,  1986, p.
137) for this purpose.

The most   favored  procedure   currently  involves  passing  the  contactor exhaust
gases   through   a   catalytic   ozone   destruction   unit,   which   contains  a
proprietary catalyst  system based  on  manganese  dioxide.    Excess  ozone  is
converted  to  oxygen,  which     may   be  discharged  safely  to  the  ambient
atmosphere.

Ozone  contactor   exhaust  gases   are  treated  in  this  manner,   rather   than
drying or   recirculating   through  the  ozone  generator  for  economic  reasons.
It is  more cost-effective to dry  ambient  air  than to  clean  and  redry  the wet
air  exiting  the  ozone contactor.


                                       105

-------
Materials of Construction

Care  must  be  taken in selecting  materials of  construction for those  portions
of  the  ozonation  system  in   direct  contact  with  either  "dry"   (before  the
contactor)   or  "wet"  (after  the  contactor)   ozone-containing  gas.     While
reinforced   concrete  is  an  appropriate  material  for  ozone  contactors,  the
ozone-containing gas  piping  system  should  be  304-L  or 316-L  stainless  steel
for dry and wet services, respectively.


Monitoring  the  Ozonation System Operation

Equipment  should  be provided  to  monitor the  operation of the components of
the  system.   The  minimum  degree  of instrumentation, all  of which  can  be
provided by the ozone equipment  supplier as part of  the  package units, is as
follows:

o   Gas pressure and temperature  devices at key  points  in the  air preparation
    system.  Simple pressure gauges and  mercury  thermometers will  suffice.

o   Continuous  monitoring  of  the  dew  point  measuring  device   to  determine
    the  moisture content  of the  dried  air  feed  gas  to the  ozone  generator.
    High dew point indications  should be  designed to sound an alarm and shut
    down  the generator.   Equipment  for  calibration of  the  dew point  monitor
    should be provided  as well.

o   Means  of  measuring  inlet/discharge  temperatures  of the  ozone  generator
    coolant  medium  (water  and/or  oil,  or air)  is required, as is  a  means of
    determining  whether  coolant  is   actually  flowing  through  the  generator.
    An  automatic  system  shutdown   should  be   provided  if  coolant  flow is
    interrupted or  if its discharge pressure exceeds specified limits.

o   A  means  of  measuring   flow  rate,  temperature,  pressure,   and ozone
    concentration  of  the  ozone-containing  gas   discharged  from the ozone
    generator is  required  to determine  the  ozone  production  rate.

o   A means of measuring the  power input  to the ozone generator is required.


Ozone Equipment Suppliers

Table  XXIV is a  listing  of  the  major  suppliers of  corona discharge ozone
generation  equipment  in  the   United  States  (Source:    International  Ozone
Association,  Pan American Committee, Norwalk, CT).   Each  of these suppliers
provides  the total  ozonation   system,  including  electrical  power supply,   air
preparation   equipment,  contacting  apparatus,  and  contactor  off-gas  destruc-
tion  devices,   as  well  as   the   appropriate    monitoring   and  automation
equipment.
                                       106

-------
       TABLE XXIV.  MAJOR  U.S. OZONATION SYSTEMS
                       SUPPLIERS
        Name

  BBC Brown Boveri Inc.


  Capital  Controls Co., Inc.


  Emery Chemicals Inc.


  Gould & Eberhardt Industries


  Griffin Technics Corp.


  Hankin  Environmental Systems


  Infilco Degremont Inc.
  Mitsubishi, Electric Sales,
   MEDAMA,  Inc.

  Ozone Research and Development
  Corporation

  PCI Ozone Corporation
  Trailigaz Ozone of America
  Welsbach Ozone Systems Inc.,
   Division of Polymetrics, Inc.
     Address

1460 Livingston Ave., North
Brunswick, NJ  08902

Box 211, 3000 Advance Lane,
Colmar, PA  18915

4900 Este Avenue, Cincinnati,
OH 45232

P.O. Box 190, Sutton Road,
Webster, MA  01570

178 Route 46, P.O. Box 330,
Lodi, NJ  07644

71 Route 206  South, Somer-
ville, NJ 08876

P.O. Box 29599,  2828
Emerywood Parkway,
Richmond, VA  23229

5757 Plaza Drive, Cypress, CA
90630-0007

3840 North 40th Avenue,
 Phoenix, AZ  85019

One Fail-field Crescent, West
Caldwell, NJ   07006

1 Jenkintown Plaza, 211  West
Avenue, Suite  210, Jenkintown,
PA  19046

101 Nicholson  Lane, San Jose,
CA  95134
                  ix.   Costs of Ozonation Systems

                            Equipment Costs

These  will  include  estimates  for  ozone  generation  equipment  and  ozone
contacting  systems,  both  of  which   are   supplied  by  the  ozone  systems
manufacturers.  Water  supply  systems treating 0.5  mgd and less will require a

                                     107

-------
daily  ozone  generation  capacity from  three to 21  pounds,  and will be  able  to
dose  ozone  at  average  levels of up to 3-5 rag/L.   At  these  production levels,
ozone normally  will  be  generated from  dried  air, not  oxygen,  in  order  to
avoid the costs  of oxygen  generation, recovery  and recycle equipment.

As mentioned  earlier,  the  new  540-mgd  Los  Angeles  drinking  water  plant
uses  oxygen  as  the feed  gas  to  the ozone  generators.   A new German ozone
generator,  currently  marketed  in  the  U.S.  by  Capital   Controls   Co.  (the
CAPOZON  system) is  designed to  generate ozone  from pure  oxygen,  in small
diameter  dielectric  tubes.   Because of  this new design, this system  is  claimed
to  be capable  of  generating  unusually  high concentrations of  ozone  (14%)  in
oxygen.    This  system  may  find  application  for  drinking  water   treatment
plants in the future.

Ozonation  equipment  to be purchased includes the following:

    o     air  preparation equipment  (drying and chilling)
    o     ozone generator
    o     ozone contactor
    o     ozone destruction  unit
    o     instrumentation and controls

For  generation   of  large  quantities   of   ozone  (100  Ibs/day   and   higher),
approximately $1,200  per  pound of ozone generation capacity  per day  wul be
required  to  procure  the air  preparation, ozone  generation,  and  contacting
equipment.   Ozone destruction  and  instrumentations and  controls will  be  in
addition  to this  figure.

For  smaller  quantities  of  ozone,  costs  will   be  higher,  but  will   vary
significantly from site  to site.
For  the   small   production   quantities   of ozone  required  by   small  water
treatment  plants   serving   less  than  10,000  persons per   day  (three   to  21
pounds per  day),   all  items  except the  contactor  can  be  assembled  into  a
single skid-mounted unit.   If  the  contactor  selected  is  the  turbine  type,  it
also can be  included in  the skid-mounted assembly unit.

Diffuser  contactors  for small  ozonation systems  generally  are constructed  of
polyvinyl  chloride  (PVC)  pipe standing  on  end,   or  of  fiberglass  reinforced
plastic  (FRP)   tanks.    A  contact  chamber  containing   diffusers  should  be
approximately  18-ft high,   providing a  water  depth  of  16-ft  and  a  detention
time  of  10-20  minutes,  to attain the appropriate CT value  required  for 99.9%
inactivation   of  Giardia  lamblia cysts   and/or  99.99%  inactivation of  enteric
viruses.   These  conditions  will  maximize  the  transfer  of  ozone  from  gas
phase to aqueous  solution  when employing diffuser contacting systems.

Tables  XXV and  XXVI list  equipment   cost  estimates  obtained  from  two
ozonation systems  suppliers in  1982, for  various daily ozone generation rates.
Ozone supplier  A  (Table  XXV)  provided estimates  for ozone dosages provided
estimates  for ozone dosages  of 3  and  5  mg/L  at  water flow rates of  500,000
gal/day,  350,000 gal/day,  and  180,000  gal/day.    This  breakdown shows  that
equip-ment   costs   for    air   pretreatment   and   ozone   generation   capacity
available from this  supplier depend  upon the dosage required at a particular

                                       108

-------
       TABLE XXV.   COSTS OF OZOHATION EQUIPMENT FOR SMALL MATES SUPPLY SISTER (Company A - May 1982)


Size of water supply                  500.000 gpd               350.000 gpd             180,000 gpd

Maximum dosage of ozone  (mg/L)
at peak flow                          53              53              53

Daily ozone requirement  (Ibs)        21         14             14         7              75
Contact chamber diameter  (14 ft
high, 4 compartments, 4 dif-
fusers, Derakane fiberglass
reinforced plastic)                   6  ft       6 ft           5 ft      5 ft           4 ft     4  ft

EQUIPMENT COSTS

Air preparation + ozone
generation unit
Contact chamber with diff users
Monitoring Instrumentation
$31
$11

,500
,500

$25
11

,000
,500

$25
10

,000
,200

$22
10

,000
,200

$22
9

,000
,900

$19,500
9,900

1) Ozone in generator product
2) Ozone in ambient plant air
3) Ozone dissolved in water
4) Dew point monitor in air
   preparation unit
                                    $15,000     15,000         15,000    15,000         15,000    15,000

Ozone Destruction Unit


TOTAL EQUIPMENT COSTS

Power requirement kWh
$6,700
(10 cfm)
$64,700
13.3
5,000
(7 cfm)
$56,500
10.1
5,000
(7 cfm)
$55,200
10. 1
4,200
(3 cfm)
$51,400
5.0
4,200
(3 cfra)
$51,100
5.0
4,200
(3 cfm)
$48,600
3.65

-------
      TABLE XXVI.   COSTS OF OZONATION EQUIPMENT FOR SMALL WATER SUPPLY  SYSTEMS  (Company  B  - May  1982)
Flow Rate 0.1 mgd 0.2 mgd 0.3 mgd 0.4 mgd
Maximum ozone dosage ,
(mg/L), at peak flow 3333
Daily ozone requirement
(Ibs/day) 3 6 7 12
Equipment Costs LP** HP*** LP HP LP HP LP HP
Air preparation +
ozone generator* - $17,500 §33,200 $30,200 $38,500 $35,500 $43,000 $40,000
Power requirements
(kWh/lb of 03
generated) 10.5 20 10.5 13.5 10., 5 13.5 10.5 13.5
Ozone contactor
with diffusers $8,500 $12,000 $16,000 $21,000
Ozone Monitors $4,000 $4,000 $4,000 $4,000
-f 0$ generation**** or 2,000 or 2,000 or 2,000 or 2,000
- O3 room or contact
chamber exhaust 2,200 2,200 2,200 2,200
- dew point 3,500 3,500 3,500 3,500
0. 5 mgd
3
14
LP HP
$49,800 $46,800
10.5 13.5
$29,000
$4,000
or 2,000
2,200
3,500
TOTAL EQUIPMENT
  COSTS         -        $33,700-  $54,900-  $51,900-  $64,200-  $61,200-  $73,700-  $70,700-  $88,500-  $85,500-
                          35,700    52,900    49,900    62,200    59,200    71,700    68,700    86,500    83,500

   * includes air preparation, ozone generation, ozone destruction and system controls.
  ** air preparation unit includes air filters or separators, compressor delivering air at 8-12 psig to a
     refrigerative cooler and a dual tower desiccant dryer.
 *** same as LP air preparation system, except compressor delivers air at 80-120 psig.   HP system takes less
     space requires less maintenance, but requires more energy.
**** $4,000 instrument IB an automatic, continuous reading in-line aonitor.
     $2,000 instrument is not automatic and utilizes wet chemistry.

-------
water  flow  rate.    In addition,  the  size  (and  cost)  of the  ozone  destruction
units  required  also  varies,  as  does  the  power requirement to  operate  the
total ozonation  system.

Ozone Supplier  A  can  provide  four monitors with  his  system.    All  are
optional,  but a!3  are  recommended  for  optimal  performance  and minimal labor
and downtime.  These will monitor:

o  the dew  point in the air preparation unit,
o  ozone output of the generator,
o  ozone in the plant ambient air (in  case  of leaks),
o  dissolved ozone  residual in the water.

The cost  of these  four monitors is  constant at  $15,000,  regardless of  system
size in the  range  shown in Table XXVI (5  to 21 pounds per  day).

Table   XXVI   shows  similar  data  for Ozone  Supplier  B.    In  this  case,
equipment  costs  are  presented  for  water  flows  of  100,000, 200,000,  300,000,
400,000, and 500,000  gal/day.   Average ozone  dosages are  taken as  3 mg/L,
and  the   daily  ozone output   required  varies  from   3   Ibs/day  for   treating
100S000 gal/day  to 15 ibs/day for treating 500,000  gal/day.

Ozone Supplier  B   offers  two types   of  air  preparation   equipment,  however,
and estimates   are  presented  for each.  One  type  operates  at  high pressures
(80-120 psig),  and  the other  at low pressures  (8-12  psig).   The  high  pressure
air  treatment  units are lower  in  capital   cost,  but  require  more energy  for
their operation.

Ozone Supplier  B  does  not   normally  provide  a  residual  dissolved  ozone
monitor,  but   offers  two  types  of  monitor   for  ozone  output  from   the
generator.    The automatic,  m-line  continuous   reading  monitor  costs $4,000;
the  non-automatic  monitor requires wet  chemistry  determinations  to  .develop
data  at some  period of  time  after   the  sample  has  been  taken,  and costs
Therefore,  cost  data  presented  in  Table  XXVI  vary  by   the   differences
between  costs  for  high  and  low  pressure  air  preparation  equipment,  and  by
the costs of the two ozone generator monitors.
In  1987,  Ozone Supplier  A  estimated  that  the cost  for  a  5-lb/day ozone
generator, sufficient  to  treat 180,000 gal/day at an  applied  ozone  dosage of 3
mg/L,   to  be   $20,500.     This  confirms  that  costs  for  ozone   generation
equipment in  1983 are still current, for the smaller water treatment plants.


                              Installation  Costs

Costs  for Installation of  the  ozonation  equipment  include  labor and  material
costs  for piping water  to and  from  the ozone generators  (if  they are  water
cooled),  for piping   ozone-containing air to the contactor  chamber,  for piping
water  to and  from  the  contactor,  and  for piping  contactor off-gases  to and
from   the  ozone  destruction  unit.     Electrical  wiring   costs  also  must   be
considered  in  these  costs.    Ozonation  equipment  suppliers contacted advise

                                        111

-------
that  for  production  of up  to  about  30  Ibs/day  of  ozone,  installation costs
will  be  roughly the  same, and will  average  15%  to  25%  of  the  equipment
costs of the  largest  units  estimated in  Tables XXV and XXVI.   The actual
figures for the two  equipment suppliers then  become:

      Supplier            Cost of Equipment     Installation Cost
      	           (500.000 gpd plant)        (15%  to 25%)

      A                      $64,700           $ 9,705 - $16,175

      B                      $85,000           $12,750 - $21,250


                                Housing  Costs

The  power  supply,  air preparation  equipment,  ozone  generation equipment,
and  turbine  contacting  units  can  be installed  rela-tively easily,  in  areas  on
the  order  of  10 x 17  feet.    However, diffuser contacting  units  are tall  (18
feet) and  bulky, and normally  are installed  outside existing  buildings  (above
ground)  or  underground  inside buildings  being  constructed.    Alternatively,  a
170 ft2  Butler  building can house the ozonation  system,  except for  above-
ground  diffuser unit.  Such a building costs about $6,000.


                    Construction Costs for Ozone Contactors

Construction  costs  for  ozone  contactors are  based upon  concrete  contactors
which use  serpentine   flow  in  an  upflow/downflow   configuration.     Ozone
dissolution   equipment is located at  the bottom of  the downflow  pass,  thus
providing  countercurrent  flow.    The   conctruction costs  include  the  ozone
dissolution   equipment and manways which can be  used  for  access  to  the
interior  of the  contactor.   Figure  24 shows  the  construction  costs  for ozone
contactors  having volumes between  2,000 and 18,000  gal  (Gumerman  et  al.,
1986).


          Construction Costs for Ozone Generation and Feed  Systems

Construction  costs  for  ozone  generation and   feed systems using  air  as  the
feed  gas  include  air   preparation  equipment,  the  ozone  generator,  ozone
destruction  equipment for contactor  off-gases, and a  building  to house  the
equipment.    For oxygen feed  installations,  the construction costs include  the
ozone  generator,   ozone  off-gas  destruction   equipment,  and   the   building.
Ozone  generators  and   the  contactors  are  housed  in  a  sealed room,  which
contains  an  ozone  leak  detector.    In the  event  of  an ozone  leak,   this
detector  will  activate an alarm  and  the building  ventilation system,  and  shut
down the ozone  generator.

Piping is included  for  the ozone generator cooling  water,  ozone, and water
to and  from  the  ozone contactor.    Off-gases from  the ozone  contactor  are
destroyed using a combination thermal/catalytic  destruct device.


                                       112

-------
   CO
   O
   O


   O

   u

   c

   (0

   O
   O
1(
i
6
5
4
3
2
1
6
8
4
3
2
1(
\
9
4
3
2
1C
9
4
3
2
1<
I
1
),000,(






000, 0<





30.000







).000







)00
)00






00


































































































































































i





























**>




























MMM




























*




























*






\



r-
_











!
J,














^





-i






^1





!
I
i



I
i




J_J—




















J-^^
*> ^




































































1




































































































































2 349 9789 2 3 4 9 9 789 2 3 4 9 9 789
00 1000 10.000 100.000
CONTACTOH VOLUME • gal
1 	 » ' 	 '1 • '
1 10 100
                        CONTACTOR VOLUME
Figure 24.    Construction cost for ozone contactors (Gumerman et al, 1986).
                                    113

-------
The building  ventilation  system  consists of  an  intake  louver  and  motorized
damper  on  one  side  of  the  building,  and  a  fan,  ductwork,  and  motorized
damper  on the opposite side of the  room.  This system  is designed  to  change
the  room  air once  every  two  minutes,  when  activated.    The   ventilation
system   can   be  activated   by   the  ozone  detector,  a  switch  activated  by
opening  the  building door,  or  a  switch  located  on the  building  exterior.   A
shutdown  switch  for  the   ozone  generator is  also  located  on  the  building
exterior.

Construction  costs are  presented in  Figure 25 for ozone generation and feed
systems (Gumerman  et  al.,  1986).


                      Operation and Maintenance Costs

Operating  costs  for  ozonation  systems vary,  and depend  upon  a number of
factors:

o   method of air preparation,
o   method of cooling the generator (water or air),
o   if water-cooled, the amount of refrigeration required for cooling water,
o   method of contacting,
o   dosage  of ozone required,
o   pumping  of generator coolant,
o   method of contactor off-gas destruction.


Air  Preparation:    High pressure  versus low  pressure,  versus  sub-atmospheric
pressure  desiccant system with or without addition  of air chiller.


Ozone  Generator  Cooling:    Air  versus water.   If  water,   the  amount  of
cooling  required.     In  northern  climates,   water   produced  at  the  plant
generally  is  cold enough  to serve as  the generator  coolant  the year  round.
In  southern   climates, generator  cooling  water must  be  refrigerated most, if
not all, of the year.


Method   of   Contacting:      Diffuser   contactors   require   no  added  energy.
Ozone/air  mixtures  normally  are  generated   under  a  sufficient  pressure  to
overcome  the  head  of  16 feet  of water.     On  the   other hand,  turbine
diffusers  require  energy for their operation,  but  take  up  much  less  space
than   diffuser  contactors.      For   disinfection   with   ozone,  however,   the
appropriate  reaction  time  must be  provided  after initial  contacting  has been
achieved to  attain  the  appropriate CT value to  assure  inactivation  of  Giardia
lamblia cysts and/or enteric  viruses.


Contactor   Off-Gas   Destruction:      Thermal   versus   catalytic   destruction.
Operating costs of these techniques vary.
                                        114

-------
        JOO.OQO.OOO
                         MAXIMUM OZONE FEED RATE - Ib/day
                1.0
10
                                                       100
                         MAXIMUM OZONE FEED RATE • kg/day
Figure  25.    Construction  cost   for   ozone  generation   and  feed  systems
             (Gumerman et al.,  1986).
                                     115

-------
Maintenance  material  requirements  are  for  periodic  equipment  repair  and
replacement  of  parts.    Air  preparation  systems  contain  air  prefilters  which
must  be  replaced frequently.   Tube-type  ozone  generators  normally  are  shut
down once per year for cleaning of the tubes  and other general maintenance.
This  can  require  several  man-days  of  time, depending  upon  the  number of
ozone generators  in  the system.   Spare parts normally consist  of  replacement
tubes, which  can  be broken during  cleaning, or  which can deteriorate  after
years of  operation  at  high voltages,  or more rapidly  if the air is improperly
treated.

Labor  requirements  are   for   periodic  cleaning   of   the   ozone  generation
apparatus,  annual  maintenance  of  the  contacting   basins,  and  day-to-day
operation of the generating  equipment (average 0.5  h/day).

Operating and  maintenance costs for equipment of Ozone Suppliers  A and B
are  summarized  in  Table   XXVII.   Also  included are  building heating  costs
(which  are taken to be the same up to 0.5 mgd) and  costs for maintenance
materials  and  O&M  labor.   There  are no  chemicals  costs related  to  ozone
generation,  except  for  periodic  changing  of  desiccant  in  air  preparation
systems (normally after 10 years of use).

Electrical  energy  is  a  major  component of  operating  costs, representing  26%
to  43% of total  O&M  costs at  small  plants (0.1 mgd)  increasing to  59% to
65%  at the larger plants (0.5 mgd).

Building  energy costs  (which  are approximately  the  same  for  all  small  size
plants  up  to  0.5  mgd)   include   energy  costs  for  heating,  lighting,   and
ventilation.   Labor  costs  (which are  independent  of  the  plant sizes  listed-
0.1 to  0.5  mgd)  account for 54%  to 70%  of  total O&M  costs at  the small
plants, but only 30% to 36% at the 0.5  mgd  plant.

Figure   26  shows  operation   and   maintenance  requirements  for  ozone
generation,  feed,  and  contacting systems  ~  building  energy,  process energy,
and  maintenance  material  for   systems  feeding  5  to  250   Ibs/day  of  ozone.
Figure  27 shows  O&M requirements for  labor  and  total O&M cost.   Both
figures are taken  from  Gumerman et al. (1986).


          Summary Statement Regarding  Costs for Ozonation Systems

Because  of  the  many  differences  in  methods  of  air  pretreatment, ozone
contacting,  contactor  off-gas   destruction,  monitoring,  and   other  operational
parameters, equipment  costs given  above  should  not  be considered  as more
than  general guidelines.  Vendor  quotes should  be obtained at the time ozone
is  being considered by the small water supply system.

It  should  be  noted  that vendor  quotes obtained  for  estimating purposes  are
likely to  be  somewhat  higher  than  firm bids made to specifications.   This is
because  the  market  for ozonation systems  currently is quite competitive,  and
suppliers  usually  bid  their  best prices when  responding  to  clear   specifica-
tions.
                                       116

-------
TABLE XXVII.  OPERATING AND MAINTENANCE COSTS FOR SMALL OZONE SYSTEMS APPLYING  3 mg/L DOSAGE
Water Flow
Rate (ngd)
Supplier A
0.18
0.35
0.50
Supplier B
0.10
0.20
0.30
0.40
0.50
Electrical
Building*
6,
6,
6,
(High
6,
6,
6,
6,
6,
570
570
570
Pressure
570
570
570
570
570
Energy
Process
6,661
12,775
51,611
(kWh/yr)
Total x
13,231
19,345
58,181
Air Preparation)
21,900 28,470
29,565
34,493
59,130
68,985
36,135
41,063
65,700
75,555
$0.
$
$1
$4
$1
$2
$2
$4
$5
07/kWh -
926
,354
,073
,993
,529
,874
,599
,289
Maintenance
Material
$120
200
300
$120
120
200
250
300
Labor x
(hrs/yr)
250
250
250
250
250
250
250
250
$10/hr -
$2
2
2
2
2
2
2
2
,500
,500
,500
,500
,500
,500
,500
,500
Total Cost
($/yr)
$ 3,546
4,054
6,873
$ 4,613
5,149
5,574
7,349
8,089
      Estimated  from data of Hansen et  al.  (1979)

-------
     100.000    10.000.000
                                           10                  100

                                AVERAGE OZONE FEED RATE • kg/day
Figure 26.    Operation  and  maintenance requirements for ozone generation,
             feed,   and   contacting  systems  --  building   energy,   process
             energy, and maintenance material (Gumerman  et al., 1986).
                                      118

-------
1,000,000 1.000.000
"|F^ 	 ai i :£=FF3=
a
9
4
3
2
1(
9
6
9
4
i 3
* *
ta
§ 1C
< 4
H
O 3
K
2
11
T
9
4
3
2
1C
I 8
6
9
4
3
2
10.000 1(
6
9
4
3
2
»t
>.
1,000 £ 1(
^ CC *
"* O








W







—
«
mm




LJ
sS













/
^
rfi
^








^
uan










T01
t



jg
*
1"
*









V




EX



^'
?'
\









*T


































PAL O&M COSTS
CLUOING CHEMICALS]
I 1
OXY(
3.S kw
M^^tf—

ff
* 8

3



r=
IR F
kwl
3£f
h/l
EE
i/IE
4 F
I
D
i C
:E
3;
)Z
H
'C
-
O
D
>r
h
N
4E
f
E"
IR FEED_ ' |-
kwh/ib OZONE





f
I

















ABOR













































1



































1



FIELD DATA LEGEND
A « LABOR - STRASBURG
2 3493 7S9 2 3490 789 2 3496789
10 100 1000
AVERAGE OZONE FEED RATE • Ib/day
I . • • , . , o j r i ,.,,,, . . .
1.0 10 100
                           AVERAGE OZONE FEED RATE • kg/day
Figure  27.
Operation and maintenance  requirements  for  ozone  generation
feed,  and  contacting  system  --  labor  and  total  O&M  cost
(Gumerman et al, 1986).

                        119

-------
              c.    Ultraviolet Radiation

Much  of this discussion is  taken from U.S. EPA,  1983, and 1986.

The effectiveness  of  UV  radiation  as  a  bactericide  and  virucide  has  been
well  established.    It  is   a  physical   disinfecting  agent   compared  to  the
chemical  disinfectants,  chlorine, chlorine dioxide, ozone,  and  monochloramine.
Radiation  at a wavelength  of 254 nm  penetrates  the  cell wall and is  absorbed
by  the  cellular  nucleic  acids.   This can prevent  replication and cause  death
of  the cell.   Since UV radiation is  not  a chemical  agent, no  toxic  residual
can be  produced.   Although  certain chemical  compounds  can  be  altered  by
the  radiation, the energy  levels used  for disinfection are  far too low  for this
to be  a  significant cause for concern.

Major advantages of  UV  radiation are its  simplicity,  lack  of  impact on the
environment  and  aquatic  life,  and  minimal  space  requirements.   Required
contact  times are  very short,  on  the  order pi  seconds rather  than minutes.
The equipment  is  simple  to  operate  and  maintain, but  fouling  of  the quartz
sleeves  or  Teflon  tubes  must  be  dealt with  on  a  regular basis.    Fouling
normally   is  handled   by  mechanical,  sonic,  or   chemical  cleaning.     High
concentrations  of  suspended  solids,   color,  turbidity,   and  soluble  organic
matter in  the water  can react with or  absorb the  UV  radiation,  reducing the
disinfection performance.                                                    <


                    i.     General Description  of the UV Process

Disinfection   by  UV   radiation  relies  on the  transference  of  electromagnetic
energy from a  source  (lamp)  to organism cellular  material (specifically the
cell's  genetic  material).    The  lethal  effects  of this  energy  result  primarily
from the  cell's  inability to  replicate.   The  effectiveness of  this radiation  is  a
direct  function  of the  quantity of  energy,  or dose,  absorbed by  the  organism.
This   dose  is  described  by  the   product  of  the   rate  at  which  energy   is
delivered,  or  intensity,  and  the  time   to  which  the organism  is  exposed  to
this intensity.

The primary artificial  source  of UV energy,  at  present,  is  the  low pressure
mercury  arc lamp.    It  is  almost universally  accepted  as  the  most  efficient
and  effective  source   for  disinfection  systems  application.     The   primary
reason  for  its  acceptance  is  that  approximately 85%  of  its energy  output  is
nearly  monochromatic  at  the  wavelength  of  253.7 nm, which  is within the
optimum  wavelength   range  of  250 to  270 nm  for  germicidal  effects.   The
lamps  are long  [standard lengths are typically 0.75  and  1.5 m (2,5 and 4.9  ft)
arc  lengths]  thin  tubes [typically  1.5  to  2  cm  (0.6  to  0.8  in)  in diameter].
The radiation is  generated  by striking  an  electric arc  through  mercury vapor;
discharge  of the  energy  generated  by  excitation  of  the  mercury results  in
the emission of the UV radiation.

The   UV    demand   of   a  water   is   quantified   by   a   spectrophotometric
measurement  at  the   key  wavelength  of  253.7  nm;  this   expresses   the
absorption  (or  transmittance)  of  energy  per  unit  depth.    The   output   is


                                        120

-------
absorbance   units/cm,   or  a.u./cm.     The  percent   transmittance   can   be
determined  from this unit by the expression:

         %  Transmittance =  100 x 10-(a-u-/cm)

The  term  most  often  used  for   design  purposes   is  the  UV  absorbance
coefficient, a, expressed in base e:

    UV absorbance coefficient, a  = 2.3 (a.u./cm)

The unit for a  is cm'l.

A  second major concern  is  the provision  of adequate exposure  time  to  the
microorganisms  in order  to  meet the  dose  requirement at  a given  intensity.
The  key  is  to  have  plug   flow  through  the  system, such that  each  flow
element resides in  the  reactor  for  the same amount  of  time.    Perfect  plug
flow  is  not  possible  to achieve,  of course.  Some dispersion will exist,  such
that  there  will   be   a   distribution  of   exposure  times   about  the  ideal,
theoretical   exposure   time.      A   design   objective   is   to  minimize   this
distribution.

The  basic premise  to  understand is that the UV  radiation  must  be  absorbed
before  it  can  have  a  disinfecting   effect.    Photochemical   damage occurs  to
the  deoxyribonucleic  acid (DNA)  macromolecules  which  interferes  with  the
ability of the  cells  to  replicate.   Cell  death  following  UV radiation  is almost
entirely attributable  to  the  photochemical damage of these compounds.

On  the other  hand,  if the  amount of  radiation received by the  organism is
not  a  lethal  dose,  but only  damaging,  photoenzymatic repair can occur,  and
the  effects  of  the UV  radiation  can be reversed.   The phenomenon  has been
termed  "photoreactivation".     This  repair  mechanism  is   unique   to   UV
radiation, but  is  not universal, and  there is  no  clearly defined  delineation of
characteristics  which  suggest  which  species would  have  the  ability  to repair
and which would  not.

The  enzymatic  mechanism  generally   involved  in  photoreactivation   requires
subsequent  (or  concurrent) exposure to  light  at wavelengths  between  300  and
500 nm.   Such  light   is available  in  sunlight  and in  most  incandescent  and
fluorescent light sources.

Therefore, to  assure that disinfection  occurs with no chance for  photoreac-
tivation,  attention  must be paid to  designing  the  UV disinfecting system to
provide  sufficient UV  dose for an  appropriate  period  of time.   As  a general
rule  of thumb  (U.S.   EPA,  1986,   p.  215),  if  a  3-log reduction  of micro-
organisms is required  to meet the  disinfection  criteria, the  system should  be
designed  to  provide   a  4-log   reduction,   to   account  for  the  effect  of
photoreactivation.

Table  XXVII lists some of the  major  suppliers of UV disinfection equipment
for water treatment systems.
                                       121

-------
  TABLE XXVII.  MAJOR SUPPLIERS OF UV DISINFECTION
  ^	EQUIPMENT       	


        Name                          Address

  Aquafine Corp.                  25230 W, Ave Stanford,  Valencia,
                                  CA  91355

  Aquionics Inc.                   21 Kenton Lands Road,  Box 18395,
                                  Erlanger,  KY  41018

  Atlantic Ultraviolet              250 Fehr  Way, Bay  Shore, NY
                                  11706

  Ultra Dynamics  Corp.           1631 Tenth St., Santa Monica, CA
                                  90404

  Ultraviolet Purifications          299 Adams St., Bedford  Hills, NY
                                  10507

  Ultraviolet Technology Inc.       8930 Osage Ave., Sacramento, CA
                                  95828
                   ii.    UV Disinfection  System Design

In  all,  the  design  of  a  UV  system  must  accommodate  a  few  simple
considerations:

o  satisfy the  UV demand of the water,

o  maximize the  use of the UV  energy being delivered by the lamps,  and

o  provide the conditions which encourage plug flow.

UV  lamps can be suspended outside  the liquid to be treated  or  submerged  in
the  liquid.   In either  design,  the intent is  to get  the  energy into  the liquid
as efficiently  as  possible.   Typically,  if  the  lamp  is  to  be submerged  in the
liquid,  it  is inserted into a quartz sleeve to  minimize  the cooling  effects  of
the water.  Lamps can be  placed  in  the liquid with the lamp perpendicular  to
the  direction   o!  water  flow.     Other  configurations  may  have  the  lamp
parallel to the flow,  or the lamp  may  be suspended  above the flowing liquid.
As the lamp  emits radiation, the  intensity will attenuate as the  distance from
the  lamp increases;  this  is simply due  to  the  dissipation  or  dilution  of the
energy  as  the   volume  that  it occupies increases.    A  second  attenuation
mechanism  involves  the  actual  absorption  of the energy by  chemical  con-
stituents  contained in the  water.   This, analogous to  the chlorine  or ozone
demands,  is  the "UV demand" of the water.


                                      122

-------
The  key points to  be addressed  when  evaluating  or specifying  the  design of
a  UV reactor  are  as follows:


1.  Residence  Time  Distribution  fRTD).    This  should  be  constructed  at  a
number  of flow conditions  for  an existing system.   It should  also  be required
when  specifying commercial systems.   The  RTD  provides  key information  on
the actual or anticipated hydraulic behavior of a reactor.


2.  Plug  Flow.   As a guideline,  an aspect  ratio  greater  than  15 should  be
incorporated  into  the  reactor  design.    Maintenance of  plug  flow  within  a
reactor  will  be  influenced  by the  approach  and  exit conditions.   The design
should  have  minimal  disturbances  at  the inlet  and exit  planes  of  the  lamp
battery.   Directional changes in the  flowpath would best  be  made outside of
the lamp battery.


3.  Dispersion  Number:   A key  goal is  to  minimize  the  dispersion number, d.
A  design  goal should  be to  have  d between 0.02  and  0.05.   This  would  be
representative  of a  plug  flow reactor with low  to moderate dispersion.'  This
can be  accomplished  by  increasing the  product  of  ux (u  = velocity  of water
as  it travels   through the  reactor,  in  cm/sec;  x  =  the  average  distance
traveled  by  an  element  of water while  under  direct exposure to  UV  radia-
tion,  in  cm), even  in a system with a  relatively high dispersion coefficient.

The  designer  should  be  aware,  however,  that  extended  lengths  and  higher
velocities  will   cause   higher  head  losses.     In  certain   situations,   some
adjustment  of  the  dispersion number  may  be  necessary  in  order  to  meet
specific head loss requirements.


4.  Turbulence:    Radial   turbulence  is  necessary  due  to  the   non-uniform
intensity  field.    The  reactor  design should  induce  an  estimated  Reynold's
Number   greater  than  6,000  at  minimum  flow.    If  possible,  it would  be
beneficial  to   confirm  the  laminar/turbulent  flow  transition  velocity  by  direct
head  loss measurements on  the lamp battery.


5.  Head Loss:  Direct measurements should  be  required  for full-scale modules
or  scaleable   pilot  units  as  part   of   commercial  equipment  specifications.
These should  be  determined over  a wide velocity range  and should exclude
entrance and exit losses.
6. Effective  Volume:   Maximal use  of the  reactor  lamp battery is essential
to  keep  the  process  cost-effective.    This  will  be  related  directly  to  the
reactor's  inlet  and  outlet  design.    The   goal  must  be  to  have  equivalent
velocities  at  all  points  upon  entering and  upon  exiting  the  lamp batteries.
Stilling  walls  (perforated  baffles),  and  weirs  should  be  incorporated   into
reactor   designs  to  assure   this.     Guidelines   for   specifying  commercial


                                       123

-------
equipment  should  require  the  ratio  0/T to  be  greater  than  0.9 and/or  the
ratio  of  tp/T  to be greater  than  0.9.


                    iii.   Estimating the Average Intensity in a UV Reactor

The second element  of UV  dose,  after  time,  is the intensity of  energy during
the  exposure  time.   Intensity is the  rate  (or flux)  of delivery  of photons to
the  target.   In  UV process  design  models,  the  rate  of  bacterial mactivation
is  described   as   a  function  of   the  intensity.     By  this   fact,  it  becomes
important   to  be  able  to  quantify  the  intensity in  a  given  system.    The
intensity in  a reactor  is a function  of  the  UV  source (output), the  physical
arrangement  of  the  source  relative  to  the  water  (the  arrangement  of  the
lamps and their  placement  in  or  out  of  the liquid),  and  the  energy  sinks
present  which will attenuate  the  source  output  before it  can  be  utilized  for
disinfection processes.

Lamps   used  in   UV    disinfection  systems  generally   have   lengths   of
approximately  0.9  and  1.6  m.    The  arc   length  defines   the  active,  light-
emitting portion  of  the  lamp  [0.75  m  and   1.5  m  (2.5  and  4.9  ft.,  resp.),
respectively].   The  diameter  of the  lamp  is   small,  typically  1.5 and 1.9..  cm.
The  lamp  envelope  is made  of fused quartz or  other  highly  transparent (to
the 253.7 nm wavelength) glass material, such as Vycor.                     ;

In  the  quartz systems, the  individual lamps  are sheathed   in  quartz  sleeves
only  slightly   larger  in  diameter   (2.3 cm)   than  the  lamp,  and  the  entire
lamp/quartz bundle is submerged in  the  flowing liquid.   In systems where  the
water  does  not  contact  the  quartz  or  lamp  surface,  separate  conduits  carry
the  waters.    The  conduits are translucent  to  the UV  light, with  the lamps
placed near the outside conduit wall.


                    iv.   Water Quality Considerations  in  the Design of a  UV
                         Disinfection System

Without  question,   a  major   element   in  the   effective   design  of  a  UV
disinfection   system   is    a   clear    and    concise    understanding   of    the
characteristics  of  the water to  be  treated.    These  are directly  related to  the
degree  of  pretreatment the  material  will receive  before  the  disinfection  step
and  will   affect   the  sizing  and   performance  of  the  disinfection  system.
Necessary   pretreatment  ranges  from  very  minimal,  as  in   the  case  of  a
groundwater  which   requires  no   treatment   other  than   disinfection,  to  full
conventional treatment of a polluted surface water.

The  four  water  quality   parameters   which  most  affect   the  design   or
performance   of   a   UV   disinfection   system  are   the   flow   rate,   initial
microorganism  density,   suspended   solids    (or   turbidity),   and   the   UV
absorbance of the water at  the point of application of UV radiation.
                            i

1. Flow  Rate:  The   flow   rate  is  set  by   design  of  the  main  plant   and
projections  of  the  hydraulic  load  to the  plant.    In  evaluating  the  design

                                        124

-------
requirements  for the  disinfection  process, some  consideration  should be  given
to  the  equalization   effects  of  the  treatment  processes  before  disinfection.
This can  have an effect  on the sizing of the  UV system.

Flow estimates should be for the design year  of the plant.  There should  be
some  knowledge  of the  progressive  increase  in  the flows  through  the  design
life  of the plant  in  order  to  determine  if  the  system  can be  phased in  by
the  addition  of modules  as the  demand increases.   Some consideration  also
should  be given to  the hydraulic load to the  unit.

For disinfection,  average  flows are  not  critical  to the  design  sizing;  rather
they   are   important  to  estimating   average  utilization  of   the  system  for
operation  and  maintenance  needs.    Peak  flows  should be  used  for  sizing,
particularly reflecting diurnal variations.


2.  Initial Microorganism  Density:  The  performance  of a UV disinfection  unit
is related directly  to  the  initial density of  the  indicator organisms.   This is
not   a  parameter  which  is   generally  monitored  at  a  treatment  plant,
particularly one already  employing  chlorine  for  disinfection.   In  the case  of
disinfection by  UV, however, it  is  critical.   Performance is given  by the  log
of  the  survival  ratio,  N/Ng,  or  by  the  number  of  "logs   the  density  is
reduced.
3.  Suspended   Solids   (Turbidity);     From   the  development   of   the  ,UV
disinfection  model,   it   is  clear  that  the   occlusion  of   bacteria   in   the
particulates will have  a  significant  effect on the  design  of a  UV system.   It
is recommended  (U.S.  EPA,  1986) that the  turbidity  measurement be used as
the primary indicator to quantify these particulates.


4.  UV Absorbance:    The  one parameter which  is solely in  the venue of  UV
disinfection is  the "UV  demand   of the water.   Some  organic and inorganic
compounds in  the  water  may absorb  energy  at the  253.7  nm wavelength.
This  absorbance  will  affect the  intensity  of the  radiation within  the  reactor;
in  specific  design situations,  the  level  of absorbance  will  affect  the sizing of
the system and possibly  the  configuration  (spacing)  of the lamps.    The  final
product  of these  calculations  is  the  average  nominal intensity  as   a function
of the UV absorbance coefficient.
                   v.    System  Design  and  O&M  Considerations for the  UV
                         Process

The design basis and the process elements which  are key  to  the design of  a
UV system are:

o   the hydraulic behavior of the  unit,

o   calculation of the intensity in  the  reactor, and


                                        125

-------
o  generation of the appropriate water characterization data.

In this section  are presented  other  peripheral  topics which the designer  (and
operator)  must  consider.     These   factors  will   affect  the  operation  and
maintenance  of the  system,   and  the  overall  economics.   Specifically,  these
factors  include  the elements  which  affect  the  reactor  intensity (lamp output,
lamp aging, and  attenuation  of intensity  due  to  fouling  of reactor surfaces),
and  methods  for monitoring for lamp  aging and  unit fouling.


                 Factors Affecting UV Intensity  in a Reactor

UV  Lamp  Output:   Output  at  any  given  time will be influenced  by  lamp
temperature  and   by   the   voltage   potential  applied   across   the   lamp.
Additionally,  output  at  the  resonant  frequency will  always  degrade with  time
of  operation  due  to  any  number  of  "aging" factors.    In  the  submerged
systems,  it  is  not  practical under  most  design  conditions  to control  the  lamp
temperature.    In  the  non-contact  systems,  such as  the  tubular  arrays,   it  is
possible   to  maintain  the  lamps  at   their  optimum  wall  temperature   by
controlling  the  temperature  of the ambient air surrounding the  lamps.   This
is being  practiced  currently  in commercial  applications.    Heat  given off.  by
the  lamp  ballasts  is  circulated into   the  lamp  reactor in  cases  where heat is
required;  otherwise  fans  vent the  reactor  with  cooler  outside  air.    These
operations are controlled thermostatically.


Voltage:   Radiance  is  a function  of the  arc  current.   This can  be  exploited
by   adjusting  the  voltage,   in  order  to  vary   the  output  of  the   lamp.
Decreasing  the  voltage will  result  in  a decrease  in  the  current.   Such  a
control  mechanism  has  been installed  at  full-scale  facilities  as  an energy
conservation  measure.   During periods  of  low UV demand, the  lamps are
"dimmed" by slowly  turning  the lamp supply voltage  down.   This results  in a
reduction in  the  power draw  of the  lamp.   Generally, the  lamp  intensity can
be  reduced  to  levels no  less  than 50% before the lamp  current  becomes too
low  and the lamp will begin to flicker and eventually turn  off.


Lamp Aginp:   A number  of factors  combine  to  effectively age  a lamp and
limit its useful life.   These  include  failure  of  the   electrodes,  plating of the
mercury to  the interior lamp  wall (blackening), and solarization  of  the  lamp
enclosure  material  (reducing   its  transmissibility).   These  all  cause   a  steady
deterioration  in  the  lamp's output  at the 253.7 nm  wavelength, such  that  its
output  at  the end  of the lamp s life  can be 40  to 60  percent  of its nominal
output.                                                                    ,

The   output  of the  lamp  throughout  its  life is  affected primarily  by  the
extent of blackening  and solarization  of  the  glass  tube;  the actual life of the
lamp is  governed  by  the  condition  of  the  electrodes.   Germicidal  lamps are
typically  of  the hot  cathode,  type;    these  will  deteriorate progressively  with
increasing  number  of  starts.    Thus,  the  lamp  life expectancy  generally  is
rated according  to  the number of times  the  lamp  is started, or the burning
cycle.   The  lamp  life  normally  cited  by most manufacturers is  7,500 hours,

                                        126

-------
based on  a  burning cycle  of eight hours.   The  average UV output  at this
point is estimated to be  70%  of the lamp output  at  100 hours (note  that the
nominal output  of the low  pressure mercury  arc  lamps  represents  its output
after a 100 hour  "burn-in" period).


                        Monitoring the  Lamp Intensity

The  procedure for  monitoring  the  lamp  intensity  is  by comparison with  that
of  a new lamp,  but  after  the  100 hour  burn-in  time.   The first step  is to
measure the  intensity, at a  fixed  distance, of three to five  new  lamps which
have been  burned  for   about   100 hours.   The  average  of the  five  then
becomes the  benchmark to determine  the  relative  output of the  lamps in use
(percent of  new  lamp  average).   Each  lamp  should  be tagged  and  given  an
I.D.  number; this allows  direct  monitoring  of  individual  lamps and allows the
operator to  keep an appropriate mix of lamps in a system and to know when
to discard  a lamp.

The  same procedure is  used to monitor the  transmittance of  a  quartz sleeve.
In  this case,  a single  lamp is used.   First the intensity  is measured  with  and
without  a  new,   clean   quartz  sleeve  in place  over  the   lamp.     Similar
measurements  then  are taken  with  the unit's quartz sleeves  and  compared to
the  transmittance of the  new quartz.  This can  be done  before  and  after \ the
quartz is cleaned.


                   vi.   System    Design    Considerations   for    Effective
                        Maintenance

An  overriding  concern  in  the  proper  maintenance  of the  UV reactor for
effective  performance  is  to  keep   all surfaces  through  which  the  radiation
must pass  as clean  as  possible.   The  effects  of surface  fouling  on  energy
utilization  efficiency  are  critical,  and  very often  can be  pointed  to  as  the
reason  for  non-performance  of  a  particular  system.    Other  concerns  relate
primarily to  the  accessibility  to UV reactors and to keeping  adequate  records
to control replacement cycles and maintenance  schedules.


Reactor Maintenance

The  most  reliable method  to determine   if a  reactor  is  becoming  dirty  and
requires  cleaning  is by visual  inspection.   The  unit  should  be  drained  and the
surfaces  observed for fouling.    In open   systems, this  can  be  done rather
conveniently  and  quickly.   Reactors which  are  sealed  vessels  can be  difficult.
These designs  should accommodate such   visual inspections  by  incorporating
large portholes or manways in the reactor shell.

Generally the  surfaces  of submerged  quartz  systems   contacted  by  the water
will  become  coated  by  inorganic scale, very much  like boiler scale.   This will
be  especially  the  case in areas  where there is  hard  water.   Additionally, the
inside  surface  of the  quartz and   the  outer  surfaces  of  the   Teflon tubes


                                       127

-------
eventually  will  develop a  grimy  dust  layer, primarily  from  airborne  dirt  and
water vapor.

Fouling  of   the   reactor's   internal   surfaces  will   be  signaled  by   reduced
performance  efficiency,  or   by reductions  in  the  intensity  measured  by  in-
line probes.    While  these  may  provide  some  signal  of fouling,  it  is  still
necessary to be able to inspect the surfaces  physically.

It  is  appropriate  to  completely  overhaul   the  reactor,  cleaning  all  interior
surfaces,   and   determining   the  lamp  outputs   and   quartz   (or   Teflon)
transmittances.   Each  lamp  is  removed  from  the  reactor and washed  with a
mild  soap  solution, rinsed,  and  swabbed  with an  alcohol (isopropyl)  soaked
rag  (cheesecloth).   Then   the  interior   surfaces of  the quartz  sleeves  are
cleaned  by  the  same  procedure  by  using  a  gun-barrel  type cleaning rod to
swab the  interior surfaces.    At  the  same  time,  each  lamp, which  is  tagged
with an  I.D.  number, is  measured for  relative  output.   Those which are  below
a specified  level  are  discarded  and  replaced with  new lamps.   These new
lamps  also are  tagged with a number.   In  this manner, each  lamp  can  be
traced on  the basis of operating  time and  output.   A reactor lamp inventory
then  can  be  mixed  and controlled  to  maintain a  minimum  average  output
level.

