PRE-PUBLICATION COPY
     NATIONAL OR6ANICS RECONNAISSANCE SURVEY FOR
     HALOGENATED ORGANICS IN DRINKING WATER
           WATER SUPPLY RESEARCH LABORATORY
METHODS DEVELOPMENT AND QUALITY ASSURANCE LABORATORY
         NATIONAL ENVIRONMENTAL RESEARCH CENTER
           OFFICE OF RESEARCH AND DEVELOPMENT
         U,S, ENVIRONMENTAL PROTECTION AGENCY
                 CINCINNATI, OHIO
                    APRIL 1975

-------
                                                          PRE-PUBLICATION COPY
NATIONAL ORGANICS RECONNAISSANCE SURVEY FOR HALOGENATED ORGANICS IN DRINKING WATER
                                          by

                                    James M. Symons
                                    Thomas A. Beliar
                                    J. Keith Carswe11
                                    Jack DeMarco
                                    Kenneth L. Kropp
                                    Gordon G. Robeck
                                    Dennis R. Seeger
                                    Clois J. Slocum
                                    Bradford L. Smith
                                    Alan A. Stevens
                           Water Supply Research Laboratory
                Methods Development and Quality Assurance Laboratory
                         National Environmental Research Center
                           Office of Research and Development
                          U.S. Environmental Protection Agency

                                   Cincinnati, Ohio

                                     April 1975

-------
NATIONAL ORGANICS RECONNAISSANCE SURVEY FOR HALOGENATED ORG.ANICS IN DRINKING WATER*









Introduction



     On Friday, November 8, 1974, Russell E. Train, Administrator of the U.S.




Environmental Protection Agency announced that he was ordering an immediate




nationwide survey to determine the concentration and potential effects of




certain organic chemicals in drinking water.  On that date Train said, "What




we learn from this National Reconnaissance Survey will tell us how widespread




and serious the situation is that we found in the study of the New Orleans




drinking water supply."




     On December 16, 1974, President Ford signed into law Public Law 93-523,




"The Safe Drinking Water Act."  Section 1442(a)(9) of this Act states "The




Administrator shall conduct a comprehensive study of public water supplies and




drinking water sources to determine the nature, extent, sources of, and means




of control of contamination by chemicals or other substances suspected of being




carcinogenic.  Not later than 6 months after the enactment of this title, he




shall transmit to the Congress the initial results of such study, together with




such recommendations for further review and corrective action as he deems




appropriate."




     Finally, on December 18, 1974, Administrator Train named the 80 cities




to be included in the National Organics Reconnaissance Survey (NORS).
*Submitted to the Journal American Water Works Association for publication.

-------
                                       - 2 -
Objectives



     NORS had three major objectives.  One, was to determine the extent of


                                                      *+@
the presence of the four trihalomethanes -- chloroform   (trichloromethane),

                    *@                      *g           *@
bromodichloromethane  , dibromochloromethane  , bromoform   (tribromomethane),



and 1,2-dichloroethane , and carbon tetrachloride   in finished water.   A



second objective was to determine what effect raw water source and



water treatment practices have on the formation of these compounds.  The



third was to characterize, as completely as possible using existing



analytic technology, the organic content of finished drinking water produced



from raw water sources representing the major categories in use in the United



States today.  This paper discusses the results of NORS that are related


                                                                         123
to the first two objectives.  Future papers will discuss objective three  ' '



Selection of Cities



     For the study of the formation of chlorination by-products, 80 water



supplies (Table 1) were chosen to participate in the NORS in consultation with



State water supply officials.  These 80 supplies were geographically  distributed,



some in each of the USEPA's 10 Regions, see Figure 1.  The supplies were chosen



to represent as wide a variety of raw water sources and treatment techniques



as possible.
* - Selected as a possible chlorination by-product,

+ - Selected because of suspected effect on health.

@ - Selected because of presence in previously sampled finished waters.

-------
         - 3 -
d
                      >-

                      u
o
<

z

o


u
LU

o
ex
                      O



                      I!
                       • CD
                      to y
                      O

-------
                                        - 4 -
                                       TABLE 1

                               Water Utilities Studied
Region I

1.** Lawrence Water Works
     Lawrence, Massachusetts
     Merrimack River*
2.   Waterbury Bureau of Water
     Waterbury, Connecticut
     Wigwam and Morris Reservoirs
     Morris Treatment Station

3.   Metropolitan District Commission
     Boston, Massachusetts
     Quabbin and Wachusett Reservoirs
     Norumbego Treatment Station

4.   Newport Department of Water
     Newport, Rhode Island
     Reservoirs
     South Pond Reservoir Treatment
       Plant #1

Region II

5.   Department of Water Resources
     New York, New York
     Croton Reservoir

6.   Puerto Rico Aqueduct and Sewer
       Authority
     San Juan, Puerto Rico
     Lake Carraizo
     Sergio Cuevas Water Treatment Plant

7.   Passaic Valley Water Commission
     Little Falls, New Jersey
     Passaic River

8.   Toms River Water Company
     Toms River, New Jersey
     Ground
     Well #20

9.   Buffalo Water Department
     Buffalo, New York
     Lake Erie
10.  Village of Rhinebeck Water Dept.
     Rhinebeck, New York
     Hudson River

Region III

11.  Philadelphia Water Department
     Philadelphia, Pennsylvania
     Delaware River
     Torresdale Plant

12.  Wilmington Suburban Water Corp.
     C1aymont, Delaware
     Red Clay and White Clay Creek
     Stanton Plant

13.  Artesian Water Company
     Newark, Delaware
     Ground
     Llangollen Well Field Plant

14.  Washington Aqueduct
     Washington, D.C.
     Potomac River
     Delacarlia Plant

15.  Baltimore City - Bureau of
       Water Supply
     Baltimore, Maryland
     Loch Raven Reservoir
     Montbello Plant #1

16.  Western Pennsylvania Water Company
     Pittsburgh, Pennsylvania
     Monongahela River
     Hays Mine Plant

17.  Strasburg Borough Water System
     Strasburg, Pennsylvania
     Ground

18.  Fairfax County Water Authority
     Annandale, Virginia
     Occoquan River Impoundment
     New Lorton Plant
*The name of the utility is listed first, followed by the city name, the name
of the raw water source, and the name of the treatment plant sampled, if the
utility has more than one treatment plant.
**The same location numbers will be used throughout the paper.

-------
                                       - 5 -
19.  Virginia American Water Co. -
       Hopewell District
     Hopewell, Virginia
     Appamatox River

20.  Huntington Water Corp.
     Huntington, West Virginia
     Ohio River

21.  Wheeling Water Department
     Wheeling, West Virginia
     Ohio River

Revion IV

22.  Miami-Dade Water and Sewer
       Authority
     Miami, Florida
     Ground
     Preston Plant

23.  Jacksonville Dept. of Public
       Works
     Jacksonville, Florida
     Ground
     Highlands Pumping Station

24.  Atlanta Waterworks
     Atlanta, Georgia
     Chattahoochee River
     Chattahoochee Plant

25.  Owensboro Municipal Utilities
     Owensboro, Kentucky
     Ground

26.  Greenville Water Department
     Greenville, Mississippi
     Ground
     Water Plant Well #2

27.  Tennessee American Water Company
     Chattanooga, Tennessee
     Tennessee River

28.  Memphis Light, Gas and Water Div.
     Memphis, Tennessee
     Ground
     Malloy Plant

29.  Metropolitan Water and Sewerage Dept.
     Nashville, Tennessee
     Cumberland River
     Lawrence Plant
*Resampled after GAC changed.
30.,  Commissioner of Public. Works
      Charleston, South Carolina
      Edisto River
      Stoney Plant

Region V

31.   Cincinnati Water Works
      Cincinnati, Ohio
      Ohio River

32.   Chicago Dept. of Water and Sewers
      Chicago, Illinois
      Lake Michigan
      South District Water Filtration Plant
33,   Clinton Public Water Supply
      Clinton, Illinois
      Ground

34.   Indianapolis Water Company
      Indianapolis, Indiana
      White River and wells
      White River Plant

35.   Whiting Water Department
      Whiting, Indiana
      Lake Michigan

36.   Detroit Metro Water Department
      Detroit, Michigan
      Detroit River Intake at head of
        Belle Isle
      Waterworks Park Plant

37a.  Mt. Clemens Water Purification
      Mt. Clemens, Michigan
      Lake St. Clair

37b.* Mt. Clemens Water Purification
      Mt. Clemens, Michigan
      Lake St. Clair

38.   St. Paul Water Department
      St. Paul, Minnesota
      Mississippi River

39.   Cleveland Division of Water
      Cleveland,  Ohio
      Lake Erie
      Division Filtration Plant.
40.   City of Columbus
      Columbus, Ohio
      Scioto River
      Dublin Road Plant

-------
                                       - 6 -
41.  Dayton Water Works
     Dayton, Ohio
     Ground
     Ottawa Plant

42.  Indian  Hill Water Supply
     Cincinnati, Ohio
     Ground

43.  Piqua Water Supply
     Piqua, Ohio
     Swift Run Lake

44.  Mahoning Valley Sanitary District
     Youngstown, Ohio
     Meander Creek Reservoir

45.  Milwaukee Water Works
     Milwaukee, Wisconsin
     Lake Michigan
     Howard Avenue Purification Plant

46.  Oshkosh Water Utility
     Oshkosh,  Wisconsin
     Lake Winnebago

Region VI

47.  Terrebonne Parish Waterworks
       District #1
     Houma, Louisiana
     Bayoulafourche

48.  Camden Municipal Water Works
     Camden, Arkansas
     Ouachita River

49.  Town of Logansport Water System
     Logansport, Louisiana
     Sabine River

50.  City of Albuquerque
     Albuquerque, New Mexico
     Ground

51.  Oklahoma City Water Dept.
     Oklahoma City, Oklahoma
     Lake Hefner
     Hefner Plant

52.  Brownsville Public Utility Board
     Brownsville, Texas
     Rio Grande River
     Plant #2
*
 Resampled
53.   Dallas Water Utilities
      Dallas, Texas
      Elm Fork, Trinity River
      Bachman Plant

54.   San Antonio City Water Board
      San Antonio, Texas
      Ground

Region VII

55a.  Ottumwa Water Works
      Ottumwa, Iowa
      Des Moines River

55b.* Ottumwa Water Works
      Ottumwa, Iowa
      Des Moines River

56.   Clarinda Iowa Water Works
      Clarinda, Iowa
      Nodaway River

57.   Davenport Water Company
      Davenport, Iowa
      Mississippi River

58.   Topeka Public Water iupply
      Topeka, Kansas
      Kansas River
      South Plant
59.   Missouri Utility Company
      Cape Girardeau, Missouri
      Mississippi River

60.   Kansas City Missouri Water Dept,
      Kansas City, Missouri
      Missouri River

61.   St. Louis County Water Company
      St. Louis, Missouri
      Missouri River
      Central Plant

62.   Lincoln Municipal Water Supply
      Lincoln, Nebraska
      Ground

Region VIII

63.   City Water Department
      Grand Forks, North Dakota
      Red Lake

-------
                                       - 7 -
64.  Denver Water Board
     Denver, Colorado
     Marston Lake
     Marston Plant

65.  Pueblo Board of Waterworks
     Pueblo, Colorado
     Arkansas River
     Gardner Plant

66.  Huron Water Department
     Huron, South Dakota
     James River

67.  Salt Lake Water Department
     Salt Lake, Utah
     Mountain Dell Reservoir

Region IX
68.  City of Tucson Water and Sewers
       Dept.
     Tucson, Arizona
     Ground
     Plant #1

69.  City of Phoenix Water and Sewers
       Department
     Phoenix, Arizona
     Salt and Verde Rivers
     Verde Plant

70.  Department of Supply and Purification
     Coalinga, California
     California Aqueduct
74.   San Diego Water Utilities Dept.
     San Diego, California
     Colorado River Aqueduct
     Miramar Plant

75.   San Francisco Water Department
     San Francisco, California
     San Andreas Reservoir
     San Andreas Treatment Plant

 Region X

76.   Seattle Water Department
     Seattle, Washington
     Cedar River Impoundment
     Cedar River System

77.   Douglas Water System
     Douglas, Alaska
     Douglas Reservoir

78.   Idaho Falls Water Dept.
     Idaho Falls, Idaho
     Ground

79.   City of Corvallis Utilities Div.
     Corvallis, Oregon
     Willamette River
     Taylor Plant

80.   Ilwaco Municipal Water Dept.
     Ilwaco, Washington
     Black Lake
71.  Contra Costa County Water Department
     Concord, California
     Contra Costa Canal and San Joaquin River
     Bollman Plant

72.  City of Dos Palos Water Dept.
     Dos Palos, California
     Delta-Mendota Canal

73.  Los Angeles  Department of Water and
       Power
     Los Angeles, California
     Van Norman Reservoir

-------
                                     - 8 -
Procedure
     Engineering evaluation of treatment facilities.   At each of the 80 sites
chosen for study, engineers from the USEPA Regional Office visited the water
treatment plant and evaluated the facilities. They collected basic information
on the raw water source and treatment facilities.   In addition to this
information, these engineers also determined the dosage of various water
treatment chemicals used and their points of application.
     Sampling for trihalomethanes, carbon tetrachloride, 1,2-dichloroethane,
Because the six compounds chosen for study were known to be volatile, a sampling
procedure was chosen that would provide for minimum loss of the six compounds
from the water to the atmosphere while the sample was in shipment or awaiting
analysis by the purging technique (see below).
     The containers chosen were glass 50-ml "Hypo-Vials"* sealed with Teflon-
faced "Tuf-Bond" discs, both available from Pierce Chemical Co., Rockford, 111.
Before use, the glass vials were capped with aluminum foil and muffled at
400ฐC for at least 1 hour to destroy or remove any organic matter interfering
with analysis.  With aluminum foil still in place, the bottles were packed
(along with sufficient discs and aluminum seals to secure the discs in place,
labels, and re-usable ice packs) in an insulated container and shipped to the
appropriate regional office for sampling.  Sufficient materials were provided
for taking three raw- and three finished- water samples.
     In the field, the vials were filled, bubble-free, to overflowing so
that a convex meniscus formed at the top.  The excess water was displaced
as the disc was carefully placed, Teflon side down, on the opening of the vial.
The aluminum seal was then placed over the disc and the neck of the vial
and crimped into place.  A sample taken and  sealed in this manner was
*Mention of commerical products does not constitute  endorsement  by USEPA.

-------
                                     - 9 -
completlyheadspace-free at the time of sampling.  Usually a small bubble would


form during shipping and storage, however.


     The samples collected from the 80 locations from late January to end of


April 1975, werelabeled appropriately, repacked with the frozen ice packs


in the original insulated container, and returned via air mail to the EPA's


Water Supply Research Laboratory in Cincinnati.  Upon receipt at the


laboratory, the samples were refrigerated until analyzed.


     Analytic methods for chloroform, bromodichloromethane, dibromochloromethane,


bromoform, carbon tetrachloride, and 1,2-dichloroethane.  The sample concentration


procedure chosen for the initial step of identification and measurement of the

                                                                                4
six volatile halogenated organics was essentially that of Bellar and Lichtenberg .


In this procedure, the sample is purged with an inert gas that is passed, in


series, through an adsorbent material that traps and concentrates the organic


materials of interest.  The organics are then desorbed from the trapping


material by heating under a gas flow and transferred, thusly, to the first


few millimeters of a cold gas chromatography (GC) column.  Separation


(chromatography) is then carried out with temperature programming.


     During this survey, only single column GC was routinely performed, mostly


because of the shortness of time for completion of the NORS.  A high level of


confidence that proper identifications were made was attained by use of the


Hall Electrolytic Conductivity Detector operated in the specific halogen mode.


Further assurance of proper identifications was given by supplementary analysis


of nine raw and nine finished duplicate water pairs (from selected locations)


on a second column using a microcoulometric detector operated in the oxidative


halogen mode.  Finally, gas chromatographic-mass spectrometric (GC/MS)


analysis confirmed analysis of 15 of the finished water samples.

-------
                                    - 10 -
     The glass purging device and stainless-steel traps used in the analyses




were fabricated exactly according to Bellar and Lichtenberg .   The adsorbent




material used in the trap was Tenax-GC, 60/80 mesh (Applied Science, State




College, Pa. or Alltech Associates, Arlington Heights, 111.).




     The chromatograph used for analysis was a Varian Model 2100 with one inlet




modified to the general configuration of Bellar and Lichtenberg's desorber




Number 1.  The column used for separating the six compounds was 12-ft x 2-mm




I.D. glass, paJcked with Tenax GC, 60/80 mesh.  The column effluent was connected




via a stainless-steel transfer line to a Tracer Model 310 Hall Electrolytic




Conductivity Detector (Tracer, Inc., Austin, Texas)to detect and measure the




compounds.  This detector was chosen as the most suitable for the immediate




needs of the survey.




                Blank water and water used for dilution of standards was




prepared by purging distilled water with helium until no interfering peaks




could be detected by use of the complete analytical procedure.  Stock




standards were prepared, with dilutions, in 95% ethanol of the test compounds.




The appropriate final aqueous dilution was made by injecting 1 to 10 yl




of an appropriate stock standard directly through the valve on the 5-ml sampling




syringe  (see below) into a blank water sample contained therein.




     The sealed water sample, as received from the field, was heated to 25 C




in a water bath.  Just before the actual analysis, the entire disc-seal




combination cap was removed with a "Dekapitator" (Pierce Chemical Co.).




Duplicate aliquots from the sample were taken as follows:  A glass 5-ml Luer-Lok




syringe  (plunger removed) was fitted at the tip with a closed Luer-Lok one-way




brass stopcock.  The water sample was poured into the back of the barrel

-------
                                     - 11 -
of the syringe until the barrel was completely full.  The plunger was then


quickly inserted into the barrel in such a way as to eliminate air bubbles.


The valve was opened momentarily.  The plunger was depressed to the 5-ml mark to


expel excess sample, whereupon the valve was again closed.   Only one of these


aliquots was routinely analyzed; the duplicate was simply stored in this


configuration until the success of the first analysis was assured.


     The syringe assembly containing the aliquot to be analyzed was connected


to the Luer-Lok needle that was inserted into the sample inlet of the purging


device (the needle was never withdrawn from the septum).  At ther. time of


analysis, the valve was opened and the sample was expelled from the syringe


b> depressing the plunger.  After this, the valve was closed until purging


was complete.  After purging, the water (to be discarded) was removed by


reversing the above procedure.


     The technique of purging the sample and desorbing the trap contents onto

                                                                              4
the GC column were carried out exactly as described by Bellar and Lichtenberg.


Purging was for 11 minutes with a helium gas flow of 20 ml per minute.


Desorption was for 3 minutes at 180ฐC with a flow of helium through the trap


onto the GC column of 20 ml per minute (in addition to the carrier gas flow).


At this time, the GC column was at room temperature.


     To separate the compounds, the column was first quickly heated to 95ฐC,


followed by a 15-minute hold, and then programmed at 2ฐC per minute to a final


temperature of 180 C with a helium carrier flow of 20 ml per minute.  Conditions


for operation of the detector were those recommended by the manufacturer for


optimum performance in the halogen mode.


     Compounds were identified according to retention time (measured from


beginning of the hold at 95ฐC) and quantified by comparing peak heights with those


of standards prepared at similar concentrations.

-------
                                     - 12 -
     Retention data and the range of minimum quantifiable concentrations (MQC)

encountered for the six compounds during the survey are summarized in Table 2.