In  similar  fashion,  the quartz  should  be  monitored  for transmittance.   It ;may
be  cumbersome,  however, to  remove all  the quartz  from  a system.    It is
recommended  instead  (U.S.  EPA, 1986,  p.  233)  that a  representative  fraction
of  the quartz sleeves be  monitored;  10-15% of the quartz inventory  would  be
sufficient.    The  same  quartz  should  always  be monitored;  these would  be
considered  as  representative  of  all  quartz in  the  system.    If  the  tagged
quartz begins  to show marked deterioration due to  aging and  wear,  it  then
may be  appropriate  to  broaden  the  monitoring  and  to  begin  replacing  the
quartz sleeves.    The  replacement can  be  accomplished gradually.    As  with
new lamps,  eventually  there will  be  a   mix  of old  and new  quartz in  the
system.    There  is little  experience  in  determining  the  effective  life  of  the
quartz sleeves;  certainly  it will vary  by  site, but  generally should be between
four and seven years.

In  Teflon  systems,  the lamps are removable on racks; they should be  cleaned
and monitored  in  the  same  manner  as the  quartz  systems.   The  Teflon  tubes
should be  cleaned  on  occasion; this  can  be done by swabbing the tubes with
soapy  water/alcohol.   A  non-abrasive  material should  be used.   Each  tube
also should  be  monitored for  transmittance,  just as  with the quartz  sleeves.
This  may  not  be as  straightforward,   however,   because  of  their  limited
accessibility  and   problems  in   getting   direct  measurements   with   a   UV
radiometer/detector.
                   vii.  Costs for Disinfection With UV Radiation

                             Construction Costs

Table  XXVIII summarizes costs developed  in 1978  by Hansen  et  al.  (1979).
By 1982, these costs had increased about  15%  (U.S. EPA, 1983) for the size

                                       128

-------
         TABLE XXVIII.  CONSTRDCTION COSTS FOR ULTRAVIOLET LIGHT DISINFECTION
Plant Capacity (gpd)
Cost Category
Excavation and
Sitework*
Manufactured
Equipment
Concrete*
Labor*
t~) Pipe and Valves*
SO
Electrical and
Instrumentation*
Hous ing*
SUBTOTAL
Miscellaneous and
Contingency*
14,400
$ 60
800
250
110
60
430
1.500
3,210
470
28,800
$ 60
1,125
250
170
150
430
1,500
3,885
560
187,200
$ 60
4,485
250
250
350
430
1.500
7,225
1,010
374,400
$ 60
8,685
250
300
450
430
1.500
11,675
1,580
748,800
z$ 80
17,365
280
400
750
480
1.800
21,155
2,830
1,123,200
$ 110
26,050
300
500
1,000
480
2.000
30,440
4,060
TOTAL            $3,680          $4,445          $8,335         $13,255        $24,085          $34,500
Data from Hanaen et al.  (1979).

-------
ranges  listed.    Data  presented  are  for  single  and  multiple   UV  lamp
disinfecting  units ranging  in  water  throughput capacity  from  14,400 gal/day
to 1,123,200 gaVday.

All  UV  generating  units   are   quite  compact;   for  example,  the   1,123,400
gal/day unit  occupies  an  area  of less  than  24  ft2.    Costs  listed  in  Table
XXVIII  include equipment  costs  of  the UV  units, and  the  related  costs  of
piping, electrical equipment,  equipment  installation,  and  a  building  to  house
the equipment.

Figure   28  shows   construction   cost  curves  for  UV  disinfection   systems
(Gumerman et  ah, 1986).


                      Operating  and Maintenance  Costs

These are  shown  in Table  XXIX (for  1979)  for the  same  size plants  as  in
Table XXVIII.    Process energy  is required  for  the mercury  lamps  operating
inside  of  the   UV  generating  units.    Continuous   24-hr/day  operation  is
assumed,  with  only  occasional shutdown  to clean  cells  and replace UV lamps
which have become weakened by  lengthy  use.   Building energy requirements
are for heating,  lighting,  and ventilation.

Maintenance  materials are  related  to  the  replacement  cost of  the  UV lamps,
which usually are  replaced after operating  continuously  for  about 2,000  hours
(about  eight  months).    Labor requirements are related to occasional  cleaning
of  the quartz  sleeves  which  surround the  mercury  vapor lamps, and  periodic
replacement of  the weak UV bulbs.

It  is  noteworthy that  replacement  bulb costs at  the  smallest plant  (14,400
gal/day)  are only about 9%  of  the  total  O&M  costs, whereas  at  the largest
plant  size  (1,123,200 gal/day), replacement  bulb  costs  are about 48% of the
total  O&M costs.    This  reflects  the  fact that  the larger  UV generating  units
contain a greater number of UV bulbs per unit  volume of water treated.

Figure   29  plots   estimated  labor   requirements  for   the   operation   and
maintenance  of  UV  systems   (Scheible et   al.,  1985).    Figure   30  shows
operating   and   maintenance  requirements  for  UV   disinfection  systems-
building  energy, process  energy,  and  maintenance materials  --  treating  5,500
gal/day to 1 million gpd.   Figure 31  shows O&M requirements for labor and
total  O&M cost.  Figures  30  and 31  are taken from Gumerman et  al. (1986).


         2.    Secondary  Disinfectants

There are  disinfectants  which  are added  in   small dosages  to waters which
have  been  treated  with  primary  disinfectants.   The  purpose  of secondary
disinfectants  is   to  maintain  a  small  residual  disinfectant in  the  distribution
system.
                                       130

-------
1
7
5
4
3
2
9
I
8
5
4
*» 3
CO
O
0 1C
M>r- w *-J
rfOliOOHiSNOO
*



4


10
1
3.000.C





000.00





10.000







1.000







00
100





0







































































•^




























.Ml




























I-*""




























^




























^




























+



























5



























"



























ml





































•••i^H

















^




























X



























^j




























4



























i*T



























di
i •












2 349 8788 2 3 4 9 9 789 2 3 4 8 8 788
000 10.000 100,000 1,000.001
PLANT CAPACITY - gal/day
                  10
                           1  I
        100                 1000

PLANT  CAPACITY • m^/day
Figure 28.     Construction  cost  for  ultraviolet  light  disinfection  (Gumerman
              et al, 1986).
                                       131

-------
                         TABLE XXIX.  OPERATION AND MAINTENANCE SUMMARY FOR ULTRAVIOLET LIGHT DISINFECTION
to
Plant Flow
Rate (gpm)
14,400
28,800
187,200
374,400
748.800
1,123,200
Energy (kWh/yr*
Building
10,260
10,260
10,260
10,260
12,310
13,340
Process
440
800
5,260
10,510
21,020
31,540
Total x
10,700
11,140
15,520
20,770
33,330
44,880
$0.07/kWh -
$ 749
780
1,086
1,454
2,333
3,142
Maintenance
Matl.*($/yr)
$ 100
140
600
1,120
2,250
3,300
Labor
(hr/yr)
24
24
24
30
36
42
Total Cost
x $10/hr = ($/yr)
$240
240
240
300
360
420
$ 1,089
1,160
1,926
2,874
4,943
6,862
                 *   Data from Hansen et al.  (1979).

-------
           1000

            BOO

            600


            400
            200


          "5"

          I 100

          I  BO

          £  60


             40
             10
                      30
                           60
                                Approximate Numtwr ol Limpi (1.5m Arc)

                                 120     300   600    1200
                                                           3000   6000
I
I
I
                                              NoM:  Labor Bl»d on
                                                  Ton! Sytum KW.
                              I  I  I
                                                                    I
                                 10    20     40  60

                                     Svtltrn Si», Total KW
                                                    100
                                                          200
                              400 600  1000
Figure 29.     Estimate of labor  requirements  for  the  O&M  of  UV  systems
               (Scheible et ah, 1985).

In  point  of  fact,  there  are  three  secondary  disinfectants  in  prevalent  use:
chlorine, chlorine  dioxide,  and  monochloramine.    Considerations  pertaining  to
chlorine  as   a   secondary   disinfectant  are   essentially  the   same  as  were
discussed earlier for chlorine  as  a primary disinfectant.


               a.    Chlorine Dioxide

Chlorine dioxide  (C1C<2)  is  an  unstable, greenish-yellow  gas, explosive  in  air
in  concentrations   above   4%.      Because   of  this   instability,  it  is   always
generated  in  solution,  on-site,  and  is  used  immediately without  storage.    As
long  as  care  is  taken  to  keep chlorine  dioxide  in  solution  and storage  of
solutions is  avoided,  there will  be  no explosion  hazards.   Chlorine dioxide  is
readily soluble  in  water and is decomposed by sunlight.
                                          133

-------
     10.000
1,000,000
                                      MAINTENANCE MATERIAL
                                         PROCESS ENERGY
                                       S&^SSftJHH
               T     2
                1000
          345 67SS
                  10.000
2  3  4 S 0 ?8®[     1
           100.000
3 4 86788
   1.000.000
                                 PLANT  CAPACITY • gel/day
                       10
  I
100
                                              1000
                                  PLANT CAPACITY  • m3/day
Figure 30.     Operation  and  maintenance  requirements  for  ultraviolet  light
              disinfection   --   building   energy,   process   energy,   and
              maintenance material (Gumerman et al, 1986).
                                      134

-------
    JOQ.OOQ
               |      2343 eras      2   3  4 s e 719      2   3 4  s«7Si

               1000             10.000              100,000         1.000.000

                                  PLANT CAPACITY - gal/day
                                          T
                          TT
                       10
       100                1QQO

PLANT CAPACITY • n»3/day
Figure 31.    Operation  ar\d  maintenance  requirements   for  ultraviolet  light
             disinfection  --  labor  and total  O&M  cost (Gumerman  et  al.
             1986).                                                        *'
                                      135

-------
The material is a more  powerful oxidizing agent  and a  better biocide than  is
chlorine.   In  addition, when  chlorine  dioxide  is  prepared  in  the absence of
excess  free   chlorine,  its  use  will  not  produce  trihalomethanes,  or  other
chlorinated   organic   by-products   of   current   public   health   concern.
Additionally,  chlorine  dioxide   has  been  used   in   pretreatment  to   oxidize
phenolic  compounds  and  to   separate  iron  and   manganese  from  organic
complexes which  are  stable to  chlorination.    No  oxidation  of  bromide  ion to
produce  hypobromous  acid has been observed  using chlorine  dioxide  free of
excess free chlorine.

Distribution  system  residuals  of  dissolved  chlorine  dioxide  are  longer-lasting
than   those  of chlorine,   because  there  is   no  reaction  with   ammonia  or
formation of chlorinated  organic  materials.    Additionally,  chlorine dioxide  is
not known to impart tastes and odors to water, as  does chlorine.

Recent  health  effects  studies  have shown  (NAS,  1987)   that  chlorine dioxide
produces hematological effects  in  both  humans and  laboratory animals.    For
these   reasons,  the  U.S.  EPA  (1981)  currently   advises   that  the   total
concentration  of  chlorine  dioxide  and  its  decomposition  products  (chlorite
and chlorate ions)  be maintained below 1  mg/L.    This allows a  maximum
applied dosage  of  1.2-1.4  mg/L (Werdehoff &  Singer, 1986).

Gaseous  chlorine  dioxide  has a strong,  disagreeable  odor,  similar to  that of
chlorine  gas, and  is  toxic to  humans  when  inhaled.   It is  detected by  the
human nose at concentrations between  1.4  and  1.7%.  When present at 4.5%
concentrations,  it  irritates  the  respiratory mucous membranes  and may  cause
severe  headaches.    At concentrations below  6%  in  air,   it  may  be  compared
with  chlorine   with   respect  to  its  toxicity   (Masschelein,  19/9).    Eventual
intoxications  appear  by  local  irritations of  the  nervous  system,  ocular  and
respiratory   mucous   membranes,  without  substantial  resorption  or  systemic
poisoning  (Ehrlicher,  1964).    There  are no  cumulative effects  in cases  of
repeated  exposure  (Haller  & Northgraves, 1955).


                   i.    Generation of Chlorine Dioxide

For  drinking water  treatment,  chlorine  dioxide is  generated  from  solutions of
sodium chlorite, NaClO2-   This material, purchased  as a  solid (80% NaClO2)
or  most normally,  as  a  25%   aqueous  solution,   is  treated  with  aqueous
solutions  of  chlorine   or  hypochlorous  acid),  and  sometimes  with  a  strong
mineral acid such as sulfuric or hydrochloric acids.
Three processes  used in  water treatment  plants  for  the generation  of C1C>2
employ  (a)  gaseous  chlorine,  (b)  sodium  hypochlorite  solution  and  mineral
acid,  and (c) mineral  acid.  Each process is summarized below.
1.  Gaseous Chlorine

This  is  a two-step  procedure, beginning  with  the  formation of  hypochlorous
acid upon dissolution of gaseous chlorine into water:


                                        136

-------
         C12   +    H2O  	>    HOC1    +    HC1

       chlorine    water        hypochlorous  hydrochloric
                                    acid           acid

These  intermediate  products  then  react  with  sodium  chlorite  to  produce
chlorine dioxide:

 HOC1    +   HC1   +  2NaClO2  ~->   2C1O2  +  2NaCl + H2O

hypochlorous  hydro-       sodium           chlorine   sodium    water
   acid      chloric        chlorite            dioxide   chloride
               acid

The  overall result of these reactions  is summarized by  the equation:

         C12   +  2NaClO2 —->      2C1O2 + 2NaCl

      chlorine     sodium           chlorine     sodium
                  chlorite            dioxide     chloride

According  to  the stoichiometry  of  this equation, one mole of chlorine  rfeacts
with  two moles  of  sodium chlorite  to produce two moles  of  chlorine dioxide.
In water supply  practice, some molar excess  of  chlorine actually  is  employed
so  as  to   insure  conversion  of  the maximum  amount  of  chlorite ion  to
chlorine dioxide.    Therefore,  the  recommended  ratio  of reactants is  two
moles of chlorine per mole of sodium chlorite.  On  a weight  basis,  1.57  parts
of chlorine gas are  added per  part  of NaQO2 (calculated  on a  100%  solids
basis   when  solutions  of  sodium   chlorite   or  80%  solids   materials-  are
employed).

Under  these conditions  of excess chlorine gas being  added, the  product  C1O2
solution  also  will contain  an  amount  of hypochlorous  acid/hypochlorite  ions.
These can  react with THM  and TOX precursor materials to produce THMs
and TOX materials.
2. Sodium Hypochlorite and Mineral Acid

This also is a  two-step  process, in which  sodium  hypochlorite solution  reacts
with hydrochloric  acid  to  form  hypochlorous  acid,  which  then  reacts  with
sodium  chlorite to form chlorine  dioxide:
     NaOCl    +   HC1  -->   NaCl    +   HOC1

     sodium   hydrochloric    sodium   hypochlorous
     hypo-       acid        chloride       acid
    chlorite
                                      137

-------
 HOC1

 hypo-
chlorous
 acid
HC1   + 2NaC102 -->  2C1O2
                    2NaCl
          hydro-
          chloric
          acid
          sodium
          chlorite
        chlorine   sodium
         dioxide   chloride
                                                               H2O

                                                              water
In  this procedure,  as  in the gaseous  chlorine  and sodium chlorite  procedure,
excess  chlorine  is  utilized  to  insure conversion  of  the  maximum  amount  of
chlorite ion  to chlorine  dioxide.   The C1O2 solution  so produced also  will be
able  to  form  some THMs  and  TOX  because  of  the presence  of  excess
hypochlorous acid.


3.  Mineral Acid

This  process  involves  mixing  a solution of  acid with  a solution  of  sodium
chlorite, and the reactions are as follows:
5NaClO2

 sodium
 chlorite
     4HC1 -->   4C1O2
           SNaCl
  hydrochloric
     acid
chlorine    sodium
dioxide   chloride
                                                       2H2O

                                                       water
When sulfuric acid is the  strong  mineral  acid, the  reactions can  be  depicted
as follows:
 10NaClO2

   sodium
   chlorite
                                8C1O2  + 5Na2SC>4 + H2O + HC1
              sulfuric
               acid
                 chlorine
                  dioxide
           sodium
           sulfate
                                                     water   hydro-
                                                             chloric
                                                               acid
The  exact  ratios  of  reactants  will  depend  upon  which  mineral  acid  is
employed for the production  of chlorine dioxide.

When  generated   using   mineral  acid,   excess   chlorine   is   not   required.
Consequently,  solutions of  chlorine  dioxide  prepared  in  this  manner do not
contain  free  residual  chlorine,  thus  no   more   than  trace  amounts   of
chlorinated   organic   by-products   can  be   synthesized.     In   addition,   pure
chlorine  dioxide  does  not oxidize  bromide  ion to produce hypobromous  acid,
which  then  can  produce  brominated  organic  derivatives.   With excess  free
chlorine  available, however,  bromide ion  can  be oxidized in  water to  form
hypobromous acid.

In all three  cases, the appropriate  aqueous  solutions of reactants are metered
into   a chlorine   dioxide  reactor  (a  cylinder  containing  Raschig rings,  glass
beads,  or  hollow  glass  cylinders)  where  intimate  mixing  of  the  reacting
solutions occurs  (see  Figures 32  and  33).   The  size  of the  reactor and the
residence  time  of the  reacting  solutions  are  such   that  after   a  few seconds,
the  solution   exiting  the  reactor  displays   a  strongly yellow  color of  chlorine
dioxide.   This solution then is  pumped  directly into the  water  to be treated.
                                       138

-------
                  Water flow mew
       4 - 20 nA
        signal
                    o
             CIO,
I
                                               Chlorine dioxi<$»
                                               gentrating lower
                                 Hatch ig
                                 ring*







c t
I >
}
c >
c >
"" V"
f,^^
--*



1 stroke
                                        Mixer  Sodium
                                              chlorite
                                              tolution
                                                                               Point of
                                                                               application
                                                   Injector water supply -
Figure 32.     Schematic   diagram   of  an  automatic  feed,   automatic   flow-
               proportional  chlorine  dioxide  system:   generation  from  chlorine
               and sodium chlorite  (courtesy Capital Controls Co., Inc.).


In  this manner,  solutions of  chlorine  dioxide  are  generated  as the  material  is
required  and  used  immediately,  without  storage.    Appropriate  metering  and
control instrumentation  can  be  installed  with  the  chlorine  dioxide  reactor so
that  the  generation  and addition  of  chlorine  dioxide  is  paced  by the   flow
rate  of  the water to be treated.    As a  result, the  unit  operates  without the
need for  constant manual attention.
                    ii.    Oxidation-Reduction  Reactions of Chlorine Dioxide

At   neutral  pH   (=  1)  chlorine  dioxide  dissolves  totally  in  water  without
hydrolysis  reactions.   On  the other  hand,  when performing  its  function  as a
chemical   oxidant  or  disinfectant  in  water,   chlorine  dioxide   is   chemically
                                         139

-------
 reduced.   One  of the reaction products  is the  chlorite  ion, from which  CIO?
 was formed initially:
         C1O2     +     e-

        chlorine   reducing
        dioxide      agent
                                   (C102)'

                                   chlorite
                                      ion
                                  Chlorin* dioxidt
                                  gerurtting toww
Figure 33.    Manual  feed  equipment  arrangement  for   generating  chlorine
              dioxide  from  sodium  hypochlorite  solution  and  mineral  acid
              (U.S. EPA, 1983).
In strongly  acid  (  pH < 2) or  strongly alkaline (pH ^ 11) solutions, chlorine
dioxide  undergoes disproportionation (self oxidation and reduction):
2C1O2  + H2O --->   (C1O2)-
   chlorine
   dioxide
           water
chlorite
   ion
(cio3)-

 chlorate
     ion
                                                   2H+
hydrogen
   ion
                                       140

-------
These  disproportionation   reactions  are  accelerated  by   the   presence   of
hypochlorous  acid  and  hypochlorite  ion.   Therefore,  when  chlorine  dioxide is
generated  in  the  presence of excess chlorine and at pH below 2 or above  11,
chlorite and  chlorate ions will be formed as well


                   iii.         Establishing a Chlorine Dioxide Residual

Because  the  cost of  chlorine  dioxide   is higher  than that  of  chlorine,  and
because  of   the   hematological  effects   upon  both  humans  and  laboratory
animals, only small dosages  of  chlorine  dioxide  (maximum  residuals of 1  mg/L
total  of C1C«2,  chlorite, and  chlorate ions) currently are recommended  by EPA
for drinking  water treatment (U.S.  EPA,  1981;  1987a).   However,  a primary
advantage of pure chlorine dioxide is that it  does  not  react with  ammonia  or
with  THM precursors  to produce  THMs, or with  TOX  precursors  to  produce
TOX, as  does  chlorine.   This means that if the water  has  been pretreated  to
remove  most  of  the  oxidant-demanding  constituents  (or   does  not  contain
them  initially),  the  total  C1O2  dosage  can be  used to  provide  primary   or
secondary  disinfection at  the  recommended  1 mg/L maximum  oxidant residual
level.
                                                                           t
In  laboratory studies, Werdehoff & Singer (1986) have  shown  that  about 70%
of  the  C1O2  added  is converted to  chlorite ion.  Therefore  a chlorine dioxide
dosage  of 1.2  to  1.4 mg/L  is the  maximum  practical dosage  in order not  to
exceed  the recommended  1  mg/L  residual  of  total oxidant.   This  value has
been  confirmed by Lykins & Griese  (1986)  in  pilot  plant and full-scale  plant
studies   at  Evansville,   IN   (see   Case  History  --  Evansville,  IN,  Section
VIII.C.1).

Another major  advantage  of  pure  chlorine  dioxide  is that it does not oxidize
bromide  ion   to   produce  hypobromous  acid.     This  means  that  bromine-
containing THMs  and  TOX derivatives  will not  be formed  when this  material
is  used for oxidation/disinfection.
                   iv.   Factors   Affecting    the    Efficiency   of   Secondary
                        Disinfection With  Chlorine Dioxide

Chlorine   dioxide   is   a   more   effective   disinfectant   than   chlorine  or
hypochlorous  acid.   Because  it does  not react  with water, ammonia,  bromide
ions,  or  most  organic  nitrogen  compounds,   it  is  not  "wasted"  in  extraneous
reactions  of this  type.   It  is  less sensitive  to changes  in  pH (except  at  very
low  and  very  high  values),  maintaining  its  disinfection  capabilities  over  the
pH range of 6 to 10.

On the other  hand, because  it is a  powerful oxidizing agent, chlorine  dioxide
can  and  will react  with  oxidizable impurities contained in  a raw  or  treated
water.    Thus,  it is important to ensure  that  oxidant-demanding components
of the  water have  been  removed to as  low a level as  is feasible, consistent
with the costs involved, before  chlorine dioxide is added.
                                       141

-------
                   v.    Chlorine Dioxide Systems Design

Table   XXX  is  a  partial  listing  of  current  suppliers   of  chlorine  dioxide
equipment.    Several  different  types  of generation  equipment  are available,
which   vary  depending   upon  the  supplier,  but   also  upon  the  generation
process  chosen  (gaseous chlorine  versus  sodium hypochlonte  plus  acid, versus
acid plus sodium  hypochlorite, for  example).   Pertinent aspects  of each  type
of  chlorine  dioxide generation equipment will be  discussed in this subsection.

  TABLE XXX.   PARTIAL  LISTING OF CHLORINE DIOXIDE
                   EQUIPMENT SUPPLIERS

              Name                      Address

  Capital Controls Co., Inc.              P.O. Box 211, Colmar,  PA
                                        18915

  CIFEC                               10  Avenue  de la Porte
                                        Molitor, F-7500, Paris, France


  Clow Corporation                     408 Auburn Ave., Pontiac, MI
                                        48058

  Fischer &  Porter Co.                  County Line Road, Warminster,
                                        PA  18974

  Olin  Water Services Co.               9393 W. 110th St.,  Overland
                                        Park, KS  66210

  ProMinent Fluid Controls              1005 Parkway View Drive,
                                        Pittsburgh, PA  15205

  Rio  Linda Chemical Co., Inc.          410 North Tenth Street,
                                        Sacramento, CA  95813

  Wallace & Tiernan Division            25  Main St., Belleville, NJ
  Pennwalt Corp.                       07109
1. Gaseous Chlorine Plus Hvpochlorite Solution

All three  of  the chlorine  dioxide  generation  procedures  described earlier  are
in general  use  in  the  United States  at  this time  for  application  in  small
water  treatment  facilities.    The  most  commonly   used procedure  involves
addition   of  gaseous  chlorine  to  sodium   chlorite  solutions.    A  schematic
diagram of chlorine/sodium  chlorite equipment is shown in Figure 32.

The  gaseous   chlorine  procedure   is particularly  applicable  when  a gaseous
chlorination system  already exists   at the  treatment  plant.   The  reaction by
which chlorine  dioxide is generated is as follows:

                                      142

-------
           +    2NaClO2  —->   2C1O2  +     2NaCl

  chlorine       sodium         chlorine         sodium
                 chlorite          dioxide         chloride

This  equation  indicates  that  71  Ibs  (26.48  kg)  of  chlorine  mixed  with  a
solution  containing  181 Ibs  (67.5  kg)  of  100% NaQC>2  will  produce  135 Ibs
(50.36 kg)  of chlorine  dioxide.   However, the  ratio of reagents  recommended
by  most  suppliers of  chlorine dioxide generating equipment  is  1:1  by weight.
This  means  that  more  than double  the stoichiometric  amount  of  chlorine
required  by the above equation  is utilized.   This excess  of chlorine  over that
required  results  in  a faster reaction  rate  and  insures  a more  complete
conversion of chlorite ion  to C1C>2.

However,  because   a   large  excess  of  chlorine  is  employed,   solutions  of
chlorine   dioxide  prepared   by   this  technique  also  will   contain  some  free
available   chlorine,   mostly   as   hypochlorous  acid.    This  amount  of  free
chlorine  will  produce  some  quantity of  trihalomethanes and other  halogenated
(TOX) materials.


2. Hypochlorite. Chlorite,  and Mineral  Acid

The  production of  chlorine  dioxide  using  sodium  hypochlorite  solution  with
sodium  chlorite  and  strong mineral  acid  is illustrated  in  Figure  33,  and  is
well suited to  most  small  water systems.   Dosages of each  chemical  can be
derived from  the equations given  earlier.

In this  system, all three  reactants  are  in  solution.   Utilization  of strongs acid
increases  the  conversion  of  chlorite  ion  to  chlorine  dioxide.    Solution  feed
pumps of equal capacities  can  be used by adjusting  the solution  strength  of
each  of  the  reactants.   Thus,  the chlorine dioxide production  and addition
rates  can  be  paced  by the flow rate of the water  being  treated and/or by its
secondary  disinfectant demand.

Sodium chlorite is  available in 55-gal  (208.33  L) drums, either as a solid  (80%
active NaClO2) or as  a solution containing 30%  NaClO2  (33% solids).   If not
used  directly  from  the drum, sodium chlorite  solution is  stored  in  polyfvinyl
chloride)  (PVC) or fiberglass tanks and  transferred  by means  of PVC,  rubber,
or  Tygon  tubing  systems.    Diaphragm  pumps  incorporating   PVC  as  the
material  in contact  with  the solutions  are used  for pumping sodium  chlorite
solutions.    Provision  must be  made for immediate  washdown of any spills  of
the chemical.   This  precaution  is generic  to  all  chlorine  dioxide generating
systems.


3. The CIFEC  System for Generating Chlorine Dioxide

A schematic  diagram   of this  system,  developed  in  France, but  in  use  at
several  U.S.  water  treatment plants,  is  illustrated  schematically  in  Figure 34.
The   system  produces  C1O2  from   gaseous   chlorine,   in  high  yield,  and
containing little excess  free chlorine.

                                       143

-------
                                                                       00 2 EXIT
                VMUUH UMC Qf CHLORINE
CHLORINATOR
                                     EJECTOR WITH CHECK WIVE ASSEMBLY
                                               SOOIUM CHLORITE METERING PUMP
                       ELECTRIC VALVE
    Figure 34.    Schematic of the CIFEC chlorine  dioxide generating  system.


    Gaseous  chlorine  is  passed  into  water  which  is  circulated  continuously  in
    what  is  referred   to  as  an  "enrichment  loop".     Under  these  conditions,
    dissolved  chlorine  (hypochlorous  acid) concentrations  become  higher than  can
    be achieved in a  single  pass.   As a  result,  the pH  of the  hypochlorous  acid
    solution is  lowered to  below  4.   This solution then  is  pumped into the C1O2
    reactor  along with  a solution of  sodium  chlorite.   As long  as the  pH  of the
    hypochlorous  acid   solution   is  below  4.0,   conversion  of   chlorite  ion   to
    chlorine dioxide  is  significantly higher than  the single pass  method  employing
    elemental  chlorine gas.   Therefore, chlorine dioxide  is produced  which  is  free
    of significant quantities  of free chlorine.
                                            144

-------
4. Rio  Linda Chlorine Dioxide Generator

Figure  35  shows  a  schematic   diagram  of  a  newer  acid/chlorite  chlorine
dioxide  generator marketed  by Rio Linda Chemical Co., Inc.   Chlorine dioxide
is  generated   by  addition  of  dilute  hydrochloric  acid  to  sodium  chlorite
solution.   The  novel  principle  of this  generator is  the  mixing of  acid  with
concentrated sodium chlorite  solution  just  before  the  two  solutions reach  the
reactor.   The  two  solutions are  brought  together in an  eductor by  means  of
a vacuum  created  by  water flow  through  the  eductor.    Such  a  system
eliminates a pump and allows  the  system  to occupy a smaller space.
                              FLOW DIAGRAM
                       REACTION
                       COLUMN
                                                 FLOW RATE
                                                  METERS
                            SODIUM
                           CHLORITE
                            SUPPLY
Figure 34.     Schematic of the  Rio Linda  Chemical  Co. acid/NaC!O2 chlorine
              dioxide generator.


                   vi.   Miscellaneous Comments

Because  several  types  of  chlorine dioxide equipment are available, as  well  as
three processes  for  its production,  it  is  considered  inappropriate  to  attempt
to  provide  detailed  instructions  in  this  document   for  the  preparation  of
chemical solutions  and feed  rates.   However, the  small  water utility  choosing
to  install  chlorine  dioxide   generating  equipment  can  have  confidence  that

                                       145

-------
each   equipment   vendor  will  provide   detailed  recipes  for  preparing  and
metering  the  appropriate  solutions  to his chlorine dioxide  reactor  so as  to
produce  an  aqueous  solution  of  chlorine  dioxide  of  known  and  constant
concentration for  addition to the plant  process water.

A  final  point to  be   noted  is  that the  currently  recommended  maximum
concentration  of  total  oxidants (chlorine  dioxide,  chlorite  and chlorate ions)
of  1  mg/L means that  a  water utility processing  0.5  mgd  (1,893  m3/day) and
dosing  a  maximum  of  1  mg/L  of chlorine  dioxide  will  require  a  maximum
ClO^  production  rate   of  only  4  Ibs  (1.82 kg)/day.    Smaller  systems will
require even  less C1C«2.   At such low dosage levels,  two of  the three  vendors
of  chlorine  dioxide  contacted  recommend  that  their  units  be  operated  inter-
mittently,  collecting  C1C»2  solution in  an enclosed holding  tank  for  metering
into the  water being  processed.    This  is because at the  low flow  rates  of
reactant  solutions,  mixing  is  less  efficient  in  the  chlorine   dioxide  reactor.
Consequently,  conversion of  chlorite  ion  to  chlorine  dioxide  is  less  efficient.


                   vii.   Costs of  Chlorine Dioxide  Generating Systems

Hansen  et al.  (1979)  summarized costs  for the  generation  of  chlorine dioxide
from  equal  parts  of  2.4%  sodium  chlorite solution, 25% sulfuric  acid  solution
and  1%  sodium  hypochlorite  solution.    Suppliers   contacted  in  1982  had
changed  the  design  of  their  generation  systems  for  small water  supply
systems to use 33% hydrochloric acid rather than 25% sulfuric acid.

Equipment costs estimated by Hansen  et  al. (1979) assumed  the use of a dual
head   diaphragm  pump  for  simultaneous  addition  of hypochlorite  and  acid
solutions,  and  a  single  head  pump for  the  addition  of  sodium  chlorite
solution.   Detention  time  in  the chlorine dioxide reactor  is  estimated at  12
seconds,  and the  generating  equipment  costs are  assumed  to  be  constant  up
to 50 Ibs/day of chlorine dioxide.

At  a  maximum  chlorine  dioxide  dosage  rate  of 1  mg/L,  a  1  mgd water
treatment  plant  would  dose  8 Ibs/day.    At  the  same  dosage  level,  a  2,500
gal/day water facility  would require only 0.2 Ib/day of chlorine dioxide.


                              Equipment  Costs

Quotes  were  obtained  (U.S.  EPA,  1983)  from  three  suppliers  of  chlorine
dioxide generation equipment sized  so as  to prepare  C1C>2 at  the  rate of 8
Ibs/day  (for a  1  mgd water treatment  plant).    These  are shown  in Table
XXXI.    Supplier A's  recirculating  loop  system  (CIFEC)   is  the  highest  in
equipment price;  their lowest  cost  unit was  priced at $34,000  in  1982.  This
unit  operates   with   a   special   recirculating  pump  designed   to  handle
hypochlorous  acid  below pH  4, plus a sodium chlorite solution pump  and  all
necessary  instrumentation   to   allow   automatic   operation,   with   shutdown
provisions  in the event of cessation of water flow.
                                       146

-------
TABLE XXXI.  1982 VENDOR  QUOTES  -- CHLORINE  DIOXIDE
                GENERATORS
Vendor
Recircu-
lating loop
Supplier A
(French)
C102
production
capacity
flbs/davl
1-10
space
reauired*
2x3x6 ft
Reactants
Cb gas +
NaClO2 solu-
tion
Unit Cost
$34,000 (1
rate, adjust
manually)
dt^rft f\f\/\ /*\
automatically)
(prices delivered to New York)
Supplier B     4          3.5x4x1.5
                          ft (wall-
                          mounted
                                      automatically)
                                     HC1, NaOCl
                                       solutions
Supplier C   14-140
Supplier C   14-140
                          4x3x1.5 ft  Cb gas +
                                      Nad
                          37.5x27x
                          6.5 in.
                                          1O2 soln.
                                      same
                          4x3x1.5 ft  HC1  +
                                      NaClO2 soln.
                          37.5x27x     same
                          6.5  in.
                                                      rates (adjust
 $41,700 (3
rates (adjust

 $25,000
(installed)
 $ 4,320**
(floor mount)
 $ 3,600**
(wall mount)

 $ 4,320
(floor mount)
 $ 3,600
(wall mount)
**
     all units  require additional space for  solution  tank(s).
     this unit  requires a chlorinator for  operation,  which is not
     included  in price estimates.
  The next  lowest  in  price  is  the system  from  Supplier B,  which  generates
  chlorine  dioxide  from  33%  hydrochloric acid,  12%  sodium  hypochlorite, and
  25% sodium chlorite  solution.   This  unit  cost  $25,000  (installed) in 1982, and
  includes  three  solution pumps, water  flow  rate  detector, and  switches  to shut
  down  the  unit if  the  water  flow  stops.    This  unit  is  wall  mounted and
  requires  3.5  x 4  ft of wall space,  plus floor space  for drums of the  three
  chemical  solutions  used  to  feed  the  generator.    For  volumes  of  C1O2
  sufficient  to  treat  flows  in  communities of  5,000 and  2,500  population, this
  unit is  said  to  be  capable  of continuous  operation, with no loss in efficiency
  of  conversion  of  chlorite  ion  to  chlorine  dioxide.   However,  to supply  the
  needs  of systems  serving  as  few as  25 persons,  the  unit would have to  be
  operated  intermittently,  with  C1O2  solution  being stored  in   a holding  tank
  for  later metering  into the water.
                                         147

-------
Supplier C provides two types of chlorine dioxide  generators  for  small water
supply  systems.   One  uses  acid/sodium chlorite,  the  other  uses chlorine gas
and  sodium  chlorite.    These  units cost $3,600 in  1982,  if wall-mounted,  and
$4,320  for  a  floor-mounted  cabinet.   The  single  size  unit  offered  by  this
supplier is  designed to  generate  up  to 140  Ibs/day.   In order  to produce  8
Ibs/day  or  less,  a  small water  utility  would  have  to  install  a  holding  tank
and  operate  the  generator intermittently.

The  chlorine  gas/sodium  chlorite  generator  of  Supplier C  requires  a  gas
chlorinator  to feed  chlorine  gas.   Therefore,  in  new plants  considering use
of  this type  of  equipment,  the  cost  of a chlorinator must  be  added  to the
cost  of the  chlorine  dioxide generator.    In  existing plants  currently  using
gas  chlorination, the  chlorinator  already is in place  and would  not represent
additional equipment cost.

Because  equipment  quotes  for   generating  chlorine  dioxide  vary  so  widely,
water  treatment personnel  are  advised not  to try  applying  past equipment
cost  estimates.   Technology  for  generating  and  applying chlorine dioxide  is
changing  rapidly (as  opposed to  technologies  for  addition  of  gaseous  or
aqueous chlorine),  and  new suppliers  enter the  market  from  time to time.   It
is more advantageous  to seek quotations  from the  various suppliers as  to the
various methods for  generating C1O2-   Select  the  methods  most appropriate
to the  specific  water treatment plant, then determine what  piping  and wiring
will  be needed to install the equipment selected.

Figure  36  (Gumerman  et  al.,  1986)  shows   that  the construction  costs for
chlorine dioxide  generating and  feed  equipment  are  constant for  feed  rates
of up  to about 45  Ibs/day.


                      Operation and  Maintenance Costs

Hansen  et  al.  (1979)  concluded  that,  in  general,   O&M costs  for generating
QO^   are   independent  of  the   quantities  generated.     Process   energy
requirements, which  are for metering pumps  and mixer for  preparing  chlorite
solution from  solid  sodium chlorite,  are  estimated  at 1,240  kWh/yr.    Energy
requirements  for 40 ft^ of building  space  to  house  the  equipment would be
4,100 kWh/yr,  resulting in total energy  requirements  of 5,340 kWh/yr.   Main-
tenance  material requirements  would  be   for  minor  equipment  repair  only,
amounting to  about $100/yr.   Labor is required for  preparation  of  solutions
and  periodic  maintenance  of the equipment.   Annual  labor requirements  are
estimated to be  1 h/day, or  365 h/yr.

Annual O&M  costs of  $4,124/yr (based  on  $0.07/kWh power  cost  and $10.00/h
labor cost) are summarized  in Table XXXII.   Figures 37 and  38 (Gumerman et
al.,   1986)  show  the   estimated  O&M   requirements  for  chlorine   dioxide
generation  and  feed  equipment     -  building  energy,  process  energy,  and
maintenance  material  (Figure  37)  and labor and total O&M cost (Figure 38).
                                       148

-------
    (0
    O
    o


    i

    o

    «E
    H
    (O

    O
1.
7
6
S
4
3
2
1<
8
9
4
i
3
2
1(
i
a
9
4
3
2
t(
fl
9
4
3
2
10
0.
300.000




DO.OOO



»n™ 	

WOO






)00

































_























_











U-
T~~ i rrn






i
l
!!
I »
i




^ ? ll
_ 	 ...I,,..., .
777"^


s

0


!
i
1




















































J,i -K,. ^




















'
f~

























































'r













































L,


























hh
















































•MB








•••




























___ i




































•••


























^B •!


























2 3 496789 2 3 456789 2 3 49878
1 1.0 10 1
MAXIMUM FEED RATE • Ib/day



























9
00
              0.1
        1.0

MAXIMUM FEED RATE • kg/day
10
Figure 36.    Construction   cost  for  chlorine  dioxide  generation   and  feed

             equipment (Gumerman  et al., 1986).
                                      149

-------
  10,000
10.000.000
                                             FIELD DATA LEGEND
                                             A « MAINT. MAIL.  - VILLA RICA
                                               • TOTAL ENERGY  ~ VILLA RICA
                                             C * TOTAL ENERGY  - FAYETTEVILLE
                                                 MAINTENANCE MATERIAL
                                                       PROCESS  ENERGY
                                                       BUILDING ENERGY
                   2   34$ 8788      2   3  4 8 i 7S9J     2   3  4 5 §788

                                 1.0                  10                 1QO

                               AVERAGE FEED  RATE - kg/day
                    I
                  0.1
                   	I                     I
                          1.0                  10
                   AVERAGE FEED RATE • Ib/day
Figure 37.    Operation  and  maintenance  requirements  for chlorine  dioxide
             generation   and  feed  equipment  --   building energy,  process
             energy, and maintenance  material (Gumerman et al.,  1986).
                                     150

-------
(0
O
O
o
H
1,000,000 1,000.000
1
0
5
4
3
2
1
5
4
3
2
1(
1
4
3
2
11
T
0
S
4
3
2
1C
'- f
0
5
4
3
2
00.000 1
9
: •
0
s
4
3
- 2
.s
),000 . i(
• K T
5 8
j 3
4
3
2
)00 1C
0
S
4
3
2
10 1C
I



.=-~__


00,000





3,000




"»••••
>00







)Q


















sam
















Lj-


















BH


















Bl
















,
z±h









MB •




























aaMBBM^



























ri&n



























HB



























ma



























••









- F


A
Ji







TO
Ci>

~T


•*"










I ELD DATA LEGEND
• LABOR - VILLA RICA
. LABOR - FAYETTEVILLE







TAL O&fc
(CLUD)N



^£0^
^P«-








*/








1 CC
3 C

onJUs
/&







»w
/
/
















>ST
HEM
^E
LmJ— -




Lt




»,e
J
T\
iff
/
f

















c






0















Al






R















.S]















2 343 9789 2 3 4 S 6 789 2 3 4 9 S 7S9
1.1 1.0 10 100
AVERAGE FEED RATE • its/day
                   0.1
         1.0                  10
AVERAGE PEED RATE • kg/day
 Figure 38.     Operation  and  maintenance  requirements  for  chlorine  dioxide
              generation  feed  equipment  -  labor  and  total   O&M  cost
              (Gumerman et al., 1986).
                                      151

-------
  TABLE XXXII.  OPERATION AND MAINTENANCE SUMMARY FOR
                  C1O2 GENERATING AND FEED SYSTEMS
        Item              Requirements*           Cost

  ELECTRICAL ENERGY

   Process              1,240 kWh/yr x $0.07  =   $  86.8

   Building             4,100 kWh/yr x $0.07  =   $ 287.0

             TOTAL   5,340 kWh/yr x $0.07  =   $ 373.8

  MAINTENANCE MATERIAL                        $ 100/yr

  LABOR                   365 h/yr   x  $10.00   =   $3,650

   Total Annual O&M Cost                                $4.124

   *    based on estimates of Hansen  et al.,  1979



                             Chemical Costs

At  a  production  rate  of  only  8  Ibs/day  (maximum  for  a  1 mgd  water
treatment plant  at a  1 mg/L applied chlorine dioxide dose),  chemical  costs
are not  as  significant  as  pumping  costs.   Nevertheless, 1982  chemical  costs
were as follows:

        gaseous chlorine          $0.47/lb
        sodium chlorite            1.55-1.65/lb
        HC1                      0.10/lb
        NaOCl                    0.93/gal (15% solution)

Prices for chlorine  and sodium chlorite  in 1987 were about 30% lower.

A chlorine  dioxide production rate  of 8 Ibs/day  equates to  2,920  Ibs/yr.   If
the chemicals cost for  C1O2  is  arbitrarily  assumed  to  be  $l/lb,  a  1  mgd
water  treatment plant  can  expect  to  pay  about  $2,920  in  addition  to the
annual  O&M  costs.   On  the same  basis,  chemical  costs at  a  2,500  gal/day
plant  would  be   $2,920/400  =  $7.30/yr for  producing chlorine  dioxide  at
$1.00/lb.


             b.    Chloramination

Chloramines  are  formed  when  water  containing  ammonia  is chlorinated,  or
when   ammonia  is   added  to   water  containing   chlorine   (hypochlorite  or
hypochlorous  acid).     This   is  accomplished  currently  by   adding  gaseous
ammonia  (purchased  as the  anhydrous  liquid,  NH3,  in  150 Ib cylinders for
small  water  treatment systems) directly to the water,  or  by  adding  a solution
of ammonium sulfate, (NH4)2SO4 (purchased in 100 Ib bags, 98% pure; 25%
available NH3).
                                      152

-------
                   i.    Chemistry of Chloramination

Three  chloramine  compounds can  be  produced,  depending  on  the ratios  of
chlorine and ammonia which are utilized:

        NH3   +    HOC1  —->  H2O  +   NH2C1

       ammonia  hypochlorous   water   monochloramine
                   acid

        NH2C1   +   HOC1  —>  H2O  +  NHC12

                                         dichloramine

        NHC12   +   HOC1  —->  H2O  +  NC13

                                             nitrogen
                                             trichloride

The distribution  of the  chemical species of  chloramines is a  function  of  pH
and of the  amount  of chlorine added.   For example,  in the pH range  of 7 to
8  and  a chlorine  to  ammonia  weight ratio  of 3:1, monochloramine  is  the
principal  product.   At  higher chlorine to  ammonia ratios  or  at  lower  pH
values  (5  to 7),  some dichloramine  will be formed.   If the pH drops below 5,
some  nitrogen  trichloride  (often  erroneously  called  "trichloramine)  may  be
formed.   Formation  of  this  compound  should be avoided because  it  imparts
undesirable  taste  and odor to the water.

Figure  39   (National   Academy   of   Sciences,   1980)   shows   the   relative
percentages  of monochloramine and  dichloramine produced  as  the pH changes,
for different weight ratios of  chlorine  to ammonia.    At  a pH value of about
5.7,  approximately  equal  amounts  of mono-  and  dichloramines are  present in
solution.

Care  also should be taken  not  to  exceed chlorine  to ammonia  ratios  of  5:1.
This is the  ratio existing  at  the peak  of the breakpoint  curve,  above  which
all  of  the ammonia  will  have been removed,  chloramines  will be absent, and
free residual chlorine will  be present.


                   ii.   Establishing a Chloramine Residual

Generation  of  chloramines   is  conducted  on-site,   in  solution,   as required,
simply   by  adding  the  appropriate  amount   of  chlorine  to  waters   already
containing  ammonia,  or   by  adding ammonia to  waters  already  containing
chlorine,  then  allowing a  short holding  time  to  be  certain  that  the chemicals
have  had  time  to  react  with  each   other  to   form chloramines.     Usually,
chloramine-forming  reactions are at  least 99% complete within a  few minutes.
                                      153

-------
                                      pH
Figure 39.     Proportions   of  mono-   and   dichloramines  in   water  with
              equimolar concentrations of chlorine  and ammonia  (NAS,  1980).


However, the  National  Academy  of  Sciences  (NAS,  1987) has  recommended
rejection  of   the   first  approach  (adding  chlorine  to  ammonia-containing
water).   Furthermore, when using the second  approach, the NAS recommends
addition  of sufficient chlorine  to produce  a  slight  residual of  free  chlorine
above  that required  to  oxidize  nitrogen  (particularly  the  organic  nitrogen
compounds), followed by addition  of ammonia to form monochloramine.
                   iii.   Chloramination System Design

Ammonia is available as the  anhydrous gas (NH3),  as a 29%  water solution
(aqua  ammonia),   or  in powdered  form  as  ammonium  sulfate  (NH^SO^
Gaseous  ammonia  is  supplied  in  150-lb  (68.1  kg)  cylinders (for small water
systems), similar to gaseous  chlorine.   Aqua ammonia  is  supplied  in  55  gallon

                                      154

-------
f208.33 L)  drums.   Ammonium sulfate  is  available in  100-lb  (45.4  kg) bags
(98% pure, 25% available ammonia).

Gaseous ammonia normally  is added to  the  treated  water using systems  and
equipment similar to those  used  for  gaseous chlorine.   Aqua  ammonia  and
ammonium  sulfate solutions  are  handled  using  systems  and  equipment similar
to   those   for   sodium   hypochlorite   and  calcium   hypochlonte   solutions,
respectively.  Aqua ammonia  is basic  and odorous,  but is non-corrosive.

Sizing of the  treatment  facility must take into  consideration  the intended 3:1
chlorine/ammonia  ratio.