                                 TABLE 2


                   Chromatographic Retention and Sensitivity Data

                                             Minimum Quantifiable
                            Typical          Concentration (MQO)*, yg/1
Compound	Retention Time (min.)  Range Observed During Survey
Chloroform
1 , 2-Dichloroethane
Carbon Tetra chloride
Bromodi ch 1 oromethane
Dibromochloromethane
Bromoform
20.3
25.8
27. 7+
31.8
41.2
49.7
0.1
0.2
1.
0.2
0.4
1.
- 0.2
- 0.4
- 2.
- 0.8
- 2.
- 4.
*2% scale deflection.
+ Broad peak not completely resolved from 1,2-dichloroethane.


These retention times were typical; they varied slightly with aging of the

columns and significantly with installation  of a replacement column.  The

MQC was not constant throughout the study because of various changes in normal

operating parameters. No attempt was made to standardize the MQC; operational

parameters were simply adjusted to the optimum for any given day.

     Confirmation analysis.  As noted above, to add confidence to the routine

analysis for the six chosen volatile halogen containing organics, replicate

samples from selected locations were subjected to reanalysis by the Methods

Development and Quality Assurance Research Laboratory for quantitation on a

second GC-Detector system and for qualitative analysis with a GC/MS

system.  Table 3 shows the sampling cities for these confirmation samples.

-------
                                      - 13 -




                                      Table 3
       Cities Whose Samples Received Quantitative and Qualitative Confirmation
 Quantitative Confirmation




 2.  Waterbury, Connecticut




 7.  Little Falls, New Jersey




16.  South Pittsburgh, Pennsylvania




30.  Charleston, South Carolina




51.  Oklahoma City, Oklahoma




60.  Kansas City, Missouri




65.  Pueblo, Colorado




71.  Concord, California




79.  Corvallis, Oregon
Qualitative Confirmation




11.  Philadelphia, Pennsylvania




22.  Miami, Florida




30.  Charleston, South Carolina




31.  Cincinnati, Ohio




41,  Dayton, Ohio




51.  Oklahoma City, Oklahoma




55.  Ottumwa, Iowa




58.  Topeka, Kansas




60.  Kansas City, Missouri




71.  Concord, California




72.  Dos Palos, California




76,  Seattle, Washington




79.  Corvallis, Oregon
       The quantitative analysis was similar to that described by Bellar and




  Lichtenberg.  The following details describe the specific procedure.   All  samples




  were stored at 4ฐC until just before analysis.   Five ml of each sample was




  purged for 11 minutes with nitrogen flowing at  20 ml/minute.   The purging  device




  was maintained at 19 C.   The sample was introduced into the purging device at




  4 C.  Therefore,  as the  sample was purged it warmed up to 19ฐ C at an unknown




  rate.   The sample was concentrated using a trap packed with 18 cm of Davison




  silica gel, grade 15, 35-60 mesh.   Desorption took place for 4.0 minutes at



  200ฐC.

-------
                                    - 14 -
     An Infotronics Model 2400 gas chromatograph equipped with a Dohrmann




microcoulometric detector (halide specific mode, oxidative) was used to




perform the analyses.  A stainless steel column packed with Porasil-C coated




with Carbowax-400, 100/120 mesh, 6-feet long, 0.1-inch I.D. was used to




perform the separations.  Nitrogen flowing at 50 ml/minute was employed as the




carrier gas. The column was programmed over the following conditions:  Desorb




into column for 4 minutes at <30ฐC; heat column to 50ฐC and hold 1 minute;




and program column to 175 C at 8 /minute.




     Under these conditions the limit of detection for the materials of




interest were:  chloroform, 0.05 yg/1; bromodichloromethane, 0.1 yg/1;




dibromochloromethane, 0.1 yg/1;  bromoform, ^5 yg/1; 1,2-dichloroethane, 0.1 yg/1;




and carbon tetrachloride, 0.05 yg/1.  Methylene chloride was routinely detected;




the limit of detection was 0.05 yg/1.  Although other unknown halogenated  organics




were detected, their concentrations were below the limit of detection for GC/MS




identification.  By calculating relative retention times, the same unknown




organohalides were found to be present in many of the water supplies tested.




     The qualitative analyses were performed on samples treated as described  above




for quantitative confirmation with a Varian aerograph 1400 gas chromatograph



interfaced with a Finnigan 1015C quadrupole mass spectrometer controlled by




a System Industries 150 data acquisition system.  A glass column packed with




Porasil-C coated with Carbowax-400, 100/120 mesh, 6 feet long x 2 mm I.D.




was used to perform the separations.  Helium at 30 ml/minute was used as the




carrier gas.  The column was programmed under the following conditions;




Desorb into the column for 4 minutes at <30ฐC; hold at <30 C for 1 minute;




heat column to 100ฐC and hold for 3 minutes; and program to 200ฐC at 8 /minute.




The mass ispectrometer was operated in the following mode:

-------
                                     - 15 -
     Mass range scan    20-350
     Integration time   12
     Samples/AMU        1
     Total run          30 minutes

     Some of the qualitative analyses were made on a different spectrometer

operated under slightly different conditions ,  Because a larger sample

volume was purged in the latter case, these analyses had better sensitivity.

     Analytical methods for general organic parameters.    Nonvolatile total

organic carbon (NVTOC) vas determined on an instrument made by Phase

Separations Ltd., United Kingdom.  Samples are acidified with nitric acid,

purged with nitrogen gas for about 10 minutes to remove carbon dioxide, then

pumped into the instrument at a constant rate of 0.6 ml/minute for about 10

minutes.  During this time the nonvolatile oranic carbon is thermally oxidized

to carbon carioxide (CCL) at 920ฐC on copper oxide, then reduced to methane (CH.)

at 450ฐC on nickle in a hydrogen atmosphere.  The methane produced is measured

continuously with a flame ionization detector.

                                                                    6-'
     Ultraviolet absorption.  See the method of Dobbs, Wise and Dean  '.

     Fluorescence.  The rapid fluoremetric method (RFM), as described by

Sylvia , and a fluorescence emission scan (EmFS) were performed.  In this

latter determination, the excitation and emission slit widths are 12 nm and

16 nm, respectively, and the aqueous sample is excited at 310 nm, with the

fluorescence emission recorded between 370 nm and 580 nm.

     Quality control.    Accuracy. To test the accuracy of the method used by

the Water Supply Research Laboratory during the survey, the Methods Development

and Quality Assurance Research Laboratory prepared a pair of "unknown" standard

mixtures in the following manner:  Two different stock solutions, each

containing all of the compounds of interest, were prepared by injecting a

known volume of each material into a volumetric flask containing 90 ml of

methyl alcohol.  After all of the compounds were injected into the flask,

-------
                                    - 16 -
the mixture was diluted to volume (100.0 ml) and mixed by inverting.  Two-



hundred microliters of the stock solution was then dosed into 1.0 liter



of super-Q water (Millipore Filter Company) and mixed by inverting two



times.  One-half of the dosed water was then transferred into a 500-ml



separatory funnel.   Several,  60-ml vials were then filled with the mixture



and promptly sealed with Teflon septate.  The samples were stored at 4 C



until delivery to Water Supply Research Laboratory.  The blank, Sample D-4,



contained only super-Q water.  The calculated concentrations of the dosed



mixtures, D-2 and D-3, are listed with the analytical results in Table 4a.



     Precision.  To test variability of results during a typical day of analysis,



two series of 5 replicate samples were preapred as 10 discrete samples in the



same manner as standards were prepared throughout the survey. One series



was at low concentrations, the other at high.  All of the samples were



analyzed exactly as described above to determine the concentration of the



six halogenated organic compounds.  Spiked concentrations and relative standard



deviations   (o/X    ) are listed in Table 4b.
                avg.

-------
                                    - 17 -

                                   Table 4 a

                           Determination of Accuracy

                             Concentration (yg/1)
                               1,2-     Carbon   Bromo-    Dibrorao-
Sample
D-2 (True value)
D-2a (WSRL lab)
D-2b (WSRL lab)
D-2a (MDQARL lab)
D-2b (MDQARL lab)
D-3 (True value)
D-3a (WSRL lab)
D-3b (WSRL lab)
D-3a (MDQARL lab)
D-3b (MDQAHL lab)
D-4 (Blank - WSRL
D-4 (Blank - MDQA
Chloroform
74.6
63
65
* 60.8
76.3
59.6
46
46
* 53.6
58.6
lab) 0.2
RL lab) 0.1
Dichloro
ethane
10.1
9.
10
9.5
9.8
5.0
6.
5
4.8
3.6
1.
0.2
Tetra-
chloride
9.5
9.
8
7.9
5.9
6.4
5.
6
6.2
3.6
NF+
NF
dichloro-
methane
39.6
39
40
35.3
37.6
23.8
22
23
21.1
19.1
NF
NF
chloro-
methane
23.8
23
23
17
15.2
19.0
14
18
13.3
11.5
NF
NF
Bromo-
form
40.4
40
38
48.5
44.9
23.2
18
24
24.1
29.4
NF
NF
*Two weeks elapsed between the duplicate analyses at the Methods Development
  and Quality Assurance Research Laboratory.

+ - None found.

-------
                                     - 18 -






                                    Table 4b





                          Determination of Precision
Low Concentration Hi^h Concentration
Spiked
Cone.
Compound (pg/1)
Chloroform
1 , 2 - Di chl or oethane
Carbon tetrachloride
Bromodichloromethane
Dibromo chl oromethane
Bromoform
2
1
2
2
2
4
Rel.
a(%)
6
5
14
5
10
20
Spiked
Cone. Rel.
(ng/1) a(%)
18 7
* *
* *
20 7
30 13
30 12
*Not determined at high concentrations.








     In summary, these precision and accuracy data indicate these analyses were



satisfactory.

-------
                                    - 19 -
Results

     Source and treatment information.    The following percentages indicate

the different categories of sources studied in this investigation:

                       Source                %_

                       Ground                20

                       Lake or reservoir     33

                       River                 47

                       Mixed                  0

The percentages for the various types of treatment practiced by the utilities

in the Survey were:

                  Treatment                            _%_

                                                        100
Disinfection

  Chlorination


  Ozonation


Raw water chlorination

Raw water ozonation

Polyelectrolyte used

Powdered activated carbon used

Granular activated carbon

Softening

  Precipitative

  Zeolite

Taste and odor control practiced
                                                         99 (At some place in the
                                                          treatment system)

                                                          1 (The only treatment
                                                             practiced)

                                                         75

                                                          1

                                                         22

                                                         25

                                                         10

                                                         25

                                                         22

                                                          3

                                                         38

-------
                                     - 20 -





A study population of about 36 million was covered in the Survey.   At 60




locations practicing raw water chlorination the following percentages of




systems employed the indicated dosages:




                   Dose, mg/1           %
0-1
1-2
2-3
3-4
4-5
5-6
6-7
7-8
8-9
>10
Unknown
15
19
12
14
14
12
4
2
2
2
4
In 86% of these locations, the raw water    chlorination dose was between 1




and 6 mg/1.




     The percentages for chlorine residual (free, combined, and both) at all 80




locations were:




             Combined residual, mg/1
0-0.4
0.4-0.8
0.8-1.2
1.2-1.6
1.6-2.0
2.0-2.4
2.4-2.8
Unknown
60
20
6
4
4
1
4
1

-------
                                     - 21 -
                   Free Residual - mg/1         %
                        0-0.4                  41
                        0.4-0.8                19
                        0.8-1.2                 4
                        1.2-1.6                20
                        1.6-2.0                 4
                        2.0-2.4                 8
                        2.4-2.8                 3
                        Unknown                 1
                 Free and Unknown Combined
                 Residual	
                 Total less than 0.8 mg/1      16
In general, rather low residuals were present in the finished waters studied,
and at 16% of the locations, less than 0.8 mg/1 of free, plus combined chlorine
residual was recorded.
     Raw and finished water data.  From the data summarized in Table 5, the six
selected compounds measured in the raw water at the 80 locations were seen to
be nondetected (30 locations) or present in very low concentrations. One
location.   (#35) was receiving water pre-chlorinated by others and this water
did contain some chloroform, bromodichloromethane, and dibromochloromethane.
The water before chlorinationw as typical of other raw waters, however.  The
nonvolatile total organic carbon determination was made on each sample and these
data are listed in Table 5.  Some of these data may be in question because of
suspended solids rir.some of the raw water samples.  The ultra-violet absorption
and fluorescence data were considered unreliable because of the presence of
suspended solids in the samples, and were not reported.
     Table 5 also summarizes all of the data on finished water quality from
the 80 locations.  The ultraviolet absorption and fluorescence data for finished
waters are not presented at this time, but will be discussed later in this
paper.  The -..range of each measurement is also noted at the end of Table 5.

-------
- 22 -





























a
•P

0
X
in p
-H
O rH
rS 3
c3 cy
E-


p
O -d
-ป o
ซ J3
Cs - X! ,
..H P 00
rH Q CD 3.

r< 0 00
03 <4H 3.



O 1 i)
S 0 C
O h cd
!H O XI i-l
XI rH -P ^^
•rt XI 0> 00
a o B 3.



I
O
I o C
O rH CS
1 fj X ^

(H -H 
•H C

4-* &
^ i — i









C^ \O CNJ %y\ T~H O
CO --( CX] (M C\a CM






tu
Ctj 2; Cti CM Cti PH
S5 fe; v fe 2


CM
Ctj tU &J O fej H,
te; 2: t^ v s* ^


fez fe ^ feS





+ 0 ^0
fcj Ct, • P<4 ft,
So ft o ft z







00
^ fcj O fej *
V Oi fe; I-H ft O







Ci rH li| tO Ct4 Tt
V CT1 ft CTV ft




c
0
•H
Wl r-v
[/) C
•H 5
g  -H fi
?-i O c/3 f-l 0)
0 P g
S 3 P  Q (1)

S: o
rH
/— ^
r— 1
•M
S
rH
OH

' 4-*

O
S'


' O
f-f

^Qf-^ate;
servo ir '

f JJt'cd '

1 ^Jt^Q '
Q' ' C '

4^>CL
' ll
' 21 ' U5 '
1 v_^'
,
•^-









d in
N5 CM







fcj tlซ
ft z



fe S5


feg





0>
fed









**







fc, CM
ft CM







tn
(U
o


0
to

05
V
1

o

p
c

s
tt
CO
PH
Q

.
trt









d O
CM CM







fej tL>
ft Z



fe z


fej CM
ft






fe s








Ptj C7)
ft CM







Ci I?
V
•p
• c
xj a
•ง ^
'j.
fH ^

S g

W c3

*O' e.
fl c_3

queduct
s Water

ฃป
o H
p 00
r< r<
&4 '•"•'

.
^D









^0 OI
t*O r-(







C^l CM
V V


00 CM
C5 O
V V


fez






fcj CM
E5








fei \O
fe; rH







C) LD





c
0
•H
E/)
t/1


S
o
U
h


0
•H
CO
t/)
o
IVJ rH







t, M.




ซ ^


fez






fe iH









ft l-H







• o>

^

i
ti
cd
o.
0)
Q


0)
p
cd

inebeck

S

m
o


Oj
rH
rH
•H
^

.
O

-------
- 23 -

















a

ctf
Q
rH
0)
4->
3
O 13
4i O
ซ 43
Q to
< — > rt *i> *H

13 cij CJ
P Q a ซ
G -H
o x a
-— ' -H 0? PL,

in co to H
3 •ซป
<1> Cy rH
rH tO Hi
43 rH O rH
0) 
(3
li

(3 I
o a

a  O/
E 3.



E
o

o
rH
O I-H
I-H ^*~
43 0










'"" *

cซ
S
1 '
43 P
3 C
CO n)
oo O
C -P

rH 3
S '-'

CN
•rH






O CN 00 CM
O O ^ l-H







ct, tu (t, u<
ft Z ft Z



CN tO
It, O fe, O
ft V ft V




fJH rH fe, r^
ft v fe: Z



fe. fc.
ft rH ft CM







fe, fe,
ft 0 ft 00








eg
" "I  U
3- rH


O CS

Cd rH Cd
CD *t -H
rH 13
3 p rt ^^
CO C > 4->
CD I-H C
XI-H X ct)
p ex. tn I-H
•H CO,
U 0 G
rH 
rH rH





•>
eg O
Ci O







fe z




fe, 0,
ft Z




fe, B,
ft Z



fez







PCJ tU
-5; 2








i-H
PJ^I o
85 v




g
tn
X


ฃ_,

4-)
n)

43
OO
I
S
OO
43

E
CO

i-H






CV
•SH







fe




By
fe*




fe



fe







ci
V








e\]
^
V




X
4->
•H
O
43

3

!H

n)
X
p
0
rt

'rt
n,

00
l-H






r-~
CM







tu
z




B,
z




ง


VD
O
V







vO









^
<0












( 	 ^
rH
O-
orton
j

1









CXl CN
\H O







k~, tu
ft: 2




CM b
ft: ^




fe^



00
fed







fern









ฐฐ.
0 0




X
G
n!
&
O
u

rH

4->
CD
4J
ง0
• H
U rH
•H P
rH U)
IO .H
1ฐ
1 r-H
Cti rH
•H 0>
C 2
•H 0>
bO g,
fn O
•H X
>

rH






c^l o
C\0 rH







^Ji tO



t^Q ^~
Qj O
V V




fez-



fet.







fcป \o
jS ^









ro
*~i CN











B1
o
u
rH
IU
g
P
00
K
•H
P
X

O
CN

-------
- 24 -




















CD

CD
a
!H
CD
•P
cD
S

(3 'O
H^ CD
!i C
f-^ CD CB -rt

•V CD ,3
p Q ^ to
ง X 3"H
U 4-> Q O


in o] to H
O CX JH
rH (0 OJ
rCl .M S) 'H
o3 1) *ti P
ฃ-, 4-ป *--l DO
rt t3 CD
CO G
a: C -H
o •*ป
Z CD
s.s
JS
TJ
4i C
CO O
A o
TJ CD
&4 CO



































*
CD
O G
H 0
CD rl
rH CD
•H U
CD O "^
rH ซH CM
o c e
> cD
G 00
O rl
Z O
CD
•o
C 1 -H
O CD rl
43 rl 0 rH
rl P rH ^
CD CD X M
U H 0 3.
t
g
O CD
1 rH G
X CD rH
CN O ,C -•ซ.
-.H P OO
rH Q CD 3.


1
0
S 6 rH
rl O Ot
ca  ID d,

CD rt C ฐ ง
4J ^ cd CD CX
CD rH rH
^ 0 CX i-H tO
-n .H TI
oo CD G > G
C Q 0 G CD
.H IP O rH

"^ S 0 ^ 00
(D CD rl O -rl
| g &, r? ฃ
^ ,
rH CN tO
CN ซCN CN







CVJ 01
1-1 O








1*4 B,
fe: Z




PO
d p-1
v 2:






^ Z



fe: CN







feS







CXI
v to















c
cD
t/) i-H
JH
O 0

rl .*~!
0 O
P O
CD O
S 43
U
CD cD
P P
C P


^ ^

•zf
CN







CV O
T-1 CN








1*4 CU
E5 Z





1*4 B.
fe; Z






fe; to



fe! rH







I*, 0
fe! CN








f*. tO
S: rH







to
0
• H
P
• H
rH
•H

3
•— 1
CD
• H
CJ
• H
C
3
S

O
rl
O

to
c
CD
a
.
LO
CN







^0 T-) vJD
CO rt r-H O








Pt( [it 64 [i,
^ rS ^ 2




CN -^t
fJL| O IJL| O
fe; V r^; V






C^ rH fcj [i,
S5 V ^ 2



r-
rXj Ct( *
J3 to S; o







S; vO fe: Oi








to 05
. I-~ .0
O rH Q) tO




X

CD

g
P O
G U

€ rl

rl P

0
ง
M O

•H (/)
> t/1
C OJ
OJ C
0) C
rH 0>
U E-
*
vO t^*-
CN CN







C^l M
Q) O








tt4 U~
fe: Z





fez"






feg



fen







1*4
fe! CN







CXI
V O




ฃ>
-H
Q

r4
0
P
CD
S

T3
g

to
CD



P P

00 cD
•H rH
iJ CX

-------
- 25 -
























nt
4->
cd
Q
(U
4-"
*
ซ "0
4i ID
/-^ nt 01 ซH
4-> -W B,
t3 cd O
4-> O S t/>
C -H
o x a
CJ 4-> Cj ID

in cd to H
3 '^
ID O1 rl
I-H to nt
JO M O rH
cd o '^3
H 4-> r~i bO
cd C3 fl>
eg c
Di K .H
0 -fi
Z fl>
•i j
kq
•ซ
+ป e
to O
P. 0
"pi Q)































rH
cd
O C
t~ 0
rH Cd
•H U
4J i-H
cd o -^
0 e E
> cd
R bo
O to
Z 0
U
C 1 -H
O Cd rl
& to O I-H
to 4-1 rH \
ni (U J3 M
U H 0 3
l
O
rl

rH C
' .C at I-H
CM U ft --.
-> .H 4-ป bO
rH Q <0 3.