A 25% to 30% solution  of ammonium  sulfate in water is  prepared in a plastic
or  fiberglass  container   and   added  to the  water  by  means  of  a  chemical
metering  pump.    Solutions  of ammonium  sulfate are  stable, but  are acidic,
and   therefore  can  be   corrosive   to   some  metals.    Materials   which  will
withstand  dilute  sulfuric  acid also  will  easily  resist  any  possible  corrosion
effects of dilute ammonium sulfate solutions.
                   iv.   Costs for Chloramination

Generation   of  chloramines  requires  the  same   equipment  for  chlorination
(gaseous  or  aqueous  hypochlormation)  plus  equipment  for  the  addition  of
ammonia  (gaseous or aqueous).   Costs for chlorination equipment and  for  its
operation   and  maintenance  have  been  presented  earlier.    In  this  section,
costs  for addition  of ammonia are presented.

During  January   1983,  costs for  liquid  ammonia  were  $0.40/lb  (in  150-lb
cylinders),  $0.70/lb of  contained ammonia in 28%  solution  (purchased in  55
gal drums),  and  $0.51/lb  for  solid  ammonium sulfate (purchased  in 100  Ib
bags), in the Washington, DC - Baltimore area.

Cost  calculations  given  below are  based on  the following reaction of chlorine
gas and ammonia  to produce monochloramine:

              NH3  +   C12  -—>   C1NH2  +   HC1

In addition,  the  calculations  assume a  dosage  of  2.5  mg/L  of chloramine  as
the   secondary    disinfectant.     This   is   the   maximum    level   currently
recommended by  the EPA to be dosed into  water  supplies (U.S. EPA, 1987a).
It  is  further  assumed  that the  chloramines  will  be  produced  by  adding
ammonia  to  water already  containing  free  available  chlorine.    Finally,  cost
calculations  are  based  upon  chlorine  added  as  the gas.  By  using  previously
described   calculations   involving   solutions   of   sodium   hypochlorite   or  of
calcium hypochlorite,  the  amounts of  these chlorinating  agents   required  to
produce monochloramine can  be calculated readily.
                                       155

-------
1.  Costs for Monochioramine From Anhydrous Ammonia + Chlorine  Gas

A  2,500 gal/day water  treatment  plant will  require  0.05  Ib/day  (23.66  g) of
monochloramine at  a  dosage of 2.5 mg/L.   According  to  the stoichiometry of
the above  equation for the generation of monochloramine from ammonia  and
chlorine,  each  gram-molecular  weight  (weight  of  one  mole  of  compound
expressed in grams) of monochloramine  will require 1  gram-molecular weight
each of ammonia and  chlorine.  Thus:

  17 g NH3   +  71 g C12  —->  51.5 g C1NH2  +  36.5  g  HC1

Since 23.66  g of  monochloramine  are  required  each  day,  the corresponding
amounts of ammonia and chlorine required are:

   Ammonia:   (23.66/51.5) x 17  =   7.81  g/day

   Chlorine:  (23.66/51.5) x 71  =  32.61  g/day

Dividing the  grams  of  each  reactant  by  454  (the   number  of  grams  per
pound),  gives a  daily  requirement  of  0.017  Ib  of ammonia  and  0.0715  Ib of
chlorine.   Multiplying  each  of  these  figures by 365   days  yields  the  annual
number  of  pounds  of ammonia  and  chlorine required.   Finally,  annual costs
for  each  are  calculated  by  multiplying  the   annual  requirements  by  the
current  costs:

   Ammonia:    0.017  Ib/day x 365   =   6.205 Ib/yr x $0.40/lb =
$2.48/yr

   Chlorine:   0.0715 Ib/day x 365  = 26.10 Ib/yr x $0.47/lb =
                                                  $12.27/vr

The  sum of these two numbers ($2.48   +   $12.27) = $14.75,  total  annual costs
for the  2,500 gal/day facility.

A  1  mgd facility will  require 400 times the  amounts of chemicals at the same
2.5 mg/L dosage, therefore:

   $14.75 x 400  =  $5.898. annual chemical costs.
2.  Costs for Monochloramine From Aqua Ammonia  + Chlorine Gas

From  the  preceding calculations,  the  2,500 gal/day facility will  require 6.205
Ibs/yr  of  anhydrous (gaseous) ammonia.   If the source  of ammonia  is  28%
aqueous ammonia, the calculation of costs is as follows:

   1 gal of 28% ammonia  weighs 8.34 Ibs and  contains 8.34 x 0.28 =  2.34  Ibs
   of ammonia.

   6.205 Ibs/yr  ammonia requires 6.205/2.34 = 2.55 gal/yr of aqua  ammonia.
                                       156

-------
At  $0.70/lb of ammonia contained in aqua ammonia, the annual cost of aqua
ammonia  is:

   6.205 Ibs  x $0.70/lb  =   $4.35/yr.

The annual  cost of  gaseous  chlorine  is  $12.27/yr,  therefore  the  total  annual
chemical costs are:

   $4.35   +   $12.27  =   $16.62/vr.

The 1  mgd  water  treatment  plant will  require  400  times  the  amounts  of
chemicals:

   $16.62/yr  at the  2,500  gal/day  plant x  400 =  S6.648/vr at  the 1 mgd plant.
3.  Costs for Monochloramine  From  Ammonium  Sulfate + Chlorine Gas

One  pound  of  ammonium  sulfate  contains  0.2576  Ib  (28.76%)  of  available
ammonia.   The  2,500  gal/day plant using  solid  ammonium  sulfate will require
6.205  Ibs  of anhydrous ammonia annually.   To  obtain  this amount of  available
ammonia  requires  6.205/0.2576   =  24.09  Ibs/yr  of  ammonium  sulfate.    At
$0.51/lb,  the  2,500 gal/day  plant will  require:

   24.09 Ibs/yr x $0.51/lb  =  $12.29/yr  +   $12.27 for chlorine =  $24.55/yr.

The 1 mgd  facility will  require 400 times  as much  chemicals,  or:
   $24.55 x 400  = $9,820/yr.


              3.    Oxidants

                   a.    Potassium Permanganate

Although   potassium   permanganate  is   not   recommended  as   a  primary
disinfectant  for coping with Giardia cysts or enteric viruses, it can be  quite
effective  as  a  non-halogenated,  preoxidizing  agent  for  removal  of   many
tastes  and  odors, colors,  iron,  manganese,  sulfide, nitrite,  and  many dissolved
organic materials.   As  such,  the use of potassium permanganate  in place  of
prechlorination  can  be  viewed  as  a  procedure  to  control  disinfection  by-
products.     The  major   supplier in  the  United States  is Cams  Chemical
Company,  P.O.  Box   1500,  LaSalle,  IL   61301  (815-223-1500).    Literature
describing this  versatile is  readily available from this supplier.

This  chemical  is shipped  as  a  solid,   dark  purple  in  color.    It  is readily
soluble  in water, and  solutions  of  1-4%  in  water are  recommended for  the
treatment of potable  water  supplies.   For example,  to  make  a 4%  solution
0.33  pound  of solid  potassium  permangante  are  added  per gallon of  water]
and the mixture  agitated  15-30 minutes using a 750-1000 rpm  mixer.
                                       157

-------
Addition of  permanganate  solutions  to the raw  water or in  the  rapid  mix is
recommended  for  water  treatment  applications.    This  is  because  as  the
permanganate  ion  performs  its  oxidative  work,  it  is  converted  into  lower
valent   manganese   compounds,   which   are   insolble.      Thus   potassium
permanganate is always added ahead of filtration.

To feed  1  mg/L of permanganate to the raw water using  a 4% solution, the
metering pump should be  set  to dispense 9.5 mL/min  for  every  100 gal/min
of flow.

Figure  40   shows   construction  costs  for  potassium  permanganate  systems
feeding  1   to  250  Ibs/day.    Figure  41  shows  operation  and  maintenance
requirements for potassium  permanganate  feed systems  — process energy and
maintenance material.   Figure  42  shows O&M  requirements  for  labor and
total  O&M  costs.    These three  figures  are taken  from   Gumerman  et  al.
(1986).
                                       158

-------
       (O
       O
       u

       z

       2
       H
       o
       D
       cc
       z
       O
       o
Tf*]
i
e
5
4
3
2
1(
^^•v
6
S
4
2
7
6
S
4
3
2
I
1(
5
3
2
1C
•
000.00






JCLOOfi
p^





I 000







Hiaiiinrre
)00







)0
0




















HBBB





























wo




























0*




























pi




























1C




























0a •>




























.--^




























<*"




















P0P


















































•**



UBMEK







I?



ra




















i

i
i
0^



























TT^P8^^^
^!^3L_ •


















™»»














P»







r^jn_L_
.^^T^

~


























i





















































































I






"T"














2 34S67S9 2 3 4 5 S 7Si 2 3 4 S 6 7SS
.0 10 100 1000
MAXIMUM FEED RATE • ib/day
1.0 10 100
                            MAXIMUM FEED RATE • kg/day
Figure  40.    Construction  cost   for  potassium  permanganate   feed  systems

             (Gumerman  et al.,  1986).
                                     159

-------
 UJ
 111
 u
 z
 <
 z
 Ul
 I-
 z
 "<
 2
10
jj
e
s
4
3
2
1(
I
6
S
4
3
2
1(
7
8
S
4
3
2
T
6
S
4
3
2
0.
00 U
: 1
N- 6
S
4
3
2
>0 _M
i- 1 i 1 1 1 1 II
• kwh/yr
|^ r» u * « o»~4!C







E


*'
































E MATERIAL
\^f\







NERGY
i rfi
^^
^^





















-^~
_^Jra








































































U™i— .









































































j
— -L
1





nn,




2 345 6789 2 3 4 S 6 789 2 3 4 i S 789
.0 10 100 1000
AVERAGE FEED RATE • Ib/tiay
"T~' ' ' \ ' • • ' J " " ' | • « P r i i v . , , .
1.0 10 100
                                AVERAGE FE£D RATE - kg/day
Figure  41.    Operation   and   maintenance   requirements   for    potassium
             permanganate  feed  systems  --  process  energy and maintenance
             material (Gumerman  et al., 1986).


                                     160

-------
      100,000    100,000
   V)
   o
   o

   2
   •0
   O
                                                   TOTAL O&M COST
                                                   (EXCLUDING CHEMICALS)
                         3  456789      2  3 496739      2  3 45370®
                                   10                 100              1000

                                 AVERAGE  FEED RATE • Ib/day
                      1.0
                            I
                            10
 I
100
                                 AVERAGE FEED RATE • kg/day
Figure 42.
Operation   and   maintenance   requirements   for    potassium
permanganate  feed  ststems  --   labor  and  total   O&M  cost
(Gumerman et al.,  1986).

                        161

-------
VIII.    CASE EXAMPLES OF EMERGING TECHNOLOGIES

Several  examples   of  the   use  of  disinfectants  other  than   chlorine  are
presented  in  this  section.    The   objectives   of   the   different  approaches
described  are  (1)  to attain  the  desired  degree  of  disinfection  and  (2)  to
minimize  the formation  of disinfection  by-products.    Some  of the  examples
illustrate  changes  in  disinfection/oxidation  techniques  to  cope  with  problems
in addition  to disinfection by-products,  e.g.,  taste and odor,  color, algae,  etc.


Case  histories  presented include  examples  of the use of ozone as a primary
disinfectant   for  controlling  Giardia   lamblia  cysts  at  one  plant  and  as  a
preoxidant for lowering THM levels at another.
Another  case  history involves  a  surface  water treatment  plant  which  has
installed  UV for  primary  disinfection, followed  by   chlorination  for  secondary
disinfection.

Other   examples   involve  uses  of  chlorine  dioxide   for  preoxidation  and  for
post-disinfection,   of  chloramines  to  lower  THM  concentration,  and   of
combinations   of   chlorine   dioxide   and   chloramine   to    lower   THM
concentrations.
   A.    OZONE CASE HISTORIES

         1.    North Andover, Massachusetts^

              Ozone Disinfection for Giardia lamblia

              a.    The Problem

During  early  1986, 18  cases of Giardiasis  were reported in the  North Andover
area.    On  May  1,  residents there were  instructed to boil their tap water for
at  least  five  minutes while  public   health  officials  sought   to  locate  and
correct  the  problem.

North   Andover's  water supply,   Lake  Cochichewick,  was  found   to  contain
Giardia  cysts  in  samples  taken 4n April  1986.   Local officials  confirmed  that
the lake's  watershed   has  a  large  muskrat  population,  and  there  has  been
concern  that   residential  septic  systems  were  not  working  well   enough  to
prevent harmful effects in  the lake.
           author  is indebted  to  Messrs. Ross  Hymen  and  Paul  Anderson of
the  Massachusetts  Department  of  Environmental  Quality  Engineering,  Mr.
Tom  Boshar  of  Lally  Associates,  the consultants  on  this project,  and  to  Bill
Nezgod  and  Dr.  Carl  Nebel,  of PCI   Ozone  Corporation,  suppliers  of  the
ozonation equipment, for providing information for  this  case history.
                                       162

-------
At  the  time of  the  Giardiasis  outbreak,  treatment of  North  Andover water
supply  involved  pumping of  Lake  Cochichewick  water  through two pumping
stations  into  the  distribution  system  without  filtration,  but with  chlorination.
Over the  years, high humic  concentrations have  resulted in the  formation  of
a  significant scale m the pipes, containing a  significant  amount of  biofilm  as
well.

Heavier chlorination overcame  the  immediate  Giardia problem, but caused a
second   water   quality   problem.    Descaling   in  the  old  distribution  system
piping  released  coliform  organisms  into  the water  supply.   The  presence  of
coliforms caused  the State  DEQE to  continue the  rtboil  water"  notice  until
the coliform organisms could be shown to be absent.

In  addition, THM  levels of  the  heavily chlorinated water  rose to above  200
Aig/L.    Also, residents  began  complaining   about  high chlorine concentrations.
Ozonation was investigated and was  found to have  promise.

Total coliforms  in Lake Cochichewick raw  water normally are  between  50  and
500  per 100 mL;  raw water turbidities  normally  are between  1 and 2 NTU.


              b.   The  Interim Solution

In  early  October, 1986,  two  ozonation systems  (ozone  generators, contactors,
and  exhaust gas  destruction)  began operating  at the  two  Lake  Cochichewick
pumping  stations  in response  to  the  emergency  situation.    Two  discrete
ozonation  systems, one  capable of  generating  50  Ibs/day  of ozone,  the  second
capable  of generating 150 Ibs/day,  were  rented from  the ozonation  equipment
supplier for a  period   of time.   Later, when the  efficacy  of ozonation  was
proven,  the town purchased  and  installed  two  150-lb/day  ozonation   systems
in the two pumping stations.

Therefore, as an interim solution,  the  lake water is being  treated with ozone
at the  two  pumping  stations.   Each  pumping station handles  2.5  to   3 mgd.
At  four  points  in the distribution system, chlorine  is   added  to  provide  a
residual  disinfectant.

The state DEQE  provided  emergency  funding  of $2.5 million,  which provided
for  the  ozonation systems,  for connecting  pipelines to  the three neighboring
communities, for relining pipes  with  cement,  and replacing water  mains.

Rental   fees for  the  two  ozonation  systems   totaled  approximately  $90,000.
The  two 150-lb/day ozonation  systems were purchased for $325,000.   This fee
included  the air  preparation  system,  ozone  generation,  diffuser  contactors,
and   ozone  destruction  equipment,   plus   the  appropriate   instrumentation,
including a monitor for  measuring  residual  ozone at the  outlet of the  contact
chamber.

These  two  ozonation systems were  installed  as a stop-gap  measure  to  control
Giardia  cysts until  a proposed $10.5 million  water 12-mgd plant is designed
and   constructed,  and   which  will  provide  complete   treatment,   including
filtration and ozonation  (in mid-1989).

                                       163

-------
The two  ozone contacting  chambers (each  10  ft  wide x  20  ft  long, with  a
16-ft  water depth)  are  designed  with baffles, to  provide  each contactor with
five complete  mix  ports,  with ozone being  applied  equally  in  each  section.
Thus,  there is plug  flow  throughout the  ozone  contactors.   Applied  ozone
dosages are 5  mg/L.

At  the  outlet of  each contact  chamber,  the  concentration of  ozone  lies
between 0.9 and  1.0  mg/L.   System  designers assume  that  the average  ozone
concentration   in each contactor is  0.5  mg/L.    Total  residence time of  water
in  the  ozone  contact chamber  is  10 minutes  at  full  pumping  rate  (summer
time).   During winter, with lower  water demand,  pumping rates are  reduced
by  50%,  thereby  doubling the residence  time in  the contactors to  20 minutes.

Water temperatures of the lake vary from about 5°C  in winter  to  just  under
20°C  during summer.    Therefore,  the  appropriate  CT  values listed in  Table
IIA for  99.9% inactivation  of Giardia lamblia range from  3 to  1.5 mg/L-min.
Assuming  the  average  dissolved  ozone  concentration  of  0.5  mg/L  carried
through the 10 minute contact  period, the summertime CT value  attained is  5
mg/L-min.

During  winter, when  the  contactor  residence   time  is 20  minutes,  the  CT
value attained is  10  mg/L-min.   Both of  these  CT values  are well in excess
of  the  3  to 1.5 mg/L-min  required to guarantee 99.9%  inactivation  of  Giardia
lamblia and 99.99% inactivation of viruses.

No  filtration   is  provided  during  the  current   interim  period.    Instead,  the
community  is  placing total reliance  on  ozone   for  primary  disinfection  plus
secondary disinfection  with chlorine.


              c.     The Results

After  approximately  90-days   of  ozone  treatment,  the State  DEQE  uncon-
ditionally  lifted  the  boil  water  order,  which  had been   in   effect for  nine
months.   Not  only are   Giardia  cysts absent  from  the North Andover  water
supply,  but  also coliform  organisms.

In  addition,  several  additional benefits  have been  obtained  as a result  of
installing  ozonation.   Prior  to ozonation being  installed, THM values were in
the range  of  8 to  120 /ig/L.   Since ozonation has  been installed, THM values
now are in the range of 1.1  to 2 A^g/L.   In  addition,  the  color of the treated
water has  improved significantly (65% to 95% lower).   Finally, taste and  odor
levels in the finished waters are greatly improved.


              d.     For The Future

As  indicated   earlier,   a   new  12-mgd conventional  treatment  plant, including
ozonation  and  granular  activated  carbon   adsorption,  has   been  designed.
Preozonation   will   be applied  before  the  rapid   mix.    After  dual  media
filtration,  GAC  adsorption  is   incorporated,  followed  by  post-chlorination for
residual.    In  addition  to  providing  an  adsorption  capability,  the  GAC  step

                                       164

-------
also  will  allow  biological  decomposition  of  the  easily  biodegraded  organic
fractions of the water present at that treatment point.

Bids  were let for this plant in the summer  of 1987;  the plant is  scheduled to
be  on-line by mid-1989.


              e.    Note - Sturgeon Bay. WI

A  second plant  is  installing  ozone  specifically for controlling  Giardia lamblia
contamination in  Sturgeon Bay,  WI.    This  is  a  groundwater plant,  without
filtration, located  on  a  Lake  Erie  shore, which  has been shown  to contain
Giardia  cysts.  Water flow at  this treatment plant is 100,000  gal/day.   Water
temperature  is about 10°C the year round.

Two  4-minute  residence  time ozone  contactors  are being  installed  through
which the  water  flows   sequentially.    After  contacting, the   ozone-containing
water will be stored in  a  filling  well.   Thus, the  total  ozone  contact time is
estimated to  be  10 minutes.    Dissolved ozone concentrations  at  the outlet of
both  contactors will be adjusted to 1  mg/L.

Thus the CT value in the second ozone contact chamber  will  be 4 mg/L-min
(4  minutes  times  1 mg/L), plus  a conservatively  estimated  2 mg/L-min  in  the
first  chamber (0.5  mg/L times 4 minutes),  plus  an  additional  1  mg/L-min in
the  filling  well,  during   ozone  decay  from  1  mg/L.    The   total  CT  value
designed  is  7 mg/L-min.   Table IIA  requires a  CT value of only  2.5 mg/L-min
for 10°C.

The  retrofitted  ozonation  system is  expected  to  be operating  in  April/May
1988.
         2.    Kennewick,  Washington (Cryer, 1986) -
              Preozonation For THM Control

              a.   The Problem

Prior to  1977,  the City of Kennewick  had been drawing  essentially  all of  its
municipal water supply from  a system  of  five Ranney  collectors  located  along
the  Columbia  River,  followed by  chlorination.   When  initially installed,  these
were capable  of producing approximately  20 mgd; however,  their  output had
deteriorated  to about  15  mgd by  1977.  By  1978, the  maximum daily system
demand  essentially  had reached  the capacity of  the  Ranney  system.   It was
determined  that direct utilization  of  the Columbia  River would be the  only
reliable source of supply for long term development.

A consequence  of  this decision  was  the need to  provide consider-ably  more
treatment to  achieve  the  same  or better finished water quality.   Thus  a  pilot
plant  study  was  undertaken  to   test   alternative  water  treatment  processing
steps.    This   study  included  the  use  of  preozonation and   of  post-filtration
GAC adsorption, in  addition to  conventional  and  direct  filtration  procedures.


                                        165

-------
Average  values  of the  water quality  parameters of  the raw  Columbia River
are:
        TTHM                   1
        TTHMFP*              136   /ig/L
        TOC                     2.4 mg/L
        No.  Particles         11,650     /mL
        Particle Volume      160,700    nanoL/L
        Turbidity                  1.7 NTU

        Chlorine contact  time:  1 week
              b.    Pilot Plant  Study  Results

Both  the  preozonation  and  coagulation/filtration  steps  provided approximately
30%  reduction in  levels  of TTHM  formation potential and 10%  reduction  in
TOC  levels.    The  combined  processes  gave  approximately  60%  reduction  in
levels   of   TTHMFP   and   20%   reduction   in   TOC   levels.       The
preozonation/coagulation/filtration   processes   was . determined   to   be
operationally   equivalent  to  activated  carbon  adsorption  for  the   removal  of
THM precursors;  it was  also  determined to be the most cost-effective method
of treating organics in the Columbia  River water supply.


              c.    Plant  Design

Approval  was given for  a  30  mgd water treatment plant to be  constructed  in
four  stages  of  7.5 mgd  each,  based  on  the following  processing sequence:
preozonation/flash   mix/coagulation/flocculation/filtra'tion   and   post-chlorina-
tion.    Design  criteria   for  the  preozonation facilities   are  given in  Table
XXXIII.

At  the  point of  application, ozone dosage  rates were 1.5 mg/L (average)  and
4.0 mg/L  maximum.    The  contactors provide  10 minutes  of  detention time.
Raw  water  total  coliform  levels  are  consistently  less than 50  per  100  mL.
Raw water turbidities are in the  range of 1.5 to 2.0 NTU.


              d.    Operational Experiences

                   i.     General

The  new  7.5 mgd  treatment  plant  currently is  operated  from May  through
October of each  year, when system demand  exceeds  10 mgd.   For  the balance
of the year,   water demand is  about  8 mgd.   This can be  satisfied by  the
Ranney   collector  system,   which  operates  at  lower   cost  than  the   new
treatment  plant.
                                       166

-------
TABLE XXXIII. DESIGN CRITERIA FOR KENNEWICK WATER
            TREATMENT PLANT PREOZONATION FACILITIES
            (Cryer, 1986)
U.S.
Item Units
Plant Capacity
Design Flow mgd
cfs
gpm
Ozone Contact Basins
No. of basins
Detention min
time @ design
flow
Basin dimen- ftxft
sions-inside
Av. water ft
depth
Basin volume ft-3
gal
Total basin ft3
volume
Metric
Initial Ultimate Units Initial Ultimate

7.5
11.6
5,200

4
10


14x8

16
1,792
13,400
7,168


30.0
46.41
20,800

16
10


14x8

16
1,792
13,400
28,672


ra3/day
rn3/sec

28,300
0.33
m3/min 19.70


min


m x m

m
m3

m3


4
10


4.2x2.4

4.8
50.75

203


113,300
1.33
78.62

16
10


4.2x2.4

4.8
50.75

812

Chemical Feed Rate (Max. Dosace (3) Design Flow"!
Ozone plant Ib/day
influent
Chemical Feeders
Ozone Ib/day 2

250


x 125

Chemical Storage Capacity (Max
Ozone No. 03
generators
Total Ibs
2

250
1,000


3 x 250
2 x 125
. Dose (a)
5

1,000
kg/day


kg/day

113


57

454


3 x 113
2 x 57
Design Flow")
No. O3

Total kg
2

113
5

456
                           167

-------
                   ii.    Water Quality

Kennewick's  new  Columbia River  Water  Treatment Facilities  came  on-line  in
1980.  Generally,  the City and the service  area customers  have been satisfied
with  the  quality   of  water   provided.    Treated   river  water  has   a  higher
customer  acceptance,  judging by   the  limited  number  of  complaints,  than
Ranney collector water alone.

THM  analyses  of  the  new  treated  water  and  chlorinated  Ranney  collector
water  indicate  that  the  plant   has  been  able  to  maintain  the  TTHM
concentration  below  the   current  100  /ig/L standard, while  the  quality   of
water  from  the Ranney  collector  system  remains  very  similar to  the levels
determined  during  pilot plants studies, for  raw^  .Columbia  River^ wate.r.  The
City's  Ranney  collector wa^rs" have  had "an  average THM'concentration   of
approximately 107  Aig/L, while the  ozone treatment plant  water  has averaged
approximately 14 /xg/L.


                   iii.   Applied    Ozonation   Dosages   -    Dissolved   Ozone
                        Residuals

Applied  ozone  dosage rates  have  ranged  from  1.7 to 2.5  mg/L.  Until 1983,
ozone residual  levels  were maintained  at  approximately 0.5  mg/L exiting  the
contactor.   In  1983,  the  City installed  a  dissolved ozone  analyzer  to  control
the   ozone   dosage,  which  has  resulted   in   lowering  the   dissolved  ozone
residual  concentrations  to  0.1  mg/L,  thus  saving ozone,   and  still  control
biological growth in the  filters and basins prior to chlorination.


                   iv.   Ozone Equipment Operational Experience

Operationally, the  ozone generation equipment  has  performed very well.  The
compressors   have  required   only   preventive  maintenance.     The   ozone
generators have required  the replacement   of  only  three  burned  out tubes
during  the  first  six  years  of  operation.     The  major  maintenance  problem
appears  to  be  tube fouling which  was  found to be a result  of high moisture
in   the   feed  gas.     This   situation  was  caused   by   two  extenuating
circumstances.   After several  years  of  operation,   it  was  discovered that  the
refrigerant air  dryer  unit had  developed  a  small leak  which   reduced  the
effectiveness  of the air  preparation system.    It was then determined that  the
absorptive   medium   in  the   desiccant   drier  must  be   replaced  when   its
regeneration  capacity  is  reduced to  40%  of its originally  specified  capability.
Cleaning of dielectric tubes has become an annual maintenance  procedure.

The  only  other   significant  operational   problems  concerned  the   ozone
contactors.    Excess foaming  and  scum  production can  occur during  spring
and  late  summer  (algae  destruction,  primarily).    This   may  require  the
installation of surface skimmers  and  froth spray equipment.   In addition,  the
stainless  steel tubes holding  the  ceramic  diffusers  corrode  after about  two  to
three years  of  use, and must  be  checked  and  occasionally  replaced  when the
diffusers  are cleaned.
                                       168

-------
                    v.    Costs Of Ozonation

 The  Kennewick  facility is  able  to  produce  ozone  at  approximately  $0.20/ib.
 However,  Cryer  (1986)   notes   that  the  ozonation  system  has  never  been
 operated  at   100%  of  the  high voltage  output,  and  that ozone  generation
 efficiency  has  averaged   approximately   16-17   kWh  per   pound  of  ozone
 produced.

 Based  upon  the City's very  limited  experience in  operating  the  facility during
 periods  of time  when  the  ozonation   system   has  been  off-line,  the  plant
 operator  believes  that  the  addition  of  ozone  prior to  flocculation/filtration
 results in a savings in  flocculation chemicals of approximately 10%.

 Currently,  the  treatment   plant  is  producing finished  water  at  a  cost  of
 approximately $14.10 per million gallons ($0.141/1,000 gallons).


               e.    Implications  of  Proposed  Surface  Water  Treatment  Rule
                    Disinfection Conditions

 During  the May/October  period of  plant operation,  the  temperature  of the
 waters  treated  is approximately  20°C.   With a  10  minute ozone  contacting
 time,  and  assuming an  average  residual  ozone concentration  of  0.5 mg/L (as
 was the  case until  1983),  the CT value  of  5 mg/L-min was  being  obtained,
 compared with a value of  1.5 mg/L-min required (Table  IIA)  by the  proposed
 SWTR for 99.9% inactivation of Giardia lamblia  cysts and 99.99%  inactivation
 of enteric  viruses at  20°C.   On  the  other hand,  because  the  Kennewick  plant
 filters   (thereby  providing  2-logs  of Giardia inactivation),  only a  single  log  of
 additional  Giardia  inactivation  is required.   This   means  (Table  Iffi) a CT
 value  of less  than  0.7 mg/L-min, which  corresponds to an ozone contact  time
 of only 1.4 minutes at  the  same residual  ozone concentration.

 After  filtration,   gaseous  chlorine is   added.    A  residual  of  0.6  mg/L  free
 chlorine  is  maintained for 60  minutes  at pH  of  approximately  8.0.   The
 contribution of  chlorination to the required  disinfection  CT value  thus  is  36
 mg/L-min.  From Table  IIA,  the  CT  value for chlorine at pH 8.0  and 20°C is
 101 mg/L-min.

 Therefore,  maintenance  of  the  0.5  mg/L  ozone residual  for  10  minutes
 provides  over 300%   of  the  required  disinfection  capability,  without  reliance
 on  the    post-chlorination.     In   addition,  post-chlorination   provides   a
 supplemental 36% disinfection credit (36/101).

 However,  since   1983   the  ozone  residual concentration  leaving   the  ozone
 contact  chambers  has been  reduced  to  0.1 mg/L.     Assuming   this ozone
 concentration   is   present  throughout  the  10  minutes  of  ozone  contacting,  a
 CT value  of  only 1  mg/L-min is obtained, versus  the 1.5  mg/L-min required
by  Table  IIA.    This  means  that  only  67%  of the  disinfection capability is
 provided  by the  preozonation step  to achieve 99.9%  inactivation  of  Criaidia.
but 130%  of  the inactivation required  to obtain  90%  additional  inactivation
because  of the  filtration capability.   The  60  minute  chlorine contact  time  at
pH 8.0  (0.6  mg/L)  provides an  additional  36%  of the   required  disinfection.

                                       169

-------
This means that the  plant  currently provides at least  100% of the  amount of
disinfection  stipulated  under  the  proposed  Surface   Water  Treatment  Rule,
disregarding the contribution of filtration.


   B.    UV RADIATION  CASE HISTORY

         1.    Ft. Benton.  Montana 2

              UV Radiation for Primary Disinfection

              a.    The Problem

The City of Ft. Benton,  MT takes water from  the Missouri River  and  treats
it  in a  20-30 year  old  filtration  plant.   This  plant  needed to  be upgraded.
Costs for installing a new  filtration  plant were deemed  to be excessive.


              b.    The Solution

A new 2 mgd peak flow treatment plant was installed in  mid-1987.   Water is
drawn  through  Ranney  collectors  installed  20-25  feet  below  the  river  bed.
This  allows  the  river  bed to  filter  the  raw  water.    Turbidities  of  water
entering the treatment plant average  0.08 NTU.   No  Giardia cysts  have been
found in the filtered river  water.

Plant  intake  water  is treated  with   UV  radiation  for  primary disinfection,
then  chlorinated  for  residual.     If  chlorination were  to be  relied  on  for
primary  disinfection, a 2-hour contact  time  would be required  by the  State
authorities.   Even a  30-minute  contact time  would require two-stage pumping
with long concrete  pipes to  provide sufficient contact time.

After primary  UV  disinfection,  the water is  chlorinated to a  1 mg/L  residual
with  no  contact  time.    The water being  chlorinated  is  quite clean,  in that
the  applied chlorine dosage is just over  1  mg/L.  The pH  is about 7.4 and
the water temperature  is approximately 20°C year around.

The entire water treatment process is housed in  a 32 x 32 ft building.
            author  is  indebted to  Fred  Zinnbauer  of Aquionics,  Incorporated,
Erlanger,  KY (supplier  of  the UV radiation  equipment)  and to Gary Swanson,
Project  Engineer  at  Robert  Peccia Associates,  Helena, MT,  consultants  for
this project, for information on this plant.
                                       170

-------
              c.    UV  Radiation Conditions

The UV disinfection system  consists  of six irradiation chambers,  two  control
cabinets  with  alarms,  chart  recorders,  relays,  hour-run  meters,  lamp  and
power  on-lights,  six  thermostats,  electrical  door  interlocks,  mimic  diagrams,
and six UV intensity motors measuring total  UV output.

Four  irradiation  chambers  are  on-line  at  all   times,  with  two   chambers
reserved  for backup.   The  alarm  system  is  tied  into  the  automatic butterfly
valves  to provide  for  a  fully  automatic  backup  system.    Each  irradiation
chamber  contains  one  2.5  kW mercury  vapor,   medium  pressure  arc  tube,
generating UV radiation at 253.7 nm.

The initial  UV  dosage is  41,000 jiW  sec/cm2 at  maximum  water flow  (1,650
gal/min)  through  each  irradiation  unit.    Expected  arc  tube  life  is   4,500
operating  hours  providing  a  minimum  UV  dosage  of 25,000  /iW sec/cm^.
These  conditions  are designed  to  reduce  concentrations of  E. coli  organisms
by  a minimum of 5-logs (1(P reduction).

The  UV  irradiation  system  is  interfaced   with  the  telemetry  control  system
activated  when  the  low  tank  pump  start  set-point  is activated  on the  set-
point  controller.    Sufficient  warmup time  is  maintained before  pump  startup
is initiated.  The UV system  is deactivated  once  the  high tank  stop set-point
is  activated   on  the  set-point  controller  and   the  pump  stops  running.
Sufficient time delay is built into  the  system  to  ensure  that water flow  is
completely stopped before the UV system is  deactivated.

A fully automated backup  system is  provided.   Each  bank of  three  irradiation
chambers has  two  units  on-line at  all  times, with  the third  unit serving as
backup.  In  the event  that  the UV  intensity  drops below  acceptable  limits
(20,000 /tW sec/cm2)  in  any  of the chambers,  the  automatic butterfly  valve
will  close,  stopping  flow  through  the  chamber;  the  automatic butterfly  valve
on  the  standby  unit will  open.   The  alarm system  also  is activated  if  UV
intensity  drops below  acceptable limits  in  any  of the  chambers.    The  UV
alarm   system  is  interfaced  with   the   automatic   dialer   and  alarm system.
Replacement of  the  UV lamps is quite  simple,  requiring no  more than  a few
minutes.


              d.    Costs

Total  equipment  costs  for  the 6-unit   UV  irradiation  system with butterfly
valves  was $74,587.


              e.    Operating Experience

The new  Ft.  Benton  water  treatment plant  started  up  in July   1987;  no
operational data are yet available.
                                       171

-------
   C.    CHLORINE DIOXIDE CASE HISTORIES

         1.    Evansville. Indiana (Lykins & Griese, 1986)
              Predisinfection for THM Control

              a.    The Problem

Because of  the  November 1979  amendment to  the  National  Interim  Primary
Drinking  Water  Regulations (U.S. EPA,  1979), the Evansville  (Indiana) Water
and  Sewer  Utility  was  faced with  reducing the  levels of  trihalomethanes in
their  finished   drinking   water.     At  the   time,   raw   water  was  being
prechlorinated  and  treated with  alum  prior to  primary settling,  treated  with
lime  (to  pH =  8), passed through secondary  settling, fluoridated, rapid  sand
filtered,  then post-chlorinated  before passage into the  clearwell.   The  treat-
ment  process  was  conducted  in  two  separate  30  mgd treatment lines  (total
production 60 mgd), and THM levels exceeded  the THM standard of  100 /ig/L.
Prechlorination  doses  averaged 6 mg/L.   Distribution  system  residence  time
averages three days.


              b.    Pilot Plant  Study

With  the  assistance   of  a  cooperative  agreement  from   the Environmental
Protection  Agency  (CR811108 -  Sept.   1983),  the  Evansville  utility  initiated
research to  evaluate  the use  of  chlorine dioxide  in  a  100-gal/min pilot plant
adjacent to  the  full-scale  plant.   One  train of  the  full-scale  treatment plant
served as the control for the pilot plant study.

The  pilot  plant  study  was  conducted  in two  phases,  optimization,  and long-
term.    Chlorine dioxide  was substituted  first  for  prechlorination,  then  for
post-chlorination.    GAC  adsorption  also  was  evaluated  prior  to  the  post-
chlorination step.


                   i.     Optimization Phase

When  predisinfection   was  eliminated   and  the  pilot  plant  effluent   was
disinfected  and  stored  three  days  (to  simulate  distribution  system  residence
time),  TTHM   concentrations   averaged   141  ju,g/L  with   chlorine  post-
disinfection  (2.5  mg/L  residual  chlorine concentration),  and   1.2 /ig/L  with
chlorine  dioxide post-disinfection (1.9  mg/L residual).

Predisinfection  with chlorine  dioxide to  maintain  a  residual  through  the  pilot
plant  did  not  increase   the  THM   concentration   and   provided  adequate
disinfection.   The C1C>2 residual decreased  from  4 to  0.3  mg/L through  the
pilot plant.


                   ii.     Long-Term Evaluation

In this phase  the  pilot plant  procedures were evaluated during  each season
of the  year to  determine  the extent  of seasonal effects.   In this  phase of

                                        172

-------
the  study,  predisinfection  with  chlorine  dioxide  (1.1  mg/L  average  applied
dosage)  reduced the  amount  of THMs  formed  during  post-chlormation  by
approximately  60%.   The  concept  of  post-disinfection with  chlorine  dioxide
was abandoned  because of  the  difficulty in  maintaining a disinfectant  residual
in  the  distribution  system  (three  days average  residence  time),  keeping  in
mind  that  the   current  EPA   recommendation  is  that   total  oxidants  from
application  of chlorine dioxide   should  not exceed  1 mg/L  in  the distribution
system.   The  average chlorine  residual  concentration in the  clearwell was  2.1
mg/L.


              c.     The Full-Scale Plant

Based on data  obtained  from  the  pilot plant  study, Evansville  utility  officials
concluded that the  MCL  for  TTHMs  would be  exceeded if the then-current
treatment  was  not  altered.   Changing the  primary disinfectant from chlorine
to  chlorine   dioxide  was  judged  to  be  the  most  effective   procedure  for
control at the  least cost.    A separate building was  constructed  to  house  the
chlorine   dioxide   generation  facility,      One   portion  contains  the   C1O2
generator, a  second portion houses two 1-ton  cylinders of chlorine.

This  installation  is  capable  of  generating  14.24 Ibs  of C1O2 per hour, which
can be  divided  in  any  proportion  between  the  two halves  of the  treatment
plant.    Gaseous  chlorine  and  25%  NaClC>2  solution  are   delivered  to  the
chlorine  dioxide  reactor under partial vacuum generated by  an eductor.   Both
reagent  flows  are  controlled by flow-rate  meters,  and  the system  is designed
to  shut   down  if  the  eductor   water  supply fails  or  if  chemical feed  lines
break.

This  production  rate has  been  achieved  with  over 95%  conversion  to  C1C>2
over the first  18 month period of operation.


              d.     Operating Experience

Chlorine  dioxide  began   being   added  as  a  predisinfectant   prior  to  any
additional chemical  treatment in  August  1983.   The C1O2 produced is  divided
evenly between  the  two   treatment  lines.    During the  first five months  of
use,  various   CIO? dosage  levels  were  used  to  determine   the  resultant
reductions  in  THM  concentrations, to  gather data  on what  percentage of  the
C1O2  dosage  would appear as  total  oxidant residuals  in the  finished  water,
and to review the general  operation of  the  entire system.

During   this   period,  only  one  major  problem   was  encountered.     Brass
corporation  cocks used  to  connect the PVC  C1C>2 feed lines  to  the raw water
influent  piping   were  oxidized  by   the   concentrated   disinfectant.    This
oxidation  and subsequent  leaking  of  C1O2 solution  resulted  in  temporary
disruption  of  the  new treatment  technique.    The  problem was  resolved by
sliding  a  section  of PVC  pipe   through  new corporation  cocks  into  the  main
stream of the raw  water  lines.   This modification permitted  the  PVC  piping
to  serve  as  an   inductor  while  preventing  direct   contact   of the   brass
corporation cocks with  concentrated C1O2 solution.

                                       173

-------
Since the  implementation  of the  treatment  change,  total  oxidant levels  from
chlorine  dioxide  have  been   maintained  consistently  below   the   1.0   mg/L
recommended by  the U.S. EPA.   With  an average  applied C1C>2 pretreatment
dosage  of  1.2  mg/L,  total oxidant  concentrations in  the  finished water  have
averaged  0.5  mg/L.    These data  show  that  approximately  42%  of  the  QO£
dosage is present  as  total oxidant.

Since  the   installation  of  predisinfection with  chlorine  dioxide  at   Evansville,
annual TTHM  levels in  the distribution  system have been  maintained  between
50 and  80
              e.    Implications   of  the   Proposed   Surface   Water  Treatment
                   Rule

Raw  water  total coliform levels at Evansville are  in the  range of  3,400 to
5,400 per  100 mL.   The filtration  system provides  2-logs  of Giardia and  virus
inactivation.    However,  because   of  the   higher   raw  water  total   colifonn
levels,  EPA  recommends  (U.S.  EPA,   1987d)   that  primary  disinfection  be
provided  to attain  99.9% inactivation of Giardia and 99.99%  inactivation of
viruses, in  addition to those provided by filtration.

Chlorine dioxide  predisinfection  at Evansville  employs  a   dosage  of  1.2  mg/L,
followed by chlormation  in  the clearwell at  an  average  concentration of  2.1
mg/L.   Although  the  contact  time after  adding C1O2  is  not  stated  in the
reference,  attainment  of  the   required  CT  value   of 21  mg/L-min  at  20°C
(Table  IIA)  and  an  average ClO^  concentration of  1.0 mg/L means  that  a 21
minute  contact  time  would  provide  99.9% inactivation  of  Giardia  cysts  and
99.99% inactivation  of  viruses,  independently of the added  degree of  disinfec-
tion  which  can   be   credited  to  post-chlorination.     Since  the  average  total
oxidant (from chlorine  dioxide) concentration  in  the  plant finished  water  is
0.5 mg/L,  an  actual  CT value   of  10.5  mg/L-min  would  be  expected in  a 21
min  contact  time.    This represents  about 50%  of  the required  degree of
primary disinfection.

With  respect  to post-chlorination,  at  an average  2.1 mg/L chlorine  residual
in the  clearwell  at  pH  8 (CT  value  =   101 mg/L-min at 20°C)  means that a
contact time  of  48  minutes would  be required to  provide  99.9% inactivation
of  Giardia  cysts and  99.99%  inactivation   of viruses.    If  only  a  30-minute
clearwell  contact time  is  provided, still  63%  of the  required  disinfection  is
provided  by   chlorine  in the   clearwell.    Therefore, it  is   likely  that  the
required amount  of  disinfection, even  with  the  high  level of raw water  total
coliforms  present, is  attained  from  both  the chlorine  dioxide  and  chlorine
disinfection  steps.
                                       174

-------
         2.    Hamilton.  Ohio (Augenstein,  1974; Miller et al.,
              1978; U.S. EPA, 1983)
              Primary/Secondary Disinfecting Wi^ Chlorine
              Dioxide

              a.    The Problem

 In  1956  a  15  mgd  lime  softening  plant  began  operating  in  Hamilton,  OH
 treating  groundwater  (18  wells, 200  ft  deep).   Chlorine was used  as  the  sole
 disinfectant when  the  plant  started  operating.    However, because of  customer
 complaints  of  chlorinous  tastes  and  odors,  a  change  was  deemed necessary.
 Chlorine  dioxide was  tested,  and  in  1972  this material  was  substituted  for
 chlorine for primary and secondary disinfection.


              b.    The Treatment Process

 Hamilton's  treatment process  in  1987  is as follows:    aeration, lime  addition,
 flash  mixing,  sedimentation,  recarbonation  (with food  grade  CO?),  filtration,
 fluoridation   (sodium   silicofluoride),   disinfection   (with   C1O2J,   clearwell
 storage.    Raw water  turbidities  are  below  1  NTU,  and  raw  water total
 coliforms are  less than 1 per  100  mL.


              c.    Generation of  Chlorine Dioxide

 Chlorine  dioxide is generated  by  mixing  37%  aqueous sodium  chlorite  and
 aqueous  chlorine  in  a ratio  of  1:1 by  weight  (2 Ibs of  each  reagent  per
 million gallons  of  water  to  be  disinfected).   This provides  an  applied  C1O2
 dosage  of  0.25  mg/L.    Chlorine  gas  is  delivered   to  the   site in   150-lb
 cylinders.  Aqueous NaClO2 solution  (37%) is delivered in 200-lb drums.

 The   chlorine  dioxide  generation   system   consists  of  one   plant-fabricated
 reactor vessel for  C1O2  production,  one peristaltic  pump for NaClO2  solution,
 two  chlorinators  (one  serves  as  standby).     Two  150-lb   liquid  chlorine
 cylinders  are  positioned next  to the  chlorinators.   The  weight of  the cylinder
 contents is  measured by  a scale.   Switchover  from one cylinder to the  other
 is manual.   PVC  tubing is specified between  the chlorinator and C1O2 reactor
 vessel; heavy  Tygon tubing transports the  NaClO2 solution  from the  drum to
 a  small  plastic day-tank and  to  the  reactor vessel.  This  Tygon tubing loses
 its  rigidity and  must be  replaced  after about  a month of use.    The  semi-
 transparent  day tank allows  visual inspection  of  the  level  of  liquid  NaClO2
 and thereby   enables the  operator to maintain  an  acceptable  suction  head  on
 the peristaltic pump.

The chlorine  dioxide reactor  vessel is constructed  of  Schedule  80 PVC  piping,
 18  inches  high and approximately 6  inches  in  diameter.  The  vessel is filled
with PVC rings, 1-inch in  diameter.   The  chamber is  opaque  except for  the
sight   glass  mounted  in-line   on   the  discharge  piping.    A  white  card  is
positioned behind the sight glass for better observation of the C1O2 color.


                                        175

-------
Applied  dosage  of  C1O2 in the  clearwell is  0.25  mg/L.   The residual  leaving
the clearwell  is  approximately  0.15 mg/L, and  is 0.10  mg/L  at  the extremities
of the distribution system.


              d.   Effects of Installation of Chlorine Dioxide

Prior to  installation of  C1C>2,  customers  had  complained  about  brown stains
(iron)  during washing  of clothes.   When  C1C«2 treatment was initiated,  brown
slimes were  loosened  from  the  mains.   Plant  personnel  then  flushed  the
distribution   system.      Shortly   thereafter,    the   brown   stain   problems
disappeared, and complaints ceased.   Plant  personnel attributed  the source of
the  problem  to  crenothrix  and  leptothrix  bacteria  (iron  bacteria)  that  had
been present in the  extremities  of  the distribution  system before introduction
of chlorine dioxide.

Since 1972,  C1O2  has been the  sole disinfectant in Hamilton's  drinking  water.


              e.   Costs  for Chlorine Dioxide

In  1977,  the  costs for C1O2 in  Hamilton were  determined (Miller et al., 1978)
to  be about 3.60/capita/year.    Chlorine  and  NaQO2  together  cost  $6,540^rr
(1977).     Total chemicals   costs  in   1977  for the  finished water  averaged
190/1,000  gal;  the  fraction due  to   CIO?  was considered   negligible.    The
operating  and   maintenance  costs for   CIO?  generation  at  Hamilton were
estimated  to  be less  than $50/yr  in  1977.   The plant-fabricated  C1C>2 reactor,
piping,  hardware,  and  installation were  estimated  by  the  plant  supervisor  to
cost  around  $400  in  the  1977 market.    Installation  was  done  by plant
personnel.   The peristaltic  pump for  NaClC>2  solution cost less  than  $200
(1977).    To this  must  be added  the  cost of the  two  chlorinators  ($600),
which were already  on line at the  plant.


              f.    Implications  of the Proposed SWTR  CT Values

Although the Hamilton raw water is groundwater,  and therefore  probably  will
not  be  subject  to  the requirements of the proposed  Surface  Water Treatment
Rule (U.S.  EPA,  1987a),   it   is  interesting  to  consider  the  effects  if such
disinfection requirements  as listed in  Table IIA  were to  be  levied  on  this
water supply system.