O
E E rH
rl O b

1

E O C4
grl Cd i-H
0 rC -^
•H f. CD 3
Q 0 E

1
go
1 0 C
O rH (3
0 0 4-> ~-
to -H ID M
oa *o 6 3.



B
0

S
0 rH
l-H \
CJ 3.








'v
T—
0
c

'c
p.


a
^
e CD
Z B

X
4-> 4->
-H C
r-H C
•H rH
4-> (X
O * — '
^^
eo
ซ

CO
o

fi


^
r-H

t>O rH O> LO IN. t^
C<] r- 1 T-H iH Cs. \Q





fcj
fe
fc^ (X PC, Ei,
oo CM fe; SE fe: 2:
V V




^ r^ ^"
Pt| O ft| O Cti U,
^: v fe v fe: 2:


fcj
fe
Cti Pu fej [iซ fei Pu
S Z S3 Z S3 Z



R^
fej Cij fej B,
E; rt Es t S3 Z



10
S3 rH Ej rH ES 0






Qj
C^" c\3
UQ ป .
Q> •** V rH V "*
t/1 fj
*-ป c
g), rt
> rH
$ ฐ"'
T3' * ^

งs 1
"I U to W

•ฐ Ji .B
S 'H
T) i
c S



^f
to
-^ CM
0ปJ
t-J Co
Co
r?g
5 o
4i
^c?(ง




Q 0) i/> to CM
CM l-H rH CM rH







fel ttj B, Pt, B,
fe: S3 Z S; Z





E; E; Z ki o




fefefe feง




fiq UH fej
co 65 2 S: tO




CO
. to
i-t V O ^ Oป







CQ
T-H Lfl
to . • d  O V rH



jj
C
E
•P
fH
+J cd

4-J fH HO
•H 4J Cd
ฃ <" *



LO VO
to to
Q> CO
1
J *
itj

Q) S
e, ?
•2 ^
0


Q> ^ EN, tS" O^ ^" C\l OO
CX]rH toปH IN.1^ C<]rH







fcj tl, fci tt, C^PU 64 PU
ง2: fe:^: fejrZ fe;Z




Tf ^3
ti^O Q>IX P^Un C^tLi
^v v.2 ^2 te: s?:




fe(tU fejtL, fej fe f*4.X,
S S ^ SI งrZ rS2




fej Et| Ctj Pt|
I^fN fe:(N r^CN r^^J"
V V



PCJ fej fej Cti
^;yD fe;tO E^r- fe;Cn







CXI
O^ r>
C^ rH • O ^t Ctj OO
^rH d^O Vrf fe; i— I



C C
0 0
•H *H
4J 4J r4 ' — *
cd Cd ID 4-1
00 4J C
•H -H X Cd rf
>H 'H >H CU

S 3 0 0 C
a, a, x o
4-> C -H
rH rH 3 O 4-1
cu n>  rH
4-1 -H -H
I/I VI Cd Q BH
c c *
d)O TJ C
E E I-H CO
m  4-1 4-1 rH Q
S S CO CJ ^
. .
cd J3 • >
r-- f-- oo 01
to to to to











Co tO
to CM







CtB,






feS




feg




feo'
V



feoo








r-H^)-
.to
OrH










, — \
c
1/1 cd
3 rH

3 -a
l-H Cd
0 0
cj oi

O -H
i-H
X r^
4-> 3
•H O
CJ '-'


O


-------
- 26 -

















CD
4-"

m
 O1 IH
rH CO CD
,Q r< ^ rH
CD CD -t* 3
E-. p ซS M
CO ซ CD
S H-i f!
to d
PCI S *H
0 -S
z  CD
C 60
O IH
Z 0
C 1 -rt
0 CD rl
,0 f-l O rH
IH P rH -V,
cD fl> ,C W
CJ E-i 0 3.
g
O O
' SSrH
CN 0 J3 -~-
ซ.H (-> M
•H CI  *^



0
O

O rH
rH "^
U =






l^t
1
r"
Ct
•r
|

ฃ
i]
x
•M 4-*
•H C
rH 0
•H rH
4-> O.
.D ^~-








Oif^- COO) Q)C-J C\,rH ^[^
^ O Q O CO ^" ^J* *O C\J i-H






f-C^ Ol Pt^tU Ptj PH rJ^U-. Ctj U-


(N tN
E^O I^Z fe; V ^rZ fe;V
V




^^t fefe S^Z lงง Sง fe



^^ ^^ ^.0 ^ ^ro
V







l^oo t^r^ fejTH ,^LO i^r--






Csa CX]
fcj Ci Pt, to fcj o d
t^OO VIO f^-H fejOO VOl
•M ' — *
O 4->
*ซH S
f-l TO
•4-> rH
VI (X
'H
Q C
- x .2
& ctj cD
& 4-> en O
3 ^ -g -gifl
tf> X CD O -H

"r( S "p. W ^
o ซ-> p. x mo.
S: ^— % cti 3 Q> CO
•p ^ C/l rH 4J 0)
IH (3 --1 cD 3
ci) CD I-H h CD S C
p rH rH Cy ^> CD
CD (X .H 4J O >
^ X cc) bo C)<
CD S C X
gs in -H 3-p
to 03 Co C Oj M
•P -M -H 13 O S CtJ
X -P TJ CT rC ซH S
nj O ฃ •!-< co -HO
Q N— / hH a- s SK

ป
rH (N tO ^ Lf)






tj-j ฃfj
Qj O
*O tO ^J1 fNJ t^ I/) Cvi LO • •
'yitO LQtO fOiH ^tO ^^
V V





r*X|ฃ^ ^ fe ^ P- fej tL, fJH [I-.


 W t/1
r^ rX X

o o
S ฃ rH
rH 
-------
- 27 -
























ca
cd
a
fH
CD
ca
3=
O T)
+i CD
•H
/ — ^ ca cjj ซH
t 4J -4i [L,
•a ca o
4-ป Q S yi
C -H
o x a
u *j 5 CD
'~' rH *• X*
i/j ca eo H
3 -ti
00- fn
rH CO ca
J3 f-l O rH
ca o M 3
H 4-> r-i bO
ca Q QJ
c/) C
Cฃ C -H
O -ri
Z CD
•rl ij

^

CO O
_f, O
&, I/)




































*
"ca
^j
0
t-
CD
•H
4->
ca
o
c
o


C3 i
o ca
•8 Z
a CD
u H
i
o

o
CN O
ซ-H
*-• Q

1
i
S
fH
CO


O 1
& o
0 fH
fH O
rQ H
s-s


1
o
fH
' O
O rH

O CJ
fH -H
ca -o
















































e
0
fH
U

0
'c
bo
f-l
0
CD
•H

O
rH

0



Q>
ca
rC
4J
CD


g
O


S
ca
.ฃ
4-1
i





i
(-<
CD
e



g
o
->
IH -M
rt ' -H
(X rH
0 -H  O ' -H
4-> -H --H
ClJ rH -H
x -M a, c
4-> C H cti
•H ni  <—i
U rH rH /-% 4^ CL,
D- rH CN rt
rt -H =tt 3: C
S r-* > OJ
O0 ui -P t/i S
^J C , c e nj rC
"— * 1> O <— ' rH Oj
XX rH DH flJ CQ
O ^ OQ ^^ Q ^


i— t CN tO
LO 4/> LO
, — \
OJ /-N
rH D
t/j rt
tn
LO
t--- LO
t-- \
f-n r-^
fN ^f
'**"' ^~^


LQ LO T-~1 fO Oj "^f" t-O O ^3 *^" Mt CN
C^ O ^Ji CN MI CN fO fO Co ^" ^Q CN






&4 tL, ft)U- fejtii fcjLL, Ct, tU fej






Ct) [In fej PH i<4 Eli fe| tU tt-( O fe( fc




fei t^tLi fJ^Cii Ct,eu t^tu 64


^0
Ctj ^fe Ptj PU fcj &jO fe|d







ฐ0
CTi
fcj • Ptj PH Ptj tU Ccj O,t Etj Ol CO
,5:0 ^ 5c 5:2 fe^rH ^oo vto








Cx] CO
CN . OO • ** \H
Ctj ป O • Ctj Qi QO • 00 * OO
!^;O VO l^rH V^t CiOO CbOO





T3

cti X
O tn rH
CQ X OH
rH X QJ
H O p .3
a) 3: rt co
+•* D-i
CS rH e ^
3: v, (n CD O I)

X f-i rH rt rt
4-ป O O S J-< S
•H S 3 
0 4-> P 0 3= -H C
•H aj ni rH ^ rt
C S 3: -P 3 *-i
O cti M CX D-
•4-* oJ flj T3 O
C S 3 C p^ rt X
< ง g -H C ^ P
3 3 rH  -M Oj > P, O
+-ป 4-> i-H OJ O W
CO O O U C3 H "— '
B 4
cd ,0 • • •
^j- to Ln \o r- oo '
l/J LO LO LO t/) LO











LO \ฃJ Mi 01
Ml tO fO rH






Piu tij ULi
^ CN fe; 2





So' !ฃ




feง feg



fecN fe X
H3 4-J
-H
•H U

3 
o ca
Ul t/1
I/I (5
•H ca
S X


Cl O
1/3 \O

-------
- 28 -




































*"^* rt
• 4-1
T) rt
4-> Q
c
0 X
CJ p
rH
LO rt
3
CD O"
rH
42 fH
rt o
H P
CO
CO
01
o
z






















































rt
•p
Q

^
CD
rt
S

O -O

13 43
•H
S, 13
Q) -H
•fi FTI
Q
^ in
.H
y CD
Qq a,
.cot?