Chlorine  dioxide  is  added  to  the Hamilton water in applied  doses  of 0.25
mg/L as  it enters the  clearwell.  The  water  temperature  is  about 20°C year
round, and the  pH  is  9.4  to  9.5.   Hamilton's  first customer is  located about
0.5  mile  from   the  plant.    Thus  there is very  little  contact  time  in  the
distribution system.

The plant  filters  efficiently, and therefore only  1-log  additional  inactivation
of  Giardia  cysts  and  2-logs inactivation  of viruses  need  be  provided  by  the
chlorine  dioxide.   Table  IIB shows  that  at  15°C, a  CT  value  of 9 mg/L-min
will provide the required degree  of disinfection.

                                        176

-------
The Hamilton  clearwell  holds 500,000 gallons.   During  periods  of high  water
use, water  is produced at the rate of 19  mgd.   In periods of low  water use,
only 8 mgd are  produced.   Thus  the  contact  time in the  Hamilton  clearweiJ
ranges  from  30  to  90  minutes,  at  the  high  and  low  production  rates,
respectively.    Assuming   that  the  average  concentration  of  C1O2  in  the
clearwell  is  0.2 mg/L, then  the  CT values provided  are 6  and  18 mg/l^min,
respectively.

Thus  the  current  disinfection  conditions  using  chlorine   dioxide  meet  the
proposed  CT   requirements,   except  at  the  most  rapid  production  of water,
when  only  two-thirds  of  the disinfection  capability is provided  by CIO?  ui
the  clearwell.   However,  simply  increasing the  C1C>2 dosage by  one-third  (i.e.,
to  0.32-0.33 mg/L) during periods  of  highest  flow will provide  the additional
amount of  disinfection  required by the proposed  SWTR.


         3.    Galveston. Texas (Myers et al.,  1986)

              Preoxidation  With CIO?
              Post-Disinfection With C1O2  + Chloramine

              a.   The Problem

The Galveston  County Water Authority (GCWA) owns  and operates an  18 mgd
capacity water  treatment  plant in Texas  City,  TX, currently (1986) producing
12  mgd of finished  water from  the Brazos  River.  Raw water  characteristics
include high color,  variable  turbidities  (68-111  NTU), high organic contents,
high   iron   (2.7-3.8  mg/L),   and   seasonally  high  algae  content,   sometimes
reaching  levels  of 5,000  blue/green  algae/mL.   Such  high  algae  and  organic
contents  create the  potential for  unpalatable   tastes  and  odors  to   develop
during treatment and distribution.

Total  trihalomethane  formation  potentials  for  Brazos River water,  measured
periodically  during  September 1983 through  April  1984  (20-day chlorination
period),  were   in  the  range  of  400  to  650  /u,g/L.    Finished  water  THM
concentrations  measured  during the same  period were  in the range  of 180  to
350  ug/L.    During   periods  of  intermittent   raw water  algae  blooms  and
associated  high organic carbon contents,  numerous consumer  complaints  were
received regarding  tastes and odors.

These   problems  prompted  an  investigation   of   alternative   disinfectant
strategies.


              b.   The Original Treatment Process

The  treatment  process  included  addition  of  cationic  polymer  for   primary
coagulation,  lime  for  pH  adjustment,  prechlorination   for taste   and   odor
control,  and ferric  sulfate   as  a  flocculant   aid  prior   to upflow  reactor/-
clarifiers.      These    provide  the  dual   functions    of   flocculation   and
sedimentation.    Dual  media  filtration  follows,  then   disinfection  (chlorination


                                       177

-------
to a  2.4-5.0 mg/L total  available chlorine residual  in  the finished  water),  and
fluoridation.
              c.    Study of Alternatives

Chloramines and chloride  dioxide  were  selected  for further study  and  pilot
plant  testing.     Ozone   was   eliminated   from   further  study   because  of
implementation   considerations.    Also,  ozone  does  not  maintain  a  residual
during  distribution,   and  therefore  would   require  the  use  of  a  secondary
disinfectant  anyway.    Chlorine  dioxide   for   both  prepxidation  and   post-
disinfection achieved  the  goals  of  GCWA better than  did  chloramines  (poor
preoxidant).

Preoxidation/post-disinfection   combinations   were   studied   in  the   following
sequences:

Phase  1)           Chlorine/chlorine

Phases 2a & b)    Chloramine/chloramine

Phase  3)           Chlorine dioxide/chlorine

Phase  4)           Chlorine dioxide/chlorine dioxide

Phase  5)           Chlorine  dioxide/combined chlorine + chlorine  dioxide,  and

Phase  6)           Chlorine dioxide/combined  chloramine  +  chlorine dioxide.


                   i.    Study Results

Phase  1:  Chlorine - Chlorine

Finished  water  exhibited  intermittent algae-related  taste  and  odor,  and  THM
levels  were in  excess  of  federal  standards  (350  /tg/L);  however  bacterial
quality was excellent.  This approach was abandoned.


Phases 2a and 2b:  Chloramine - Chloramine

Phases  2a  and  2b  employed  chlorine:ammonia  weight  ratios of  3:1  and  7:1,
respectively.   Although  THM  levels were  lowered  to  about  60  /ig/L, bacterial
counts  for coliforms  demonstrated  that   confluent  growth  was   occurring  on
several  cultures,  indicating  that inadequate  residual  was  being maintained in
the  distribution   system  (3:1  chlorine:ammonia   ratio).    This  approach  was
abandoned.

The  experiments   were   repeated  using   the   7:1  chlorine/ammonia   ratio.
Acceptable bacterial  quality  was  achieved,  but  numerous  taste   and  odor
complaints  were received  during this period.   This  approach was abandoned.


                                        178

-------
Phase  3:   Chlorine Dioxide  - Chlorine

Chlorine   dioxide  was   installed  in  May  1984,  using  a  generator  with   a
conversion  efficiency  (chlorite  ion   to  C1O2)  of approximately  80%>.    Alter
C1C«2  preoxidation,  the clarified water showed  no traces of THMs.   No taste
and odor  complaints  were  received, despite very  high  raw  water algae  counts
(up to  5,000  blue/green  algae/mL).   However,  finished  water  THM   levels
sometimes persisted above  100 /ig/L.

In  November  1984,  a  chlorine  dioxide  generator  with  a  98+%   conversion
efficiency  was  installed,  and  this phase  was continued for an additional two
months.    THM levels averaged 102 /ig/L), and no  taste  and  odor  complaints
were received.  This approach  was abandoned.


Phase  4:   Chlorine Dioxide  - Chlorine Dioxide

This  approach of  using chlorine  dioxide  both  as  the  preoxidant  and  post-
disinfectant was authorized on an  interim  basis  by the  Texas Department  of
Health  in  early  March 1985  with  the  stipulation  that   a   maximum   C1C«2
residual   of  1.0  mg/L  be  maintained  and   that  finished water  quality  be
monitored throughout the distribution system.

Finished  water THM  concentrations during the test period averaged 60  /ig/L,
and  finished  water  turbidities  were  the  lowest  of   any  of the   alternative
disinfectant  scenarios.    Bacterial  counts  generally  were  excellent,   but   inter-
mittent  elevated  counts   were  noted   at  three  different   points  in  the
distribution system:    at the  clearwell,  and  at locations 2  and  5  miles  from
the plant.

Additionally,   the  bacterial   species  distribution  also  changed  upon   C1C>2
disinfection,  both  at  the   plant  and  in  the  distribution   system.    Bacterial
counts  displayed  a  shift  from  orange   to  yellow-staining  gram  negative  (-)
rods   to   white-staining  gram  positive   (+)   rods,   similar   to   slime-forming
Bacillus  (sp.).

This approach,  although promising, also was abandoned.


Phase  5:   Chlorine Dioxide  - Combined  Chlorine + Chlorine Dioxide

Phase  4  was  repeated  except that chlorine  was applied  in  conjunction with
ClO^  until the  bacteria could be  identified.   Excellent bacterial  quality was
obtained.   The bacterial plate counts remained  at  or  below  the  guideline  of
500  colonies  per  100  mL for  all monitoring locations.   The  shift  in bacterial
species  distribution  continued  as  the  plate  counts decreased,  so  that over
95% of  all colonies  examined were either yellow  gram  (-)  rods or  white gram
(+)  rods.

Finished  water THM  levels rose to  an  average  of 81 /ug/L.   This approach
was  rejected,  in favor of Phase  6.
                                        179

-------
Phase 6:  Chlorine Dioxide - Combined Chloramine  + Chlorine  Dioxide

This  phase  was  implemented  on  December  6,  1985.   With this  post-disin-
fection  combination,  finished water THM  levels  averaged  68 /x,g/L,  and  the
bacteriological  quality  remained excellent.    No  coliforms  were  found  in  the
clearwell  or  in  the  distribution  system  during  this   phase,   and  bacterial
counts have  ranged from <  1  to  30 colonies  per 100 mL.   Of  those  colonies
identified,  over 95%  were  the white-staining,  gram (+) Bacillus (sp.)  variety
with less than 1% belonging  to  the  orange-staining gram (-)  type.


              d.   The Results

Preoxidation   with   chlorine  dioxide   and  post-disinfection  with   C1O2   in
combination  with either  chlorine  or chlorammes  has  provided   effective taste
and  odor  control,  maintained  an  active  disinfectant  residual   and  mitigated
THM formation (to a current level  of about 68
Chlorine  dioxide  preoxidation  has  proven  to  be  an  excellent  algicide  and
biocide.     The  effectiveness  of  C1O2  in   removing  algae  in  flocculation/-
sedimentation  has  been  demonstrated  in  a  decrease  in  filter  fouling  and
improved finished water turbidities since implementing its use.

Odors   experienced   at  the   flocculating   clarifiers  and   taste   and   odor
complaints  in  the  distribution  system  have  been  reduced.   The  unexpected
benefit of  improved treatment  plant performance  has been  maintained for at
least   the   first seven   months  of  operation  of the  treatment  modifications.
Finished  water turbidities  decreased from  an  average  range  of 0.3  to 0.65
NTU  during  1983 (chlorine  pre- and post-treatment)  to  0.09 to  0.2  NTU in
1986.

Bacterial   counts   using  the  membrane  filter  method   for  coliforms   have
declined  to excellent levels.   Since  the advent of C1O2  post-disinfection,  all
bacterial  counts  obtained  from  samples  collected  at  the  GCWA  distribution
point  have been below 30  colonies/100 mL, and all  but  two  samples  have
been  below 10 colonies/100  mL.   This  indicates a substantial residual present
at  that  distribution  system  point.    The  bacterial counts  obtained  from  the
GCWA distribution  system continually are below the  guideline  of 500/mL,  and
often  are  below  5  colonies/100  mL, indicating the  maintenance  of  a  good,
active  residual.

The Galveston  County  Water  Authority has  successfully implemented  the  first
chlorine  dioxide disinfection program  in the State of Texas.
                                       180

-------
         CHLORAMINE CASE  HISTORIES

         1.    Bloomington. Indiana  (Singer, 1986)

              Prechlorination. Post-Chloramination

              a.    The Problem
Bloomington  takes its  water  from a  lake having  TOC  levels  of 4-6  mg/L.
This  water  is  treated  by  conventional  coagulation  with  alum,  flocculation,
sedimentation,  and  filtration  through  pressure  filters.    Prior  to  September
1984  the  utility  applied  chlorine  to  the  raw  water and  again  ahead  of the
pressure   filters.     Average  chlorine  dosages   were   1.8   and   1.0
               .                                         .
respectively.    Typical  average TTHM concentrations  reported  by the  utility
as part of its quarterly compliance monitoring program were as follows:
         March  1983         77 /ig/L
         June  1983         168 jig/L
         Sept  1983         142 ^tg/L
         Dec    1983         50 /xg/L
         March  1984         79 /ig/L
         May   1984        113 jtg/L

It is  clear that the current 100 jug/L  TTHM running  annual average standard
was being exceeded.


              b.    The Solution

In  September  1984,  the  Bloomington  water  utility  changed  from   post-
chlorination  to  post-chloramination.   An average 0.54 mg/L  of ammonia  was
applied along  with  1.5  mg/L  of chlorine  ahead of  the  pressure filters.    The
desired residual chlorine concentration  leaving  the  plant  of 1.0 mg/L of  free
chlorine was  changed to 1.5 mg/L of combined chlorine.


              c.    Performance

Quarterly  compliance   monitoring  data  for   TTHMs   subsequent   to  the
changeover are as follows:
        Sept.  1984          47
        Dec.   1984         24 /tg/L
        Feb.    1985         43 /*g/L
        April  1985          56 jug/L
        Aug.   1985         57 jug/L

Clearly, TTHM  concentrations have  been brought  down  to  about  50  /ig/L
from well over 100 ju,g/L

Table  XXXIV summarizes THM  and  TOX  (Total  Organic Halide)  data  for
samples  collected  as  major   points  in  the  treatment  system  when  chlorine

                                       181

-------
alone was used for  pre- and  post-treatment.   The data show  that TOX  levels
increased in parallel with increases in TTHM levels.  Table XXXV summarizes
similar data taken using combined chlorine residuals.   These data show that
although  the TTHM  formation  ceases  after  the  addition  of  ammonia,  the
production  of TOX continues,  although at  a  greatly  reduced rate.    This
illustrates  that   as   EPA  sets  MCLs   for  additional   halogenated  organic
materials other  than THMs, utilities  opting for post-chloramination must  plan
to determine the makeup of their TOX fraction.

  TABLE XXXIV. SUMMARY OF THM DATA AT BLOOMINGTON,
                 INDIANA  WITH FREE CHLORINATION 8/16/84
                 (Singer, 1986)
   Sampling Point
Residual Free       TOC  TTHM  TOX
Chlorine,  mg/L      mg/L   ju,g/L   ju,g/L
   Raw Water

   Settled Water
     0.25
4.3

3.6
                             1
       23
48    127
Filtered Water
Dist. System #1
Dist. System #2
1.0
1.8
0.65
2.4 81 205
110 291
151 363
  TABLE XXXV.  SUMMARY OF THM DATA AT BLOOMINGTON,
                  INDIANA WITH POST-CHLORAMINATION 8/26/85
                  (Singer, 1986)
   Sampling Point
  Residual
Chlorine,  mg/L
TOC  TTHM
mg/L  /Ag/L
                                                        TOX
Raw Water
Settled Water
Filtered Water
Dist. System #1
Dist. System #2
—
Trace, Free
1.2 Combined
1.0 Combined
0.9 Combined
4.1
2.8
2.8
—
—
0
53
55
52
57
17
94
91
115
116
Since   switching   to   post-chloramination,  the  utility  has   experienced  no
adverse  effects  in operations  or  in  finished  water  quality.   According to
distribution   system  monitoring  records,  the  microbiological  safety  of the
water  has been maintained.
                                    182

-------
         2.    Philadelphia. Pennsylvania (McKeon et al., 1986)

              Pre-Chlorine Dioxide + 1-Hour  Prechlorination
              Post-Chloramination

This  is  a case  history which  is  still evolving,  but one which has  resulted  in
modifications   to  attain   current   regulatory   and  water  quality   objectives.
However, the  types  of studies which have  been conducted by  the  Philadelphia
Water  Department  point  to  the  directions  this utility probably  will  take  in
response to continued  tightening of  federal  regulations and  changes  in water
quality.


              a.   The Problems

The  Baxter  Water Treatment Plant  (formerly  called  Torresdale),  was  built  in
1960 on the  site of a  slow sand   filtration  plant  that  had  been  in  service
since  1903.    The  current  plant is   rated  at 282  mgd, and  supplies   potable
water to  over  800,000  people, from  the  Delaware   River  by   conventional
treatment.

Chemicals  fed  routinely  include chlorine,  ferric  chloride or  ferrous   sulfate,
lime, fluoride,  and  ammonia.    Powdered  activated carbon is  fed  on  demand
for  control  of  taste and odor  events,  and  chloride  dioxide  is fed  currently
for  control  of  THMs  and  tastes  and  odors.    The  C1O2  feed  system  was
incorporated  in the  original  plant  design  in  the 1950s   to  oxidize phenolic
compounds  found  in the watershed.   Since  that  time,  the sources  of  phenol
have been  eliminated and the CICb   feed system  has  found use as part of the
program to reduce concentrations of trihalomethanes.

Prior to  1976, the  Baxter  plant practiced  breakpoint  chlorination  at the  raw
water basin and maintained  free  chlorine  in  the  distribution system.    From
raw water  chlorination  to the consumer,  a total of 96 hours  of  free  chlorine
contact time normally elapsed,

Initial analyses  for THMs in  1978 showed peak concentrations  above 300 /ig/L
with  a  running  annual average  of  140 /ig/L.   In light  of  these  results, the
Philadelphia  Water  Department  began  to  reevaluate  its disinfection strategies.
The  selection  of an alternative disinfectant  strategy  at  the  Baxter  plant not
only  must take  into consideration  bacterial  kills,  but  taste  and odor   control,
suppression  of algae, corrosion control, residual duration,  and economics.

This  case   history  reviews   Philadelphia's   experiences   over  ten  years   in
modifying  chlorination   practices,   and   the    problems   which  surface  with
reduction of chlorine contact  time.


              b.    Process Modifications

Initial efforts  to reduce  formation   of  THMs  concentrated  on  reduction  in
free chlorine  contact times.    From  the original  96-hour  free  chlorine  contact
time, today  (1986)  chlorine contact time has been  reduced to 1-hour,,

                                       183

-------
                   i.    Chloramination  of Finished Water (Reduction of Free
                        Chlorine Contact Time from 96-hours  to  24-hours)

Chloramination  of the  finished water was  initiated in  1976.    Ammonia  was
added to  convert  free  chlorine  to monochloramine,  to  yield  a longer  lasting
residual  in the distribution system, to improve  the organoleptic  properties  of
the water, and to reduce  the corrosion  rates  associated  with the use  of  free
chlorine.   Adequate  disinfection  was  assured  by  maintaining  a  2-3  mg/L  free
chlorine  residual throughout the plant  treatment process.

Trihalomethane formation  within  the  treatment  process  was  reduced by 40%
under summertime conditions  in  comparison to  the ultimate (96-hour) THMFP
reached  previously  (from 231 to 174 ju-g/L).


                   ii.   Reduction in  Chlorine  Treatment at Raw Water Basin
A gradual phasing  out  of chlorine  addition at  the  raw water  basin inlet took
place  from  1976 through 1979.    Whereas  in  1975,  60  Ibs  of chlorine per
million gallons  of  chlorine  was  added  at the  intake, by the  end  of  1979,
chlorine  addition was  limited  to  20-30  Ibs/million  gallons  of water.    From
1977 to  1979, the  marginal  chlorination of the  intake  followed  by five  hours
of  free   chlorine  contact time in  the  floe/sedimentation basins  and   filters
produced summertime  average  THM  concentrations  of 200  /Ltg/L  and  yearly
averages of  140  /ig/L.  While this  treatment  regimen produced positive results
in the control of algae  blooms in the  raw  water  basin  and minimization of
taste and odor  events,  its debit was high formation of  THMs.  The decreased
use  of chlorine  in  the raw  water  basin  also had  the effect  of reducing the
amount  of chlorine loss due  to evaporation,  and thus the quantity  of chlorine
used.
                   iii.   Utilization  of  Chlorine Dioxide  at Raw  Water Basin
                        Inlet (5  Hours Free Chlorine Contact  Time)

Beginning  in  1980,  routine  use  of chlorine  at  the raw water  intake  basin  was
abandoned.   In  its place,  chlorine  dioxide was added in dosages  between  0.5
and  1.0  mg/L.  Summertime THM values thus were  reduced from  200 /ig/L to
140  /ig/L.   This  treatment was sufficient  to  control  algae,  and  thus tastes
and  odors,  at  all  times   except during  the  spring  algae  bloom.    For  that
period  of time,  breakpoint  chlorination  of  the  intake  water  and/or  100-200
Ibs  of powdered activated  carbon  per million  gallons  is required  to  eliminate
vegetative tastes and odors.


                   iv.   Installation  of  a New  Chlorine Application Point (One
                        Hour Free  Chlorine Contact  Time)

In the fall  of 1980, a  chlorine  application  point  was  installed  in  the "applied
to  filters"  channel  which  allowed  for  increased   flexibility   in   the  use   of
chlorine.    Free  chlorine  contact time  was reduced  from  five hours to one.

                                       184

-------
Chlorine  was  added  at  the  rapid  mix  to  barely  achieve  breakpoint  and
provide  a residual  which  dissipated  within  a few  minutes.    Chlpramines  were
carried  across  the  floe/sedimentation  basins.    Sufficient  chlorine  then  was
added  at the  new  application  point  to achieve  a  free  chlorine  residual of
1.5-2.0 mg/L.   This residual  was  converted to chloramines  one  hour later as
the water flow  left the filter building.

This  treatment regimen  gave  adequate  control of  taste and  odor,  except for
the   previously  mentioned outbreaks  associated  with  spring  algae  blooms,
which  forced  reversion  back  to  free  chlorine  at  the  intake.   THM levels,
with  only one  hour free chlorine  contact time,  resulted in summertime values
averaging 100 /tg/L  with  a  yearly running average of 60
                   v.    Ten Minutes Free Chlorine Contact Time

 A plant trial was carried out  in  November  1982 utilizing a  10-minute chlorine
 contact  time.     Results   indicated  that  satisfactory  disinfection  could   be
 achieved  with  only  chloramines  carried  through the  floe/sedimentation  basins
 when  water  temperature was  below 60°F.    This strategy was  initiated  on  a
 plant-scale  in  December   1982.     Adequate  disinfection was  achieved,  but
 periodic taste  and  odor results  were  less  than  satisfactory.    Medicinal  and
 chemical  taste  complaints  were  received and  several  large  taste  and odor
 events occurred following storms.

 THM  results  under  the reduced chlorine  treatment  reduced  running  annual
 average concentrations from 60 to 50 ^tg/L.

 The trial  was  terminated  in  December 1983  because the  disinfection  scheme
 did  not   adequately   address  the  taste  and  odor  qualities  of  the  water.
 Treatment regimen returned to 1-hour free chlorine contact time.


              c.   Economics

 Realization of 70% reduction in THM  concentrations  over the ten year  period
 was  obtained at  minimum  cost.   The  1978 disinfection  cost was  $5.01/million
 gallons;  in  1986  the  cost  was $9.53.   Converting the   1986 costs  into 1978
 dollars, the  1986  adjusted  cost  is  $5.52/million  gallons.    Cost  increases were
 minimized   because   the  reduced  chlorine   contact   times   resulted  in   less
 evaporative losses  of chlorine,  which netted a  20% decrease  in  the  amount of
 chlorine  utilized.    This partially  offset  the  increased costs incurred  by  the
 use of chlorine  dioxide.


              d.   Operational Improvements

 Free  chlorine contact  times  at the Baxter plant have been reduced from  96-
 hours  in  1975   to   1-hour  in  1986.  Trihalomethane   concentrations  in  the
 finished  water  have  been  reduced  from  an estimated annual average of  220
jig/L  in  1975 to  60  /*g/L  in  1986.   Treatment  strategy  is most critical  during
 the spring  and  summer  months when vulnerability to  taste  and  odor events  is

                                        185

-------
high and  THM formation  within  the  treatment  process  is  accelerated due to
warm water temperatures.


              e.    For the Future

The near  term goal of the  Philadelphia Water  Department is to  reduce  the
annual running  THM concentration  average below  50 £tg/L, a  15% reduction
from  current  (1986) levels.  This  can  be achieved at minimum  expenditure by
installation  of  a  pH  adjustment  point  at  post-treatment.    The  existing
treatment scheme calls for raising the raw water pH to  8.4 at the  rapid  mix
and carrying   this  high  pH  through  the  distribution  system  for corrosion
protection.   Addition  of a pH adjustment  point  at  post-treatment  will  allow
for a pH  of  7.5  to be used  through  the floe/sedimentation basins  and filters,
with pH  adjustment to  8.4 after  chloraminatipn.   Plant  scale  trials  using  this
treatment regimen yielded  a 20% reduction in THM formation (to about 40
If EPA lowers  the  THM  MCL to  below  50 /itg/L,  ozone  and/or granular
activated   carbon  become   the   likely  alternatives   at   the  Baxter   plant.
Extensive laboratory  and  pilot plant evaluations  (Neukrug  et  al,  1984)  have
developed   conceptual   full-scale   plant   designs   incorporating   these   two
treatment techniques.

Estimated  annual  amortized  capital  and  operating  costs  for  ozone  at  the
Baxter  plant,  spread  over the lifetime  of  the  equipment,  are estimated to be
about  $50/million  gallons.   The  associated costs for GAC post-contacting (15-
minute  empty  bed contact time) with  a  75-day regeneration frequency would
be  about   $212/million  gallons.    This  design  configuration   is  capable  of
producing THM concentrations of less than 10 /ig/L.


IX.      SUMMARY RECOMMENDATIONS FOR DISINFECTION STRATEGIES
         AND  FOR  THE  CONTROL  OF  DISINFECTION/OXIDATION  BY-
         PRODUCTS

Consideration  of  all  of the  preceding  discussions,  including the  case  studies,
leads  to the  following  general and  specific  conclusions and recommendations:

   A.    For Disinfection

1. From review of EPA's  proposed  Surface Water Rule (U.S. EPA, 1987a) and
   the  National  Academy  of Sciences recommendations  (NAS, 1987), ozone
   and  chlorine  should  be  viewed  as  the primary  disinfectants  of  choice for
   surface  waters  and  ground  waters  directly  influenced  by   surface  waters.
   However, chlorine can  be  used  only when  the  THM  and  TOX  formation
   potentials  are  sufficiently  low to meet  current  and  projected disinfection
   by-product  regulations,  and  to insure  that  no  other  halogenated products
   are  formed  in quantities  sufficiently  high  that   they  will  be  subject  to
   regulation.   UV radiation  can be  used  as  a primary  disinfectant,  but for
   groundwater  only,  since  it  does  not  provide   the  required  amount  of
   inactivation of Giardia cysts for surface  water treatment.

                                       186

-------
2. For groundwater  disinfection,  ozone,  chlorine,  and UV radiation are  the
   primary  disinfectants  of choice,  with  the  same restrictions  on  chlorine  as
   noted  above.

3. Monochloramine  is not recommended  by  EPA or  by  NAS  as  a primary
   disinfectant because of its  very  high  CT  values   with  respect  to  Qiardia
   cysts  and  because inactivation  of Giardia  cysts with  chloramine  does  not
   guarantee  inactivation of viruses.

4. NAS  also recommends against  the  use of preformed monochloramine and
   against  marginal   chlorination   as  methods   of   introducing  chloramines.
   Organic    N-chloramines,   which   are   even   weaker  disinfectants    than
   monochloramine,   and   which  cannot  be   distinguished   from   inorganic
   chloramines  by  current field  analytical techniques,  are formed  by  these
   chloramination techniques.

5. For  surface  or   groundwater,  chlorine is  the  most  effective   secondary
   disinfectant,   provided   the  water  is  sufficiently  clean  that   significant
   quantities  of  THMs   and   other  halogenated  organic compounds  are  not
   produced.

5. If  an  MCL is  set for chlorine  dioxide at  1.0 mg/L,  this  material can be
   considered  for  primary  disinfection,  and  definitely  can  be employed  as  a
   secondary  disinfectant.    A  ClO^  dosage  of  1.2-1.4  mg/L  will  produce
   residual  oxidant (C1C>2  + chlorite  ion  + chlorate ion) levels of  ~  1 mg/L.

6. On the  other  hand,  if an  MCL for chlorine  dioxide  is  set at  0.2  mg/L
   (Table   VII),   use  of   C1O2 as   a   primary  disinfectant  will  be  severaly
   restricted  to   exceptionally   clean  waters,  because  the   maximum  applied
   dosage   then  would be  equivalent  to  about  0.3   mg/L.     Its  use   as  a
   secondary  disinfectant   also  will   require a  high  quality finished  water  in
   order  to  provide  a stable,  detectable residual  throughout the  distribution
   system.

7. If an  MCL is  set for monochloramine  at  2.5  mg/L,  it  can  be  used as  a
   secondary  disinfectant.    However, if  an MCL  is  set  at 0.29 mg/L  (Table
   VII),   its  use  as  a secondary  disinfectant,  while  not  eliminated, will be
   limited to  high quality  finished waters.

8. If EPA  sets low MCLs for both  C1O2 and chloramine, then chlorine may
   become  the  only,  practical  secondary  disinfectant  available.   In this   event,
   the water will  have  to be  pretreated more  efficiently  in order  to  lower
   TOX  precursor  levels.

9. The National  Academy of  Sciences  (1987)  recommends that  when   chlor-
   amine  is  used  as the secondary  disinfectant,  chlorine should  be   added
   first to  produce  a slight residual of  free  chlorine  above that required  to
   oxidize nitrogen, followed by addition  of ammonia  to  form monochloramine
   and limit THM  formation.

10.  CT values for different disinfectants  employed in  a  treatment process are
   additive.    For example, preoxidation with  ozone  or  chlorine  dioxide  for  a

                                       187

-------
   non-disinfection application nevertheless  provides  some  CT  benefit  which
   can  be added  to  that  obtained  in  the  primary  and  secondary  disinfection
   steps.     When  attempting  to   take   disinfection   credit  for   ozone,   a
   measurable  residual  of  dissolved  ozone must  be  maintained  for  a specific
   contact time  in the  treatment  plant.    Preferably,  dissolved  ozone  should
   be  monitored  at  two points,  and the  detention  time between then  should
   be   known  as  accurately  as   possible.   This  will  allow  more  accurate
   depiction   of  CT  values  than  those   calculated  from  measurement  of
   residual ozone at  a single point.

11.  When  using  ozone  for  disinfection,  contact  time  must be defined in  terms
   of  detention  time  in the  ozone  contacting  chamber  or subsequent  ozone
   reaction  chamber.    This is because  ozone  residuals dissipate  rapidly  after
   application  of ozone  to water  ceases.   To  calculate the disinfection  credit
   of  ozone under such conditions  requires making  assumptions  regarding the
   period of  time over  which the  ozone  residual  concentration  is  established
   and maintained (if only a  single  ozone  contactor is employed).   Satisfying
   the  ozone  demand of  the  water  and  attaining  the  desired  residual  of
   ozone in  a  first contact  chamber, followed  by a  second chamber  in  which
   the  desired  residual ozone  is  maintained,  is   a  more quantitative and
   reliable disinfection design approach.

12.   Changing  disinfection   procedures usually  must  be  made  with  attention
   being  paid  to overcoming other  water  quality  problems,  such  as  control  of
   color,  taste  and  odor,   iron  and manganese,  synthetic  or  volatile  organic
   contaminants, etc.

13.   Whenever ozone  is employed,  for  oxidation or disinfection purposes,  it
   should  be  added  in   the  absence  of   chlorine,   chlorine  dioxide,  or
   chloramine.   This  is because  ozone  oxidizes these three  chemicals   (very
   rapid oxidation of C1C>2),  and  both  reagents are  mutually destroyed,  to  no
   benefit to  the water treatment  process.   When  primary  disinfection with
   ozone is  followed  by   chlorine,   chlorine dioxide,  or  chloramine  as  the
   secondary  disinfectant,  it is important  that  no  residual  ozone be present,
   particularly  in the case of chlorine dioxide, for  the  same reason.


   B.    For Controlling Disinfectant/Oxidant By-Products

14.   Currently,   halogenated   by-products   are   the  only   compounds   being
   regulated  or selected for regulation in the next  few years.

15.  Attention  to  minimizing  the  use of chlorine during disinfection  will also
   minimize the production  of  disinfection by-products.

16.   Conventional   physical   water   treatment    processes   (e.g.,  coagulation,
   flocculation,  sedimentation,  filtration)  should  be  improved  to  remove  or
   minimize  concentrations  of organic   precursors  of  disinfection  by-products
   prior  to  the   addition   of chlorine.     Preoxidation  and  activated carbon
   adsorption   are treatment  steps  which  can  be  effective   in  this  regard.
   Ozone,   potassium   permanganate,   hydrogen  peroxide   can  be  effective


                                       188

-------
   preoxidizing  agents.   Chlorine  dioxide  also  is  an  effective preoxidant, if
   its use at 1.2-1.4 mg/L applied dosages continues to be  allowed.

17.  When  relying  on  monochloramine  as the  secondary  disinfectant,  close
   attention  should be  paid  to the TOX  levels  and  the halogenated  organic
   products  which are  contained  in  this  group parameter.    TOX  levels  are
   known  to increase  slowly  in  distribution systems  employing  chloramine  as
   the secondary disinfectant.

18.  Most current  preoxidation applications of ozone are  designed to  provide
   only  a trace residual of  ozone  at  the exit  of the  contact  chamber.    In
   plants  currently   employing   ozone   for   preoxidation   using  a   single
   contactor,  simply  increasing  the  dissolved  ozone  residual  to  the  level
   appropriate  for   the   designed  ozone  contact  time   and   the   water
   temperature  can  provide   at  least   the  major  portion,  if  not all,  of  the
   primary  disinfection  required  to  guarantee  inactivation  of Giardiq  cysts
   and  enteric viruses.
X.            REFERENCES

Akin, E., et al., 1987, 'The  (U.S. EPA)  Office  of Research and Development
   Health  Research  Program  on  Drinking  Water  Disinfectants and  Their By-
   Products:  An Issue  Paper Prepared for a  Science  Advisory Board  Program
   Review".

American Water  Works  Association, 1973, Water  Chlorination  Principles  and
   Practice. AWWA Publication No.  M20,  p.  12.

American  Water  Works  Association,  1985,   Water  Treatment Plant  Design
   (Denver, CO:  Am. Water Works Assoc.)

Angehrn, M,  1984, "Ultraviolet Disinfection of Water", Aqua 2:109-115.

Augenstein,   H.W.,   1974,  "Use   of  Chlorine  Dioxide  to   Disinfect  Water
   Supplies", J.  Am. Water Works Assoc. 66:716-717.

Christman,   R.F.; Johnson,  J.D.;  Pfaender, F.K.;  Norwood,  D.L.;  Webb, M.R.;
   Hass,  J.R.;  Bobenreith,  M.J.,  1980,  "Chemical  Identification  of   Aquatic
   Humic   Chlorination   Products",   in  Water   Chlorination.  Environmental
   Impact  and  Health  Effects.  Vol. 3.  R.L.  Jolley,  W.A.  Brungs,  and  R.B.
   Gumming,  Editors  (Ann Arbor,  MI:   Ann Arbor Science  Publishers,  Inc.,
   1980), pp.  75-84.

Condie,  L.W.,   1986,   "Toxicological  Problems   Associated   with   Chlorine
   Dioxide", J. Am.  Water Works Assoc. 78(6):73-78.

Cotruvo, J.A.;  Vogt,  C.D.,   1985,  "Regulatory   Aspects  of  Disinfection",  in
   Water Chlorination:   Chemistry.  Environmental  Impact,  and Health Effects.
   Volume  5.  R.L.  Jolley, RJ.  Bull, W.P. Davis,  S.  Katz,  M.H.  Roberts, Jr.,
   and  V.A. Jacobs, Editors  (Chelsea,  MI:  Lewis  Publishers,  Inc.,  1985), pp.
   91-96.

                                       189

-------
Cryer,  E.T.,   1986,  "Preozonation  at  Kennewick,  Washington, Case  History",
   presented  at  Intl.  Ozone  Assoc.  Workshop  on Drinking  Water  Treatment
   With  Ozone, Perrysburg, OH, April, 1986.

Duguet,  J.P.;  Brodard, E.;  Dussert,  B.;  Mallevialle, J.,  1985,  "Improvement  in
   the  Effectiveness  of  Ozonation  of  Drinking Water Through  the Use  of
   Hydrogen Peroxide", Ozone:  Science & Engineering 7(3):241-258.

Ehrlicher, H.,  1964, Zentr. Arbeitsmed.  Arbeitsschutz 14:260.

Glaze, W.H.;  Kang, J.-W.; Aieta, M., 1987, "Ozone-Hydrogen Peroxide Systems
   for Control  of Organics  in Municipal  Water  Supplies",  in  Proc.  Second
   Intl.  Conf.  on The Role  of Ozone  in  Water  and  Wastewater Treatment
   D.W. Smith  &  G.R. Finch,  Editors  (Kitchener,  Ontario, Canada:    TekTran
   International Ltd.,  1987), pp. 233-244.

Great  Lakes  - Upper Mississippi  River Board of  State Sanitary  Engineers,
   1985, Recommended Standards for  Water Works  -  1985  Edition  (Albany,
   NY:  Health Education Service, 1985).

Grunwell, J.;  Benga,  J.; Cohen, H.; Gordon,  G.,  1983, "A  Detailed Comparison
   of  Analytical Methods for Residual Ozone Measurement",  Ozone  Science  &
   Engineering 5(4):203-223.

Gumerman, R.C.; Bums,  B.E.;  Hansen, S.P., Small  Water  System  Treatment
   Costs (Park Ridge, NJ:  Noyes Data  Corporation, 1986).

Haller, J.F.; Northgraves, W.W., 1955, Tappi  38:199.

Hango, R.A.;  Doane, F.;  Bollyky, L.J.,  1981,  "Wastewater Treatment for ReUse
   in  Integrated  Circuit   Manufacturing", in  Wasser  Berlin   '81.   5.  Ozon-
   Weltkongress  (Berlin,  Federal  Republic  of Germany:   Colloquium Verlag
   Otto  H. Hess, 1981), pp. 303-313.

Hansen,  S.P.;  Gumerman,  R.C.; Gulp, R.L.,  1979, "Estimating  Water  Treatment
   Costs.  Volume 3.  Cost  Curves Applicable  to 2,500  to  1  mgd  Treatment
   Plants",  U.S.  EPA  Report  No. EPA-600/2-79-162c (Cincinnati, OH:   U.S.
   EPA, Water Engineering Research Laboratory, 1979).

Hoffman, J.;  Eichelsdorfer, D., 1971, "On  the Action of  Ozone  on  Pesticides
   with Chlorinated Hydrocarbon Groups in Water", Wasser 38:197-206.

Laplanche,   A.;  Martin,   G.;   Tonnard,  F.,  1983,  "Ozonation  Schemes   of
   Organophosphorus  Pesticides.   Application in  Drinking Water Treatment",
   in  Sixth Ozone World  Congress  Proceedings (Norwalk, CT:   Intl. Ozone
   Assoc.,  1983), pp.  94-95.

Lykins,   B.W.,  Jr.;   Griese,   M.H.,   1986,  "Using   Chlorine  Dioxide  for
   Trihalomethane Control", J.  Am. Water Works Assoc.  78:88-93.
                                      190

-------
 Masschelein,  W.J.,  1979,  Chlorine  Dioxide:    Chemistry  and  EnwoamentaJ
    Impact of Oxychiorine Compounds (Ann Arbor, MI:   Ann  Arbor Science
    Publishers, Inc., 1979).

 Masschelein, W.J.,  1982, "Contacting  of Ozone  with Water  and Contactor Off-
    Gas  Treatment",  in  Handbook  of Ozone  Technology  and  Applications.
    Volume One. R.G. Rice & A. Netzer, Editors (Ann Arbor, MI:   Ann  Arbor
    Science Publishers, Inc., 1982),  pp.  143-224.

 McKeon,  W.R.;  Muldowney,  JJ.;  Aptowicz,  B.S.,  1986,  "The Evolution of  a
    Modified Strategy to  Reduce  Trihalomethane Formation", in Proc. Annual
    Meeting (Denver, CO:  Am, Water Works Assoc., 1986), pp. 967-997.

 Miller, G.W.;  Rice, R.G.;  Robson, C.M.; Scullin,  R.L.; Kiihn,  W.;  Wolf,  H., 1978,
    "An   Assessment   of  Ozone  and   Chlorine   Dioxide   Technologies  for
    Treatment of  Municipal  Water Supplies", EPA Report  No.  EPA-600/2-78-
    147.

 Morris, J.C.,  1967, "Kinetics  of  Reactions  Between  Aqueous  Chlorine and
    Nitrogen  Compounds",  in  Principles  and Applications  of Water Chemistry.
    S.D. Faust &  J.V.  Hunter, Editors (New York, NY:   John  Wiley & Sons,
    Inc., 1967), pp. 23-53.

 Myers,  G.L.;  Thompson,  A.;  Owen,  D.M.; Baker,  J.M.,  1986,  "Control  of
    Trihalomethanes  and  Taste  and  Odor   at   Galveston  County  Water
    Authority", in  Proc.  Annual  Meeting  (Denver,  CO:    Am.  Water  Works
    Assoc., 1986), pp. 1667-1675.

 National  Academy of  Sciences,  1980,  Drinking Water   and Health.  Vol!   2
    (Washington, DC:  National Academy Press, 1980), Chapter 2.

 National   Academy  of   Sciences,   1987,   Drinking   Water   and   t^ealth.
    Disinfectants   and  Disinfectant  By-Products.   Volume   7  (Washington,  DC:
    National Academy  Press, 1987).

 Neukrug,  H.M.;  Smith,  M.G.; Maloney, S.W.;   Suffett,  I.H.,  1984,  "Biological
    Activated  Carbon  -   At  What   Cost?",  J. Am.  Water  Works  Assoc.


 Rachwal,  A.J.;  Bauer,  M.;  Chipps,   M.,  1987, "Ozone's Role  in  Biological
    Filtration  Processes",  presented at  Second   Intl. Conf.  on  The  Role  of
    Ozone  in Water and Wastewater Treatment, Edmonton,  Alberta, Canada.

 Rice,   E.W.;  Hoff,  J.C.,  1981,   "Inactivation   of   Giardia  lamhlia  Cysts  by
   Ultraviolet  Radiation",  Applied &  Environmental Microbiology, 42(3):546-
   547, Sept. 1981.

Rice  R.G.,  1980, "The  Use  of Ozone to Control Trihalometfianes  in  Drinking
   Water  Treatment",  Ozone  Science & Engineering  2:75-99 (1980).

Rice,  R.G,  1987a, "Rationales  for Multiple  Stage Ozonation in Drinking Water
   Treatment Plants", Ozone: Science  & Engineering 9(l):37-62.

                                      191

-------
Rice,  R.G.,  1987b, "Applications  of  Ozone  in  Soft  Drink  Bottling  Plants",
   presented  at  34th  Annual  Meeting,  Soc.   Soft  Drink  Technologists,  Las
   Vegas, NV.

Rice,  R.G.;  Gomez-Taylor,  M, 1986,  "Occurrence of  By-Products of  Strong
   Oxidants  Reacting  with  Drinking  Water  Contaminants  -  Scope  of  the
   Problem", Environmental Health Perspectives 69:31-44.

Schalekamp,  M., 1986,  "Pre-  and  Intermediate  Oxidation  of Drinking  Water
   with   Ozone,   Chlorine,   and   Chlorine   Dioxide",  Ozone:   Science   &
   Engineering 8(2):151-186.

Scheible,  O.K.;   Kreft,  P.;  Venosa,  A.D.,  1985,  "Demonstration  of  Process
   Design  and  Evaluation Procedures  for Ultraviolet Disinfection",  Summary
   prepared   for  U.S.   EPA,  Water   Engineering   Research   Laboratory,
   Cincinnati, OH,  Contract No. 68-03-1821.

Scheible,  O.K.;  Casey, M.C.;  Forndran,  A., 1986, "Ultraviolet  Disinfection  of
   Wastewaters   from   Secondary  Effluent  and  Combined  Sewer  Overflows",
   EPA Report No.  EPA-600/2-86/005   (Cincinnati,   OH,   U.S.  EPA,  Water
   Engineering Research Laboratory).  NTIS No. PB86-145182.

Seeger,  D.R.; Moore,  L.A.;  Stevens,  A.A., 1984,  "Formation  of Acidic Trace
   Organic  By-Products From  the  Chlorination  of Humic  Acids",  U.S.  EPA
   Report  No. 600/D-84-159; NTIS Report  No. PB84-201722.

Singer,  P.S.,  1986, "THM  Control  Using  Alternate  Oxidant  and  Disinfectant
   Strategies:   An  Evaluation",  in Proc. 1986 Annual Conference (Denver,  CO:
   Am. Water Works  Assoc.), pp. 999-1017.

Sontheimer, H.,  1985,  in  "Trends  in Ozonation:  Roundtable Discussion", J.  Am.
   Water Works  Assoc. 77(8):30.

Stevens,  A.A.; Moore, L.; Dressman,  R.C.; Seeger,  D.R.,  1985,  "Disinfectant
   Chemistry  in  Drinking Water  -  Overview  of  Impacts  on Drinking  Water
   Quality", in Safe Drinking Water:   The  Impact of Chemicals on A Limited
   Resource.  R.G.  Rice, Editor (Chelsea,  MI:   Lewis  Publishers,  Inc.,  1985),
   pp. 87-108.

Stevens,  A.A.; Miltner, R.J.;  Moore,  L.A.;  Slocum, C.J.; Nash, H.D.; Reasoner,
   D.J.;  Berman,  D.  (1987a), "Detection  and  Control  of  Chlorination  By-
   Products in Drinking  Water", in Proc.  Conference on  Current Research in
   Drinking Water Treatment  (Cincinnati,  OH:   U.S.  EPA,  Water  Engineering
   Research Lab., 1987).

Stevens,  A.A.;  Moore,  L.A.;  Slocum,  C.J.;  Smith,  B.L.; Seeger, D.R.; Ireland,
   J.C.,  1987b,  "Chlorinated   Humic  Acid Mixtures   Establish  Criteria  for
   Detection  of  Disinfection  By-Products  in Drinking Water", Drinking  Water
   Research   Division,  Water   Engineering Research   Laboratory,  U.S.   EPA,
   Cincinnati,  OH.
                                      192

-------
Stevens,  A.A.; Moore, L.A.;  Slocum, C.J.;  Smith,  B.L.;  Seeger,  D.R.;  Ireland,
   J.C.,   1987c,  "By-Products   of  Chlorination  at  Ten  Operating Utilities
   presented   at  Sixth  Conference  on   Water   Chlorination:   Environmental
   Impacts   and  Health  Effects,  May  3-8,   1987,   Oak   Ridge  Associated
   Universities, Oak Ridge TN.

U.S.  Environmental Protection Agency,  1979,  "Control of Trihalomethanes in
   Drinking Water; Final Rule", Federal Register 44(231):68623-68642.

U.S.  Environmental Protection  Agency,  1981, "Technologies  and  Costs  for the
   Removal   of  Trihalomethanes   from  Drinking   Water"  (Washington,  DC:
   Science and Technology Branch, Criteria and Standards  Division, Office of
   Drinking Water, Dec. 9, 1981).

U.S.  Environmental   Protection  Agency,  1983,  "Microorganism  Removal  for
   Small  Water  Systems",  Report No. EPA   570/9-83-012  (Washington,  DC:
   U.S. EPA, Office of Drinking Water, June 1983).

U.S.  Environmental   Protection  Agency,  1985,  "National   Primary  Drinking
   Water Regulations:  Synthetic  Organic Chemicals,  Inorganic  Chemicals  and
   Microorganisms; Proposed  Rule", Federal  Register 50(219):46935-47022,  Nov.
   13, 1985.

U.S.  Environmental  Protection  Agency,  1986,  "Design  Manual:    Municipal
   Wastewater  Disinfection",   Report  No.  EPA/625/1-86/021  (October  1986).
   U.S.   EPA,  Office  of  Research  and   Development,  Water  Engineering
   Research   Laboratory,  Center   for  Environmental   Research   Information,
   Cincinnati, OH  45268.

U.S.  Environmental   Protection  Agency,   1987a,  "Water   Pollution  Control:
   National   Primary   Drinking   Water   Regulations;   Filtration,   Disinfection,
   Turbidity,    Giardia   lamblia.  Viruses,  Legionella.   Heterotrophic Bacteria;
   Proposed  Rule", Federal Register 52(212):42177-42222.

U.S.  Environmental Protection Agency,  1987b,  'Technologies  and Costs  for
   the   Removal   of  Microbiological   Contaminants   from  Potable   Water
   Supplies",   Draft  Document,   available   from  Office  of  Drinking   Water,
   Criteria and Standards Division.

U.S.  Environmental Protection  Agency,  1987c, "Concept Outline:    Development
   of  Best  Available   Technology  Criteria",  Science and Technology  Branch,
   Criteria  and  Standards  Division,  Office  of Drinking  Water,   March  31,
   1987.