(^
CO Ctf
?) T— f
•ซ* ?
t-J oO
ti <1>
G


o
Q) p;
.C 'H
ij
13
4i ฃ
CO O
.5 %
•fi H)
BH CO

























*
0)
p
0 G
E- 0
 rt
Pi bo
O rH
Z O
T)
C 1 -H
O rt rH
42 rH O rH
rH 4-> rH -^.
rt  -^
fH -H CD b<
03 T) E 3.




B

<2
S
O rH
rH \
45 M
U 3









S
0


<
!

t

Q)


2
X
p P
•H
rH
•H rH
P Cu
•3 ^_









Ml O Mi rt ซJ CM
(0 CM r-f rH Oj LO







fei pLi &j tU p,.. [T
fe r2 fe ป2 fe *Z




to rt
ซ [j^ B. [i^ B*
O O ES Z 85 Z






S3 "v E; *v S- ง







^ to B; rt fe fe






|2 fe^ ^rH









l-tj LO Ctj Ctj





X X
SrH
a
& ง•
o co
u h
rH IU
0 P P
4J rt C
• rt ^ a>
s • e
rH 4J
X rt rH
•P /-v & rt
C P -H P,
3 C O m
o rt -H Q

(X 3 rH

•H rH P
3 rt B rt
O rH rH ^
J 4J O
13 0 X
• S F3 p
p CJ -H -H
03 v-^ J CJ
rH CM tO
vฃ) ^ VO







LO LTJ
C\3 (Nl Ci O
OI-- co^o ..  .• OO
ซlrH ^rH v-|rH V~(O yy I-H iH







Ct. PL, Ctj PL, fcj tU Cti Lu fej PL, fcj [I*
r^ r^ fe -2 fe S S IZ fe ^Z fe ^




^1
fej FTi CJU| PM Pt| tU Q pi, Ct| UL| fcj PL|
r^ 2 ^ 2 "^ <2 V S "^ t*Z *^ I"S






fejtL, r-XlUl fc, CtjtU r-t^tO fej^
^2 -5:2 fe;00 1^ Z ^I-H ^v







ftj Ct( (N Etj CTi Ctj fe) Ctj t^"-
^tO !^v -~5^f l^CO ^C^J t^i-H





CO
fi^ O [<| rij I-H titj ^J" fej O Ct| LO
fe;rH fe;(N l5;i-H S;nH fe;v .5:^H








OJ CS] T~S CM C\3
Oi -< VCM fe;to Cioi vv vc.n
**

I ฐ


rH "
* S
m W
CO P ryi

S 1 -3 ง
s s t: ง
CU Ci) rt t, CD
•o S B & B S
rH a rH 0 tj *
a a S
O /— - 
CDrHfnrHfH S n OG
•pa. rt a. iu H^ 43 rt
rt O -P CD r^rHO-rH
3e c ca n rt x ^* a.
o (US rt tij SH .
fH4JOC ^J a-POCD
CDCOrHTJG C T)
>fH43fH O P K!rt XfH
G icl 0) rt fH rH H-l-H p CD
OS 3CJ 3 rt .*Jo..H>
Q ^-- CL. -—s X CO Q*-^ CJ -— •
rt' w VO 1--' 00 0>'
vO \O vD \O vO \O









IN. rt
^3 CM







fej Bi
^ Z





fcj Bj
^ Z






fe; CM







Ctj L/l
S3 rH






!g r^








CM
V rH
C
o
•H
p
rt
o
-H
•H
d,

T!
ง
X
rH
OH

CO
MH
O

p


-------
- 29 -










































/—*ป rt
•4->
"O 03
4J 0
ง X
U P
-— ' -H
rH
in cd
3
0 cx
rH
.0 rl
cd 0
H P
cd
in
OS
O
Z




































































cd
p
a
rl
0
•p
cd
s

ซ -o
•4-i 0
a x;
Ci in
•H
rl C
Q) -H
-ri B.
^ CO
•H
ซ 0
ft; p.
X
_co H
"*1 rl
co cd
O rH
s a
ซ 0
ซ-
C -H
"ti
CD
CO E
C -rl
•ti ปJ
O
-a
•w B
to O
r, 0
•ti 0
CM CO









































*
rH
cd
) i
0 B
H 0
X>
0 rl
rH Cd
•H CJ
P rH
cd O -~.
rH *H W
ง ง B
B CO
0 rl
Z 0
0
•a
B 1 -H
0 Cd rl
ฃ1 rl O rH
Tl P rH \
cd 0 X bo
U H U 3
I
o
rl
O 0
1 3 Cd rH
CM O pE ^^
--H p bo
rH Q O 3.


1
o
B B rH
rl O M
CQ <4H 3

1
O 1 CD
6 o c
O rl Cd rH
ri 0 ,C -^
,h rH 4J bO
•H .C 0 3
o u B

1
o
, rl 0
1 o 3
o rH a
e X X rH
O O 4-> ~~.
rl 'H 0 (X
CO 13 E 3


B
0
cfl
S
O rH
rH \
X -•


















Ml O>
to rH







fel B,
ft Z






ft Z






fe -1
ft V



ttj
ง vO







fo
Ci 00
V r~*






OQ
. rH
o to





p
1
p
rl
cd
OH .
0
Q
rl
0
P
cd
S
X
p
s
3
O '—*
c_> P
c
cd cd
•P rH
in cx
o
CJ E
cd
cd B
rl rH
P rH
E O
O CQ
tj ^


rH
t-^


















MiCTl cxito O^OO tO^O
MirM T-HrH C\1CN1 r-)rH







ttjU. fejtl- CtjB. Pi,B,
ftZ ftZ ftZ ftZ






fez fez fez fez





00
PL fcg BH fcj PM O
งt^ Sz gto Sv



Ctj ^ Ptj Cti <7l fej
ftto ftto fern S;^-







fcj to Uj fej 0 fe, tn
ftLO ft^O fttO ftrH






1^-1
CtjrH Cif\J fj^CM fcjt-H
ft\o vto ftin ft-*





rl
0
P 3
E ฐ
0 (X
e
P t3 p p
rl B OH E ^
cd cd 0 0 p
PH . Q BE
0 rl p cd
Q 0 Irt rl rH
P  • :b p cd
OP cd 0
rH (1 rl ^^ 3 rl
cd. 4> 0 P H
(X Q P E O
cd cd o to
Ul CO S --H IO Cd
O 0 (X -H 0
Q *-H Q O rl
0 00 rl E T3
<4H Of 0 Cd Cd B
O fi 'H E rl <
< Q cd B.
X rl E
P <5 fi -H E Cd
•H a cd s cd to
CJ i-J CO ^-^ CO ^-^


CM to Tt m
it- r^ r-^ r-


B
0
p
t/l
X
co
p
-H
Q
<4H
0
TJ
a
PJ




o^ 01 MI oo
C> O to CNl







Ct, B, Ct. B.
ft Z ft Z






fsx fe z






fc, B, fc, B.
ft Z ft Z


^f
fc, B, tt, 0
ft Z 6> V







a> oo
Cu • fe, .
ft o go






ซ
d in fcj o
V rH ft •*













4J
B
0
B
p e
rl 0
cd p
ฃX  P
cd cd
s s

0 CO
rH Cd
P rH
P bo
cd 3
0 O
CO Q


^o r--
i-^ t^













w

Oi
t-H

LO tO C^ ^f LO rH '
Qi O T-H O CV tO ^

Q!
V


.
M
Pi, B. Pt, BH PL B,
ง Z ft Z S Z '
<^
V



to
,
fe z fe z fe z w


^-



fe, B. PL B, fc, B, fe
ft Z ft Z ง Z *=



fe, fe, B. fe, fe
ft to ft z ft in ft





CQ
Q
fe, fc, fe, m '
ft to ft to ft to _

d
V


Oi
CXI O
t-H r-^
Q fe, <ฃ> . *O '
V CM ft CM Cl rH
r—

ci
V



4->
. B
> 0
•H B
Q P
P rl
B "> cd
0 0 CX
B *H 0
•P -P Q
rl -H .,,
Cd '-' rl ":
0. -H 0 S
0 P P g
0 = g a
rl CO
0 -rl rH
P rH cd
Cd rH ^ CX
S Cd P TH
> B CJ
tO rl Cd -H
rH O rH B
rH CJ CX 3
cd S
B, HH rl
O O O
O rH O
X X X cd
cd 4J cd 2
TJ -H H rH
rH CJ ^— ' rH


00 Cft O
t-~ r- oo












CM
CM
rH

1

in
o
o

V






.
& to
Z rl
4->
cd
3

^0 |
^

B. e
Z ^
•^ O
10

r.1 C
S -H
i s
z ii
ts
ง ^
s .a
;• S


^4-1
O

g
O
s^'S
^j in to
rt E 3
, ฐ -ฐ
1 -H B
B, y R
K cd o
^- rl
4-> 0
E P
i 0 0
O rH
S &
-H ฐ B
Z ฐ 8
"> 0 B
. T-H -H
J.-S'"
'"'•HO
' t4H
? -H 0
v p to
E 3
cd cd
3 0
cr cu
XI
3 rl
B o
•H rl
E rl
• H 0
B
rl -H
O
t4H 0
XI
CM
X
0 cd
X3
a) cd
H p
cd
CD T3
0
CO rl
^-^ 0
p
TJ cd
ง *
O 3
tp cd
rl
0
E 0

Z 0
CO
1
B.
Z *

-------
                                  - 30 -
     Confirmation samples.  Quantitative.  The data presented in Tables 6 and




7 show good quantitative confirmation of the results of the routine analysis




of the six selected compounds in the raw and finished waters in the 80




locations.  Because of the increased sensitivity of the confirmation




method used, analysis by that technique often produced a low measurable




concentration where the routine method did not find the compound.  This is




not an inconsistency.  The differences between the concentrations of the




routine and confirmation analyses in a few cases are not considered to be




significant.




     Qualitative.  The data in Table 7 show that the compounds quantitated




by the routine analysis were the correct compounds.  In no case did the




routine analysis ever quantitate a given compound that later had a negative




confirmation by GC/MS when the most sensitive GC/MS method was used.




This did happen occasionally when the less sensitive GC/MS method as




described in this paper was used.  In a few cases, because the sensitive




GC/MS method  used a larger sample for purging, this technique would




detect the presence of a compound when none was found by the routine




procedure.  These are not inconsistencies.

-------
- 31 -





























^o


to CM r-^ to
[i, . O ...
2 O 
PU| PU fT, ["]• ( fT4
222 B, 2 2
^-* ' ^-_^ ^^ V^^ / \^_ /







/*~*\ /•> t~\ /"%
PU UH PL, PL, PL,
^ rS r2I PH ^ ^
v^_^ ^ ^/ ^^ ^_/ \^_r/





/-^
/— -\ /— \ CN /-^
J"T l prH f_Tt • |*T |
222 B, o 2
x^-' ^— ' p?^ v^_x \ป^





"H **"** /^"N /"™\ <*~^
* CM I-H cn -^J-
p^, • • to • •
52: o o • o o
vป— ' ^—/ O ^— ' '"— '

-p
r^
o
•H
+J
O S
0 0
C 2
c
0
u to
rH
•> rH
X rt
?H PM
3 X
Xt 0 0
0 P rH
•P -P 0
rt -H I-}
s J

, ป
,CJ t-



f — S
• u7
UH 0 2
^^ V^ ^-^




r— \
rH / — \
• Pt,
tw o 2;
^ vซ/ v 	 t






/— \ /— \
pLi PU
pLt ^ ^
ฃS *"— ' v^-/







/•"N /"*\
ry^ PL
^ S 2/







,r— > /— ^
fT ( p^
UH 2 2
55 v-/ v^





^— ^ /— ^
CM rt
to
• rH O
O ^— ' v— '






ซ\
^f^J
W)
ฃ_t
3 rt
XI -H
w C
•p rt
+J >
•H rH
OH X
(/)
ฃg
3 0
O OH
CO

.
\o
rH




BH
|y|







B.
2








B,
2









B,
2







UH
2






CM
O
V






Xi
•p
3
o
^0

r,
C
O
•P

0
rH
rt
Xi
u

.
o
to



/ — \
UH
2
v — <





/ — \
UH
2
^ — '






f — \
P-,
2
V 	 '







ff*^
B.
2
V 	 /






f~~^
UH
2
^ 	 '





/ 	 \
to
o
^ — /













rt
r^
•H
rH
O
rt











UH
2







B.
2








UH
2









UH
2







UH
2







UH
2



•
rt
rH
^
O

•\
X
•p
•H
u

rt
g
o
•s
rH
f^l
o

.
rH
LO



B,
2
V 	 /




/ — \
LO
.
i— i
^ — i






i — >
UH
2
**-S







/—^
B,
2
**— . '






/ 	 V
B,
2
' — ^





/_ ^
LO
o
< — '





























f — \
UH
2
N 	 '





f^—^
UH
2
V 	 X






/ — \
UH
2
< — i







t — \
UH
2
V 	 1






f-~\
UH
\ — t






f — N
UH
2
^ — t



























-------
-  32   -
                                         r?














/•— >
••
*0
•p
c
o
U
v_*ป

^Q

0
rH
XI
rt
H







































O
E~











i
o

o

r*
<3

i
o
rH
O
r-H
•6
• H
T3





























0
rt
0
•H
^
o
r-H
rC
0
o
MH
O
S
0
rH
CQ




fl^
C
rt

4->
0
G



0
C
rt
+->
Q>
€
O

O

O
rC














C
O
•H
4->
rt
O
o
ป_3


/-^ f-^
tU UH
2^5 ง


U* UH
fc 2 2 UH




r-~\ f~\
PL, PL,
UH 2, 2, UH








/™\ •*•"•*
UH U,
U. 2 2 U-
Hป S^^y \^ / *jgป







/•""N /•"""*
UH tU
PL, Z 2 U.
l-jy. y ^/ t^^/ ^J

_ _
UH UH •
tu 2 2 0
2 ^^ ^-^ V



O
O "0
•S. rt

^ o
X -H
P O
•H U
U
•\
l/> O
rt TH
W XI
pj fl>
rt r*
^ DH


• •
o to
vO ^O


,_^
1

/"— \
UH
S




/"N
UH
2








^—^
UH
2^








/-^\
UH
\W*


UH
S


























/— v
1


UH
S




/"^
UH
^r
VHM/







^-^
UH
S








/••"v
UH
V.^

-^*s
UH

^M/

























/— % /-N
UH UH
UH 2 2


UH UH
UH 2 2
2 **-> ^



f — \ /-~\
UH tU
UH 2 2
2 v — ' v— '







.T*"^ ป^*^
UH UH
UH 2 2
2,1 ^"^ v*—/







tO /"^ /^™N
• UH UH
O 2 2
v *•— .^ ^-^

/ — v '^
tO UH UH
• 5, S





•
t(H
•H
TH
rt
U

•\
•d

0
o
C
o
u


•
rH
t^ซ



UH



UH
r^-t"





UH
2









tu
2








UH
2



UH
2


C
0
bO
V
f-H
o

ป*
in
*H
i-H
r-H
rt
^
0
U


•
CTi
t^-


^^
UH

^-N
i-H
•
o
v^_/



/^-\
UH
Z
V— /







/^-\
tu
2
V 	 /







/ — \
2
*^— '

in
•
o
^— ^

























,^-N
UH
S

i-H
•
O
\ป— '



/— -x
UH
J2J
v_/







/— -\
UH
^
^— '







/— \
UH
V—^

(N
•
o
V 	 '























                                    ฃ
                                    C
                                    o
                                         rt
                                         o
                                        •8
                                        _] 0
                                        x; P
                                         o
                                         tH X
                                         rt X)
                                         s  I
                                         x.o
                                         ^ri
                                         C/3

                                         rH -H
                                         a>  w

                                         t^^
                                         S  rt

                                         oS
                                         x;
                                         p  c
                                         x.S
                                         XI P
                                             rt

                                         a>
       x
       rH
       O
                                         o  C  rt
                                         4-1  O  rH
                                         rH  O  O
                                           i-J
                                         V) -H
                                         •H  P  X!
                                         w  rt  o
                                         X P
                                                rH
                                                rt
                                                a>
                                                t/>
                                         rt'  rt  0)
 rt  P
 c  C
 X
 rH
 rt
 S .
                                                O
                                             6  3
                                             O  w
                                         O

                                         4-1
    0 X
    tH P
    rt -H
       rH
       rt
 0 >—' o*
 tH
 rt  t/i *"C3
   •H  rt
•O  wi  nj
 O  0
•p x: -P
 w  p  rt
•H  C  CU
TH  0  S

 rt  rt  O
•P  PHTH
 rt      0
*X3  Cซ  ^
   •H  0
P     Q
 !/>  rt
 fH  P  W
•H  rt t3
4-1 T3  O

 0  0 -P

-------
                                                       - 33  -
CD
i—I

•ง
H






1 — t
CD

•H
4->
cd
•H

pj
cd
^j
o"
CD
•H
1 >
cd

•H
i-H
3
o;
5-4
0%
,p
cd
5:

TJ
CD

I/)
•H
(3
•H
PL,

(4.4
O
_i
Co
4->
cd
Q
t/)
•H
X
r- 1
cd
Ci
**
o
•H
4->
cd
S
•H
^4-4
s
o
u




















i
o
O CD
i rH f3
1 43 cd
CN CJ 43

r-l Q CD



CD
f3 1 -H
O Cd rH
43 (H O
M 4-> iH
cd O 43
U t- 0

P
O

o
6
o
H
CO

/ — s
1 — 1 1
~~^ O 1 CD
bfl S O P5
3- O in cd
f-i O 43
f3 43 rH 4->
•H -H 43 CD
Q 0 S
cd
^J
cd i
13 O
ฃH (U
r-l IOC
r-l O r-l Cd
< E 43 43
v-/ O 0 4->
rH "H CD
CQ -O g



S
O

O
rH
O
1— 1
u



















c
o
•H
4_)
cd
u
o






CN ^— \ /— X
• tin PL,
0 Z Z





1 	 v 1 	 V
LO LO
• •
CN O O
W ^v>/ V ^



/•"^v ("""N
[T^ rr t
rH )x^ ^
V *ป— ' v*— '






(gj
•JC f~*\ f~~\
\O [XH ft,
• ^ 2H
O v— ' ^— '






/— ^ ^— ^
LO \O
•Jc
O LO LO





+ C? c7
•K
^ rH rH
O^ vQ vO
^^-/ v-^



•P
O
•H
40
u
CD
G

o
U

ft
X
fn
3

J^
O
4->
cd



.
CN




y*-^ /— •%
/— \ /— x tO CN
CN PL, PL, •
• Z Z * PL, 0 0
O ^^ — ' v— ' ^O Z v^ ^ J
V



/^"^ /"~s /*— \ /^ ^
rt tO tO i-H
*
CN O O BH U, O O
v *— ' ' — ' Z Z ' — ' % — '



/— N /~~"t /*— N /^"*V
PL, PL, * PL, PL,
PL, Z Z PL, PL, Z Z
Z **— J **— ' Z Z v~*/ "^^ '