U.S.   Environmental    Protection   Agency,   1987d,  "Guidance    Manual   for
   Compliance With   the Filtration and Disinfection Requirements for Public
   Water  Systems  Using  Surface  Water  Sources",  draft  dated  October  10,
   1987.    Science  and Technology Branch,  Criteria   and  Standards  Division,
   Office  of Drinking Water, U.S. EPA,  Washington, DC.
                                      193

-------
U.S. Environmental  Protection  Agency,  1988a, "Drinking Water;    Substitution
   of Contaminants  and Drinking Water Priority List  of Additional Substances
   Which  May  Require  Regulation  Under   the  Safe  Drinking  Water  Act",
   Federal Register 53(14):1891-1902.

Weil, I.;  Morris,  J.C.,  1949,  "Kinetic Studies on the  Chloramines.   I.   The
   Rates  of  Formation  of  Monochloramine,  N-Chloromethylamine  and  N-
   Chlorodimethylamine", J.  Am. Chem. Soc. 71:1664-1671.

Werdehoff,  K.S.;   Singer,  P.S.,  1986,   "Effects  of   Chlorine  Dioxide   on
   Trihalomethane and Total  Organic Halide  Formation Potentials and  on the
   Formation  of Inorganic  By-Products",  in  Proc.  Annual Conference (Denver,
   CO:  Am.  Water Works  Assoc., 1986), pp.  347-364.
                                      194

-------
              DISINFECTION / BY-PRODUCTS WORKSHOP

                   SURFACE WATER PROBLEM

I.   RAW WATER PARAMETERS
Source                 :     Unprotected Watershed
Turbidity               :    5-10 NTU
Total Coliforms         :    500 - 5,000 per 100 mL
THMFP                  :   350 /tg/L
TOXFP                  :   500 jig/L
TOC                    :    5-10 mg/L (mostly humics)
[NH4+]                  :    0.1 mg/L
[organic-N]             :    0.5 mg/L
[Br]                  :    0.1 mg/L
C\2 Demand              :    7  mg/L
C1O2 Demand            :   2 mg/L
KMnC«4 Demand            :    4 mg/L
Ozone Demand           :   3 mg/L

II.  TREATMENT OBJECTIVES
Disinfection
TTHMs                  :   100 /ig/L

III.     REGULATORY FACTORS
(C102  + C1O2- + C103-) MCL    :    1.0 mg/L
Chloramine MCL              :     2.5 mg/L
[Residual Disinfectant]      :   detectable residual

-------
               SURFACE WATER TREATMENT PROBLEM





                            Modification # 1





I.   RAW WATER QUALITY - Same



II.  TREATMENT OBJECTIVE - TTHMs  =  50 /tg/L



III.  REGULATORY FACTORS



(C1C-2 + C1O2' + 003-) MCL   :    0.5 mg/L



Chloramine  MCL              :    1.0  mg/L



[Residual Disinfectant]       :   detectable residual







               SURFACE WATER TREATMENT PROBLEM





                     Modification  # 2



I.   RAW WATER QUALITY - Same



II.  TREATMENT OBJECTIVE - TTHMs  =  25



III.  REGULATORY FACTORS



(C1O2 + C1O2- + C1O3-) MCL   :    0.2 mg/L



Chloramine  MCL              :    0.25 mg/L



[Residual Disinfectant]       :   detectable residual





3/30/88

-------
              DISINFECTION / BY-PRODUCTS WORKSHOP
                     GROUNDWATER PROBLEM

I.   RAW WATER PARAMETERS
Source                 :    Aquifer - Not Directly Influenced
                                by Surface Water
Turbidity              :    0.3 NTU
Total  Coliforms         :    0-100 per  100  mL
THMFP                  :    200 /tg/L
TOXFP                  :    300 /ig/L
TOC                    :   3-5 mg/L
[NH4+]                 :    0.1 mg/L
[organic-N]             :    0.5 mg/L
[Br]                  :    0.1 mg/L
Iron                  :    0.6 mg/L
Manganese              :    0.30 mg/L
C\2 Demand              :    4 mg/L
C1O2 Demand             :    1.3 mg/L
KMnO4 Demand            :    2 mg/L
Ozone Demand            :    3 mg/L

II.  TREATMENT OBJECTIVES
Disinfection
TTHMs                   :    100 /ig/L

III.     REGULATORY FACTORS
(C1O2 + C1O2- + C1O3') MCL   :   1.0 mg/L
Chloramine MCL               :    2.5 mg/L
[Residual Disinfectant]       :    detectable residual

-------
                GROUNDWATER TREATMENT PROBLEM

                     Modification # 1

I.   RAW WATER QUALITY -  Same
II.  TREATMENT OBJECTIVE - TTHMs  =  50 fig/L
III.  REGULATORY FACTORS
(C1O2 + C1O2' + C1O3-) MCL   :    0.5 mg/L
Chloramine MCL              :    1.0 mg/L
[Residual Disinfectant]      :    detectable residual

                GROUNDWATER TREATMENT PROBLEM

                     Modification # 2
I.   RAW WATER QUALITY -  Same
II.  TREATMENT OBJECTIVE - TTHMs  =  25
III.  REGULATORY FACTORS
(C1O2 + 002" + C1O3-) MCL   :    0.2 mg/L
Chloramine MCL              :    0.25  mg/L
[Residual Disinfectant]      :    detectable residual

3/30/88

-------
Technical Session: Organics
John  E.  Dyksen,  Senior Project Manager,  Malcolm  Pirnie, Inc.,
Paramus, NJ
                              VII-1

-------
                    WORKSHOP ON
             EMERGING TECHNOLOGIES FOR
             DRINKING WATER TREATMENT
           ORGANICS TREATMENT TECHNIQUES
                     OVERVIEW
                   Conducted by:

   United States Environmental Protection Agency

                        and

Association of State Drinking Water Administrators
                   Presented By:

                  John-E.  Dyksen
          Senior Manager,  Water Treatment
               Malcolm Pirnie,  Inc.
               Paramus, New Jersey

-------
                                   WORKSHOP ON
                            EMERGING TECHNOLOGIES FOR
                            DRINKING WATER TREATMENT
                          ORGANICS TREATMENT TECHNIQUES
                                    OVERVIEW
                                TABLE OF CONTENTS


                                                                       Page

 I.    INTRODUCTION                                                       1

      A.    Purpose and Scope                                              1

 II.   AVAILABLE  ORGANICS  TREATMENT  TECHNOLOGIES                           2

      A.    Summary of  Available  Technologies                              2

      B.    Other Applicable Technologies                                  2

 III.  GRANULAR ACTIVATED  CARBON                                           9

      A.    Process  Description                                            9

      B.    Process  Design Consideration                                   9

      C.    Testing  to Evaluate Process Design Parameters                 11

      D.    Optimizing Design Criteria                                    13

      E.    Facility Design Considerations                                14

      F.   Applicability to Various*System Sizes                         15

      G.   Treatment Economics                                           , ^
                                                                        -Lb
IV.   PACKED COLUMN AERATION

     A.   Process  Description

     B.   Process  Design  Considerations                                  18

     C.   Testing  to Evaluate Process Design  Parameters                 19

     D.   Development  of  Design  Criteria                                20

     E.    Facility Design Considerations                                21

-------
                         TABLE OF CONTENTS (Continued)
IV.  PACKED COLUMN AERATION (Continued)

     F.   Applicability to Various System Sizes

     G.   Treatment Economics

V.   BIBLIOGRAPHY
                                                            Page



                                                             24

                                                             24

                                                             26
Table
 No.
  2

  3

  4

  5

  6

  7

  8

  9
Figure
  No.

  1

  2

  3

  4
                      LIST OF TABLES


Description

Typical Performance of PAC and Conventional
  Treatment Processes

Typical Performance of Diffused Aeration

Typical Performance of Slat Tray Aeration

Typical Performance of Ozonation Process

Typical Performance of Reverse Osmosis Process

Readily Adsorbed Organics

Poorly Adsorbed Organics

Summary of Carbon Usage Rates

Freundlich Isotherm Constants for Carbon Adsorption


                      LIST OF FIGURES


Description

Schematic of PAC Adsorption Process

Schematic of Diffused Aeration Process

Home Diffused Aeration System

Schematic of a Redwood Slat Tray Aerator
Following
   Page
     3

     4

     4

     5

     9

     9

    10

    12
Following
   Page

     2

     3

     3

     4

-------
TABLE OF CONTENTS (Continued)
Figure
No.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
LIST OF FIGURES (Continued)
Description
Schematic of Ozonation Process
Schematic of Reverse Osmosis Process
Schematic of Mechanical Aeration Process
Schematic of Catenary Grid System
Schematic of Higee System
Effect of Type of Compound on Carbon Life
Effect of Contaminant Levels on Carbon Life
Effect of Carbon Type on Facility Cost
Empty Bed Contact Time versus Percent Radon Remaining
Steady-state Adsorption/Decay Curve for Radon
Carbon Isotherm Test Results
Diagram of Dynamic Mini-Column Adsorption
Technique System
Diagram of Pilot Column Test System
GAG Facility Cost Components
Breakthrough Curves for c-1,2 DCE
Breakthrough Curves for TOG
EBCT versus Carbon Usage for c-1,2 DCE
EBCT versus Carbon Usage for TOC
Parallel Mode of Operation
Series Mode of Operation
Cost Comparison of Parallel versus Series Flow
•p /^ v I-. 	 T "i r\/~n™i
Following
Page
4
5
6
6
7
10
10
10
11
11
12
12
13
13
13
13
13
13
13
14
14

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


                          LIST OF FIGURES (Continued)

Figure                                                             Following
  No.     Description                                                 Page

 26       Cost Comparison of Parallel versus Series Flow               14
            for TOC

 27       Schematic of Pressure GAC Contactors                         14

 28       Schematic of Downflow Gravity GAC Contactors                 14

 29       Cost Comparison of Carbon Recharge Options                   15

 30       Typical Home GAC Unit                                        15

 31       Capital Cost Curves for GAC Facilities                       16

 32       O&M Cost Curves for GAC Facilities                           16

 33       Diagram of Packed Column Aeration                            18

 34       Effect of Compound on Packed Column Design                   18

 35       Effect of Temperature on Packed Column Design                19

 36       Schematic of Pilot Aeration Column                           19

 37       Mass Transfer Relationships for Packed Columns               21

 38       Packing Height versus Air to Water Ratio                     21

 39       Schematic of Typical Packed Column Facility                  21

 40       Distributor Types                                            22

 41       Schematic of Catalytic Incineration Process                  23

 42       Schematic of Vapor Phase GAC System                          23

 43       Capital Cost Curves for Packed Column Aeration Facilities    24

 44       O&M Cost Curves for Packed Column Aeration Facilities        24

-------
                               I.  INTRODUCTION

A.   Purpose and Scope
     The June  1986  Amendments to  the  Safe Drinking Water Act (SDWA) require
the United States Environmental Protection Agency (EPA) to set maximum contam-
inant  levels  (MCLs)  for  several contaminants found  in drinking  water.   The
MCLs are to be established based upon:
     1.   Health goals
     2.   Effectiveness of treatment technologies in removing the contaminants
     3.   Level of treatment that is affordable for the water supply systems
     EPA is  currently establishing  MCLs for  radon and  a  number  of organic
chemicals  occurring   in   contaminated   water  supplies.   According  to  the
Amendments, all  public water  systems  will be required to  come  as  close as
possible to  meeting  the  MCLs  by using  the  Best  Available  Treatment   (BAT)
technology.
     The purpose of this document is two-fold:
     1.   To assist water utilities in  selecting  appropriate  treatment methods
          to meet the regulations.
     2.   To assist  the  regulatory agencies in  assessing the  feasibility of
          treatment methods proposed by the water utilities.
     Provided  in  this document  is  a  review  of all  the emerging treatment
technologies for removing organic  contaminants and  radon  from drinking water.
Emphasis is given  to  treatment  methods  that have  been proven effective for
their  removals.   The  discussion  on  these  treatment methods  includes  the
process selection criteria, treatability  tests,  design considerations, opera-
tional issues  and facility costs.
                                      -1-

-------
                II.  AVAILABLE ORGANICS TREATMENT TECHNOLOGIES

A.   Summary of Available Technologies

     1.   Most Applicable Technologies

          a.   Granular activated carbon (GAC) adsorption
          b.   Packed column aeration

     2.   Other Applicable Technologies

          a.   Powdered activated carbon (PAC) and conventional treatment
          b.   Diffused aeration
          c.   Multiple tray aeration

     3.   Additional Technologies (Emerging)

          a.   Ozone/UV
          b.   Reverse osmosis
          c.   Mechanical aeration
          d.   Catenary grid (aeration)
          e.   Higee (aeration)
          f.   Resins

B.   Other Applicable Technologies

     1.   Powdered Activated Carbon. (PAC) and Conventional Treatment

          a.   Based on  the  principles of  adsorption, PAC  is  generally added
               at one  or  more  application points in  a  conventional treatment
               train  (coagulation,   flocculation,  sedimentation  and  filtra-
               tion) .  Principal design considerations are:

                    Dosage
                    Contact time (at least 15 minutes)
                 -  Point(s)  of application (1 to 3)

               A schematic of the PAC treatment process is shown on Figure 1.

          b.   Typical performances of PAC and conventional treatment process-
               es  are  presented in  Table 1.  As  can be  seen, PAC  addition
               achieves higher removals than conventional treatment alone.

          c.   Principal advantages of PAC:

                    Better organics removals than conventional treatment

                 -  Acts as coagulant aid

                 -  Taste and odor removal
                                      -2-

-------
CHEMICALS-POWDERED ACTIVATED CARBON
                                      DISINFECTANT
                                          FILTER AID
RAPID MIX
               FLOCCULATION
                                  SEDIMENTATION   FILTRATION
      FIGURE 1 - SCHEMATIC OF P.A.C.
           ADSORPTION PROCESS

-------
                                    TABLE 1

                        TYPICAL PERFORMANCE OF PAC AND
                       CONVENTIONAL TREATMENT PROCESSES
                               Conventional
                                Treatment
                                  Percent                  PAC
                                  Removal     Dosage  (mg/L)    Percent Removal

VOCs

Carbon Tetrachloride                 -          9.6-30            0-25
1,1,1-Trichloroethane                -              7              40-65

SOCs

Acrylamide                           58                13
Alachlor                            <50           4-34           36 - 100
Carbofuran                         54-79          9-25           45-75
o-Dichlorobenzene                    -            8-27           38-95
2,4-D                               0-3          11 - 306          69 - 100
Ethylbenzene                         -            8-27           33 - >99
Heptachlor                          64           11 - 97           53 - 97
Lindane                            10-20          2-34           82-97
Monochlorobenzene                    -            8-27           14 - >99
Toluene                              -            8-27            0-67
2,4,5-TP                            63          1.5 - 17           82 - 99
Toxaphene                            -            1-44           40-99
Xylenes                              -            8-27           60 - >99

Note;

      (-)  Information not available.

-------
       -  Useful for short-term,  emergency  applications in conjunc-
          tion with conventional treatment
d.   Limitations:

       -  Excessive dosages may be required  for certain contaminant
          removal.

          Limited   applicability  in  locations  which  have  certain
          constraints (e.g., hydraulic,  space,  sludge  handling) and
          also where  conventional treatment  is not  in place  or is
          not required.

Diffused Aeration

a.   Based  on bubbling  air  into a water-filled  contact  chamber
     through  a  diffuser  mechanism usually located  at or  near the
     bottom of the chamber.  Principle  design considerations are:

       -  Basin depth (5  to 10 feet)
       -  Diffuser type
       -  Air/Water ratio  (5:1 to 15:1)
       -  Detention time  (10 to 15 minutes)
          Basin geometry

     A schematic of diffused aeration process  is  shown on Figure 2.
     A  typical  home diffusion  aeration  unit  for  radon  removal is
     shown on Figure 3.

b.   Typical  performance  of diffused aeration at an  air  to  water
     ratio of 5:1  to  15:1  and  a contact  time  of 10  to  15 minutes is
     presented in Table 2.

c.   Principle advantage  of diffused  aeration:

       -  Existing basins  such as a chlorine  contact  basin  can be
          retrofitted  with  diffusers   and  converted   to  aeration
          basins.

d.   Limitations:

       -  Less  effective  than packed  column  aeration  (discussed
          later).

       -  Generally used  only where existing  basins  are available.

Multiple Tray Aeration

a.   Water falls over a series of trays  with  slats,  perforations, or
     wire  mesh.   The counter-current   flow   of  air   (either   from
                            -3-

-------
            AIR SUPPLY
INFLUENT
DIFFUSER GRID -^3


."««•"'


<.



RIO


T-
' •


c.

. „



I




•


•

*r-

-,



•

-
•
.
• — . _: 	 •_ 	 : —

'M
•'•f
.
. i
- 1,
. 5f

• V
'•'• ',
' r


                                       EFFLUENT
           FIGURE 2- SCHEMATIC OF
         DIFFUSED AERATION PROCESS

-------
   PL'MP/LEVEL -
   CONTROL SOX
RAW WATER  FROM
                   PUMP
   VARIABLE TIMER ~P~j
                                        AIR-STRIPPED RACON
                                        TO CUTSiDE VEST
                                            • AERAT'.CN TANK


                                            UCL'.O LEVEL
                                            PROSE
                                             PUMP
                                           TO S1ETERED
                                           HOUSEHOLD
                                           WATER USE
                                   AIR  CIFFUSER
                                                    HTCRCPS'EUMATiC
                                                       TANK
        (       )
       AIR COMPRESSOR
       OR PUMP
AIR FLOW
REGULATOR
         FIGURE 3-  HOME DIFFUSED AERATION SYSTEM

-------
                     TABLE 2

    TYPICAL PERFORMANCE OF DIFFUSED AERATION


                                     % Removal

VOCs

Trichloroethylene                       53-95
Tetrachloroethylene      .               73-95
1,2-Dichloroethane                      42-77
1,1-Dichloroethylene                      97
1,1,1-Trichloroethane                   58-90

SOCs

Carbofuran                              11-20
1,2-Dichloropropane                     12-79
cis-l,2-Dichloroethylene                32-85
trans-l,2-Dichloroethylene              37-96
o-Dichlorobenzene                       14-72
Ethylbenzene                            24-89
Monochlorobenzene                       14-85
Toluene                                 22-89
Xylenes                                 18-89

-------
               blowers or natural  ventilation)  removes  the  volatile organics
               in the water.  Principal design considerations are:

                 -  Tray type (wood,  plastic)
                 -  Tray height (12 to 16 feet)
                 -  Natural  or forced draft
                 -  Air/Water ratio (30:1)

               A  schematic  of a slat  tray aeration  system  is shown  on Fig-
               ure 4.

          b.   Typical performance of  slat tray aeration  at an  air  to water
               ratio  of  about  30:1   and  tray  height  of  12   to 16 feet  is
               presented in  Table  3.

          c.   Principle advantages of multiple  tray aeration:

                    Less  susceptible  to  clogging,  from  iron  and  manganese
                    precipitation,  than packed column aeration.

                    Readily  available as package units from manufacturers.

          d.   Limitations:

                    Not  as   effective  as  packed  column  aeration  (discussed
                    later)

                 -  Control   of  biological   growth  and  corrosion  may  be
                    required.

B.   Additional Technologies (Emerging)

     1.   Ozone/UV

          a.   Ozone  provides  organics removal   by  acting  as  an  oxidant,
               generally breaking  down unsaturated  bonds within  the molecules.
               Ultra-violet   (UV) radiation catalyzes  this  breakdown  reaction.
               Principal design considerations are:

                    Ozone dosage
                    Contact  time (15  minutes)
                    Radiation intensity

               A schematic of the  ozonation process is shown on  Figure 5.

          b.   Typical performance of the  ozonation process  at  a  contact time
               of about 15 minutes  is presented  in  Table 4.

          c.   Principle advantages of Ozonation/UV process:

                 -  Can combine organics removal with disinfection,  and taste
                    and odor control
                                      -4-

-------
      INLET
      CHAMBER


  DISTRIBUTOR
     NIPPLES

  STAGGERED
  SLAT TRAYS
   AIR INLET
    BLOWER
AIR
SEAL
WATER
INLET
           -Tt
     AIR
— OUTLET
'

-- u,
-T»-
•— tb 1)
'i 1 *^~
— ; . — .
•
MLJ: T
LI"- 'T.T~".r!'' i
c\y '

-»-= — -n-1
                           - BAFFLES

                           AiR STACKS
            AIR SEAL
                          WATER
                          OUTLET
FIGURE 4- SCHEMATIC OF A REDWOOD SLAT TRAY AERATOR

-------
                 TABLE 3




TYPICAL PERFORMANCE OF SLAT TRAY AERATION







 VOCs                          % Removal




 Trichloroethylene               30-90




 Tetrachloroethylene             20-85

-------
              OXIDANT
              (OZONE)
RAW WATER
                             CONTACT BASIN
•^•PRODUCT WATER
            FIGURE 5- SCHEMATIC OF OZONATION PROCESS

-------
                 TABLE 4





TYPICAL PERFORMANCE OF OZONATION PROCESS
Dosage (mg/L)
SOCs
Carbofuran 9
1,2-Dichloropropane 0.9-6
cis-1,2 Dichloroethylene 2-10
trans-1,2 Dichloroethylene 0.9-6
o-Dichlorobenzene 9
Ethylene Dibromide 0.9-6
Ethylbenzene 1.5-9
Heptachlor 17
Heptachlor epoxide 17
Lindane 0.4 - 150
Monochlorobenzene 0.4-6
Toluene 1.5 - 12
Xylenes 1.5 - 12
% Removal

100
8-22
87 - 93
100
88
8-9
47 - 95
100
26
0 - 100
86 - 98
49 - 98
54 - 98

-------
            -  Waste  disposal not required

            -  No THM formation

     d.    Limitations:

            -  May not achieve  complete  oxidation,  concern about  inter-
               mediate  breakdown products.

            -  Background matrix and presence of other  contaminants  will
               impact effectiveness.

2.   Reverse Osmosis

     a.    Based on the  application  of hydrostatic pressure to  drive the
          feed  water through  a  semipermeable  membrane  while  a major
          portion of its  impurity  content  remains  behind and is  dis-
          charged as  waste.  Principal design considerations are:

               Operating pressure (150 to 250 psi)
               Percent recovery (10  to 50 percent)
               Membrane type
               Pretreatment requirements

          A  schematic  of the  reverse  osmosis process  is  shown on  Fig-
          ure 6.

     b.    Generally,   reverse  osmosis  shows  promise  for  removing   low
          molecular  weight,  volatile  organic contaminants  and  certain
          pesticides.  Typical performance of reverse osmosis process  at
          an operating pressure of about  200  psi  is presented  in Table  5.

     c.    Principal advantage of reverse  osmosis process:

            -  It is  a proven technology  for  inorganics removal and  may,
               therefore,  be very  useful in situations  where  a  dual
               inorganic/organic problem  is present.

     d.    Limitations:

               Considerable pretreatment  may be  required,  depending  on
               the raw water quality.

               Low recovery.

3.   Mechanical Aeration

     a.    Surface or submerged  mechanical aerators are  used  to  entrain
          atmospheric air into the water by creating a  region of  intense
          turbulence  in the water.  Volatile  organics are removed  by the
          dispersed air.   Principal  design considerations  are:
                                 -5-

-------
          LEGEND

I. SOURCE WATER PUMPING
2. POLYMER FEED SYSTEM  '
3. FILTRATION
4. HIGH PRESSURE PUMP
5. pH AND SCALE CONTROL
6. REVERSE OSMOSIS
7. FILTER WASH WATER STORAGE
8. DISINFECTION-
9. FILTER BACKWASH
10. WASTEWATER DISPOSAL
I  I. CARTRIDGE FILTRATION
	EXISTING FACILITIES
     FIGURE 6- SCHEMATIC OF REVERSE OSMOSIS PROCESS

-------
                    TABLE 5

TYPICAL PERFORMANCE OF REVERSE OSMOSIS PROCESS


                                     % Removal

    VOCs

    1,2-Dich'loroethane                 15-70
    1,1,1-Trichloroethane              15-100
    Carbon tetrachloride                 95
    Trichloroethylene                   0-75
    Tetrachloroethylene                70-90
    Benzene                             2-18
    p-Dichlorobenzene                   0-10

    SOCs

    Acrylamide                          0-97
    Aldicarb                           94-99
    Alachlor                             100
    Carbofuran                         86-99
    1,2-Dichloropropane                10-90
    cis-1,2-Dichloroethylene            0-30
    trans 1,2-Dichloroethylene          0-30
    2,4-D                               1-65
    o-Dichlorobenzene                    65
    Ethylbenzene                         30
    Ethylene dibromide                 37-84
    Lindane                            50-75
    Methoxychlor                        >90
    Monochlorobenzene                  50-100
    PCBs                                 95
    Xylenes                            10-85

-------
            -  Aerator type
            -  Detention time (1 to 2 hours)
               Energy requirements

          Schematics of the two types of mechanical aerators are shown on
          Figure 7.

     b.   Performance of mechanical aerators for the  removal  of specific
          organics is not available.

     c.   Principal  advantages of mechanical  aerators appear to be:

            -  Simplicity of operation

            -  Applicable where there is a risk of  chemical precipitation
               or biological  growth  that  would tend to  clog a  packed
               column aerator

     d.   Limitations:

               High  energy requirement
            -  Long  detention time and,  therefore,  large  area  requirement
               Potential freezing of  open aeration  basins in cold climate
               Generally used for wastewater  treatment

4 .    Catenary Grid Aeration

     a.   Based on bringing the raw water  in contact  with air and trans-
          ferring the organics from water to  air phase.   Water flows down
          a column which is divided into sections by catenary  grids.  The
          shape  of  the  catenary grid  is designed so  that   a zone  of
          turbulence  is  created above  each  grid  section.    Principal
          design considerations are:

               Air/water ratio
            -  Number of grids
               Hydraulic loading rate

          A schematic of the catenary grid aerator  is shown on Figure 8.

     b.   Catenary grids can provide VOC removals  that are comparable to
          packed column aeration.  In general:

               Removal increases with increasing air/water ratio.

               Removal increases with increasing number of grids.

            -  Mass  transfer coefficient decreases  with increasing number
               of grids.
                                 -6-

-------
                   Drive
              Mechanical Surface Aerator
                 Drive
                               Compressor
                            .  . i.arapi:c
                            I '.Nl
          Turbine
          Sparger
               Suboerged Turbine Aerator
FIGURE 7- SCHEMATIC OF MECHANECAL
                 AERATION PROCESS

-------
                  DEMISTER
                  FLUIDIZED
                  ZONE
                  CATENARY
                  GRID (TYP.)
           TREATED WATER
           SAMPLE COLLECTOR
     MANOMETER FOR
     AIR FLOW RATE
     MEASUREMENT-r
BLOWER 7
                 AIR FLOW
                 DAMPER
                                                 WATER INLET
                                                    (TYP.)
 RAW WATER
 ROTAMETER

rRAW WATER
\SAMPLE TAP

 *
 TREATED WATER
 SAMPLE TRAP
   WATER FLOW
   METERING VAL

   RAW WATER
   FROM WELL
                                                TREATED WATER
                                                  TO DRAIN
     FIGURE 8- SCHEMATIC OF CATENARY GRID SYSTEM

-------
    c.   Principal  advantage  of  catenary  grid  aerator:

              Compact  design:   a  smaller  diameter  and  lower  column
              height than packed column aeration.

              Lower capital  cost   than packed  column because  of  more
              compact  design.

    d.   Limitations:

              The  scale-up procedure  from  pilot  to  full-scale is  not
              fully understood and  requires  further  development.

              The  air  flow pressure drop is relatively high, resulting
              in  larger  blowers and  higher  energy  costs  than  packed
              column aeration.

              Limited   treatability  data  are  available   for   several
              organics.
5.    Higee
     a.    The  process uses a rotating bed of packing with a high  surface
          area per unit volume.  Air is forced into the  rotating  packing
          element while the contaminated water  flows  from the center of
          the  packing  outward in  a counter  current  manner.   Principal
          design considerations are:

            -   Air/Water ratio
               Pump (Higee)  size

          A schematic of the Higee  system  is  shown on Figure 9.

     b.    Very  little  treatability  data  is  currently available   for
          organics removal.

     c.    Principal advantages of Higee  system appear to  be:

            -   Packing  requirements  are   less   than   in packed   column
               aeration for  equal removal  efficiencies.

               High  liquid  loading  rates  are possible  with relatively
               compact equipment

               Air volumes required are  reduced.

     d.    Limitations:

            -   Limited  treatability  data  are   available  for  organics
               removal.
                                 -7-

-------
                            EXHAUST AIR
   AIR IN
         BLOWER
GROUNDWATER
                FILTER
HIGEE
         PRODUCT
          WATER
                                               o
                                    PUMP
      FIGURE 9- SCHEMATIC OF HIGEE SYSTEM

-------
               Generally considered only under  special  requirements such
               as space  constraint,  height limitations  or exhaust  air
               treatments.

               Currently,   treatment  cost   is   high.   Appears   to   be
               cost-effective  only   for   high  organic   concentrations
               (greater than  1,000 ppm)  and  low flow  rates  (less  than
               100 gpm).
6.    Resins
     a.    Process uses special synthetic resins that have high adsorptive
          capacity for  organic compounds.   Principal design  considera-
          tions are:

               Resin  type
               Empty  bed contact time (EBCT)
               Regeneration frequency

     b.    Studies indicate that Ambersorb XE 340  (made by  Rohm and Haas)
          is effective in removing VOCs.

     c.    Principal advantages of synthetic resins appear to be:

            -  Shorter EBCT required than GAC
               Longer life than GAC
               Resin  regeneration on site using steam

     d.    Limitations:

               High cost  of  resins - $10 per  pound as compared  to  $.80
               per pound for GAC.

               Disposal of highly concentrated  regenerant water.

-------
                        III.  GRANULAR ACTIVATED CARBON

A.   Process Description

     1.   Based on  the  principle  of adsorption - the  transfer of a dissolved
          contaminant (adsorbate)  from a solvent  (solution)  to the surface of
          an adsorbent  (carbon).

     2.   Factors Affecting Adsorption Process

          a.   Type of  contaminant  (adsorbate)  -  Tables  6  and 7 present lists
               of readily adsorbed and poorly adsorbed organics, respectively.

                    branched-chain  compounds  more  adsorbable  than  straight-
                    chained compounds

                    increasing molecular weight increases adsorption

                    lower solubility increases adsorption

                    greater concentration,  increased adsorbability

          b.   Adsorbent

                    high degree of porosity
                    extensive internal surface area
                    affinity of adsorbate for absorbent (polar, nonpolar)

          c.   Aqueous Solution

                 -  temperature
                 -  pH
                    dissolved solids
                    other adsorbates

B.   Process Design Considerations

     1.   Key Parameters

          a.   contaminant

          b.   levels

          c.   GAG type

          d.   empty bed contact  time

          e.   carbon usage rate  - pounds of carbon  per 1,000 gallons of water
               treated

          f.   surface loading rate (2 to 10 gpm/sf)
                                      -9-

-------
                               TABLE 6

                      READILY ADSORBED ORGANICS



Aromatic Solvents

     Benzene, toluene, nitrobenzenes

Chlorinated Aroraatics

     PCBs, chlorobenzenes, chloronapthalene

Phenol and chlorophenols

Polynuclear Aromatics

     Acenapthene, benzopyrenes

Pesticides and herbicides

     DDT, aldrin, chlordane, heptachlor

Chlorinated non-aromatics

     Carbon tetrachloride, chloroalkyl ethers

High MW Hydrocarbons

     Dyes, gasoline, amines, humics



                               TABLE 7

                      POORLY ADSORBED ORGANICS

Alcohols
Low MW Ketones, Acids, and Aldehydes
Sugars and Starches
Very High MW Or Colloidal Organics
Low MW Aliphatics

-------
     g.    carbon depth (10-30 ft)

     h.    radon removal

2.    Contaminant and Levels

     a.    Effect of  different organics on  GAG design  is shown  on Fig-
          ure 10.

     b.    Effect of  contaminant levels on  GAG design  is shown  on Fig-
          ure 11.

     c.    Selected pretreatment can  lower contaminant loading  on carbon
          and thereby increase carbon life.   Example:

            -  Packed  column   for  volatile   organic   chemical  (VOC)
               reduction.

            -  Coagulation/filtration  for  turbidity and  total  organic
               chemical (TOG)  reduction.

2.    GAG Type

     a.    Low cost GAG -  low capital  cost, but more frequent reactivation
          and, therefore,  higher operating cost.

     b.    Total cost for  different GAG types are  shown on Figure 12.


4.    Empty Bed Contact Time (EBCT)

     a.    Affects capital costs
     b.    Average - 10 minutes for most organics
     c.    Calculated based on flow rate and GAG volume

          EBCT (min)  =    GAG volume  (cu ft)      7.48 _gal_
                       Flow rate  (gal per mm)         cu ft

5.    Carbon Usage Rate

     a.    Rate at which carbon capacity will be exhausted

     b.    Affects facility capital and operating  costs of the facility

     c.    Carbon usage rates  for  several  organics are shown in Table 8.
          In general,  pesticides are  better adsorbed than VOCs.
                                -10-

-------
   400
 w
 >
 u
 0.
    300
 o 200
 CQ
 cc
 o
    100
      0
EFFLUEMT CONCENTRATION 10ug/l


EBCT-10 MINUTES
                             TETRACHLOROETHYLENE
                                TRICHLOROETHYLENE
                                1,1,1,-TRICHLOROETHANE
       0  IOO 200 300  400 500600  700 800 900  IOOO


                  INFLUENT CONCENTRATION/ug/l
FIGURE 10- EFFECT OF TYPE OF COMPOUND OH CARBON LIFE

-------
   200
 u
 u.
o
CD
cc
<
o
   100
     0
       TRICHLOROETHYLENE

       10 MINUTE EBCT



EFFLUENT CONCENTRATION

          50jug/i

          10.ug/l

          1
         IOO 200 300 400 500 600  700 800 900  IOOO
            INFLUENT CONCENTRATION, oig/I
FIGURE 11- EFFECT OF CONTAMINANT LEVELS ON CARBON LIFE

-------
   70-
 £
 o
                                03 CARSON  COST  S 1 . 0 0 / I b

                                b= CARBON  COST  J 0 . I 0 / I b

                                « = C A S 8 ON  COST  1 0 .60 / I b
                                                a

                                                b
                                                c
                   1
                   4.
~2      4-      6      B      X)

 REACTIVATION FREQUENCY, MONTHS
T2
FIGURE 12- EFFECT OF CARBON TYPE ON FACILITY COST

-------
                                    TABLE  8

                         SUMMARY OF CARBON USAGE RATES
                                        Concentration  (ug/L)
                                        Influent   Effluent
Volatile Organic Chemicals  (VOCs):

  Tetrachloroethylene  (PCE)
  Trichloroethylene  (TCE)
  Trichloroethane  (TCA)

Pesticides:

  Chlordane
  Dibromochloropropane (DBCP)
  Aldicarb

Chlorinated Aromatics:

  Dichlorobenzene
  PCB (Aroclor 1016)
100
100
100
100
100
100
100
100
2
2
2
1
1
1
2
2
                     Carbon Usage
                    (lbs/1,000 gal)
0.08
0.16
0.96
0.012
0.016
0.02
0.01
0.015

-------
     6.    Radon Removal

          a.    EBCT

                 -  Requires large EBCT  (100  to 200 minutes).  EBCT  required
                    for various  removals  are presented on Figure 13.

          b.    Adsorption/decay

                    Radon that is  adsorbed on  the carbon decays continuously

                 -  Comparison of breakthrough  curves  for radon and  a  nonde-
                    caying adsorbate  is shown  on Figure 14.

                 -  Decay of  radon  acts   to continuously self-regenerate  the
                    carbon bed.

C.   Testing  to Evaluate Process Design Parameters
                                  ' *    .   .*.*.*.,       >
     1.    Isotherm Testing   , .  .      •

          a.    Useful screening  tool  for:

                 -  Determining  preliminary carbon  usage rates.

                 -  Evaluating relative  adsorbabilities of different  contami-
                    nants.

                    Evaluating the effects of  temperature  and pH on adsorption

                    Comparing the  performances of different carbons

                    Evaluating  the  relative  effects  of -other contaminants
                    present in raw water

          b.    Procedure:

                 -  Mix a  measured  weight of pulverized carbon in  water  of
                    known  organic  concentration  and  agitate  for  a  certain
                    contact time.

                 -  Measure  resultant  effluent  organic  concentration   and
                    calculate equilibrium capacity from the  amount of organic
                    adsorbed and the  known weight of  carbon in  solution.

          c.    The  relationship between  equilibrium  capacity  and   effluent
               concentration has  been found  to generally  follow  Freundlich
               isotherm relationship:

                                    /     v l/n
                                  x/m =  Kc
                                     -11-

-------
 100

  90

  80

  70


  60


  SO



  40
  30
  20
CZi
= 10
  a
I   I
  \
      20   40  °°  a°   100  120  MO  100  100  :oo
           EMPTY DEO CONTACT TIME. MINUTES
      FIGURE 13- EMPTY BED CONTACT TIME
         VS PERCENT RADON REMAINING

-------
Lu
UJ
LJ
   -100
    75
so:
      0
                           -TYPICAL GAC BREAKTHROUGH  CURVE
                           FOR NGN- DECAYING   A3SCR3ATE
                         ;—TYPICAL  STEADY-STATE,

25;
-
0 .
/
/ /
^Z— *—

* 1 I IVJ-^L. *J i u_^-^u i v > r^ J b«^
ADSORPTION/DECAY
OF RADON ON GAC


               10    [5     20    25   30    35   40

                      TIME,  days
          FIGURE 14- STEADV-STATE ADSORPTION/DECAY

                         CURVE FOR RADON

-------
          where:

          x/m    = equilibrium capacity (mg organic/gm carbon)
          c      = organic concentration (mg/L)
          K, 1/n = Freundlich constants

     d.   Isotherm constants  for  several  organics have  been reported by
          Dobbs  and  Cohen   (Carbon  Adsorption  Isotherms  for  Toxic
          Organics, EPA-600/8-80-023, April  1980).   Examples of the test
          results are shown on Figure 15 and in Table 9.

     e.   An  approximate  carbon  usage  can  be calculated  based  on  the
          isotherm equation.   For example,  assume:

          Trichloroethylene (TCE)  influent concentration  (C.) = 100 ug/L
          Trichloroethylene (TCE)  effluent concentration  (C  ) =   5 ug/L
          Safety factor (SF)               =0.75

          Isotherm equation:   x/m = Kc

          Where:

          K = 28 (mg)  (L)  1/n      (From Table 9)
                 (gin)  (mg)

          1/n =0.62               (From Table 9)

          Rearranging:

          Carbon Usage    =  Ci - Co   x 8.34   (Ibs)     (L)

          (lbs/1,000 gal)    K(Ci)1/nSF       (1,000 gal)   (mg)


                          = (100 - 5) ug     I    mg   8.34 Ibs       L
                            	L   1,000  ug X    1,000 gal X mg

                              28 ( IPO)0'62   0.75
                                 (1000)      X

                          =0.16 lbs/1,000  gal

2.   Minicolumn Testing

     a.   Used to determine:

            -  Feasibility of carbon treatment for a given water
            -  Preliminary process design criteria
            -  Rough estimate of system economics

     b.   A typical setup for dynamic minicolumn  test (DMCT) is shown on
          Figure 16.   The  procedure  involves  running  a sample  of  raw
          water through a short GAC column (about 70 mm deep) and analyz-

                                -12-

-------
                                               TABLE  9
                                  FREUNDLICH ISOTHERM CONSTANTS
                                       FOR  CARBON ADSORPTION
  COMPOUND-    Tetrachlorcethene (Tetrachloroethvlene)
   STRUCTURE:
                         Cl^       /Cl
                             c=r
                         cr
                                       MOL. Wt.
                                                165.83
FIEUNOIICH
PA«AME1E«S
K
l/n
pH
5.3
50.8
0.56
Coir. Co.l. r j g.96
INItlAl CONC. -g/l
1.0








ADSOtPTION CAPACITY, mg/g™
51
0.1 : 14.0
0.01 ' 3.9
0.001
1.1








                                                  COMPOUND:     Triehloroethene  (Trichioroethylene)
                                                   STRUCTURE:
                                                                         Cl.
                                                                         Cl
                                                                            /
                                                  FORMULA:
                                                                CyHCIi
                                                                                                  131.39
COMPOUND:   1,1,1-Trichlcroethane

 STRUCTURE:
                           Cl   H
                            I    I
                      Cl—C—C—H

                           Cl    H
FORMUIA:    C?H3C13

FBEUNDLICH
PARAMETEtS
K
l/n
Corr. Co.l. F
INITIAl CONC. mg/l
1.0
0.1
0.01
0.001

5.3
28.0
0.6?
0.99
pH








ADSOIMION CAPACITY, mg/jm
28
6.7
-'•6
0.38




~
--— 	
133.41
fKUNDLICH i
PAIAMETEIS '
K
l/n
Coir. Co«l. r
INIIIAl CONC ~,,l
1.0
O.I
	 ?.OI
O.C01
pH
5.3
2. S3
0.21



0.97 j




AOiOI'IION CAPACllY, mg/gm
2.5 {
!. 1
0.5! I
0.23




                                                                    REFERENCE: OOBBS AND COHEN (1880)

-------
   COMPOUND- — .-Te-trJf^ijorogthene (Tetrachiorcethy 1 e
OUND:   Trichloroethene (TricMoroethylene)
ma ADSORBED / gm CARBON
5 i
^ ,
<
o.o




too




to



1.0

























.xf














1
W"
ii>





k^-i






! i

i













I
i;

i .
1 I








[Till








i ' '


u^
*r! 	
i
|

s M'
^1









D! ' ' ' ' '"(J'.Ol ' ' ' ""&'.] ' ' ' '
RESIDUAL CONC (Cf|, mg/l


uf ~

i!

y
!l
ij. .
I

i
i
Hi
Y.'o












































Tft













'fd
!
•
z ,
0
S '
<
u
I '
n '

at
O ,
t/t
O
<
a '





O.C




10



1.0




0.1




























































"<










jf
*


Slboi









Y











































>





'6







—JL
If





.01







^













a






















1





-t- 	



-W 	 1
1
1


'o.'i '

	


/i









* *

	 L J.
4-L
Tnf
J j









.0
                                                                  RESIDUAL CONC (C,), mg/l
                           COMPOUND- .„..,„? tK|-Trich1oroetriane
m CARBON
mg ADSORBED / t
< '
i
i
o!
	 . — . — . ; i : .; i 	




10





! !




1
! i .



l.C

--•'
0.1




J






j^


i





1
"T i i





j



I i



i
T""^^
























































j




T
It"


L-ftp







D01 ' ' ""d 01 ' ' ' ' "o.l ' ' ' '
RESIDUAL CONC (C(). mg/l



I





*JL 	









^











H












==:







"l.O ' ' '





~T
-b







10
             FIGURE 15- CARBOM ISOTHERM TEST  RESULTS
REFERENCE: DOBBS AND  COHEN (1980)

-------
 11.1
          YtSlOM OAI tAMTI MO 1AO
11
I . u Knout
         nn ON njBjHO i«~« c • >.»«• oo
            =4=1
   btf OLAU (Aitftl COLLf CTKM CVMMO4
   WITH ADATTU) U^TUH HU*
                                                                    -QO
                                                                      \
-4-
\
$
x •
V
\
X
^
s.
1
\
$
\





Ij
t&
•IJ
iv(
$
$
•3
T\
'••i
&
t^r
***•
•\
^
x-
V
v
1
-**^
\;
s
^Y
\'
X
^X
^r*
X
                                                                                        ii trta cotm
                                                                                        ^«
                                                                                        D > i *M oo
                                                                                    - rntu woo.
                                                                            1
                                                                    I u ITAMLIU 8TUL f
                                                                             DUCAT COLUMN
                FIGURE 16-  DIAGRAM OF  DYNAMIC  MINI-COLUMN
                                     ADSORPTION SYSTEftl

-------
               ing the effluent  concentration.   Analysis of  the  test results
               is discussed in the following section.

     3.    Dynamic Column

          a.    Used to simulate full-scale GAC facility  operation and develop
               following  design criteria:

                 -  Effect of various contact times
                    Optimum carbon bed depth
                 -  Effect of hydraulic loading
                 -  Number of vessels needed
                    Type  of GAC needed
                 -  Carbon life/changeout  frequency
                    Carbon exhaustion rate
                    Contaminant loading rates

          b.    A typical set-up  for  pilot column test is  shown  on Figure 17.
               Raw water is run through GAC  columns  (about five-foot deep and
               4-inch I.D.).  Effluent  is sampled to obtain  the breakthrough
               curve (time versus concentration relationship).

D.   Optimizing Design Criteria

     1.    Optimize  key design parameters  to minimize treatment  costs.   Cost
          components shown  on Figure 18  indicate  that  key design parameters
          are:

            -  EBCT
               Carbon usage

     2.    Optimizing EBCT

          a.    Plot breakthrough  curves -  examples  of breakthrough curves for
               cis-1,2 dichloroethylene (DCE) and TOC are  given  on Figures 19
               and 20, respectively.

          b.    Calculate carbon usage rates:

               CU  (lbs/1000 gal) =  Mass of Carbon (Ibs)  x 1,000 gallons
                                   Volume Treated to Breakthrough  (gallons)

          c.    Plot  carbon  usage versus  EBCT - example  for  DCE  and TOC are
               shown on Figures 21 and 22, respectively.

          d.    Select optimum EBCT - 15 minutes for DCE,  20 minutes for .TOC

     3.   Parallel versus series  flow

          a.    Parallel   flow   (shown  on  Figure  23)    necessitates  carbon
               replacement at breakthrough
                                     -13-

-------
     FEED      PRE
     TANK- TREATMENT1
CARBON
 FEED
 TANK
CARBON COLUMNS
                                SAMPLE
                                 TAPS  -
Pretreatment tanks optional depending on the
suspended solids concentration in feed.
                                     PRODUCT
                                      WATER
                                                          BACK WASH
                                                          WATER
       FIGURE 17- DIAGRAM OF PILOT COLUMN TEST SYSTEM

-------
CARBON CONTACTOR
      (50%)
     Captiai  Cost
                                     CARBON REGENERATION
                                           (30%)
                                           CARBON TRANSFER
                                           AND STORAGE (5%)
                                     SITEWORK ETC.
                                         (15%)
                                    LABOR (10%)
     O&M Cost
                  CARBON REGENERATION
                          (75%)
                                            POWER (10%)
                                              MAINTENANCE
                                              MATERIAL (5%)
          FIGURE 18- GAC FACILITY COST COMPONENTS

-------
CT

•-*

o
i-
<
     LU
     o

     6
     o
          60
          50
          40
          30
          20
          10

EBCT -
7.2 MINUTES -



/

V
/



-
rEBCT =
15.2 MINUTES — v
\
I
1
,
,'
//

/^/" EBCT =
' -y- 16.4 MINUTES
/
J
/

j
P
j
t
                      too
                                   200
                         TIME (DAYS)



Note:   Ci = Influent Concentration

      Ce= Effluent Concentration
                                                1.O
                                                0.8
                                                0.6
                                                    Ce
                                               tO.4
                                              300
   FIGURE 19- BREAKTHROUGH CURVES FOR c-1,2 DCE

-------
V.
Cl

 E
2
LJ
O
2
O
O
                                     EBCT =  i
                                    10 MINUTES!
                                     EBCT -  •
                                    20 MINUTES
                                   _ E2CT =
                                   ;30 MINUTES!
40 MINUTES |
             40
                    80
                           120
                                   160
                                          200
                                                   i.o
                                                   0.75
                                                   0.5
               0.25
                                                        Ce
                                                        CJ
                                                 240
                       TIME (DAYS)


Note:  Ce = Effluent Concentration
       Ci = influent Concentration
   FIGURE 20- BREAKTHROUGH CURVES FOR TOO

-------
  O)
  o
  O
  o
  LU
  O
  <
  CO
  D

  Z
  O
  CQ
  DC
  <
  O
     0.25
0.2
                               BREAKTHROUGH
                            10
                                15
20
                      EBCT (MINUTES)
FIGURE 21- EBCT VERSUS CARBON USAGE FOR c-1,2 DCE

-------
 o

 O
 o
 O
 HJ
 o
 :D
 2
 O
 03
 cc
 <
 o
 BREAKTHROUGH
             10
20
30
40
50
                  EBCT (MINUTES)
FIGURE 22- EBCT VERSUS CARBON USAGE FOR TOC

-------
     INFLUENT
    EFFLUENT
      o
   z

   Q
   o
   m
   cc
   <
   o
              MAXIMUM LOADING
UNUSED CAPACITY
               BED DEPTH
FIGURE 23- PARALLEL MODE OF OPERATION

-------
          b.   Series  flow   (shown   on   Figure 24)   provides  for   carbon
               replacement at exhaustion

          c.   Optimize  between  number  of  contactors  (capital  cost)  and
               regeneration frequency (operating cost)

          d.   Cost comparison between  the  two modes are shown on  Figures 25
               and 26 for DCE and TOC,  respectively.

E.   Facility Design Considerations

     1.   Major Process Elements

          a.   Carbon contactors
          b.   Regeneration system

     2.   Carbon Contactor Configuration

          a.   Upflow

                    Can be operated in  parallel or series mode

                    Bed  allows  suspended  solids  to pass without  developing
                    excessive pressure  drop

                    Release of carbon fines in effluent stream is of concern

                    More widespread use for wastewater  treatment

          b.   Downflow

                    Can be operated in  parallel or series mode

                    Small amount of  suspended matter can be  handled by  back-
                    washing

                    Can be operated in  pressure or gravity mode

                    Pressure  contactor   (shown  on  Figure 27)   offers   more
                    flexibility as the  system can be operated  with higher head
                    loss

                 -  Gravity contactor (shown on Figure  28)  are less expensive
                    due to common wall  construction

     3.   GAC regeneration:

          a.   On-Site Regeneration - economical where  carbon exhaustion rate
               is greater than 1,500 pounds per day.
                                     -14-

-------
  INFLUENT
                               o
                     EFFLUENT
   O
   o
   _J
   2:
   o
   CQ
   CC
   <
   O
             MAXIMUM LOADING
UNUSED CAPACITY
              BED DEPTH
FIGURE 24- SERIES MODE OF OPERATION

-------
    10,000
     1000
£T
<
UJ
C/)
o
u
      100
       10
                                    /
                                  /,
                                 /.
                               /
              SERIES
/.
                            /.
                           PARALLEL
                         10
           100
                PLANT CAPACITY (MGD)




    FIGURE 25- COST COMPARISON OF PARALLEL

             VERSUS SERIES FLOW FOR c-1,2 DCE

-------
cc
<
LU
03

O

O
    10,000
     1000
      100
       10
              PARALLEL
                             SERIES
                         10
100
                PLANT CAPACITY (MGD)
     FIGURE 26» COST COMPARISON OF PARALLEL


                VERSUS SERIES FLOW FOR TOC

-------
         INFLUENT
                        I
   20,000 LB EA.
    GRANULAR
ACTIVATED CARBON
                     90p6i
                         T

   T T T

I * ^- - - - 5 o « f &f:
i^ ,-.1*4 •>   \'i
.=•:;•'
                                     ?,
             COLLECTOR

             SYSTEM
                                       TO SYSTEM —
                                                   • TREATED WATER
     FIGURE 27- SCHEMATIC OF PRESSURE GAC CONTACTORS

-------
SURFACE
WASH
 INFLUENT
        EFFLUENT-
                                                       GRAVEL

                                                       FILTER BLOCKS
                               'DRAIN
               FIGURE 28- SCHEMATIC OF DOWNFLOW
                          GRAVITY GAC COMTACTOR

-------
          b.   Off-Site Regeneration - economical where carbon exhaustion rate
               falls between 500 and 1,500 pounds per day-

          c.   Off-Site Disposal - economical where  carbon  exhaustion rate is
               less than 500 pounds per day.

          d.   Cost comparison of different carbon recharge  options are shown
               on Figure 29.

     4.   Operational Issues

          a.   Desorption

                    Due to decrease in influent  concentration

                 -  Due to chromatographic effects (displacement of an organic
                    compound by another that  is  more  favorably adsorbed)

          b.   Bacterial growth

                    Could enhance removals by  biodegradation
                    Causes clogging of the carbon bed
                 -  Causes higher plant count  in treated water

          c.   Backwash

                 -  Minimized by providing adequate pretreatment.   Backwashing
                    should be gentle in order  not to  fluidize the carbon bed.

     5.   Waste Disposal

          a.   Backwash

                 -  Carry-over of carbon fines during initial backwashing

          b.   Spent carbon

                    Regeneration on-site or off-site
                 -  Disposal in a hazardous waste landfill

F.   Applicability to Various System Sizes

     1.   Organics Removal:

          a.   <1 mgd, package pressure contactors
          b.   1 to 10 mgd,  pressure contactors
          c.   >10 mgd, gravity contactors

     2.   Radon Removal:

          a.   Small systems, home GAC units  - see Figure 30
          b.   Large systems, not economically feasible
                                     -15-

-------
   120
   100
    80

Z
o
03

<   60

O
CQ
z
UJ
O
   40
   20
     100
            CARBON REPLACEMENT
FLUIDIZED

BED REACTOR
                  MUTIPLE
HEARTH
                    1000
                     10,000
              CARBON REQUIREMENT (LBS/DAY)
            100,000
             FIGURE 29- COST COMPARISON OF

                     CARBON RECHARGE OPTIONS

-------
(tn
i




. • •
• •
. * •
. • •
• • *
• • "
. •
> * *
• * *
. • •
* * '
— . •
, • •
• • "
• • '
. • •
. • •




1


T 	



/
S

•"'•'•'*•
".*••' * "."•••


^\
2
f- -r
i



— r-"

7ntt£^h-
- -^--,



3



\
LEGEND: 1 - MAIN WATER LINE
2 • HYDROPNEUMATIC TANK
3 - CAC UNIT
4 - TREATED WATER
-3»»- A 5 • 3ACKWASH TO DRAIN


5
.^:-::i:-:::-::-:-^:-:::::l:i--:::l:l--:::l-i
FIGURE 30- TYPICAL HOME GAC UNIT

-------
G.   Treatment Economics
          Capital cost components include:

          Basic               Site Specific

          Contactors          Special sitework
          Activated carbon    Raw water holding tank (for ground water
                                systems)
          Regeneration        New/restaged well pump (for ground water
          facility              systems)
          Carbon storage      GAC contactor building

          Carbon transport    Chemical facility
          facilities
                              Clearwell

                              Finished water pump(s)

                              Backwash storage

          Operation and maintenance  (O&M) cost components include:

               carbon make-up
               labor
            -  fuel
               steam
               power
               maintenance
               laboratory analyses

          Cost curves  for different  contaminant  groups are  shown on Figure 31
           (capital cost) and Figure  32  (O&M cost).  In general:

               Pesticides -                  least costly
               Chlorinated aromatics -            I
               VOCs -                             y
               Radon -                       most costly
                                     -16-

-------
1,000,000




100,000



^ 10,000
S£
^^
frt
t-
co
o t.ooo
_J
<
H-
o.
<
o
100




1 ft
0.
























01




































J















4-k™™_




'





II

[ '

|H X K>y , ^* ^ iih Hi nil i 100 /* r s I I r i j ^ /• 1 [ 1 ' ' : - 11 • / \]// \ • lit III i - : i i •' ii h ii i M III 1000 Pt # I 1 1 "• 1 . . 1 ! i 1! • ii ! •• i 1 1 ^_ * f f 1 1 !-lT[ ; i i PLANT DESIGN CAPACITY (MGD) ._.».» ALACHLOR (PESTICIDE) — TCE (VOC) - — RAOON FIGURE 31 - CAPITAL COST CURVES FOR GAG FACILITIES


-------




10,000

•x.
V>
X 1,000
CO
H
W
O
u
5 100
08

O


10




1
0.








-
















01
































i 1



I :


j













""T~

'
1
I

1 1 "
1,
|
TT"""1
T •
I t 1
I:
i

i '
1 1
1 1
ii
i
1


1 t
!
!
I
0.













/





S
^





1













v




^i
^^
,<






i » ' t





1




'J
Xj


1

.^
^
^







11
-11
II


III
III
/
mj/
'if r
i i
1 1 !


• jr

i! j^
$'
II-





1.0







/
/




jt


+*
















f





s
















P^3
-/•
r"






<
i*~















/


:
ll


i .S
/
;-•



ii







10 .










s


r





















P^TT

jg~
ki


















i , -
i
^
IX
i^r • ^
1 ' '^^
rfi
1



I t









100






>^
S\
>|'
^ . j ,

'




!













Ii :

^ •
!••
in
•'HI


t ; ' .
ili'
i



1 I I rl
Illi



1 1 1 M
} 1
III!
1000





























I



1



1















"T
i ,
1 1
1
1


I '
i j


i •
-1?
I!


: I
1 \
1
II



i
I



!



1 1
1







i
T




T




            PLANT OPERATING CAPACITY (MGD)
	ALACHLOR CPESTCOO
	  TCE (VOC)
—'  RADON
         FIGURE 32 - O & M COST VURVES FOR
                          GAC FACILITIES

-------
                          IV.  PACKED COLUMN AERATION

A.   Process Description

     1.   Based  on the principle  of mass  transfer  from liquid  phase to gas
          phase.   Rate  of  mass  transfer proceeds   according   to   following
          equation:

               M = KT a A P
                    j_j

               Where:  M   = mass of substance transferred per unit time and
                             volume  (Ib/hr/cf)

                       K   = coefficient of mass transfer (Ib/hr/sf)
                        LJ

                       a   = effective area (sf/cf)

                       A P = concentration difference or driving  force

     2.   Driving force is the differencte between actual conditions in the air
          stripping  unit  and  conditions  associated with  equilibrium between
          the gas and liquid phases.

     3.   Henry's Law

          a.   Equilibrium  concentration  follows  Henry's  Law,   which states
               that the amount of gas dissolves in a given quantity of liquid,
               at constant temperature and total pressure, is directly propor-
               tional to  the partial pressure of the  gas  above  the  solution.
               Henry's constant calculated as follows:

                    H (dimensionless units) = (16.04)   (P) (M)
                                                  (T)   (S)

                    P = vapor pressure in mm
                    M = gram molecular weight of solute
                    T = temperature in degrees Kelvin
                    S = solubility in mg/L

          b.   A compound's Henry's Law constant indicates relative volatility
               of the compound;  high Henry's  Law  constant - easily removed by
               air stripping.

          c.   Henry's Constants for several chemicals  (at 20 C):

                                                       Henry's Constant
                                                       	(a tin)	
               VOCs

                         Vinyl chloride:                     2,985
                      -  Trichloroethane (TCA):               286
                                     -17-

-------
                      -   Tetrachloroethylene (PCE):            274
                      -   Trichloroethylene  (TCE):              155

               Pesticides

                      -   Chlordane:                              5.35
                         Dibromochloropropane (DBCP):           13.8
                      -   Aldicarb:                           2.32 x 10

               Chlorinated Aromatics

                      -   Dichlorobenzene:                      108
                      -   PCB (Arochlor 1242):                  78.6

               Radon                                         2,260

     4.    A schematic of  the  packed  column  aeration  system is shown on  Fig-
          ure  33.

B.   Process Design Considerations

     1.    Key  Parameters:

          a.   contaminant

          b.   levels

          c.   type of packing material

          d.   A:W ratio (cubic feet  per cubic  feet)

          e.   Liquid loading  rate  (gpm/sf)

          f.   Packing height  (ft)

          g.   water  temperature

     2.    Contaminant and Levels

          a.   Higher the  volatility of a  compound,  the  more  easily  it is
               stripped  by aeration.

          b.   Effect of different organics  on packed column design is shown
               on  Figure 34.

     3.    Type of  Packing Material

          a.   Packing materials are  designed to provide:

                   maximum air-water contact area
                   low  air pressure  drop
                                     -18-

-------
  INFLUENT
          EFFLUENT
                       .r/^r/V-"^/^
                       "j^-^'-'^Vx^
                            u
                                         LIQUID DISTRIBUTOR
                                      PACKING MATERIAL
                                       PACKING SUPPORT
                                         AIR  IN
FIGURE 33- DIAGRAM OF PACKED COLUMN AERATOR

-------
  100
   80
H-
LL
I
Q.  60
LJ
Q
0
Z
u
<
Q.
   40
   20
        CHLOROFORM
           TCE
          1,2 DICHLOROETHANE
                          95% REMOVAL
                              55°F
          20/1
40/1
60/1
80/1
100/1   120/1
                           A/W RATIO
  FIGURE 34- EFFECT OF COMPOUND ON PACKED COLUMN DESIGN

-------
          b.    Generally made  of  plastic  or  ceramics,  and  vary  in  sizes
               (1/2  inch to 4  inches)

          c.    Commonly used packing materials include:

                    Super intalox
                    Tellerettes
                    Tri-packs
                    Pall rings
                    Berl saddles
                    Raschig rings

     4.   Air to Water Ratio

          a.    Function of:

                 -  Contaminant type
                 -  Water temperature

          b.    Determines the  blower size and operating  cost of the system

          c.    Typical range of air/water ratio are:

                     30:1  - Highly volatile compounds  (e.g.,  TCE)
                    200:1  - Highly soluble compounds  (e.g.,  DBCP)

     5.   Liquid Loading Rate

          a.    Determines the  column diameter
          b.    Typically 25-30  gpm/ft

     6.   Packing Height

          a.    Function of:

                 -  Air/Water  ratio
                    Required removal

          b.    Determines the  capital  cost of the system

     7.   Water Temperature

          a.    Solubility  of organic  compounds  in  water generally  decreases
               with increase in temperature

          b.    Impact  of  water  temperature  on  column design  is  shown  on
               Figure 35.

C.    Testing to Evaluate Process  Design Parameters

     1.   A  schematic  of  the  pilot  column  used for  testing packed  column
          aeration is shewn on Figure  36.
                                     -19-

-------
o
o

O
<
Q_
20
2C
I C
C
c
5
55
BO
LIQUID
AIR : w;
CONTAM



LOADING RATE-3Qgpm/sf
;TER RATIO = so: i
INANT = TCE



^^'^^*'
^^



^'
^^







^'
jf
	 •*"
--— ^^

^^


_- 	 "
~. — "^



'

^^




0 75 90 95 97.5 99 99.5 99.8
Defl F REMOVAL EFFICIENCY (%)
Deg F
           FIGURE 35- EFFECT OF TEMPERATURE

                    ON PACKED COLUMN DESIGN

-------
       -ATEH FLOW
        ".ETERISG VALVE
                 HATER METER
 	rVo_v^V
       *"	i^-^t—a  /r
FROM RAH WATER      ^^
    SOURCE
             RAW WATER
             SAMPLE TAP
                                        EFFLUENT ilR

                                             i
                       EFFLUENT WATER
                       SAMPLE  TAP
                    EFFLUENT WATER
                      TO DRAIN
                                                a
                                          \
                                                   /
                                                       -HATER DISTRIBUTION  PLATE
                                                      -PACKING MEDIA
                                                        INTERMEDIATE SAMPLE TAP (TYP.)
                                                     •SUPPORT PLATE (TYP.)
                                                        IONNECTING  FLANGE (TYP.)
•TOWER SUPPORT TABLE
           AIR FLOW
           METERING VALVE

                 PRESSURE
                   GAGE
120  VOLTS
SINGLE FH«*:
15 AMPS
BLOntfi-
 •INFLUEUT AIR
              FIGURE 36- SCHEMATIC OF PILOT AERATION COLUMN

-------
     2.   Test variables are:

         a.   Hydraulic loading  rate
         b.   Air  flow rate
         c.   Air:Water  (A:W)  ratio
         d.   Type  of packing

     3.   Testing Procedure:

         a.   Water and  air  flowrates  are  varied  to  obtain  several A:W
              ratios.  For each  run,  the  column is  operated  for about  30 min-
              utes  to  achieve  steady state.  Eight to twelve runs are  recom-
              mended for  each  test.

         b.   For  each  run,   influent  and effluent  samples  are collected,
              along with  air  and water flowrates, and air and  water tempera-
              tures.

         c.   The  procedure is repeated for different packing materials.

         d.   Precautions:   In order to  ensure  reliable  data  from the  pilot
              study, the  following factors  have to  be considered:

                    Packing  material  should  be  dumped  carefully to   avoid
                    channeling  and  void  spaces  in the column.

                    The  pilot  column  should be  on  level  ground to minimize
                    channeling  or wall effects.

                    Sampling  points  should  be  located so that  the  samples
                    truly  represent the  process  performance.

                    Samples should  be  collected  in duplicates.

                -   Accuracy of the laboratory  analysis method should first be
                    verified.

D.   Development of Design Criteria

     1.   Design of  packed  column  aeration  system  involves   the following
         relationships:

         a.   Z =  (HTU)  (NTU), where:

              Z   = packing height (ft)
              HTU  = Height of  transfer  unit (ft)
              NTU  = Number of  transfer  units
                                     -20-

-------
          b.   HTU =   L  , where:
                     K aC
                      L  O

               L  = liquid flow, Ib mole/hr/sq ft
               C  = molar density of water, Ib mole/cu ft
                o
               NTU =  R
                     R-l
                         In
(x./x  )  (R-l)  + 1
                              a.  o
       R
               Where:  x./x  = Ratio of influent to effluent liquid phase
                               concentration
                       R     = Dimensionless stripping factor

          d.   (A) (H)  MW       DensityAir
                 (W)     MW .     Density
                          Air          Water

               = A (H) (7.512 x Id"4)
               Where:  A = Air Flowrate (cfm)
                       W = Water Flowrate (cfm)
                       H = Henry's Constant  (atm)

     2.   Develop mass transfer relationships as shown on Figure 37.

     3.   Select  the  packing material  with largest  K a,  and  optimum liquid
          loading rate (gpm/sf).

     4.   Estimate  optimum  A/W  ratio based on A/W  ratio  -  packing height
          relationship, as shown on Figure 38.

     5.   Nomographs  for preliminary design  of  packed  columns  have  been
          developed  by  Speece,  Niamalakhandan  and  Lee  (Nomograph   for  Air
          Stripping of VOC  from  Water, Journal  of  Environmental Engineering,
          Vol. 113, No. 2, April 1987).

E.   Facility Design Considerations

     1.   iMajor Process Elements

          a.    Column and column  intervals
          b.    Packing medium
          c.    Blower
          d.    Booster pump
          e.    Instrumentation and control

     2.   A schematic  diagram of  a  packed  column  facility  is  shown  on Fig-
          ure  39.

-------
   200
   150
V)
V.
13
o
-C
100

 90

 80

 70


 60


 50



 40




 30
      10
    40
                        TETRACHLOROETHUENE
                                           THEORETICAL

                                           2" TELLERETTES
                                        2" TRI-PACKS
            15     20        30      40    50   60  70  80 90 100

                 LIQUID LOADING RATE (gpm/sf)
      10        15      20         30     40   50   60  70  80 90 100

                    LIQUID  LOADING RATE (gpm/sf)

          FIGURE 37- MASS TRANSFER RELATIONSHIP

                            FOR PACKED COLUMN

-------
      10
20     30      40      50



 AIR TO WATER RATIO (cf :  cf)
60      70
LIQUID LOADING RATE = 30 GPM/SF
   FIGURE 38- PACKING HEIGHT VS AIR TO WATER RATIO

-------
TREATED WATER
 TO RESERVOIR
                                  EXHAU3T
                                     AIR    A
                                              INFLUENT
                                               WATER
WELL NO.6
              FIGURE 39- SCHEMATIC OF A TYPICAL

                        PACKED COLUMN FACILITY

-------
3.   Location/Site Constraints

     a.   Zoning requirements
     b.   Height restrictions
     c.   Location of air intake louvers

4.   System Hydraulics

     a.   Restaging well pumps
     b.   Flow and system pressure
     c.   Repumping to distribution system

5.   Housing

     a.   Freezing potential
     b.   Noise
     c.   Security
     d.   Equipment maintenance

6.   Column and Column Internals

     a.   Column Construction

               FRP (fiberglass-reinforced plastic)
               Aluminum
               Stainless steel
            -  Concrete

     b.   Mist eliminator

     c.   Liquid distributor

               orifice plate (shown on Figure 40)
            -  trough-type distributor (shown on Figure 40)
               orifice headers
               spray nozzles

     d.   Support grid

     e.   Packing Media

7.   Air Quality

     a.   Intake air - air-bourne contaminants
     b.   Exist air - discharge regulations

8.   VOC Emissions

     a.   Emission rate in the exit air is calculated based upon:
                                -22-

-------

     Orifice - type distributor
   Trough - type distributor
FIGURE 40~ DISTRIBUTOR TYPES

-------
          Emission rate (Ibs/hr)  = (C.  - C )  x V x  5

                                                    io7

          Where:    C.  = Influent concentration in raw water (ug/L)
                     i
                    C  = Effluent concentration in treated water (ug/L)
                     e
                    V  = Water flow rate (gpm)

     b.   Ambient concentrations

     c.   Modeling

     d.   Column modifications

            -  Height
               Air flowrate
               Exist velocity

     e.   Treatment Options:

               Thermal  incineration  of  the  organics  -  Disadvantage  of
               high energy requirements

               Catalytic   incineration  of   the  organics    (shown   on
               Figure 41)  -  Lower temperature requirements  than thermal
               incineration,   but  currently  not  effective   for  removing
               chlorinated organics at  low levels.

               Ozone  destruction  -   catalyzed   by   UV   radiation.    At
               present, limited application for vapor-phase treatment.

               Vapor  phase  carbon  adsorption   (shown  on  Figure 42)
               Appears to be  the most feasible method at  present.  Carbon
               replacement frequency can be estimated by:

                    Mass  balance  based   on  packed  column   exit   air
                    concentration and flow rate.

                    Monitoring   using    gas    chromatography   or   mass
                    spectrometry.   Samples can be collected in  bags  and
                    steel  cannisters  or  using   carbon   and   synthetic
                    resins.

                 -  Using a  combination  of mass  balance  (initially)  and
                    monitoring (towards the anticipated carbon  exhaustion
                    period).

9.   Clogging of Packing

     a.   Iron
                                -23-

-------
                                           Recirc. damper
                                           or secondary
                                           heat exchanger
Filter/
mixer
                                     ~\     Secondary
                                 n n "f—air
                                   \J  (/   addition
  FIGURE 41 - SCHEMATIC OF CATALYTIC
                 INCINERATION PROCESS

-------
                                         TREATED
CLEAtt
                         H£ATfN<3    BLOWER
 SLOWEf?
        TREATED
            FIGURE 4 2- SCH6WAT1C OF VAPOft
                       PHASE <5AG SYSTEM

-------
          b.    Solids

          c.    Biological growth

          d.    Pretreatment requirements may have to be considered for any one
               of these problems

     10.   Corrosivity of Treated Water

          a.    Problem:  increased DO, reduced CO
          b.    Solution:  increase pH; provide post treatment

F.    Applicability to Various System Sizes

     1.   Organics removal

          a.    All system sizes
          b.    Typically used for ground water systems

     2.   Radon removal

          a.    Small systems:  home diffused aeration units preferred

          b.    Large systems:   use  may be  limited  by radon emission  in  exit
               air

G.    Treatment Economics

     1.   Packed column cost components.

               Basic                    Site Specific

               Column Structure         Special Sitework
               Internals                Raw water holding tank
               Packing                  New/restaged well pump
               Blower(s)                Blower building
               Clearwell                Booster pump building
               Booster pump(s)           Chemical facility
               Piping                   Noise control installation
                                        Air emissions control

     2.   Cost curves for different contaminant  groups  are  shown  on Figure 43
          (capital cost) and Figure 44  (O&M cost).   In general:

               Vinyl Chloride                - least costly to remove
               PCE
               TCE
               Carbon Tetrachloride
               1,2-Dichloroethane
               DBCP                          - most costly to remove
                                     -24-

-------
1,000,000
100,000
-> 10,000
CO
h-
to
o 1,000
D.
u
100
10
0.
[ ' 1 r~.~7";~."r~ 	 r — ; — : — r~r
1 	 ' 	 	 r f ' •* 	 ' 	 1 	 1 — f-f-













; 1
'.i
M












1




! "

1
! 1 :
I'll




T7T — " 	 ! 	 ; — . 7 i -, , . : ; ; ;;•;"•",



II










i i
i







|:
I
1






i
i


.; :; .. ;. / , .... — : : . . :. ..j
, ,



1 "l 1 1 i : 1
• / ; • •
X,


i
i t



1 1 . 1 1
'!
-T- 	 1 i !/i* 	 Ss 	 : 	 —
i Six.''
.
Xi

1
i i S








i


11 ' i
iii
i!
i!




i


i 	 1 	 ( 	 ; . I ; i i 	 • 	 1 — i | | |













• 1


• ! , ^^
i i ^ ^
\ Jxi
,'

— , — , , , ., ; — e, — , — ^_|_





i





I












1 i
1 :
~J


^'
• : . 'jT'. JT .
I/ - -^^ ^
/
II X^X
X^^
Ssj^. 	 ,
' 1 1 ^ i ^ '
1 {
























i

f , i


X X" : • i
• X.; ,.

I i
ij


:'"




il


i
il
•'
IP






i









!




j ; • [ :
1 ;
!

1 I
1 jl










01 0.1 1.0 10 100






i




H
1!
• :" i •






~T


i



i
M











1000
                     PLANT DESIGN CAPACITY (MGD)
	__  VINYL CHLORIDE. RADON
     '  TCE
	  DBCP (PESTICIDE)
              FIGURE 43 - CAPITAL COST CURVES FOR
                    PACKED COLUMN AERATION FACILITIES

-------
   10,000
cc
<
ai
*   1,000
I-
W
O
o
08

O
     100
      10
       0.01

                          V


                 0.1
1.0
10
100
1000
                    PLANT OPERATING CAPACITY (MGD)
                                   I
                                                           IT
           —• VINYL CHLORIDE. RAOON

          — TCE

          — DBCP (PESTICCE)
                  FIGURE 44 - O & M COST CURVES FOR


                     PACKED COLUMN AERATION FACILITIES

-------
3.   Vapor Phase GAC Cost

     a.   May dictate system economics and choice of the process.

     b.   Typical costs for different system sizes  at  air  to water ratio
          of 40:1 are:

               System Size (MGD)              Capital Cost (K$)

                        1                            40
                       10                           200
                      100                         1,400
                    1,000                         9,000

     c.   Cost  includes  carbon contactors,  initial carbon  charge,  gas
          heater and installation.
                               -25-

-------
                               V.  BIBLIOGRAPHY
PAC Adsorption Processes

Aly, 0. M. and S. D. Faust, "Removal of 2,4 Dichlorophenoxyacetic Acid Deriva-
tives from Natural Waters." JAWWA, Vol. 57:221-230, 1965.

Baker,  D.,  "Herbicide  Contamination  in  Municipal Water  Supplies  in  North-
western Ohio."  Final  Draft  Report  1983.  Prepared for  Great  Lakes National
Program Office, U.S. Environmental Protection Agency,  1983.

Cohen, J. M. ,  L.  J. Kamphake, A.  E.  Lemke, C.  Henderson  and R. L. Woodward,
"Effects  of Fish Poisons on Water  Supplies",  Part 1, Removal  of  Toxic Mate-
rials JAWWA 52:12, 1551-1566, 1960.

Croll, B.  T.,  G. M.  Arkell,  and  R.  P. J.  Hodge, "Residues  of Acrylamide in
Water", Water Research.  Vol. 8:989-993, 1974.

ESE,  Alternative  Powdered Activated  Carbon Study, Prepared  for the City of
North Miami Beach, Gainesville, Florida, 1979.

Lettinga,  G.,  W. A. Beverloo, W. C. van Lief, "The Use of Flocculated Powdered
Activated Carbon  in Water Treatment". Prog. Wat.  Tech., Vol. 10,  Nos.  1-1/2,
pp. 537-554, 1978.

Miltner,  ?.. J.  and  C.  A. Fronk,  "Treatment  of  Synthetic Organic Contaminants
for Phase II Regulations", Internal Report DWRD, EPA,  December 1985.

Robeck, G. G., K. A. Dostal, J. M.  Cohen  and J. F. Kreissl,  "Effectiveness of
Water Treatment Processes in Pesticide Removal". JAWWA 57:2,  181-199, 1965.

Singley,  J. E.,  B.  A.  Beaudet,  and A. L.  Ervin, "Use of Powdered Activated
Carbon  for  Removal of Specific Organic  Compounds,"  Proceedings  of Seminar,
Controlling Organics   in  Drinking Water,  AWWA  Annual  Conference,   1979,  San
Francisco, California.

United  States  Environmental Protection  Agency  (EPA),  1982a,  Drinking  Water
Cost Equations, Municipal Environmental Research Laboratory,  Cincinnati, Ohio,
PB83-181826.

Diffused Aeration

Love,  0.   T.   and  R.   G.  Eilers,  Treatment   of  Drinking  Water  Containing
Trichloroethylene  and  Related  Industrial  Solvents,   JAWWA,  Vol.  74,  No. 8,
pp. 413-425, 1982.

Miltner,  R. J.  and  C.  A. Fronk,  "Treatment  of  Synthetic Organic Contaminants
for Phase II Regulations", Internal Report DWRD, EPA,  December 1985.
                                     -26-

-------
Ruggiero, D.  D.,   "Removal  of Organic  Contaminants from  the Drinking  Water
Supply at Glen Cove,  New York, USEPA Document,  EPA-600/2-84-029,  January 1984.

Ruggiero, D.  D.  et al., Removal of  Organic  Contaminants from Drinking  Water
Supply at Glen Cove ~New York.   Presented at the 100th  Annual American  Water
Works Association National Conference,  June 18,  1980,  Atlanta, Georgia.

Multiple Tray Aeration

Hess, A. F.,  J. E. Dyksen and G.  C. Cline, Case Studies Involving  Removal of
Organic  Chemical  Compounds  from  Ground Water,  Presented at  the  Preconference
Seminar  Concerning Organic  Chemical  Contaminants  in Ground Water at  the 1981
Annual Conference  of the American  Water  Works Association,  June 7-11,  1981,
St. Louis, Missouri.

Hess,  A. F.  J.   E.  Dyksen,  and  H.  J.  Dunn,  Control  Strategy —  Aeration
Treatment Technique,  in AWWA Research  Foundation,  Occurrence and  Removal of
Volatile Organic Chemicals from Drinking Water,  Denver,  Colorado, pp.  125-127,
1983.

Joyce, M.,  Smyrna,  Delaware Solves  a  Water  Problem,  Water and  Sewage Works,
March 1980.

Ozone/UV

Buescher, C.  A.,  J.  H. Dougherty,  and  R.  T. Skrinde,  "Chemical Oxidation of
Selected Organic Pesticides". JWPCF, Vol.  36, No.  8, 1964.

Gilbert,  E,   "Chemical  Changes  and  Reaction Products in  the Ozonization  of
Organic  Water  Constituents"  In;    Oxidation  Techniques  in  Drinking  Water
Treatment,  Office of  Drinking  Water,  Washington,  D.C.,  Report  No.   EPA-
570/9-79-020, 1979.

Hoigne,  J.  and  H. Bader, "Rate Constants  of Reactions  of Ozone with Organic
and Inorganic Compounds in Water  -  I".  Water  Research, Vol.  17,  173-183,  1983.

Legube,  B.,   S. Guyon,  H.  Sugimitsu,  and M. Dore,  "Ozonation  of  Some  Aromatic
Compounds in  Aqueous Solution:   Styrene,  Benzaldehyde,  Naphthalene,  Diethyl-
phthalate,  Ethyl  and  Chloro  Benzenes",  Ozone:    Science  and  Engineering,
Vol. 5, pp.  151-170,  1983.

Miltner, R.  J. and C.  A. Fronk,  "Treatment of Synthetic Organic Contaminants
for Phase II Regulations", Internal Report DWRD, EPA,  December 1985.

Yocum, Floyd  H.,  "Oxidation of Styrene with Ozone  in Aqueous Solution",  In:
Ozone/Chlorine  Dioxide  Products  of Organic  Material,  International  Ozone
Institute, Cleveland, Ohio,  1978.
                                     -27-

-------
Reverse Osmosis Processes

Berkau, E.  E.,  C. E. Frank, and  I A. Jefcoat, "A  Scientific  Approach to the
Identification  and Control  of Toxic Chemicals  in  Industrial Wastewaters",
AlChe Symposium Series No. 197, Volume 76, 1980.

Cabasso, I., E. Klein,  C. Eyer, and  J. Smith,  "Trace Organic Contaminants in
Drinking Water; Evaluation  of  Semi-permeable Membranes and Osmotic Pumping to
Achieve  Concentration",  Presentation before  the  Division of Environmental
Chemistry (American Chemical Society), 204-210, 1974.

Edwards, V.  and P. F. Schubert, "Removal of  2,4-D and Other Persistent Organic
Molecules from Water Supplies by Reverse Osmosis", JAWWA, October 1974.

Hinden, E.  et  al., "Organic Compounds  Removed by Reverse  Osmosis"  Water and
Sewage Works, 1968.

Malaiyandi,  M., P. Blais, V. S. Sastri, "Note:  Separation of Lindane from Its
Aqueous  Solutions by Reverse  Osmosis System".  Separation Science  and  Tech-
nology 15:7, 1483-1488, 1980.

Miltner, R.  J.  and C. A. Fronk,  "Treatment  of Synthetic Organic Contaminants
for Phase II Regulations", Internal Report DWRD, EPA, December 1985.

Perry, D. L.,  J.  K.  Smith,  and S. C. Lynch,  1981,  Development of Basic Data
and Knowledge Regarding Organic Removal Capabilities of Commercially Available
Home Water  Treatment Units Utilizing Activated Carbon,  Phase 3/Final Report,
Performed for Criteria  and  Standards  Division,  Office of Drinking Water, U.S.
Environmental Protection Agency.

Sorg, T. J.  and O. T. Love, Reverse Osmosis Treatment to Control Inorganic and
Volatile Organic  Contamination, Proceedings:   Experiences with  Ground  Water
Contamination, Annual AWWA Conference and Exposition, Dallas,  Texas,  1984.

Mechanical Aeration

Roberts, P.V. and Levy, J.A., "Air Stripping of Trihalomethanes", Presented at
AWWA Seminar, Controlling Trihalomethanes, Las Vegas, Nevada,  May 5,  1983.

Catenary Grid

CHEM-PRO CORPORATION, 27 Daniel Road, P.O. Box 1248, Fairfield, NJ 07007.

Malcolm Pirnie,  Inc., Control  of  Organic Chemicals  in Ground  Water Supply.
Report submitted  to Jamaica Water Supply  Company,  Lake Success, NY,  October
1983.

HI GEE

GLITSCH,  INC.  P.O. Box 660053, Dallas, TX 75266-0053.
                                     -28-

-------
GAC Adsorption

Dobbs, R.A. and J.M. Cohen,  "Carbon  Adsorption Isotherms for Toxic Organics,"
EPA Report 600/880-023, Office of  Research  and Development,  MERL, Cincinnati,
Ohio, April 1980.

Dyksen, J.E., Raman, K., Raczko, R.F. and Clark,  R.M.   "GAC  Treatment Costs -
Minimizing  Them",  presented at   the  AWWA  Annual  Conference,   Kansas  City,
June 14-18, 1987.

Fiessinger,  F.  and  Y.  Richard,   "International  Experience  with  Granular
Activated  Carbon."   Paper presented  at the  AWWA meeting,  Atlanta,  Georgia,
June 15-20, 1980.

Hess, A.F., "GAC  Treatment Designs and  Costs for Controlling Volatile Organic
Compounds in Ground Water".  Paper presented at the National American Chemical
Society Meeting, Atlanta, Georgia,  March 30 - April 3,  1981.

Hiltebrand,  D.J.; Dyksen,  J.E.  and  Raman,  K.   "Treatment  Alternatives  and
Associated Costs  for the Removal  of Radon  from  Ground  Water" In:   Radon in
Ground Water.  Ed:  Barbara Graves, Lewis Publishers, 521, 1987.

Love, O.T.  and R.G. Filers,  "Treatment for the  Control  of  Trichloroethylene
and  Related  Industrial Solvents in  Drinking Water," Drinking  Water  Research
Division, U.S. EPA, Cincinnati,  Ohio, February 1981, JAWWA, 74:413, 1982.

Lowry, J.D. and  Brandown,  J.E.   "Removal of  Radon  from  Ground Water  Supplies
using Granular  Activated Carbon or  Diffused Aeration",  University of  Maine,
Department of Civil Engineering,  Orono,  Maine 06469.

Malcolm Pirnie, Inc.,  "Preliminary Treatment  Designs and Costs for Control of
Organic Compounds."  Report  prepared for the Office of Drinking Water,  U.S.
EPA, Washington, D.C.,  April 1981.

McCarty,  P.L.,  D.  Argo  and  M.   Reinhard,  "Operational  Experiences  with
Activated Carbon  Adsorbers  at Water  Factory  21," JAWWA,  71:683-689,  November
1979.

Meijers, A.P.,  "Objectives  and  Procedures  for GAC  Treatment,"  JAWWA,  71:628,
1979.

O'Brien,  R.P.,  D.M.  Jordan  and  W.R.  Musser,  "Trace Crganics  Removal  from
Contaminated   Ground  Waters   with   Granular   Activated   Carbon,"    Calgon
Corporation,   Pittsburgh,   Pennsylvania.   Presented at  the  National  Ameri-
can Chemical Society Meeting, Atlanta, Georgia, March 29  - April 3, 1981.

Peel, R.G.  and A.  Benedek,   "Attainment  of  Equilibrium in  Activated  Carbon
Isotherm Studies," Environmental Science and Technology,  14:66-79', 1980.

Randtke,   S.J.  and  C.P.  Jepsen,   "Effects  of  Salts  on  Activated  Carbon
Adsorption of Fulvic Acids," JAWWA, 74:84-93, February 1982.
                                     -29-

-------
Roberts,  P.V. and R.S. Summers,  "Performance  of Granular Activated Carbon for
Total Organic Carbon Removal," JAWWA, 74:113-118, February 1982.

Ruggiero, D.D. and R. Ausubel,  "Removal  of Organic Contaminants from Drinking
Water Supply at Glen Cove,  New  York,  Phase II,"  Report No. EPA-600/2-82/027,
U.S. EPA, March 1982, and Phase I, Report No. EPA-600/2-80-198, 1980.

Schulhof, P., "An Evolutionary Approach to Activated Carbon Treatment," JAWWA,
71:648, 1979.

Schalekamp, M., "The Use of GAC Filtration to Ensure Quality in Drinking Water
from Surface Sources," JAWWA, 71:638-647, November 1979.

Sontheimer, H. ,  "Applying  Oxidation and Adsorption Techniques:  A Summary of
Progress,"  JAWWA, 71:612-617, November 1979.

Sontheimer, H., "Design Criteria  and  Process  Schemes  for GAC Filters," JAWWA,
71:618, 1979.

Symons,  J.M.,  "Removal  of  Organic  Contaminants  from  Drinking   Water  Using
Techniques Other  than  Granular Activated  Carbon Alone - A  Progress Report,"
Drinking Water Research Division, U.S. EPA, Cincinnati, Ohio, May 1979.

U.S. EPA,  "Process Design  Manual  for Carbon Adsorption," Technology Transfer,
U.S. EPA, October 1973.

Weber, W.J.,  Jr.  and M.  Pirbazari,  "Adsorption of  Toxic and  Carcinogenic
Compounds from Water," JAWWA, 74:203, 1982.

Weber,  W.J.,  Jr.  and  B.M.  VanVliet,  "Synthetic Adsorbents  and  Activated
Carbons  for  Water  Treatment:  Overview and  Experimental  Comparisons," JAWWA,
73:420, 1981.

Packed Column Aeration

Camp,  T.R.,  "Gas Transfer  To and  From  Aqueous Solutions," Journal Sanitary
Engineering Division, Proc. ASCE, 84:SA4:1701, 1958.

Cummins,  M.D.  and  J.J.  Westrick,  "Packed Column Air  Stripping  for Removal of
Volatile  Compounds."   Presented  at   the   1982  Conference  on  Environmental
Engineering,  ASCE, July 14-15, 1982.

Dyksen,  J.E.  and  A.F.  Hess,  "Aeration  Techniques for  Removing Trace Organic
Compounds  from  Drinking Water."   Paper  presented  at  the 1981 ASCE National
Conference on Environmental Engineering, Atlanta, Georgia, July, 1981.

Dyksen, J.E., A.F. Hess, M.J.  Barnes and G.C. Cline,  "The  Use  of  Aeration to
Remove Volatile  Organics From  Ground Water."   Presented  at the  1982 Annual
Conference of the American  Water  Works  Association, Miami Beach,  Florida,  May
1982.
                                     -30-

-------
Haney, Paul D., "Theoretical Principles of Aeration," JAWWA, 46:4:353, 1954.

Hess, A.F.,  J.E.  Dyksen  and  G.C. Cline,  "Case  Studies Involving  Removal of
Organic Chemical Compounds  from Ground Water."  Presented  at  the  1981 Annual
Conference of  the American  Water  Works  Association,  St.  Louis, Missouri, June
1981.

Hiltebrand,  D.J.;  Dyksen,  J.E.  and  Raman,  K.   "Treatment Alternatives  and
Associated  Costs  for  the Removal of Radon from  Ground Water"  in  Radon in
Ground Water.  Ed:  Barbara Graves, Lewis Publishers, 521,  1987.

Kavanaugh, M.C. and R.R. Trussel, "Design of Aeration Towers to Strip Volatile
Contaminants from Drinking Water," JAWWA, 72:12:684,  1980.

Kavanaugh,  M.C.  and  R.R.  Trussel,  "Air Stripping  as  a Treatment Process."
Paper presented at the AWWA conference, July 1981.

Langelier, W.F., "Theory and Practice of Aeration," JAWWA,  24:1:62, 1932.

Love, O.T.  and R.G.  Eilers,  "Treatment for the Control of Trichloroethylene
and  Related  Industrial Solvents in Drinking Water," U.S. EPA, Drinking Water
Research Division, Cincinnati, Ohio,  October 1980.

Mackay,  D.  et al.,  "Determination  of  Air-Water  Henry's  Law Constants  for
Hydrophobic  Pollutants,"  Environmental Science and Technology, 13(3), 33-337,
1979.

McCarty, P.L., K.H. Sutherland, J. Graydon and M.  Reinhard, "Volatile Organic
Contaminants Removal  by Stripping,"   Proc. AWWA  Seminar, Controlling Organics
in Drinking  Water, San Francisco, June 1979.

McKinnon,  R.J.  and  J.E.  Dyksen,  "Aeration  Plus  Carbon  Adsorption  Remove
Organics from  Rockaway Township  (NJ)  Ground Water Supply."  Presented at the
1982  Annual  Convention  of the  American  Society  of  Civil  Engineers,  New
Orleans, Louisiana, October 25-27, 1982.

Metcalf  and  Eddy,  Inc.,  "Volatile Organics Removal:  Two Ground  Water Supply
Case Histories."  Presented at the New York Section AWWA, 1980.

Munz, C. and P.V. Roberts,  "Transfer of Volatile Organic Pollutants into a Gas
Phase During Bubble Aeration,"   Technical Report No. 262,  Department of Civil
Engineering, Stanford University, 1982.

Nebolsine,  Kohlman  and Ruggiero  Engineers,  "Removal of Organic  Contaminants
from  Drinking  Water Supply at  Glen  Cove, New York,"  Interim Report on U.S.
EPA  Agreement  No.  CR806355-01,  Office  of  Research and   Development,  MERL,
Drinking Water Research Division, Cincinnati, Ohio, July 1980.

Raczko,  R.F.,   J.E.   Dyksen and  M.B.  Denove,   "Pilot  Scale  Studies  of  Air
Stripping  for  Removal of  Volatile Organics from Ground  Water."  Presented at
the  14th Mid-Atlantic  Industrial Waste  Conference, University  of Maryland,
College Park, Maryland, 1982.
                                     -31-

-------
Roberts,  P.V.   and   P.   Dandliker,   "Mass   Transfer  of  Volatile   Organic
Contaminants  During  Surface   Aeration."    Presented  at  the   1982   Annual
Conference  of  the American  Water  Works Association,  Miami  Beach,  Florida,
May 1982.

Singley,  J.E.,  A.L.  Ervin,  M.A.  Mangone, J.M.  Allan  and  H.H.   Land,  "Trace
Organics Removal by Air Stripping,"  AWWA Research Foundation,  1980.

Symons,  J.M.,   A.A.   Stevens,   R.M.  Clark,  E.E.  Geldreich,   O.T.  Love  and
J. DeMarco,  "Treatment Techniques  for  Controlling  Trihalomethanes in  Drinking
Water,"  U.S. EPA, Drinking Water Research Division,  1981.

Treybal, R.E., "Mass Transfer Operations," McGraw-Hill Book Co.,  New York (3rd
edition), 1980.

Warner,  H.P.,  J.M.  Cohen and  J.C. Ireland,  "Determination  of Henry's  Law
Constants  of  Selected  Priority  Pollutants,"  Wastewater  Research Division,
Municipal Environmental Research Laboratory,  Cincinnati, Ohio,  1980.
                                     -32-

-------
                       WORKSHOP ON EMERGING TECHNOLOGIES

                         FOR DRINKING WATER TREATMENT


                     TECHNICAL SESSION ON ORGANICS REMOVAL


                                  CASE STUDY
EXAMPLE 1:

     A municipal ground water supply has been contaminated with several
organic chemicals.  Based on the raw water quality given below, determine;

          An appropriate treatment method
          Size and process design of the treatment facility
                       ORGANIC CONTAMINANT LEVELS
                           IN LITTLETOWN WELLS
Trichloroethylene

Tetrachloroethylene

1,1,1-Trichloroethane
Notes:
     1.   Total Combined Flowrate = 225 gpm

Consider:
          Treatment objective
          Treatment options
          Treatability testing required
       -  Cost
1 units in micrograms/liter)
Well No.
10
200
1,000
750
11
200
1,000
600
14
100
200
300
                                      -1-

-------
                             TREATMENT OBJECTIVES
         VOC
Trichloroethylene

Tetrachloroethylene

1,1,1-Trichloroethane
Composite of
All Wells
160
650
500
Design
Effluent
Concentration
5
10
10
Required
Percent
Removal
97
98.5
98
Note:
     1.    All concentrations given in ug/L.

Define:

  -  What are the treatment options?

-------
                          CONTAMINANT CHARACTERISTICS
          VOC

Trichloroethylene

Tetrachloroethylene

1,1,1-Trichloroethane

       * at 20 C
Henry's Constant
       (H>*
	(atm)	

       155

       274

       288
 Adsorption
  Isotherm         Carbon
  Constants      Usage Rate
  K     1/n   (lbs/1,000 gal)
28      0.62

50.8    0.56

 2.48   0.34
0.19

0.18

 2.8
Define:
          Which treatment process will work (aeration or GAG)?

Consider:
          Volatility of the contaminants
          Carbon usage rates
                                      -3-

-------
                          PACKED COLUMN  - PILOT STUDY RESULTS
                 :3       23     40   50   60  70  80 90 100
                LIQUID LOADING RATE (gpci/sf)
                                                       30
Define:
   15     ID       33     <0   50   60  70  80 90 100
        LIQUID LOADING RATE (gpm/sf)

-  Optimum loading  rate

   Type  of packing  material



-  A:W Ratio

   Packing height
                                                                              :a     40   50   BO  ;c so  90 100
                                                                     LIQUID LOADING RATE (gpm/sf)
                                            -4-

-------
Air:Water  Ratio Versus Packing Height
      40
      35
      30
    0)
    O
    
-------
                     PACKED COLUMN PROCESS DESIGN CRITERIA
     A sample calculation for developing the process design criteria is shown
below:

     Pilot Study Results

     VOC = Tetrachloroethylene

     Packing height = 10 ft

     Column diameter = 1 ft

     Packing material = 2-inch Tri-packs
               Water      Air
     Run     Flow Rate  Flow Rate
     'Io.       (gpm)      (gpm)     A/W
                                                  PCE (ug/L)
                                             Influent    Effluent    % Removal
                 16
                           43
    20:1
100
93
     1.
          Calculation K a (Refer to workshop notes for formulas)
                       L
          a.   R = (A) (H)  x 7.5212 x 10
                      (W)

                 = 20 x 274 x 7.5212 x 10

                 = 4.12

          b.   NTU =  R
                                         -4
                                         -4
                          In
                     R-l
                              (x./x )  (R-l)  + 1
                                i  o	

                                     R
                        4.12
                      4.12 - 1
                                In (100/7)  (4.12 - 1)  + 1
                                          4.12
                   = 3.2

          c.    HTU =   2
                      NTU
 10
3.17
                                     =  3.2 ft
          d.   K a =   L
                L
                      HTU
                               16 gpm    x    1     x  60  min  x   1    ft
                                                          ~
             3.2 ft
                                                           hr    7.48  ga
                          -1
                   = 51 hr
     K  a  for  other  loading  rates  are  calculated similarly.
/ersus  loading  rate is  made.
                                                            A plot of K a
                                      -6-

-------
     2.   Calculation of Packing Height


                                          2
          Optimum loading rate = 25 gpm/ft


                          2                            -1
          K a at 25 gpm/ft  for 2-inch Tri-paks = 58 hr    (from graph)
           Li


          Assume A/W ratio = 40



          % removal required (PCE) = 98.5%



          a.   R =  (A)  (H)  x 7.5212 x .10~4

                      (W)



                 = 40 x 274 x 7.5212 x 10~4



                 = 8.24



          b.   x./x  = 	1	  =     1      = 66.7

                10   1 - % Removal     1 - 98.5

                              100             100



          c.   NTU =   8.24    hr  (66.7) (8.24 - 1) + 1

                     8.24 - 1              8.24



                   = 4.64



          d.   HTU = L  (gpm/ft )   x  60 min  x  	1_  ft



                       K a (hr"1)          hr     7.48  gal
                        L


                   = £5_ gpm/ft  x 60 min x   1   ft

                     58     -1        hr   7.48  gal
                          hr                     ^


                   = 3.5 ft



          e.   Required packing height = (NTU) (HTU)



                                       = 4.64 x 3.5 ft



                                       = 16.2 ft



     Packing heights for other A/W ratios are calculated similarly.  A plot of

packing height versus A/W ratio is made.  Optimum A/W ratio can be estimated

from such a plot.
                                      -7-

-------
                               VOC Emission Rate
     PCE  concentration in raw water (C.)  = 650 ug/L

     PCE  concentration in treated water (C )  =10 ug/L

     PCE  Emission rate (Ibs/hr)  =C.-C  xVx  5
                                   i    e
                                                io7
                                = (650 - 10)  x 225 gpm x  5
                                                         io7
                                = 0.072 Ib/hr
     Since OSHA regulations require exit air treatment only at emission rates
of above 0.1 Ib/nr, no additional treatment is required for the exit air at
this facility.
                                      -8-

-------
 •-ote :
                      PACKED COLUMN  PROCESS DESIGN CRITERIA
                              3ASED  ON  PILOT STUDY
           Maximum Flcwrate                    =

           Hydraulic Loading Rate              =

           PCE Mass Transfer Coefficient,  KTa =
                                            j_i
           A:W Ratio                           =

           Air Flow
           Packing Height
                          (1)
  225 gpm

   25 gpm/ft"

   56 hr-1

   40:1

1,200 cfia

   20 feet
           Based on 98.5% removal of PCE
                     7CC REMOVAL AT DESIGN  CONDITIONS
    VOC

TCE
PCE*
! ,1,1-Trichloroethane
    *Denotes critical compound.


Design Percent
Removal
97
98.5
98


Actual Percent
Removal
)99
98.5
98
Design
Effluent
Concentrations
(ug/1)
5
10
10
Actual
Effluent
Concentration
(ug/1)
(5
10
10
uerine:
          Treatment costs
                                       -9-

-------
                     PRELIMINARY COST ESTIMATES FOR
                     PACKED COLUMN AERATION PROCESS
Based on cost curves presented in the workshop report:

     Capital Cost (Figure 43)            =  $150,000

     O&M Cost (Figure 44)               =  $  8,000/year

     Total Cost                         =  $0.22/1,000 gallons
     (20 year amortization period
      and 10% interest rate)
                                     -10-

-------
                                MINI-COLUMN  TEST RESULTS
   200
    133
    90
    :0
    70
_  50
    50
UJ
j±  40

~   3
rt
O
CQ
CC
    ;3
                               I
               TETRACHLOROETHYLENE —7
                I          I
                                                A-TRICHLOROETHYLENE
                                                                     ,1.1-THJCHLOROETHANE
                                                                      I
                         TETRACHLOROETHYLENE
                             TRICHLOROETHYLENE i
                                                              i. I-TRICHLOROETHANE::
                              6   8  10        20        40    60  30

                                 INFLUENT VOC CONCENTRATION (ug/1)
                                                                              200
400   GOO
    Define:
              Carbon Type
                                           -II-

-------
                           MINI-COLUMN TEST RESULTS
               E   30
               xt
               CXL
               o
               z
               o
                            TETRACHLOROETHYLENE
                                                    TRICHLOROETHYLENE
                                                         s
                                                        1.1.1 -TRICHLORO
                                                        ETHANE
                   20
10
                               1000         2000         3000

                                     WATER TREATED  (ml)
                                                 4000
Define:
       -  Controlling VCC

          Carbon Usage based  on:

               Carbon usage  (lbs/1,000 gal)  = W x 8461.5
                                               V

               whero 17 = weight of  carbon (0.1 gm)
                     V = volume treated (ml)

       -  Preliminary Process Design
                                      -12-

-------
                          GAG PROCESS DESIGN CRITERIA


1.   Maximum Flowrate:  225 gpm

2.   Carbon Usage Rate:

          weight of carbon in the mini-column  (W) = 0.1 gm

       -  volume treated based on tetrachloroethylene breakthrough at 10 ug/L
           (V) = 1,500 ul

     Carbon Usage (lbs/1,000 1) = W x 8461.5
                                  V

          =  0.1 gm  x 8461.5 = 0.56
            1,500 ml

3.   Carbon Charge:

     Assume EBCT = 10 minutes

                                               3         3
     Carbon volume = 225 gal x 10 min x _1	 ft  = 300 ft
                         min            7.48 gal

     Carbon weight = 300 ft  x 29 Ibs = 8,700 Ibs

                                  ft3

4.   Contactor Dimensions:

     Assume hydraulic loading rate = 5 gpm/ft
     Contactor diameter =  225 gpm  x 4_ = 7.5 ft

                        •J 5 gpm/ft

     Carbon bed depth =     300 ft       = 7 ft
                                2   2
                         TT (7.5)  ft
                         4

     Contactor height = 15 feet
5.   Replacement Frequency:

          Replacement frequency = 	carbon charge  (Ibs) x 0.7	
                                  carbon usage (   Ibs   )  x flow  (gpm)
                                                1,000 gal

                                = 	8,700 Ibs x 0.7     = 48 days
                                  0.56    Ibs    x 225 gpm
                                       1,000 gal
                                     -13-

-------
                         GAG PROCESS  DESIGN CRITERIA
Define:
         Maximum  Flowrate


         Hydraulic  Loading  Rate


         Carbon Usage


         EBCT


         Number of  Contactors


         Contactor  Dimensions


         Depth of GAC  Bed



         Carbon Charge


         Type of  GAC


         Carbon Life
=    225 gpm

               2
=      5 gpm/ft


       0.56 lbs/1,000 gallons


=     10 minutes


       1


       7.5 ft (diameter)  x 15 ft (height)


       7 ft


=  8,700 Ibs


     Carborundum GAC 830


      48 days
          Treatment costs
                                     -14-

-------
                     PRELIMINARY COST ESTIMATES FOR
                         GAG ADSORPTION PROCESS
Based on cost curves presented in the workshop report:

          Capital Cost (Figure 31)            = $160,000

          O&M Cost (Figure 32)               = $ 25,000/year

          Total Cost                    '     = $.375/1,000 gallons
          (20 year amortization period
           and 10% interest rate)
                                     -15-

-------
                               PROCESS SELECTION
DEFINE:




          Recommended treatment process




CONSIDER:




          Treatment efficiency




          Operational factors




          Treatment costs
                                     -16-

-------
                          COMPARISON OF ALTERNATIVES
                             LITTLETOWN WELLFIELD
  Packed Column

  High removals

  Continuous operation
- VOC emissions in exit air
  less than OSHA limit

- No waste disposal
  Costs:
     Capital Cost = $150,000
     O&M          = $ 8,000/yr
     Total        = $0.22/1,000 gallons
GAG Adsorption

   High removals

   Requires down time for Carbon
   replacement

   No air emissions-
-  Disposal of backwash water and
   spent carbon

Costs:
   Capital Cost = $160,000
   O&M          = $25,000/yr
   Total        = $0.375/1,000 gallons
                                     -17-

-------
Technical Session: Inorganics
J. Edward Singley, Vice President, James W. Montgomery Consulting
Engineers, Inc., Gainesville, FL
                              VIII-1

-------
                     Emerging Technologies for Control
                         of Inorganic Contaminants

                             J. Edward Singley
I.    Introduction

      A.   Regulations

           1.  Present
           2.  Proposed


      B. Metals

           1.  Lead
           2.  Barium
           3.  Cadmium
           4.  Chromium
           5.  Mercury
           6.  Radium
           7.  Selenium
           8.  Silver
           9.  Strontium

      C.   Non-Metals

           1.  Arsenic
           2.  Fluoride
           3.  Nitrate

II.   Lead

      A.   Lead leaching

      B.   Workmanship

      C.   Control techniques

      D.   Data needs

III.  Barium

IV.   Cadmium

V.    Chromium

VI.   Mercury
VII.    Radium

        A.  Chemistry

        B.    Control techniques

             1.  Precipitative
                 softening
             2.  Ion exchange
             3.  Demineralization

        C.    Radioactivity

VIII. •  Selenium

IX.      Silver

X.      Strontium

XI.      Arsenic

XII.    Fluoride

        A.    Activated alumina
             adsorption

        B.    Ion exchange

XIII.   Nitrate

XIV.    Summary

-------
                     Emerging Technologies for Control
                         of Inorganic Contaminants
GENERAL
      Of the 12 inorganic contaminants  that  are  presently  regulated  by
primary  agencies,  the  one  most  commonly present at levels in excess of
the primary MCL is lead, Pb.  The others are less common  but  are  present
in  site  or  regionally  specific  cases.   For this reason major attention
will be focused on Pb with limited attention to the others.
REGULATIONS
      Present.
      The present regulations  on  the  inorganic  contaminants  under  the
National Interim Drinking Water Regulations cover 12 different substances:
                                         MCL (mg/L)*
           Arsenic                       0.05
           Barium                        1.0
           Cadmium                       0.010
           Chromium (total)              0.05
           Fluoride                      4.0**
           Lead                          0.05
           Mercury                       0.002
           Nitrate, as N                  10
           Selenium                      0.01
           Silver                        0.05
           Radium                         5 (pCi/L)
           Strontium-90                   8 (pCi/L)

             *as of August 1, 1987
            **Secondary MCL of 2.0 mg/L
      The  significance  of  the  twelve  varies  considerably  since their
presence and concentrations vary so much.   For example  silver  is  rarely,
if  ever,  found  unless  a  point-of-use water treatment device is used by
the consumer, whereas lead is found to  be  almost  ubiquitous  in  samples
from  the  consumer's  tap.  Radium, strontium, cadmium, selenium, arsenic,
mercury, barium, nitrate, chromium and excess fluoride are  found  only  in

-------
site-specific cases although their concentrations  may  vary  widely.   For
some  of  the above the problem of elevated concentrations is indigenous to
a section, or sections, of the country.   Thus  a  treatment  technique  may
be  of  limited  applicability.   Despite this the need is real  and must be
addressed because the MCLs must be met.

      Proposed.

      There is no doubt that  additional   inorganic  contaminants  will   be
regulated  nor  that  there  will   be  changes in the MCLs of some of those
presently regulated.  Some will increase  and some decrease.    The  proposed
regulations are shown in Table 1.

      Since  silver  is  not  found in the raw water, added in treatment or
by corrosion of the system, it will likely be  dropped  entirely.   On  the
other  hand,  the  present lead MCL of 0.050 mg/L will certainly be reduced
to 0.020 mg/L or less, unless a treatment technique  is  specified.   Other
changes in the regulations, as proposed  can be observed in the Table.

      The  presence  of  lead,  zinc,  copper and cadmiun in water entering
the distribution system is rare.   They occur because of  reactions  between
the  water  and  the  conveying system in transit from the plant to the  tap
of the consumer.  The particular  contaminant(s)  picked  up  depends upon
the  composition  of  the pipes,  valves,  meters, etc., with which the water
comes in contact.  The corrosion reactions that  occur  between   the  water
and  metallic  substances  are  electrochemical   in  nature  and  result in
converting the free metals, which  are very insoluble  in  water,  to  their
ionic  forms  which  are  usually   very   soluble.   The ionic solubility is
strongly influenced by the water quality.  It may  be  stated  safely that
"all   metals   corrode  in  direct  contact  with  water."   The  rate   is
determined  primarily  by  the  properties  of   the   metal   itself   and
secondarily  by  the properties of the water, i.e., the water quality.  The
most common metals used in transmission   mains  and  plumbing  systems  are
steel  (iron),  galvanized  steel,  copper,  lead,  brass  and bronze.  The
presence of zinc and cadmium,  and  to  a  lesser  extent  lead,  in  water
samples  comes  from  the zinc used in galvanizing pipe.  The lead may come
from  lead  services,  lead  goosenecks,   lead  pipes   (rare)   and  from
lead-based  solders  used  to join copper pipes.  In most parts  of the U.S.
the latter is the major source of  lead at the tap.

LEAD

      It is of historical interest to note that lead  pipes  were  used   in
Rome  to  distribute  the  water  from  the  Aquaducts  to the homes of  the
wealthy, which is cited by some authorities as the  cause  of  the  decline
of the Roman Empire.

      Lead Leaching.

      Lead  based  solders  used  to connect copper plumbing for many years
have been based on tin-lead combinations, the most common being   50:50  and
60:40  tin:lead.   The  recent  ban  on   lead in such solders will strongly
affect lead content in water obtained from the taps  of  homes  constructed

-------
after  the  states  implement the ban but will have minimal effect on water
samples from homes and businesses constructed prior to the ban.

      There is an extensive discussion of the  rate  of  leaching  of  lead
from  lead-based  solders  and  fluxes  in  "Internal  Corrosion  of  Water
Distribution  Systems"  published  by  the American Water Works Association
Research  Foundation  jointly  with  the  Engler-Bunte  Institute  of   the
University  of  Karlsruhe,  West Germany.  Examples from the U.S. have been
cited by Norman Murrell in testimony before  Congress.   The  concentration
of  lead  varies  with  the amount of water that has been withdrawn both in
the sample itself and before the sample, as well as  the  time  of  contact
of  the  water  with  the household plumbing.  Thus first draw samples give
the worst  case  examples  and  the  smaller  the  sample  the  higher  the
concentration  measured.   One  study  in  Scotland  by Lyon and Lenihan in
Glasgow in 1977, cited  by  Schock  and  Wagner  in  the  reference  above,
showed  that  as much as 20 ug of lead per fitting per hour could be picked
up for new  connections  in  copper  pipes  even  though  they  were  "well
made."   This  declined  to  about  1.25  ug/fitting/hr  in  4-5  weeks and
remained at that level for many  years.   Studies  at  the  Illinois  State
Water  Survey  by Neff and Schock also showed that lead can be leached from
chrome plated brass fixtures.

      Workmanship.

      There is no doubt that a major  contribution  to  lead  corrosion  is
poor  workmanship  in the formation of lead-based soldered joints in copper
pipes.  This leads to galvanic corrosion of the lead in  contact  with  the
more  cathodic  copper  and  tin.  It is likely that the lead concentration
in water in contact with poorly made joints is almost  independent  of  the
water  quality,  thus  there  is  little  that the utility can do to reduce
lead levels significantly when such  a  mechanism  predominates  except  to
provide a protective coating, such as CaC03-

      Control methods.

      Early  studies  in  1887  in Germany by Wolfhugel, identified oxygen,
carbon dioxide, water quality, age and quality of  lead  pipes,  stagnation
time,  flow  velocity  and  temperature  as  contributing  factors.   These
include  essentially  all  of  the factors recognized today as important in
lead pick-up.  These observations resulted from  a  serious  leach  problem
in  Dessau  and  the  effective  solution  was  the  addition  of limestone
(calcium carbonate).  Later soda ash (sodium carbonate) was  added.   These
treatments  increased  the  pH  thus  reducing  the free carbon dioxide and
increasing the carbonate alkalinity.  Today, one hundred  years  later,  we
use  the  same  approach when treating the same water qualities that caused
the problems, i.e., low hardness, low alkalinity, and low pH.

      Some of the classic cases in the U.S. were discussed and  studied  by
Karalekas,  et  al. starting in 1976 in Massachusetts.  They showed that pH
adjustment to a value above 8.0  using  caustic  soda  (NaOH)  reduced  the
pick-up  of  lead  significantly.  The use of zinc orthophosphate at levels
as high as 12  mg/1  showed  little  if  any  improvement.   These  effects
correlated   well  with  the  effects  in  Glasgow  ninety  years  earlier.
Figure 1 shows the solubility of lead in water.

-------
      The  use  of  various   corrosion   inhibitors   has   been   studied
extensively with the following results:

      Carbonates  and  pH.   These two are almost inseparable because of the
chemical interdependence of the carbonate  system  and  pH.    Whereas_  many
authors  have  concluded  that a major water quality factor  in a particular
corrosion situation  was  free  carbon  dioxide,  none  have  shown  carbon
dioxide,  per  se,  as a participant in the corrosion reactions.  Rather it
is the dependence upon the historical precedence of citing  carbon  dioxide
rather  than  pH  because the term pH was not used or defined prior to  1909
when Sorenson simplified the discussion of the hydrogen  ion  concentration
by  defining  a  new  term "pH".  Until shortly before that  time it was not
possible to measure the hydrogen  ion  concentration  in  aqueous  solution
accurately,  and  even  then  it  was  a laborious procedure.  On the other
hand it was possible to analyze accurately for the  "free carbon  dioxide
in the middle 1880's.

      The   shift   in   the   carbonate  system  from  carbon  dioxide  to
bicarbonate and then carbonate as the  pH  is  increased  conforms  to   the
well  known  definition  of  the various forms of alkalinity.  The relevant
reactions and equilibrium constants are:

      C02 + H20 - H2C03,pK =1.5
      H2C03 - H+ + HCOa'.pK, =6.3
      HC03- = H+ + C03=,pK2 = 10.3

The conversion of free carbon dioxide to bicarbonate is complete  at a  pH
of about 8.3 and to carbonate above a pH of 10.3.

      A  comprehensive  study  of  the  effect  of  these variables on  lead
corrosion was conducted by Schock,  et  al.  at  the  EPA  Laboratories  in
Cincinnati.   From  these  studies  they  developed chemical models of  lead
solubility as functions  of  carbonate,  pH,  orthophosphate,  sulfate   and
chloride.   They  concluded  that  orthophosphate  could  be  an  effective
corrosion  control  inhibitor  under  a  limited  range  of   carbonate   and
hydrogen ion concentrations.

      Silicates.   Although  some  authors  and  studies have suggested the
efficacy of silicates, no studies have supported the hypothesis.

      Orthophosphates.  The suggestions of  the  use  of  these  to  reduce
lead  corrosion  has  not  been  borne out in many studies performed in the
U.S. and Europe.  These studies showed that  the  solubility  of  lead   was
greater  than  predicted  by  the  models  that  had  been  developed.   See
Figure 2.

      Sulfate, Chloride and Nitrate.  These have been shown   repeatedly  to
have little effect on lead solubility.

      Organics.    Some  studies  have  shown  that  natural  organics  may
increase lead solubility.

      Chlorine.  The solubility of lead from lead-based  solders  has  been
shown  to  be  increased in the presence of chloramines as compared to free
chlorine.

-------
      Polyphosphates.   Recent studies have  shown  no  advantage  over  the
simple  orthophosphates.   The addition of zinc sulfate did not improve the
effect.

      Calcium Carbonate Films.  One of the traditional methods  of  control
of  corrosion  is  to  interpose a barrier between the metal and the water.
This can be accomplished by reference to one  of  the  indices  of  calcium
carbonate   saturation,   such   as   the   Langelier   Index   or  others.
Modifications of water quality that will initiate the  precipitation  of  a
protective  coating  of calcium carbonate is generally based on pH control,
if adequate concentrations of alkalinity and calcium ion are present.

      pH Control.  It has been shown that  pH  values  above  9.0  lead  to
reduced   lead  concentrations  because  of  the  reduced  availability  of
electron acceptors.  This has been one of the  most  effective  methods  of
control  of  lead  corrosion.   When pH control is practiced in conjunction
with alkalinity supplementation in low alkalinity  waters,  lead  corrosion
is  reduced.   Waters  of  high alkalinity may require softening, to reduce
lead corrosion.

BARIUM.

      Barium has chemical properties very similar  to  the  other  alkaline
earth  elements  calcium,  magnesium  and  radium,  though its removal  from
drinking water parallels that of the other  members  of  the  family.   Its
sulfate  is  less  soluble than calcium sulfate but coagulation with ferric
sulfate or alum is relatively  ineffective.   BaS04  has  a  solubility  of
only  2.2  mg/L at 25°C, which exceeds the MCL of 1.0 mg/L.  The solubility
product constant is about lxlO~4.  Barium carbonate  has  a  solubility  of
about  18  mg/L  at  25°C  but  is  removed, almost quantitatively, by  lime
softening at pH values over the range of 9.5 to 10.5.

CADMIUM

      This  metal  rarely,  if  ever,  occurs  in  raw  water.   It  is   a
contaminant   in   zinc   used  for  galvanizing,  thus  occurs  when  zinc
galvanizing corrodes.   Its electrode potential is very  close  to  that  of
iron   (-0.402v.  vs  -0.441v.),  thus it would corrodes only slowly if  used
alone to protect  steel.   In  contrast,  zinc's  potential  is  much  more
anodic (-0.76v.) and thus provides the desires cathodic protection.

      The  removal  of cadmium from drinking water is rarely needed but the
prevention of its  presence  can  be  accomplished  by  corrosion  control.
Figure 3 shows the effect of pH on cadmium solubility.

CHROMIUM.
      Chromium  exists  in  water  as  the  +3  or  +6  valence state.   The
chromium +3 ion is relatively insoluble as  the  hydroxide,  see  Figure 4,
but  the  +6  state is very soluble as the chromate or dichromate ion.   The
chromate ion varies with pH       , as shown in Figure 5.    The  dichromate
ion  is  converted  to  chromate at pH values above 7.  Its high solubility
prevents its removal  by  any  technique  but  ion  exchange  (anionic)  or

-------
demineralization  (reverse  osmosis  or  electrodialysis).
readily removed by coagulation or lime softening.

MERCURY.
                                                            Chromium III is
      Although mercury occurs only rarely  in  untreated  waters  and  then
      more  rarely  above  the present MCL of 0;002 mg/L, its high toxicity
                    It can occur as two mercury ions, +1  or  +2.    The  +2
form  is  the more common but still contributes little to the total  mercury
in natural water.  The predominant forms are free  metallic  mercury  above
                   organically  complexed  mercury.   Surprisingly the free
even
justifies concern.
                                              only  effective  methods  for
above  pH  5  and
metal has a solubility of 0.025  mg/L.    The
removal is adsorption on activated carbon.

RADIUM.

      Although  the  radium  concentration   in  most  of  the  major surface
supplies in the U.S. is low, as shown in Table 2,  there are  several   areas
of  the  country  where  radium  is  a  major problem,  principally Illinois,
Iowa and Florida.  Some examples of raw water quality  in  these  areas  are
shown  in  Table  3.  Treatment of a water  containing  Ra-226 plus Ra-228 at
concentrations above the MCL of 5 pCi/L can be estimated  from  calculating
the  amount  that  must  be  removed.  This can be expressed as f=l-(5/RWR)
where f is the radium removal fraction  required, and RWR is the  raw  water
radium concentration.  This is shown graphically in Figure 6.

      Chemistry.

      It  is  to  be  noted  from Table 3 that the waters containing radium
also have high total hardness (TH).   The  chemical  properties  of  radium
are  similar  to  those of calcium and  magnesium;  thus it is not unexpected
that  those  processes  which  remove  hardness  (primarily   calcium   and
magnesium salts) will also remove radium.

      "The  hardness  of  almost  all  water  supplies  is  caused  by  the
presence  in  solution  of calcium and  magnesium ions.  Other divalent ions
such as strontium,  ferrous  iron  and   manganese  may  contribute  to  the
hardness   to   a   much   lesser  degree.    The  lime-soda  process  is  a
precipitative  softening  process  which uses   the   addition   of   lime
(CaO-quicklime,   or  Ca(OH)2-slaked  or hydrated  lime)  to  convert  the
soluble bicarbonates  of  calcium  and   magnesium  into  insoluble  calcium
carbonate  and  magnesium  hydroxide.   This  is  the  removal  of "carbonate
hardness", or the calcium and magnesium ions associated  in  solution  with
the  bicarbonate  ion.   Calcium and magnesium associated with the sulfate,
chloride or other ions, ("non-carbonate  hardness"),  are  removed  by  the
addition  of  both  lime and soda ash (Na2C03) which provides the carbonate
ion  necessary  for  formation  of  calcium  carbonate.   Since   magnesium
removal  occurs  only  above a pH of about  11 at normal water temperatures,
excess lime sufficient to raise the  pH  to  11  must   be  added  prior  to
removal of magnesium as magnesium hydroxide.

      The  precipitated  compounds are  flocculated, settled, and removed as
sludge while the  clarified  effluent  is  usually  filtered  in  order  to
polish the effluent by removing residual floe particles.

-------
      The chemistry of  water  softening  is  probably  best  explained  or
illustrated  by  showing  the  chemical reactions that take place when lime
and soda ash are added to water containing  calcium  and  magnesium  salts.
The reactions in the lime-soda process, then are:


      C02 + Ca(OH)2 = CaC03  + H20                               (1)

      Ca(HC03)2 + Ca(OH)2 - 2CaC03  + 2H20                       (2)

      Mg(HC03)2 + Ca(OH)2 = CaC03  + MgC03 + 2H20                (3)

      MgC03 + Ca(OH)2 = CaC03  + Mg(OH)2                         (4)

      2NaHC03 + Ca(OH)2 = CaC03  + Na2C03 + 2H20                 (5)

      MgS04 + Ca(OH)2 - Mg(OH)2  + CaSO                          (6)

              Na2C03 = CaC03  + Na2S04                           (7)
      These  equations  show all of the reactions taking place in softening
a  water  containing  both  carbonate  and  noncarbonate  hardness  by  the
lime-soda process.  It should be noted that,  in  Equation  1,  the  carbon
dioxide  is  not  hardness as such, but in proportion to its content in the
water will consume lime and must therefore  be  considered  in  calculating
the  amount  of  lime  required.   Similarly,  in  Equation  5,  the sodium
bicarbonate or sodium alkalinity, if present, is not part of  the  hardness
but,  since  it  is included in the total alkalinity, it will  consume lime.
Equations 2  and  4  show  the  removal  of  carbonate  hardness  by  lime.
Whereas  only  one molecule of lime is required for one molecule of calcium
bicarbonate, Equation 2,  two  molecules  of  lime  are  required  for  the
removal  of  one  molecule  of  magnesium bicarbonate hardness, Equations 3
and 4.  Equation 6 shows the removal of  magnesium  noncarbonate  hardness,
shown  as  magnesium  sulfate,  by  lime.  No softening is effected by this
reaction because, for each  molecule  of  magnesium  noncarbonate  hardness
removed,   an   equivalent  amount  of  calcium  noncarbonate  hardness  is
formed.  Equation 7 shows the removal  of  calcium  noncarbonate  hardness,
shown  as  calcium  sulfate,  whether  originally in the water or formed as
shown in Equation 6.

      From these reactions it is apparent that  the  amounts  of  lime  and
soda   ash   required  to  soften  a  water  may  be  calculated  from  the
concentrations of free  carbon  dioxide,  bicarbonate  (usually  the  total
alkalinity),  magnesium  hardness,  and  noncarbonate hardness."  (Singley,
et al. 1977).

      Control Techniques.

      Soluble radium, a  divalent  alkaline  earth  metal  ion  similar  to
calcium   and   magnesium  is  also  removed  in  the  lime-soda  softening
process.   Table  4  shows  the   radium   and   total   removal   hardness
efficiencies  of  seven  lime-soda  softening  plants in the United States.
The data from Iowa and Illinois are the most reliable, in that  radium  and
hardness  analyses  were  performed  on  water  samples  taken  at the same
time.    The  hardness  data  from  Florida,  on  the   other   hand,   were

-------
reconstructed  from  plant  operating  records  on  the  same  day that the
radium samples were taken by another  agency.    This  means  that  the  two
samples  may  have been taken as much as eight hours apart.  It can be seen
from Table 4 that radium removal varied from  plant  to  plant  and  ranged
from  59-96  percent  removal, averaging 80 percent.  Figure 7 shows a plot
of  total  hardness  removal  versus  radium  removal   based  on  the  data
presented  in  Table  4.   If  the  two  points  from  Florida  plants  are
discounted, the line y = x2-86 fits the data  reasonably  well  ^nd  serves
as  the  basis  for a radium removal model  in  lime-soda plants."  (Singley,
et al.,  1977)  Despite the fact the the data show  some  scatter,  a  model
based on the data in Table 4 for the fraction  of hardness removal as

      Total Hardness Removed Fraction = THR =  f2-85
where  f   is  the  radium  removal fraction, as before.  The total  hardness
removal  fraction  for  achievement  of  a  final  radium  concentration  of
5 pCi/L  can be estimated by combining the equations and getting

      THR =  1 - (5/RWR) 2-86
This  is  plotted  in  Figure 8 for radium  concentrations in the raw water
from 16  to 50 pCi/L.  The lower limit for radium of 16 was chosen  since  a
lime-soda  softening  plant  would rarely if ever reduce the total  hardness
by less  than 35%, i.e., from  150  to  100  mg/L,  as  CaC03.   This  would
compare  to  a  radium  removal  fraction  of   about  70%, or from a RWR of
16 pCi/L down to 5 pCi/L.  A  correlation  of   radium  concentrations  with
other raw water quality parameters is shown in Table 3.

      A  major  problem  with  any  process  for  removal  of radium is the
disposal of the waste  stream(s).   In  the  case  of  lime-soda  softening
there are two streams, filter backwash and precipitated sludge.

      Another   process   for  the  removal  of  ionic  species,  including
hardness, ion exchange, is also applicable to  the removal  of  radium.    In
this  process  the  divalent radium ion is exchanged for sodium or hydrogen
ions.  Simultaneously other divalent ions such  as  calcium  and  magnesium
are  also  exchanged for sodium ions to accomplish softening.  Fortunately,
the ion  exchange materials have a greater affinity for radium  than  either
calcium  or  magnesium  since  the  concentration  of radium is always much
lower than either calcium or magnesium.  This  also provides  an  additional
measure  of  protection since the hardness breakthrough precedes the radium
breakthrough.  A study, sponsored by  EPA,   evaluated  a  radium  selective
complexer  (Dow  Chemical  Company, RSC) for removal of radium from brines.
The resin was shown to have a  very  high  capacity  for  radium  in  water
having   a  TDS  of 450 mg/L (51,000 pCi/g dry resin);  Both weak and strong
acid exchange resins were shown to remove radium  at  over  96%  efficiency
(Clark,   1987).    Conventional   cation  exchange  resins  have  exchange
capacities of 18,000 to 30,000  grains  of  hardness  per  cubic  foot.   A
diagram  of a typical ion exchange unit is shown in Figure 9.

      Since  ion  exchange for radium removal  is almost 100% effective, the
flow fraction to be treated can be obtained  from  Figure  1.   The  actual
efficiencies  for  ion  exchange  in operating plants are shown in Table 5.
They are seen to average over 95% for those  plants  where  no  operational

-------
problems  affected  the  results, thus the flow fraction can be modified to
accommodate the difference.   Figure  10  shows  the  relationship  between
radium  and  total  hardness  removal  for the plants shown in Table 4.  It
illustrates the  greater  affinity  of  the  resins  for  radium  than  for
hardness ions.

      Since  the  resins are greater than 95% effective for removal of both
hardness and radium, it is usually necessary to treat  a  fraction  of  the
water  and  blend  with  a  by-passed  untreated  fraction.   Assuming  95%
efficiency,  the  fraction  to be treated can be calculated from:  fraction
treated = Ft = 1.053(l-5/RWR).

      As with lime-soda softening  a  major  problem  is  disposal  of  the
waste  stream,  in  this  case  the  unused regenerate brine containing the
radium that was removed, diluted with the wash water.

      Another method for removing radium is reverse osmosis.   This  method
is  effective  for ionic species as well as any molecular species which has
a size greater than the pores in  the  semi permeable  membrane  used.    The
pressure  required  to  produce  a water of acceptable quality depends upon
overcoming the osmotic pressure of the  raw  water,  the  percent  recovery
required,  the  properties  of the membrane and the quality of the finished
water desired.  Studies supported by EPA (Clark,  1987)  showed  that  from
82-96% removal of radium could be accomplished.

      Radioactivity.   The  control  of  radioactivity in drinking water is
summarized in Figure 11, from Aieta et al, (1987).

      The only element presently regulated is  radium,  thus  emphasis  was
placed  on  it.   Table  6 provides more detail as to the efficiency of the
various treatment technologies.

SELENIUM.
      Selenium is chemically similar to arsenic and  is  generally  present
in  two  valence states in water Se IV, selenite, and Se VI, selenate, both
as the oxides SeOs"2 and SeO^2.   it  is  obvious  that  anionic  exchange
resins  will  effectively  remove  both  species.   Coagulation with ferric
sulfate results in about 80% removal  of Se IV  but  less  than  10%  Se VI.
Lime  softening  at  high  pH, i.e.>11.5, was only about 50% effective for
Se IV and less than 10% for Se VI.

SILVER.

      The occurence of silver in untreated water is so rare  that  EPA  has
proposed that it be eliminated from the regulated inorganics.

STRONTIUM.

      Since  strontium  is  an  alkaline  earth  metal in the same chemical
family as calcium,  barium  and  magnesium,  its  chemical  properties  are
similar.    The  technology  for  removal  of  strontium  is  the  same  as
discussed above for barium and radium.

-------
ARSENIC

      Arsenic  occurs  in water primarily in  the  tri-  and  pentavalent form,
usually as anionic oxides, As03~3 and AsO/f3  see  Figure  12  for  effect  Of
pH  on  arsenic  species.   The  treatment techniques  for  arsenic V removal
using  conventional  coagulation  with  alum   or  - ferric  salts  show  good.
removal below pH 8 by ferric sulfate (95+%) and about  90%   by  alum  at  pH
below  7.   Arsenic  III  is  only  about removed 50%  by ferric sulfate and
less than 15% by alum at  the  same  pH  values.    Lime  softening  removes
arsenic  V  at pH values above 10.5 at about  99%  efficiency but arsenic .III
at only about 80% above pH 10.5.

      Although  data  are  limited  there  is  little  doubt  that  reverse
osmosis and anion exchange will  remove  both  forms   effectively,  if  the
arsenic  is  present  complexed by organics it would be  removed effectively
by activated carbon.   Figures  13  and  14  show  the  effects  of  ferric
sulfate coagulation and lime softening (Sorg  and  Logsdon,  1978b).

FLUORIDE.

      The  presence  of  fluoride in drinking water is regulated under both
the primary  and  secondary  contaminant  limits   due  to   its  health  and
cosmetic  effects.   The  anomalous  addition  of  fluoride  by  many water
purveyors  to  reduce  dental  caries  presents  a  conundrum.    Excessive
fluoride   must   be   removed   but   lower    concentrations  may  require
supplementalion.

      The most cost-effective control treatment method has  been  shown  to
be  the  adsorption on activated alumina (Rubel,  1984).  Several full-scale
plants have been operated  in  the  U.S.  using  this  process,  which  was
patented  by  Churchill  in  1936.   Studies   at   these  plants  and in the
laboratory have shown that capacities greater  than 2,000  grains/cu.  ft.
can  be  obtained  by  optimizing  the  pH at about  5.5  units.  The major
sources of activated alumina is Alcoa  but other  sources  are  available.
It  is  possible to reduce the fluoride ion concentration  concurrently with
TDS  reduction  by  reverse  osmosis.   Rejections  up  to  90%  have  been
achieved.  Lime softening can remove fluoride  when high   pH  is  used  to
precipitate  magnesium.   Alum  coagulation will  reduce  fluoride because of
the formation of  insoluble  aluminum  fluoride.    High  concentrations  of
alum  are  required,  which  minimizes the selection of  this process unless
such alum concentrations are required for other purposes.

      The design of an activated alumina plant  is  covered  in  detail  by
Rubel (1984) in a manual written for the U.S. EPA.

NITRATE.

      Nitrate  occurs  at  levels above the MCL of 10  mg/L (expressed as N)
in many groundwater supplies in the U.S.  Until recently there  was  little
alternative .to  use  of  high nitrate water.  Now there are two techniques
for removal, ion exchange and reverse osmosis, R.O.   Both  remove  nitrate
so  effectively  that  some  untreated  water  usually can be by-passed and
blended.    Nitrate  removals  of  75-95%  have been   experienced  1n  RO
plants.  (Clark, 1987).
                                     10

-------
SUMMARY

      Although many inorganic species occur in natural   waters,   there  are
a  limited  number  of  applicable  treatment  processes for their removal.
The twelve that are presently regulated appear in the  regulations  because
of  their  toxicity  and  high incidence of occurence in raw waters used as
sources for  drinking  water.   A  brief  summary  of  the  best  treatment
process(es)  for  each of the twelve is shown in Table  7.   Although this is
not a comprehensive compilation of relative  effectiveness,   it   does   give
an overview.
                                     11

-------
                                 REFERENCES
Aieta, E.M., Singley, J.E., Trussell, A.R., Thorbjarnarson, K.W. and
      McGuire,   M.J.   (1987),  Radio  nuclides  in  Drinking  Water:   An
      Overview, Jour. AWWA 79:144.

AWWARF/EB1 (1985)  Internal Corrosion of Hater Distribution Systems,
      American Water Works Association Research Foundation, Denver.

Churchill, H.V., U.S. Patent 2,059,553 (November 3, 1936).

Clark, R.M. (1987), Treatment Techniques, Drinking Water Research
      Division, U.S.  EPA, Cincinnati (unpublished report).

Hursh, J.B. (1954), Radium Content of Public Water Supplies, Jour.  AWWA
      vol.46:43.

Rubel, F. Jr.  (1984)  Design Manual:  Removal of Fluoride from Drinking
      Water Supplies  by Activated Alumina.  EPA-600/2-84-134, Cincinnati.

Singley, J.E., Beaudet, B.A., Bolch, W.E., and Palmer.  J.F. (1977),
      Costs of Radium  Removal   From  Potable  Water  Supplies,  U.S.  EPA,
      EPA-600/2-77-073) Washington, D.C.

Sorg, T.A. (1978a)  Treatment Technology to Meet the Interim Primary
      Drinking  Water  Regulations  For  Inorganics:   Part  1,  Jour. AWWA
      70:105.

Sorg, T.A. and Logsdon, G.S. (1978b).   Treatment Technology to Meet the
      Interim Primary Drinking Water Regulations For  Inorganics:   Part 2,
      Jour. AWWA 70:379.

Sorg, T.A., Csanady,  M.,  and Logsdon,  G.S. (1978c) Treatment Technology
      to   Meet   the   Interim  Primary  Drinking  Water  Regulations  For
      Inorganics:  Part 3, Jour.  AWWA, 70:680.

Sorg, T.A. (1979) Treatment Technology to Meet the Interim Primary
      Drinking Water  Regulations   For  Inorganics:   Part  4,  Jour.  AWWA,
      71:454.

Sorg, T.A. and Logsdon, G.S. (1980) Treatment Technology to Meet the
      Interim  Primary  Drinking  Water Regulations For Inorganics:   Part 5,
      Jour. AWWA 72:411.

-------
                 TABLE 1.  PROPOSED MCLG^) FOR INORGANIC
                               CONTAMINANTS

                 CONTAMINANT                   MCLG, mg/L
                    As                         0.050
                    Asbestos                   7.1
                    Ba                         1.5
                    Cd                         0.005
                    Cr                         0.12
                    Cu                         1.3
                    Pb                         0.020
                    Hg                         0.003
                    N03                        10.0
                    N02                        1.0
                    Sw                         0.045
(!)    MCL  goal  -  as  changed  from  RMCL by SDWA Amendments, 1986.

(2)    MFL  =  million  fiber/liter.

-------
    TABLE  2.   Ra  CONTENT  OF  SELECTED  RIVERS  IN THE UNITED STATES
City Supply
  Source
Atlanta, Ga.
Baltimore, Md.
Birmingham, Ala.
Bismarck, ND
Boston, Mass.
Charleston, S.C.
Charleston, W.  Va.
Cincinnati, Ohio
Denver, Colo.
Detroit, Mich.
Indianapolis, Ind.
LaVerne, Calif.
Louisville, Ky.
Oklahoma City,  Okla.
Omaha, Neb.
Philadelphia, Pa.
Phoenix, Ariz.
Pittsburg, Pa.
Portland,  Ore.
Raleigh, N.C.
Richmond,  Va.
Sacramento, Calif.
Salt Lake  City, Utah
San Francisco, Calif.
St. Louis, Mo.
Tacoma, Wash.
Washington, D.C.
Ra Concentration
pCi/llter H-,0

     Raw
    Water
Chattahoochee R.
Gunpowder R.
Cahaba R. and L. Purdy
Missouri R.
Nashua R.
Edisto R.
Elk R.
Ohio R.
South Platte R.
Detroit R.
Fall Cr. and White R.
Colorado R.
Ohio R.
N. Canadian R.
Missouri R.
Delaware R.
Along Verde R.
Allegheny R.
Bull Run R.
Walnut Cr.
James R.
Sacramento R.
Cottonwood Cr.
Calaveras Res.
Mississippi R.
Green R.
Potomac R.
      0.017
      0.020
      0.024
      0.243
      0.014
      0.181
      0.041
      0.061
      0.077
      0.026
      0.137
      0.100
      0.084
      0.106
      1.770
      0.048
      0.027
      3.700
      0.014
      0.022
      0.033
      0.018
      0.034
      0.018
      1.080
      0.002
      0.033
        Hursh  (1953)

-------
                       TABLE 3.  CORRELATION OF RAW WATER QUALITY PARAMETERS WITH  RADIUM
                                             (Singley, et al. 1977)

DCI, IL
Peru, IL
Herschcr, IL
Elgin, IL
Lynwood, IL
Greenfield, IA
Adair, IA
Stuart, IA
Eldon, IA
Estherville, IA
Grinnel, IA
Hols^tein, IA
Webster City, IA
West Des Koines, IA
Sarasota, FL
Venice, FL
Englewood, FL
Mean
r*
ra = 0.05**
Ra2264
3.26
5.82
14.3
5.55
14.7
14.0
6.30
16.0
50.0
5.2
6.2
14.0
7.1
9.6
4.30
8.73
1.69
11.0
	

THb
286
298
404
246
849
630
710
640
400
915
368
920
530
376
460
570
360
527
0.057
0.606
Cab
153
188
226
142
620
400
450
375
246
600
205
600
275
210
	
	
	
335
0.036
0.532
Mgb
133
110
178
104
229
230 .
260
265
154
315
163
320
255
166
	
	
	
206
0.048
0.532
Fec
0.44
0.44
0.12
' 0,04
0.60
1.6
0.58
0.94
1.9
1.6
1.1
1.8
0.69
0.36
	
	
	
0.87
0.518
0.532
TDSC
1220
890
1426
350
1766
2160
1905
1770
1228
1350
784
1510
1010
1200
____ '•
_ 	
	
1327
0.176
0.532
ALKb
286
318
259
303
215
190
158
182
252
367
298
288
294
260
____
	
• 	
262
0.252
0.532
Bac
0.13
0.13
0.10
•8.7
0.09
'<0.10

-------
TABLE 4.   RADIUM226 AND TOTAL HARDNESS REMOVAL EFFICIENCIES LIME-SODA SOFTENING
                             (Singley, et al.  1977)
LOCATION


W. Des Moines, [A
8/1/74
Webster City, IA..
Clarlfier 11
8/13/74
Webster City, IA
ClarUier S2
3/13/74
Webster City, IA
Clarifier 111
2/20/75
Webster City, IA
Clarifier 11
2/20/75
IVru, IL
2/20/75
Peru, IL
2/25/75
Peru, IL 3/4/75
Mean
Elgin, IL 3/7/75
ElRin, II. 3/14/75
Elgin, IL 3/21/75
Mean
Venice, FL 8/7/75
Englewood, FL
8/7/75
CLARIFIKPS
Ra in
pCi/1
9.3

6.1


h - 1


7.8


7 . S


fi.49

5.41

5.48
5.82
7.45
5.7
3.51
5.55
8.73
1.69

Ra out
pCl/1
2.6

1.9


2.6


0.9


0.3















% RA
Remova 1
72

69


57


88


96















TH In
mg/l CaCO
376

507


507


482


482


329

278

236
289
246
243
242
244
570
360

TH Out
mR/1 CaC01
215

333


282


150


150















% Til
RFJ1
43

34


44


69


69















pH Treat-
ment
10.4

10.05


10.1


10.95


10.95


8.4

8.4

8.4

10.2
10. -2
10.2

9.7
8.5

FI1.TF.RF
RA In
pCI/l
2.6

1.9


2.6


0.9


0.3















RA Out
pCI/1
2.35

0.9


0.9


0.3


0.3
1 R.-,
P^mova I
10 '

53


65


Til In Tl! Out
ir.c/IC.-njO ! ma /I racn
215 190


X TH
REJI
12

331 262 ! 21



2RT. ' 262 7


67 j 150 ' 106
1 !
1
0


0.51

1.62

1.33
1.15
.75
.80
.71
.75
2.19
.69

150 ! 106


! 174

























180

122
147
99
112
95
102
100
166



29


29















OVERALL
7. Ra
RP-I
?5

85


85


96


96


92

70

76
79
90
86
80
85
75
59

7, TH
REM
49

4S


48


78


78


47

35

57
49
60
54
61
59
82
5'.


-------
  Table5.   RADIUM  REMOVAL IN ION EXCHANGE  PLANTS
Plant
Eldon, la
Estherville, la
Grinnell, la
Holstein, la
Dwight Corn
Inst., II
Herscher, II
Lynwood, II
SarasoTa, Fl
Ra in
pCi/L
49
5.7
6.7
12
3.26
14.31
14.69
4.3
Ra out % Ra
pCi/L Rem. •
1.9
0.3
0.2
0.5
0.36
1.31
0.41
0.70±
96
95
97
96
89
91
97
• 84
TH in TH out
mg/L/CaC03 mg/L/CaC03
375
915
385
920
286
406
848
460
'0
46
11
18
43
60
78
159±
% TH
Rem. •
97
95
97
98
85*
85*
91
65**
• Removed
± Adjusted to take account of raw water blending.
* % Hardness and %Ra removals are somewhat low due to breakthrough
  occurring prior to all samples being collected.
**%Hardness and % Ra removals are somewhat low due to incomplete
    regeneration of media as 1/2 brine  pumping  capacity was down for
    repair.

-------
TABLE 6. TREATMENT TECHNOLOGIES FOR REMOVING RADIONUCLIDES
(Aleta, et al. 1977)


Treatment
Technology
Coagulation-filtration







Lime-soda softening







Ion exchange





Adsorption




Aeration




Reverse osmosis









Radionuclide
Ra
U





Rn
Ra




U

Rn
Ra


U

Rn
Ra


U
Rn
Ra

U
Rn

Ra



U

Rn
Approximate
Process
Efficiency
percent
<25
95+


18-98



80-95
75-96
85
43-92
80
85-90
99

95+

99
99


90+
85-90


62-99
18
20-25

20-96
93+
87-96
87-98
95+

95+





Comments
High pH and Mg required
High pH (10+) and high
dosages of ferric chloride
or alum
Only accomplished in lab
studies with diatomaceous
earth filtration
N/A*
Best choice for large plants


Plant-scale results
Plant-scale results
pH 10.6-11.5
High pH, high Mg
N/A
Best choice for small plants;
cation exchangers
Brine disposal problem
Anion exchangers; largely
experimental
N/A
Experimental
Sand adsorption;
experimental
N/A
GAC adsorption
Adsorption on Fe(OH)a
Fe(OH)3 and sand
N/A
Depends on process
Depends on process
Plant-scale data
Based on eight plants
High-volume brine solution
for disposal
High-volume brine solution
for disposal
N/A











































*N/A = not applicable

-------
                               TADLE 7.  SUMMARY OF BEST TREATMENT PROCESSES FOR

                                        REMOVAL OP INORGANIC CONTAMINANT
Contaminant	Coagulation
                  Line
                Softening
                                                     Exchange
Activated
 Alumina
                    GAC	R.O. (a)
                        ..(c)
                                                                                        Bone char       +
    Da

    Cd

    Cr III

    F
+pH>8
                                     +pH 9.5-10.5
                    >10.5
                                  +

                                  +

                  Bone char       +
    Pb

    Hg (Organic!

    Ilg ( Inorganic )

    (103

    Se
                                                                            (Se IV)
                                     E pH 9.5-10.B   +
      Ho - reverse osmosis, also clectrodialysis assumed to be equivalent.
      As III can bo oxidized to As V with C12.
      •*• mcanB 90-1001 removal.
      + moans cation exchange; - moans anion exchange.

-------
200
     *mg/L,as CaC03
    8
                      PH
Figure 1. Solubility of Lead as a Function  of pH
    and  Carbonate. (Adapted from AWWARF/EBI,
    1985)

-------
                         P04
                      Concentration
minimum Pb
  atpH=7.6
Figure2. Lead Concentration as Function of pH and
        Phosphate Concentration  at a Total Alkalinity
        of  50mg/L  (Adapted from Schock)

-------
 JE

 >N
 .£>
 13
                     I    I   I    I   I
   .001
Figure3. CADMIUM SOLUBILITY AS A FUNCTION
       OF pH,

-------
C
o
c
0)
o
c
o
o
o
     1.0
    0.1
   0.01
Cr(OH),
              MCL
                        1
   I
       6       7       8        9       10      11

                            PH


  Figure 4. Solubility of Cr111  as a Function of pH

-------
TJ
CD

O
O
to
CD

"o
CD
Q.
c:
CD
CO
CD
°-   .20  ~
     0
Figures.  Effect of pH on Speciation of Chromium VI

-------
    .1OO r
CD
(T _J
_. \

> "^
o °-
E 10
^ •*-
CL o


II
O
o:
  (D
£ °
o •»-

o
D
     .80
,40
     .20
             10
             20
30
40
50
60
              Radium Level -Raw Water (pCi/L)


 Figures. REQUIRED REMOVAL  FRACTION FOR

        RADIUM

-------
  1.00
|.80
O
o
   .60
O
QC

-------
         O
         
-------
     METER
INLET
                      ION-EXCHANGE
                         UNIT
                         EXCHANGE
                         MATERIAL.
 BACKWASH
   OUTLET
         BACKWASH
           INLET
 OUTLET
RINSE
OUTLET
                                                WASH-WATER
                                                COLLECTOR
                                             I'

                                                    PRESSURE WATER
                                                     EJECTOR
                         SUPPORTING
                            BED
                                                       REGENERANT  TANK
                            TO WASTE
     Figure 9.  DIAGRAM  OF  TYPICAL  ION EXCHANGE UNIT (3)

-------
1.00
   .20
   .40      .60      .80      1.00
Total Hardness Removal Fraction
  FIGURE 10. RADIUM REMOVAL FRACTION
     VS. TOTAL HARDNESS REMOVAL
     FRACTION IN  ION EXCHANGE PLANTS
     (BEFORE  BLENDING)
    (Adapted fromS/ng/ey, eta I, 1977)

-------
4-
D
O
O)
"o
0)
Q.
•t—
c

-------
  0.40
c
o
  0.30
£  0.20
o
c
o
o

o


£  0.10
  0.00
                            As*
          Initial Concentration
               0.4 mg/L

          MCL
       8
                                10
11
                          PH
Figure 13. Effects of pH on Arsenic Removal  by

        Lime Softening (Adapted from Sorg 8

        Logsdon  1978)

-------
   0.20
    0.15
   0.10
   0.05
   0.00
                      Asm0.3mg/L
                 Initial Concentration
                     As Shown
          MCL
Figure 14. Effects of pH on Arsenic Removal by
         Ferric Sulfate  Coagulation (30mg/L)
         (SorgB Logsdoml978)

-------
                   WORKSHOP ON EMERGING TECHNOLOGIES FOR
                     DRINKING WATER TREATMENT INORGANIC
                   CONTAMINANTS PROCESS,PLANNING OVERVIEW
     The planning of appropriate treatment processes for the twelve regulated
inorganic contaminants is complex and thus difficult to adapt to a rigid check-
list format, although there are many elements in common.  It is important to
keep in mind that there are many alternatives in each case, not only process-
wise, but compatibility with existing process streams and equipment.  In many
cases the decision-making process may be considerably simplified by application
of common sense	which cannot be programmed.  Figure 1 shows many of the
elements that can be used sequentially - most of which are obvious.

-------
     CONTAMINANT
                              
-------
                      Inorganic  Contaminants

                     Process  Design  Check  List
  I   Contaminant  Identification

     A.   Historical  Records
     B.   Current  Analyses

 II   MCL Compliance  Evaluation

     A.   Present  MCL
     B.   Proposed MCL
     C.   Treatment Required

III   Plant and Process Evaluation

     A.   Physical Evaluation

         1.  Treatment Units
         2.  Appurtenances
         3.  On-site Space

     B.   Process Evaluation

         1.  Operating Parameters
         2.  Flexibility

 IV  Evaluation of Alternate Processes

     A.   Coagulation

         1.  Alum
         2.  Iron Salts

     B.   Lime-soda Softening
     C.   Ion Exchange

         1.  Cation
         2.  Anion

     D.   Reverse Osmosis
     E.   Activated Carbon

         1.  PAC
         2.  GAC

     F.   Activated Alumina

-------
   V  Optimization of Existing Processes
      A.   Feasibility
  VI  Evaluate Compatibility With Plant
      A.   Physical
      B.   Chemical
          1.  Control Contaminant(s)
          2.  Maintain Quality
 VII  Evaluate Process Charges in Existing Plant
VIII  Evaluate Technical Alternatives
      A.   Waste Stream Disposal
          1.  Solids
              a.  Hazardous
              b.  Non-hazardous
          2.  Liquids
              a.  Hazardous
              b.  Non-hazardous
      B.   Cost
          1.  Capital
          2.  O&M
      C.   Energy Dependence
      D.   Reliability
      E.   Operational Compatibility
   IX  Final Selection
      A.   Process
      B.   Plant
          1.  Modifications to Existing Plant
          2.  New Plant

-------
                    PROCESS CHECK LIST
            Evaluation of Alternative Processes
A.  Contaminant(s) Present
         1.  Identification of those exceeding MCL
         2.  Historical record of contaminant(s)
         3.

B.  Candidate Process(es) Considered
         1.  Non-treatment
                a.  Blending
                b.  New source(s)
                c.  Purchase
                      1) Raw
                      2} Treated
                d.  Bottled water
                e.  Point-of-use treatment

         2.  Treatment
                a.  Coagulation
                      1) Alum
                      2) Iron salts
                b.  Softening, lime-soda
                c.  Ion exchange
                      1) cationic
                      2) anionic
                      3) pre-treatment
                d.  Reverse osmosis  (electrodialysis)
                      1) high pressure
                      2) low pressure
                      3) pre-treatment required
                e.  Activated Alumina
                f.  Activated Carbon
                      1) PAC
                      2) GAC

C.  Compatibility With Present Process
         1.  Process reoptimization
         2.  Facilities
         3.  Retrofitting

D.  Waste  Disposal
         1.  Solids
                a.  hazardous
                b.  non-hazardous
         2.  Liquids
                a.  hazardous
                      1) concentrate
                      2) neat
                b.  non-hazardous
                      1) .dry  (concentrate  solids)
                      2) recycle portion
                      3) sewer
                      4) return to  source

-------
PROCESS CHECK LIST - Evaluation of Alternative Processes
Page 2
E.  Cost Evaluation
         1.  Processing
               a.  Capital
               b.  0 & M

         2.  Disposal
               a.  Capital
               b.  0 & M

         3.  Environmental Permits

-------
                Sample Ra Control Process
          (Applicable to Any Other Contaminants)
I.   Lime-Soda
                  Raw                         Finished
    Parameter     Mg/L    Meq/L    pCi/L      Mg/L  Meq/L  PCi/L

    TH,asCaCO^    300
    Ca,    " J    250      12.50
    Mg,    "       50       4.17
    Alk,   "      200       4.00
    TDS           400
    CO             15       3.40
    Ra                               22                     <5.0
    A.  Estimate the finished Total Hardness  (TH)
    B.  Estimate the finished  .a concentration
    C.  Approximate the pH of treatment
    D.  If the population of the service area was 15,000,
        what would be
             a.  the approximate volume of filter backwash,
                 in gal/day?
             b.  the approximate dry weight of the sludge?
             c.  the radium content of the sludge (70% dry)?
             d.  the radium content of the backwash water?


II.  Ion Exchange

     Same water, same population

     Assume 1) operating capacity of resin^
               20 kgr hardness as CaC03/ft
            2) regeneration with 10% brine
            3) salt for regeneration =
               0.30 pounds salt/kgr. hardness
               removed  (as CaCO.,) or
               6 pounds/ft. .
            4) 95% TH removal
            5) 95% Ra removal

     A.  Calculate the  fraction of the flow to be treated.
     B.  Estimate the amount of regenerant salt, per million
         gallons of finished water.
     D.  Estimate the total volume of brine to be disposed
          (backwash + rinse water) per MG.
     E.  Estimate the Ra concentration in the waste brine
          (bw + rinse).


Ill.  Reverse Osmosis
      Same water, same  population.
      Assume 1) 95% removal of all ions
             2) 95% removal of Ra
             3) blending to 5 pCi/L Ra

-------
Sample Ra Control Process
(Applicable to Any Other Contaminants)
Page 2
     A.  Calculate the TH of the blended, finished water.
     B.  Calculate the fraction of water to be treated.
     C.  Calculate the reject brine Ra concentration.

IV.  what would be the most compatible choice of treatment
     process for the following utilities?  What factors should
     be most relevant?

     A.  Utility has only wells with chlorination.
     B.  Utility has lime-soda softening plant, reducing
         Calcium hardness from 190 to 70 mg/L, as CaCO.,.
         Its source is a river with average turbidity of 8 ntu.
     C.  Utility has an ion exchange plant for softening a well
         supply from 200 to 100 mg/L as CaC03.

-------
                             SAMPLE CALCULATIONS
Lime-Soda Process

High Solids Raw Water

     TH   - 750 mg/1 @ CaCC>3
     Ca++ - 500 mg/1 @ CaC03
     Mg^ - 250 mg/1 <§ CaC03
     ALK  - 300 mg/1 @ CaC03
     IDS  - 2000 mg/1
     C02  - 11.4 mg/1 @ CaC03

1 - Desired finished water for RaQ =7.5 pCi/1
    Ra removal required - 33 percent or  .33
    TH removal required - Ra2-86
                            rem
                        = .332-86 =  -042
    Minimum TH considered practical in lime-soda process is 35 percent

    TH removal required - .35  (750) = 262 mg/1 @ CaCC>3, or 5.25 mg/1
    TH of finished water= 488 mg/1 @ CaC03 or 9.75 mg/1
                                                    1
2 - Chemical requirements
Parameter



CO 2
Ca"1"1"
Mg"^"
HC03
TOTALS



mg/1

5
200
60
366

Raw

AfJ
njgyl

0.23
10.00
5.00
6.00

Finished

meq/1

0
4.75
5.00
0

A
gxr
mgyl
\
0.23
5.23
0
6.00

Required
CaO
meq/1

0.23
5.25
'
0.75
6.23
Required
CaO
meq/1

_
-
-
-

     Lime Required = i^-23^28) =  194 mg/1


     Soda Ash Required = None

-------
 IJJ4 Exchange

 1 - Mass Balance Calculation

   To determine fraction of raw water influent blended, and  fraction treated,
   assuming 95 percent ren .ival in I/E or R/O column.

   For mass balance of Ra:

        .05 Ra0x + RaQ (1-x) = (1) (5)

        .05 Ra0x + Ra0 - Raox = 5

        x Ran  (.05 - 1) 4-  Ra0 = 5
        -.95 x Ra0 = 5 - Rao


             X = ^pVo

             x = Ra0 - 5
                  .95 Ra0

   Blending allowed  to attain minimum  level of  5  pCi/1  Ra assuming 95 per-
   cent Ra removal in I/E and R/0
   Ra Level                Fraction  Blended                Fraction Bicttderd

     50                          .053                           .947
     20                          .211                           .789
      7.5                        .649                           .351

2 - Regenerant  Chemicals  Calculations

   Assume:   1)  operating capacity of  resin  = 20 kgr hardness as CaCOo/CF
             2)  regeneration is  with a 10 percent brine concentration
             3)  salt  required for regeneration is 0.30 Ib salt/kgr hardness
                as  CaCo3  removed (6  Ib salt/CF)
             4)  95  percent TH removal  in unit (5 percent leakage)

-------
Solids Levels:
High - TH    = 750, TH_ =  .05  (750) = 37.5
——     raw          F
       TH     =  712.5 rag/1
         rem           &
Kilograins hardness removed

       712. 5 mg/lxl  grain/gal^ kgr
                    17.1 mg/l*1000 gr   '
 Salt regeneration  requirement/MG
        .0417  x  106  kgrx-30  Ib  salt _   2
              rag          kgr           '
 Volume  of  rinse water/MG,  assume  30  gal/CF resin
        41,700  kgrTHr CF 30  gal  _
             mg     20 kgr   CF  ~
 Volume of regeneration brine  solution/MG
               - 12,510 = 112,590  lb  H20

                          13,500 gal  H20


 Total volume of backwash + rinse  water/MG

 High Solids - 13,500 + 62,550 = 76,050 gal or 7.6  percent of total flow

-------
Reverse Osmosis Calculations

1 - Assumptions

    High Solids - Initial TDSi = 2000 mg/1

                  Product TDSi = .05  (2000) =  100 mg/1

                  Calcium initial, Cai = 200 mg/1


2 - Brine to Product Ratio (BPR) =
       RPR _ 1 - (TDSp/TDSj)
       B K   (900/Cai) - 1
           _ i - (100/2000) _ ^9i _  271
       BPR ~  (900/200) - 1 " 3.5 "  -271
3 - Brine Volume
       Vb = Vp x BPR =  .271 Vp Vp = Product Volume



4 - Feed Volume (Vj_)


       Vi = Vp + .271 Vp = 1.271 Vp

-------
REMOVAL  MECHANISMS
      PRECIPTATION
      CO-PRECIPTATION
      ADSORPTION
      COAGULATION
      OXIDATION
      COMPLEXATION
      ION EXCHANGE

-------
ANIONS:
         SPECIAT1ON
As (AsO^AsOi3)
Cr (CrOi2,CrO?)
F-
             Se(SeO"2SeO;2)
 CATIONS:
Ba+2
Gd+2
Pb+2
             Ag
               +1
COMPLEXES:
Hg (ORGANIC)
HgOH*
HgCL*
CdOH+
MC0°

-------
SPECIATION  PARAMETERS
  pH
  SPECIES CONCENTRATION
  COMPLEXANT CONCENTRATION
  PARTICULATE CONCENTRATION
  REDOX POTENTIAL
  IONIC STRENGTH
  TEMPERATURE

-------
         EFFECT OF pH
IONIC SOLUBILITY:

  HYDROXIDES EXCEEDS "REGS" ONLY BELOW:
  Cd  pH <11.0
  Pb  pH < 6.2
  Hg  pH < 7.0
ACIDIC SPECIES:
  As03"—HAs Oa
  As 0? - HAs 042- H2As 041- H3As 04
  CrQ;2-HCr04--H2Cr04
  Cr2a27-HCr20;-H2Cr207
  SeOj2—HSe 0"3—H2Se 03
  SeO^-HSe 04-H2Se 04
  F-HF

-------
  100
   80
I  60
tu
a:
g  40
Ul
   20
     MCL for
     0.4 mg/l
           As 0.4 mg/l
 0 As*3 chlorinated
 5JC As*3 not chlorinated
 Pilot plant tests
 XAs*5
 $£ As*3 not chlorinated
****
                                 *.-
                                       -*••-*
                      9              10
                      pH OF TREATED WATER
         ARSENIC  REMOVAL BY LIME SOFTENING
                                                             11
                12
  100  -
   80
Q
Ul
o  60
ui
Q.
   40
   20
       ,,,,,,...
       --
Lime softening
pH 10.9-11.1
A As*5
 Ferric sulfate 30 mg/1
 • As*5
 OAs*3
 Alum 30 mg/1
 QAs*5
 A As+3

                                     I
                                                            I
                 I
     0.10
                  0.5       1.0                     5        10       20
                      ORIGINAL CONCENTRATION, mg/l
          ARSENIC REMOVAL BY COAGULATION AND LIME SOFTENING

-------
CADMIUM SOLUBILITY AS A FUNCTION
OF pH.

-------
 100
                         • 111»1111 • 111 • 11 n t • I • ® 111111 • « • • • • f}9M
                                                             Illl
  80
2  60
             MCLfor
             0.03 mg/l
UJ
UJ
e_
   40
   20
           Cd 0.03 mg/l
           • Well water
           X Pilot plant tests
                                 I
                                                I
                                 9               10
                                pH OF TREATED WATER

                    CADMIUM  REMOVAL BY LIME SOFTENING
                                                               11
12
  100 -
   80
Q
UJ
2  60
   40
   20
            Cd 0.03 mg/l
            River water
            :J* Ferric su I fate
            O Alum
            Pilot plant tests
            X Ferric sulfate
                                 7              8
                                pH OF TREATED WATER
                    CADMIUM  REMOVAL   BY  ALUM
                    COAGULATION
                                                                               10
                                                        AND  FERRIC  SULFATE

-------
  100
   80
Q
UJ
   6°
           Ba 7-8 mg/l
          I Hardness
      _  X Pilot plant tests
UJ
te.
l-
UJ
y  40
HI
a.
   20
           MCLfor
           8 mg/l
                                    *
                                               _L
                              :•*«,_—
                                   W
                                                         \
                                                              tr
                                9              10
                               pH OF TREATED WATER

                   BARIUM  REMOVAL BY  LIME SOFTENING
                                11
                                                                             12
   100
   80
 UJ
 oi

 LJ
 Ul
 a.
   60
   40
    20
               MCL for
               8 mg/l
             Ba 7-8 mg/l
             :£ Ferric sulfate 20-30 mg/l
             O Alum 20-30 mg/l
                    BARIUM
                    COAGULATION
   7              8
  pH OF TREATED WATER
REMOVAL   BY   ALUM
                                                                              10
AND  FERRIC   SULFATE

-------
  1.00
•4—
o
o
   .80 ~
   .60
o
CD
O.
CO


S  .40
4—
C
CD

CD


£  .20
     O
                      5       6

                          PH
8
         Effect of pH on Speciation of Chromium VI

-------
4—

O

-------
  100  -
   80
                                         	*•"	
                                                               MCL for
                                                               0.15mg/l
UJ

I  60
UJ
C£.

UJ
   40
Ul
                                                               Ag 0.15 mg/l
                                                               • Well water
   20
                                 I
                                                I
                                9              10
                                pH OF TREATED WATER

                    SILVER REMOVAL BY LIME SOFTENING
                                                              11
         12
  100
   80
o
UJ
tt
t
UI
a.
   60
   40
                                                   Cr*3 pH 7.3-7.6
                                                   Well water
                                                   Chlorine 2.2 mg/1
                                                   s|e 6 h contact time '
                                                   O 20 h contact time
   20
     0.10
                           0.5        1.0      :
                                ORIGINAL CONCENTRATION, mg/l
10
20
                     EFFECT OF PRECHLORJNATION ON Cr+3  REMOVAL BY ALUM
                     COAGULATION

-------
100
80
PERCENT REMOVED PERCENT REMOVEt
0 g £ 8 8 8 o8oS
9""ii,,
*****»*,
«•••
0
1
Se*4 0.03 mg/l
Ferric sulfate 25 mg/l
£ River water
jj: Well water
Alum 25 mg/l
*""••£,. MCL for D River water
"»,- 0.03 mg/l 0 Well water
•••.£<,, ^^ Pilot plant tests
x '^"Xj *tf*tt Gravel pit water
VN*. ' ***«>, X Ferric sulfate
^*^^ %^ River Water
^^^ **^ + Ferric Sulfate
1 i 1 M 1 i 	 1
6 7 8 9 10
pH OF TREATED WATER
Se+4 REMOVAL BY ALUM AND IRON COAGULATION
~" MCL for 0.1 mg/l
MCL for 0.3 mg/l
_ Se+4 Well water
~" • 0.1 mg/l
A 0.03 mg/l
X Pilot plant tests
~ Se*4 0.03 mg/l
1
8
^farjj%4fr**^**f\*~^
— i 	 1 	 1 	 1 	 i 1 i 	 |
9 10 n i:
          pH OF TREATED WATER
Se+4 REMOVAL BY LIME SOFTENING

-------
  TOO
   80
2  60

o

tu
ui
U
e£
   40
   20
                Pbi
            &**—*„*•      i«

           a*6
           U •••€»•
               I

               6
• Illllllllflllll""1111
                                    pH OF TREATED WATER


                 REMOVAL OF  INORGANIC CONTAMINANTS BY ALUM COAGULATION

-------
  100
tu
u
t£
   80
S  60

o

LJ
   40
   20

r	
                                7                 8

                                   pH OF TREATED WATER


                REMOVAL OF INORGANIC  CONTAMINANTS BY IRON COAGULATION

-------
  100
   80
Q
ui
O
ui
ui
Ui
O.
60
40
   20
             0*3
                        J
     8                  9                  10                  11
                                    pH OF TREATED WATER
     FIGURE 3    REMOVAL OF INORGANIC CONTAMINANTS BY LIME SOFTENING
                                                                            12

-------
   TREATMENT PROCESS
              CONVENTION  COAGULATION
 PRINCIPAL APPLICATION
 FOR WATER TREATMENT
           CLARIFICATION OF SURFACE WATERS
                             HIGH
 INORGANIC CONTAMINANT
 TREATMENT CAPABILITY
     EFFECTIVENESS*
MODERATE
                             LOW
               Cd, Cr III, Cr VI, As V, Ag, Pb
As III, Se IV, Hg(0), Hg(l)
               Ba, F, NO. Ra, Se VI
                      3
     MOST PROBABLE
    APPLICATION FOR
  INORGANIC REMOVAL
  REMOVAL OF Cd, Cr, As, Ag, OR Pb FROM SURFACE WATERS
HIGH - GREATER THAN 80 PERCENT MODERATE - 20 TO 80 PERCENT  LOW - LESS THAN 20 PERCENT

-------
  TREATMENT PROCESS
                 LIME SOFTENING
 PRINCIPAL APPLICATION
 FOR WATER TREATMENT
             REMOVAL OF HARDNESS  FROM
             GROUND AND SURFACE WATER
                             HIGH
INORGANIC CONTAMINANT
 TREATMENT CAPABILITY
     EFFECTIVENESS*
MODERATE
                             LOW
               Ba, Ra, Cd, Cr III, As V, Pb
Se IV, As III, Hg(l), F
               Cr VI, NO , Se VI, Hg(O)
                      3
     MOST PROBABLE
    APPLICATION FOR
  INORGANIC REMOVAL
       REMOVAL OF Ba OR Ra FROM GROUNDWATERS;
       REMOVAL OF Cd, Cr III, F, As V, OR Pb FROM HARD
       SURFACE WATERS REQUIRING SOFTENING
HIGH - GREATER THAN 80 PERCENT  MODERATE - 20 TO 80 PERCENT  LOW - LESS THAN 20 PERCENT

-------
   TREATMENT PROCESS
       CATION  EXCHANGE
  ANION EXCHANGE
 PRINCIPAL APPLICATION
 FOR WATER TREATMENT
     REMOVAL OF HARDNESS
     FROM GROUNDWATERS
REMOVAL OF NITRATE
FROM GROUNDWATERS
                             HIGH
 INORGANIC CONTAMINANT
 TREATMENT CAPABILITY
     EFFECTIVENESS*
MODERATE
                             LOW
             Ba, Ra, Cd, Pb, Cr HI
 NO , Cr VI, Se
   3
  Ba, Ra, Cd, Pb, Cr II!
            As, Se, NO , F, Cr VI
                    O
     MOST PROBABLE
    APPLICATION FOR
  INORGANIC REMOVAL
      REMOVAL OF Ba OR Ra
      FROM GROUNDWATERS
REMOVAL OF NO3
FROM  GROUNDWATERS
HIGH - GREATER THAN 80 PERCENT  MODERATE - 20 TO 80 PERCENT LOW - LESS THAN 20 PERCENT

-------
  TREATMENT PROCESS
        REVERSE OSMOSIS AND  ELECTRODIALYSIS
 PRINCIPAL APPLICATION
 FOR WATER TREATMENT
   DESALTING OF SEA WATER OR BRACKISH GROUNDWATER
                             HIGH
INORGANIC CONTAMINANT
 TREATMENT CAPABILITY
     EFFECTIVENESS*
MODERATE
                             LOW
              As V, Ba, Cr, Pb, Cd, Se, Ag, F, Ra, Hg
NO , As III
  3
     MOST PROBABLE
    APPLICATION FOR
  INORGANIC REMOVAL
    REMOVAL OF ALL INORGANICS FROM GROUNDWATERS
HIGH - GREATER THAN 80 PERCENT  MODERATE - 20 TO 80 PERCENT  LOW - LESS THAN 20 PERCENT

-------
  TREATMENT PROCESS
  POWDERED  ACTIVATED CARBON
                 GRANULAR ACTIVATED
                 CARBON
 PRINCIPAL APPLICATION
 FOR WATER TREATMENT
  REMOVAL OF TASTE AND ODORS
  FROM SURFACE WATERS
                REMOVAL OF TASTE,
                ODORS,  AND ORGANICS
                             HIGH
 INORGANIC CONTAMINANT
 TREATMENT CAPABILITY
     EFFECTIVENESS*
MODERATE
                             LOW
                                  , Hg(0),
, Hg(0), Cd
Cd
           Ba, Ra, Cr III, F, NO , Ag
                         3
                Ba, Ra, Cr 111, F, NO
     MOST PROBABLE
    APPLICATION FOR
  INORGANIC REMOVAL
  REMOVAL OF Hg FROM SURFACE
  WATERS DURING EMERGENCY
  SPILLS
                REMOVAL OF Hg FROM
                SURFACE OR GROUND-
                WATERS
HIGH - GREATER THAN 80 PERCENT  MODERATE - 20 TO 80 PERCENT LOW - LESS THAN 20 PERCENT

-------
  TREATMENT PROCESS
                ACTIVATED ALUMINA
 PRINCIPAL APPLICATION
 FOR WATER TREATMENT
       REMOVAL OF FLOURIDE FROM GROUNDWATERS
                            HIGH
INORGANIC CONTAMINANT
 TREATMENT CAPABILITY
    EFFECTIVENESS*
MODERATE
                            LOW
              F, As, Se
              Ba, Ra, Cd
    MOST PROBABLE
    APPLICATION FOR
  INORGANIC REMOVAL
      REMOVAL OF F, As, OR Se FROM GROUNDWATERS
HIGH - GREATER THAN 80 PERCENT MODERATE - 20 TO 80 PERCENT  LOW - LESS THAN 20 PERCENT

-------
PERFORMANCE SUMMARY FOR INORGANIC TECHNOLOGIES EXAMINED
Inorganic
Compound
Reverse
Osmosis
                                       Removal  Efficiency
Ion
Exchange
Aeration
Carbon
Adsorption
Nitrate
Radium
Uranium
Radon
   -H-
   -H-
  •H-
                 •H-
                       +f= Excellent 70%-100%
                        % = Research Being  Conducted By  DWRD

-------
  MOST PROBABLE APPLICATIONS
              OF
  WATER TREATMENT PROCESSES
              FOR
INORGANIC CONTAMINANT REMOVAL

-------
                                                                    Generally
                                                                   effective tor
Alternatives  for reducing radionuclide concentrations in
drinking water (all  alternatives can include blending)
(Aieta, et al.  1987)

-------
    .100 r
O)
cr
(D
.80
§5

E i?-60

Q: *o
D
     ,40
c
o
*4—
O
a
     .20
        )     10    20   30    40   50    60

              Radium Level - Raw Water (pCi/L)


        REQUIRED REMOVAL FRACTION FOR

        RADIUM

-------
    1.00
c
o
•M
O
1/1
(/I
d)
 rci
-•->
 O
        16.0  20.0
 30.0              40.0


Raw Water Radium pCi/1
50.0
   Total  hardness removal fraction as a function of raw  water   radium  content
   required to meet limit of 5.0 pCi/L finished water.   (Singley,  et al.  1977)

-------
  1.00
1.80
u
o
§  .60
E
o>
rr
V)
 X =  West DesMoines, la
+ =  Webster City, la W/0 Soda Ash
©=  Webster City, la W Soda Ash
El55  Peru, i| (3dates)
A=  Elgin, II (3 dates)
•X- =  Englewood, Fl
0=  Venice, Fl
   .40
o
X
   .20
              .20       .40      .60
                Radium Removal Fraction
                                 .80
1.00
               LIME SODA PROCESS, TOTAL HARDNESS
            REMOVAL  FRACTION VS RADIUM  REMOVAL
            FRACTION.

-------
                 Raw Water
                Q
             f, + fg- 1
      f2Q
     f,Q
       Bypass
       Water
           Ion
           Exchange
           Unit
            Q
Finished
Water
MASS BALANCE FOR DETERMINING  FRACTION
OF RAW WATER TO BE TREATED

-------
 1.00
c
o

'•5
D
  -80
E

-------
                                          WASH-WATER
                                           COLLECTOR
OUTLET
                                                  REGENERANT  TANK
              DIAGRAM OF TYPICAL ION  EXCHANGE UNIT

-------
  1.00
c
o

o
o
   .80
o

E
0)
a:


E  .60
3

T3
O
o:
   .40
     .20
                I
           I
I
I
.40       .60       .80        1.00

  Total Hardness  Removal Fraction
    RADIUM  REMOVAL FRACTION VS. TOTAL HARDNESS REMOVAL

    FRACTION IN ION EXCHANGE PLANTS (BEFORE BLENDING)

-------
  you can't
 SEE THEM
SMELLTHEM
       or
 TASTE THEM
  and they can
                           Guess Who Has Started-Worryin
                           About Qualit            -
                                      *** «-W^ *ts«ra£
                                      ! a Ugt memoes. l£l£S%*fa ««,« trthalo.
                                      ^^.-^.fiSLaarEsjf*
                                       • . »• Kunla also 'm^^. .. '   .
'?»'•• l-l pSdlitS*-?* ^thy Wr'
level of PracaSSrevHi',,?'' ta»*« « WE« metha
f""' I" be loldlKSJ^f' ^ "^ «™ tectf0'
ie«' <" ^^'ft'^er1'6 ' "i* "Dr.
                          5Wa5S«2SS Stft^s5te»g
                          V^SfftSS^r^ asaK^^jSrssaS
                          v - -*. «.fc«uus 'ft?iS53?s^aS
                           % ._"      ' ,    i j/^t«..t_ ...  "flflS*

                               Known1  •    ' wulr
      •»'  •   ' 2VI^i*«^^Sfl?*««i5
      55«&a &;to <5top ^«"«
      iWs. Conducted S _.S« Dr. Kunfa «»,„.„-,	

-------
       you can't
     SEE THEM
  SMELL THEM
              or
  TASTE  THEM
     and they can

   KILL  YOU!
  The TRIHALOMETHANES (THM's) found In drink-
Ing water are organohalogen compounds that are named
derivative! of methane, formod when thrao of tha four hydro-
gen atomi have been replaced by three atom* of chlorine,
bromine or Iodine. Ten distinct compounds can be form-
ed from the possible combinations of three hydrogenated
atoms, one hydrogen atom, and one carbon atom. Current
analytical methodology applied to drinking water has thus far
led to the detection of chloroform (trlchloromethane),
bromodichloromethane, dlbrombchloromathane, bromoform
(trlbromomethane), and dlchlorolodiomathane. Monitoring
methods are currently available for the bromlneted and
chlorinated THM's, but not for the iodinated types because
of their chemical Instability. EPA has promulgated an Interim
standard of 0.1 mg/l (lOOppb) for THM's.

-------
  Would You drink this?
              \     '                     •

        ,;      CHLOROFORM
    MAGNESIUM o      o     Q LIMESTONE

                °   .1 .'•''•'     0° COUFORM
   IRON o ,0

 •'• '"•'  ' (
  Q O O o

CHLORINE
                              „ » o O O.

                                 ASBESTOS
 But You maybe!    v      !

 ft seems unbelievable, but it's true! EPA and GAO. Federal studies have
 shown that most city water treatment plants do not effectively remove cheinK
^ cal and mineral contaminates from your water... the water you drink ...cook
[with... bathe with.,. launder with. The falnSoft Classic Apollo Phase Four
                                and treats every
 want you to drlnkonly clean, sparkling, healthy.mbneysavingwater.
                     684-9411
                     PALMBEACHCOUNW
 What water was meant to be.

-------
 Guess Who Has Started 'Worrying
          >          *  i           '           >               ''•••'••.' *—•* ••

 About Quality of California Water

;>•         By SCOT J. PALTROW    ,       son" for Evian, concedes in a telephone in-
1  s«n//Reporter'o/THKWxij.STiucBT JOURNAL ' tervlpw that  Beverly IlilJs water contains.
   BEVERLY CHILLS, Calif.-Ttys wealthy fewer, than TOO parts-per million,of trihalo-
 city has long prided itself on having'a high methanes, the level the Environmental Pro-
 level of practically everything. And now it's tection ASen«y "as established as>safe.
 about to be told that it may.also have a high    Dr.  KuiUn also  concedes  there hasn't
 level of carcinogens in its water.      '   been^any tafally conclusive proof of a. link
   A campaign is underway to convince res- 'between:-the  chemical and cancer. But he
 idents that their water contains a suspected W» p"k£*J* °f on^u^,town, whose
 carcinogen. But the campaign's booster isn't wate,r JJ,ad. levels of the  chemical •signifl-
,a group of environmentalists .or concerned <*"«* higher,than 100 parts per million ex-
 citizens It is Evian Waters of France Inc.- Perienced except onaljjr high levels of blad-
 vhich  is  about to introduce  its bottled Jer and Intestinal tract cancer. "Too little Is
  .Tench Alps spring water to Southern Call- known- "^   he  adds-
  irnia.                .    •  .         /City's  Advice
 ' Evian says it will hold "a news conference „ The Evian test/purportediy shows that
.today disclosing that "an  independent lab Beverly HiHs has a higher level of the chem-
 analysis"  has found in Beverly Hills water ^ «£» °&er communities to the area. But
 significant  'amounts  of trlhalomethanes. Mel  ™om< Beuver'y ffln» dl™*°* of ;P^»c |
 which it says 'are "suspected carcinogens." f^ce- W» &<* W and ™f cpmmunitieir;
 Evian says Dr. Samuel A. Kunin. a Los An- in &* region get. their water from the same
 geles-area urologist and.surgeon,  will pres- source,.and Its chemical content should be
 ent and interpret the laboratory  results.  ^^ H« savs•$* water meets all testing
                                      requirements, and he wouldn t advise resi-
 "Too Little Is Known'                dents  "to 'stop drinking It  with  their
    But  Evlan's press conference  invitation Scotch."
 doesn't say that Evian itself commissioned    But Dr. Kuntn says levels of the trihalo-
 and paid for  the  analysis, -conducted  by methanes can vary-and that Evian's water
 Montgomery Laboratories of Pasadena.   contains lower levels of the chemical than
    Dr.  Kunln, a paid "medical spokesper-  several competing bottled waters.

-------
   (Eh=0.875)


         1.2
 O
 o
 lO
 CM
    H
    Q.
    0)
    10
    O
 UJ
 II
 X
ro
cvi
 UJ
 I!
         1.0
     0.8
     0.6
     0.4
     0.2
    LJ

(Eh=-0.356) 0
             WATER OXIDIZED  02
                             H,0
                     Pb(H20)
                     Pb
             WATER REDUCED H20
                             K
                      8

                      pH
                                     (Eh=0.64)
                                      (Eh«-0.59)
                                       10
       MODIFIED POURBAIX DIAGRAM FOR
       LEAD IN WATER

-------
200
       Solubility of Lead as a Function of pH
    and  Carbonate. (Adapted from AWWARF/EBI,
    1985)

-------
1000
            7
                      PH
       Solubility of Lead in the Presence of
     50mg/L (as CaC03) DIG. (Adapted  from
     Schock (AWWARF/EBIJ985).

-------
    ORDER OF ACTIVITY
    OF COMMON METALS
  MAGNESIUM
     ZINC
   ALUMINUM
  MILD STEEL
 WROUGHT IRON
  CAST IRON
     LEAD
     TIM           MORE
     TIN           ACTIVE
    BRASS
    COPPER
STAINLESS STEEL

-------
  2400
GO
   300
                   LEAD
COLD WATER-12/13/82
   5 ERNEST COURT
UNDER CONSTRUCTION
                  LEAD IN WATER
                        50>ig/L DRINKING
                        WATER STANDARD
          i   i   i   i   i    i      \
      0  10 20 30  40  50  60 70 80
                TIME-SECONDS

-------
200 H
LEAD
                        KITCHEN SINK
                    OFFICE  BUILDING-MELVILLE
                    PLUMBING-23 MONTHS OLD
                          f*-c.\jpi±f u.
           1234

             TIME (Minutes)

-------
0
        USEPA/SHWD LEAD SOLDER STUDY
             CADMIUM (pg/L)
             TEST SITE (32W18)
20
 40     60     80
Time-(Seconds)
100
120

-------
   LOCATION
                                         f
                                         (
                DRAW
                   NASSAU COUNTY HEALTH  DEPT. TEST
LEAD
                 PH
         HARDNESS
            mg/L
           % LEAD
           SOLDER
            AGE
LOCUST VALLEY
17000
7.0
41
60.3
PORT WASHINGTON
4400
6.8
49
61.7
MANORHAVEN
 3500
6.8
49
47.9
WOODBURY
2900
7.3
64
58.4
NORTH PORT WASH.
  930
7.0
49
50.3
NORTH HILL
  750
6.7
23
56.2
NORTH HILL
  530
6.7
23
60.0

-------
                      MINNESOTA
                   DEPT. OF  HEALTH
     46 COMMUNITIES
 LESS THAN 3 YEARS OLD
COMMUNITY

 Winona
 Randolph
 Mankato
 New Trier
 New Germany
 Prinsburg
 Red Wing
 Madelia
 Norwood
 Watertown
 Minnetonka Beach
 Barnum
 Waconia
 Young  America
 Spring  Lake Park
 Fridley
 Becker
first draw lead
   ug/L
   1500
   860
   790
   300
   270
   260
   200
    140
    130
    120
    110
     95
     80
     63
     61
     57
     54
                 25 COMMUNITIES
                NEW CONSTRUCTION
COMMUNITY

 Mankato
 Kinckley
 Rice
 Park Rapids
 Northfild
 Eveleth
 Winsted
 St Peter
 Brainerd
 Waconia
 Rushford
 Willmar
 Zumbrota
 Elk River
 New Ulm
 Backus
 Rochester
first draw lea
   ug/L
  33000
  15000
   8800
   4600
   4100
   3800
   2400
   1900
   1100
   1100
    640
    450
    400
    350
    310
    290
    150
          VS 50 ug/L DRINKING WATER STANDARD

-------
PERCENTAGE OF TEST  SITES WITH LEAD IN DRINKING WATER
     GREATER THAN 20 ug/L AT LOW pH (6.4 & LESS)
AGE OF
TEST SITE
(Years)
0-1
1 -2
2-3
3-4
4-5
6-7
9-10
15-16
20 a Older
FIRST
DRAW
100%
100%
86%
100%
86%
78%
71 %
57%
86%
10
SEC
100%
, 71 %
86%
86%
57%
44%
29%
14%
27%
20
SEC
100%
86%
57%
100%
29%
33%
14%
14%
29%
30
SEC
100%
57%
57%
71%
43%
33%
14%
14%
0%
45
SEC
100%
57%
43%
71 %
43%
11 %
14%
14%
14%
60
SEC
86%
29%
43%
71%
43%
11 %
14%
14%
0%
90
SEC
86%
43%
43%
29%
14%
11%
0%
14%
14%
120
SEC
88%
14%
29%
29%
0%
0%
0%
14%
0%

-------
PERCENTAGE OF TEST  SITES WITH LEAD IN DRINKING WATER
     GREATER THAN 20 ug/L AT MEDIUM pH (7.0-7.4)
AGE OF
TEST SITE
(Years)
0-1
1 -2
2-3
3-4
4-5
6-7
9-10
15-16
20 a Older
FIRST
DRAW
100%
80%
40%
5O%
30%
10%
20%
40%
20 %
10
SEC
90%
60%
20%
20%
10%
0%
0%
20%
0%
20
SEC
90%
30%
10%
20%
10%
0%
0%
20%
0%
30
SEC
60%
10%
10%
30%
0%
0%
0%
10%
0%
45
SEC
30%
20%
10%
20%
10%
0%
0%
0%
10%
60
SEC
20%
0%
0%
30%
0%
0%
0%
0%
0%
90
SEC
10%
10%
0%
30%
0%
0%
0%
0%
0%
120
SEC
10%
0%
0%
20%
0%
£>%
0%
0%
0%

-------
PERCENTAGE OF TEST SITES WITH  LEAD IN DRINKING WATER
     GREATER THAN 20 ug/L AT HIGH pH (8.0 a GREATER)
AGE OF
TEST SITE
(Years)
0-1
1 -2
2-3
3-4
4-5
6-7
9-10
15-16
20 a Older
FIRST
DRAW
100%
67%
30%
25%
30%
20%
10%
33%
20%
10
SEC
100%
22%
10%
0%
10%
0%
0%
22%
0%
20
SEC
60%
11%
10%
0%
0%
0%
10%
11%
0%
30
SEC
10%
11%
0%
0%
0%
0%
0%
11%
0%
45
SEC
20%
11%
0%
0%
0%
0%
0%
0%
0%
60
SEC
10%
0%
0%
0%
0%
0%
10%
0%
0%
90
SEC
20%
11%
0%
0%
0%
0%
0%
0%
0%
120
SEC
0%
0%
0%
13%
0%
0%
10%
0%
0%

-------
RESULTS  FROM HOUSEHOLD LEAD STUDY FIRST DRAW SAMPLES






             pH             % of Homes Exceeding 50ug/L







             6.2                       58.7




           6.2-8.2                     26.6




           78.2                       24.4

-------
SOLDER TEMPERATURES
NORMAL COMPOSITION %
TIN LEAD ANTIMONY SILVER
95 — 5 —
95 — — 5
30 70 — —
35 65 — —
40 60 — —
45 55 — —
50 50 — —
60 40 — —
MELTING
SOLIDUS°F
452
430
361
361
361
361
361
361
RANGES
LIQUIDUS°F
464
473
491
471
460
441
421
374

-------
                                SOLDER
        Common Usage
    Type
                                            *April 1984
Percentage   Liquid Temp.  Range  Solid Temp. Wholesale Cost
  Tin/Lead
  50/50
421° F     60°F     361° F
$4.5071 b.
        Alternates
Tin/Antimony      95/5
 Tin/Silver
  95/5
              464° F
           12° F     452° F
$8.00/lb.
473° F     43° F    430° F     $20,00/lb.
                        ASTM Standards - Part 8-B32
  *estimated  by plumbers at less than 1 pound SOLDER required per residence

-------
                US EPA/SEATTLE  STUDY

                Tin Antimony Solder

          UNIVERSITY OF WASHINGTON BUILDINGS
   BUILDING
  YEAR
OCCUPIED
 SAMPLE
ANTIMONY
 OCEANOGRAPHY
   TEACHING
  1969
MECH. ROOM
 STANDING
  <0.6
  <2.0
 SOUTH CAMPUS
    CENTER
  1969
MECH. ROOM
 STANDING
  <0.6
  <0.6
HEALTH SCIENCES
    WING T
  1973
MECH. ROOM
 STANDING
  <0.6
  <0.6
    CONDON
  1974
MECH. ROOM
 STANDING
  <0.6
  <0.6
  WATER FRONT
   ACTIVITIES
  1977
MECH. ROOM
 STANDING
  <0.6
  <0.6
  HALL HEALTH
    CENTER
  1978
MECH. ROOM
 STANDING
  <0.6
  <0.6
     LOEW
   1979
MECH. ROOM
 STANDING
  <0.6
  <0.6
  ENGINEERING
    LIBRARY
  1969
MECH. ROOM
 STANDING
  <0.6
  <0.6

-------
      COPPER  LEACHING IN FOUR  PIPE  LOOPS
 pH                   SOLDER                 COPPER
                                            (mg/L)
5.5                TIN/SILVER                 3.80
5.5                TIN/ANTIMONY               4.28
5.5                TIN/COPPER                 4.50
5.5                TIN/LEAD                   4.28

-------
• AGE OF THE PLUMBING
 CORROSIVE WATERS
 GALVANIC CORROSION
 FLUX USED

-------
         CORROSION  CONTROL
pH ADJUTMENT  ALONE
ADJUST ALKALINITY AND pH
STABILIZE WATER (CALCIUM CARBONATE  BALANCE)
ADD CORROSION INHIBITORS

-------
 SEATTLE CORROSION  CONTROL PROGRAM
     a SERVICE LIFE ECONOMICS
            oLOSS $7.8 MILLION/YEAR

     o AESTHETICS
            o RED a BLUE STAINS
            o METALLIC TASTE COMPLAINTS

     a HEALTH
            oLEAD ABOVE MCL


(Courchene, J.E. &  Hoyt, B.R "Benefits of Corrosion Treatment
in Seattle"  CA/NV AWWA Seminar, 1985)

-------
 SEATTLE CORROSION CONTROL PROGRAM
     INSTITUTED MATERIAL SELECTION PROGRAM

      n PL AST 1C PIPE

            o ALLOWED PB AND CPVC IN BOTH
             HOT AND COLD

            o REQUIRED DIELECTRIC  INSULATORS

            o ENCOURAGED TYPE K AND L COPPER
             PIPE

            o DISCOURAGED GALVANIZED  PIPE

            o < 0.2% LEAD SN SOLDERS
(Courchene, J.E. 8  Hoyt, B.P "Benefits of Corrosion Treatment
in Seattle"  CA/NV AWWA Seminar, 1985)

-------
 SEATTLE CORROSION CONTROL PROGRAM


          WATER TREATMENT PROGRAM

            TOLT
                  2mg/L CaO

                  9mg/L Na2C03

            CEDER
                  2 mg/L Ca 0
(Courchene, J.E. 8 Hoyt, B.R "Benefits of Corrosion Treatment
 in Seattle" CA/NV  AWWA Seminar, 1985)

-------
      SEATTLE  CORROSION CONTROL PROGRAM

           RWQM* OVERNIGHT STANDING SAMPLES,
               ALL CEDAR AND TOLT GROUPS
Parameter

Pb,pg/L
Cd,jjg/L
Cu, mg/L
Fe, mg/L
Zn, mg/L
Average Concentration
Before

10.6
0.76
0.28
1.21
0.60
After

4.0
0.41
0.09
0.99
0.33
^Reduction

63
46
68
18
45
%>MCL
Before

5
0
1 1
75
0
After

0.4
0
0
67
0
# Residential Water Quality Monitoring-300 residences
     (Courchene, J.E. B Hoyt, B.R "Benefits of Corrosion Treatment
     in Seattle"  CA/NV  AWWA Seminar, 1985)

-------
     SEATTLE CORROSION CONTROL PROGRAM
                     LEAD LEVELS
                  % BEFORE
                         % AFTER
SOURCE
>20
>50'
>20'
>50:
 TOLT
 12.5
 5.9
 3.9
0.6
CEDAR
9.8
 4.3
            0
 * Jjg/L"
     (Courchene, J.E. 8 Hoyt, B.R "Benefits of Corrosion Treatment
     in  Seattle"  CA/NV  AWWA Seminar, 1985)

-------
SEATTLE CORROSION CONTROL PROGRAM

   WATER QUALITY CHANGES SINCE TREATMENT
Water Characteristic

pH
Tot. Alkalinity, mg/L CaC03
Carbon Dioxide, mg/L
Hardness, mg/L CaC03
Sodium, mg/L
Tot. Diss. Solids, mg/L
Spec. Conductance, pmhos
Larson's Ratio
Cedar
Before

7.2
16
2.7
21
1.8
46
55
0.65
After

8.2
19
0
24
1.8
48
59
0.55
Tolt
Before

6.0
2.5
6.0
8
1.0
25
27
3.9
After

8.2
13.5
0
11
4.8
36
42
0.8
(Courchene, J.E. 8t Hoyt, B.R "Benefits of Corrosion Treatment
         TA/MW  AWWA
n

-------
    INLET
   OUTLET-
             I	u	u.	
r









t_
u
il
u
u
II
li
n
U
II
II
!S!
l^f M
T ~i
u i
u i
11 i
11
ti i
n i
u l
!! i
IIOUT,
II ff^ 1
J
                      PLAN VIEW
INLET/OUTLET ELEVATION
SECTION
                 LIMESTONE CONTACTOR

-------