/""^ /^^
tO ^-N CM /—%
• fTi *^ ' OH
0 Z * • 0 Z
CN ' — ' ' — ' LO O •— ' ' — '






t — \ i — \ ^ — s r— *\
>O •t LO LO
• • • •
\O G^ r^ * rH rH





*^^\ /~^ /^™\
i— 1 O l-ป i — *
. CN
rH LO * O
O"> LO tO \O rH C?i
LO v— ' >— ' OO OO v— ' v— '
•i-
X
CD
W +-
CD cd
>-3 * ft,

^ • n
CD cd 43
Z a. oo

3
 cd 43
i— 1 -H (ft
rH 43 4J
Cd PH 4J
PL, rH .H
0 CL,
CD T)
rH Cd 43

4-> -H 3
•H 43 O
j a, en


• • •
t^. rH \O
rH rH



*
CN
•
O

V





*
ft l
^





*
to








*
LO
to








*
oo
^





*
rH
rH
to








*
*
cd

•H
(H
O
rH
PL,

ft
*H
S
rt
•H
IS


ป
CM
CM




*
*sj"
tu *
^^ ^^
V






PL, *
Z CN
V



OO *
• PL,
CO p^|








CO
*
O •*








*
* IO
Ol i-H





*
LO *
Cn LO
rH *?!•






*
*
O
U -H
43
CO O

A ft
e -H
O 4J
4-> Cd
w c

-------
-  34   -





















t — \
13
•P
O
U
V — /

r-~
rH
O
rt
H
































i
O
O  !H O
H H-> f-l
Cti 0) ,C
U H 0
O
S
O
rH
CQ
1
O 1 O
6 o c
O rH CIS
(H 0 43
fl rH +J
•H rC 
1 O C
O rH OS
6 X rC
O U -P
rH -H 0)
CO T3 E
O
o
rH
O
rH
|r^
U













r^
o
•H
•P
rt
o
o
!-]




CM ••* /— v ^— \
• • ["T | pr |
0 0 Z Z IX, PH
V V v_ / v_ / g; 2;



/"^i /^^*
\O OO
*
CM CM O O PL, *
y y v_/ v— ' g to
/ — v

pi
* Z PH *
Tf \O ' — ' * — ' Z LO


/~"^ ***™\
LO Oป
* to O * *
rH O rH rH P, C7>
rH CM ' — ' ' — ' Z rH




/^~N /^™%
OO LO
• •
* to 00 * *
* OO CM CM PH OO
OO CM ^— ' ' — ' Z "*

f ~N /^~\
rH LO
*
* O LO OO *
* ^f ^t "^ • OO
00 •<* >— ' '—' O 00
ci
o
*
•> * I/)
X nS ni
O 4-1 S ")
•H -H O CS
4S U rH Cj
O X
rt
^6 rt
CO S nj
ox i AJ
•P nj 3 0)
X rH 4-> (^
CS AS H-> O
Q 0 OH


• • • •
rH i— 1 LO 00
ซJ LO LO LO






(X,
Z






f T |




PL,
Z





•jc
CM








*
OO




*
^f
CM

^
O
rS

•\
X
4_>
•H
U

10
eS
t/i
C
rt


•
O
^o




^-^
[Xi
z
V 	 1



/ — ป
to
•
O


/ — >
PL,
Z
V — /


/ — \
CM
LO

O
v — /





^ — \
\Q
•
LO
v — t

i — >
rH
•
IO
CM





















                      UH
                      Z
                            -
                           Z   Z
to   to
o   o
                 IX,    PL,
            W.   Z    Z
            Z   '—'   '—'
                                       IX,
                                       Z
                                                 (x,
                                                 Z
                      UH
                           PL,   PL,
                           Z   Z
                 PL,
                 Z
                                            to
                           tL.
                      CM   Z
                      v    v—'
     IX,
     Z
00   00

o   o
                                                 CM
                                00
                                       to
                                            LO
                                            CM
                       vO
                       CM
                      O
                      •3
                      O
                      U
                                        BJ
                                       U
                                        C
                                        O
                                       U
                      LO
                      vO

-------
- 35 -






















i — >

T)
4J
C
0
U
v — /
O
rH
,0
rt
H






































CM
ft
rH


pj
O
,0
fH
rt
U







i
o
e
0

<"•>
•rt
Q



1
O
e
o

pa






























o
o
rH
A
CJ
•rt
T3


|
rt
fH

CD
H








1
O
&H
O
rH
r^
O
1
O
H
O
rH
rฃ~
O
•rt
































CD
C
rt
fj
4_)
CD
CD
T)
•rt

O
rH
f
O
o
HH
O
e
o
fH
CQ

CD
C
rt
r<^{
•p
(U
g



0)
s
o
o
!H
O
i— i
^cj
U
















C
O
•rt
4J
rt
o
o

^-^
[i,
PL, PL, PL, Z
^ ^ j^ \ 	 /



( — s
CM
.
U, PL, PL, O
222^



fc
* PL, LX, 2
t-- 2 2 ^






/-•\
* * tL,
*3- PL, fc 2
to 2 2 '-'



^-*^
•K vO
* Cft
tO * rH
to o to *— '



^
o
•
* * * IO
rH LO VO CN
VO rH CN ^-^
*
*
(3
o -
4J C
SH bo O
•rt C M
rH -rt CD
rt rC C
U w (D
rt
W t/)
O ft -rt
rH 0) rH
rt rH rH
o, +-> rt

rt
&
o

rCn
o

to
rt
bo
0)

•rt
+J
rt
4->
•rt
rH
rt
0*

CD
^
•rt
4->
•rt
tn
o
a,

rt

W)
4J
rt
u
•rt
*T3
C
M
1

4t
rH
"0
o
4->

i

CD
^
•rt
4->
•rt

C
CD


X

T3
o>
4->
O
CD
rH
rH
O
o

rt

rt
t)
C/5
S
•x^
CJ
CD

1
*
*
ormed by the Water Supply Research Laboratory.
HH
fH
CD
PH

I/I
•rt
to
X
rH
rt
c
rt
X

i
•rt
rH
P.

CD
A
4J

B
0
^1
M-l

0)
cd

*T3
CD
4J
•rt
rH

rt
4J

T3
CD
-C^
p
tive confirmation analysis performed by the Methods
oi

•H
^
c
otj
ฃ3
f

(D
(p^
•P

ง

4-(
CD
fn
rt

^"""N

Vw-^

I/)
•H
I/)
0>
X
(^
d)
5_j
rt
cx
(3
•rt

rt
4->
rt
T3
CD
f"j
1— I
oratory.
X)
rt
nJ

fjC1^
o
rH
rt

(/)
4)
OS

o
c
rt
rH
3
t/5
t/)


X
+->
•rt
rH
rt
J3
ex

•o
Co

^J
c
CD
a
o
rH

>
CD
Q





m
TJ
CD
4->
Q,
|f

ij
jj
ctf

IJ
o

ฃ
o
•rt
4_>
rt
p;
•rt
B
0)

O
TJ

CD
s
rj
CD

d)
^
•rt
+J
rt

f .
rH

13
O'
I
•t-

-------
                                     - 36 -
Discussion



     Occurrence of organics. Trihalomethanes. The first objective of the NORS



was to determine the extent of chlorination bypproducts in finished drinking


                         8                                   9
water as reported by Rook  and Bellar, Lichtenberg and Kroner .   To meet



this objective, raw and finished water from 80 locations, representing a



wide variety of raw water sources and water treatment practices, were



sampled for the four trihalomethanes -- chloroform (trichloromethane), bromodi-



choromethane, dibromochloromethane, and bromoform (tribromomethane).



     In general,these four compounds were not fouid in the raw waters tested.



Chloroform was present in 49 locations in concentrations of less than 1 yg/1 with



the exception of Whiting, Indiana that was receiving chlorinated raw water.



Bromodichloromethane was present in 7 locations in concentrations of less



than 0.8 yg/1 with the exception of Whiting, Indiana.  Dibromochloromethane



was found only in Whiting, Indiana's raw water.  Bromoform was not found in any



of the locations tested.  Therefore, the presence of any of these four



compounds in the finished water was concluded to be caused by chlorination



practices.



     All of the systems investigated disinfected, but in one system, the only



treatment was ozonation.  All of the finished waters tested contained some



chloroform although the system described above contained less than 0.1 yg/1.



A number of finished waters did not contain bromodichloromethane, dibromo-



chloromethane, and bromoform; however, their presence was frequent enough;



to be considered widespread throughout the finished waters of the nation.



     Although the range of concentrations found for each of the four



trihalomethanes varied greatly for the type of systems surveyed, the



concentrations of each of the compounds was not evenly distributed throughout



the range, but were grouped toward the lower end of the range.  Therefore



high concentrations of these parameters occurred infrequently in this study.

-------
                                     - 37 -
Note that many groundwater supplies in the United States do not chlorinate

and, therefore, probably do not contain any trihalomethane, but that none of

these supplies were included in the Survey.  To show the central tendency

of the data, Figure 2 presents the frequency distribution of the "trihalomethanes.

Based on Figure 2, the theoretical finished water with the median concentration

(one-half of the data above and below) of each compound, would contain about

21  ig/1 of chloroform, 6 yg/1 of bromodichloromethane, 1.2 yg/1 of dibromo-

chloromethane, and bromoform below the detection limit of the analytical method

used.

     Most of the finished waters had concentrations of the four trihalomethanes

that became less in the same order as those in the theoretical "median" water

described above; this was not true in all cases, however.  The reason for

concentrations of bromoform and dibromochlormethane being greater than the

bromodichlormethane and chloroform in some finished waters is not known at this
           Q
time.  Rook has postulated that if bromide was present in a water, the chlorine

will oxidize the bromide to bromine and the higher molecular weight bromo-

compounds would be formed.  Whether this phenomenon occurred in some of the

fnished waters surveyed is not known, but no data was developed to refute

this theory.

     1,2-dichloroethane and carbon tetrachloride.  Analysis for 1,2-dichloro-

ethane and carbon tetrachloride was also made on all samples because they had

been found previously in other drinking waters and had potential health

significance.  In this Survey, these two compounds were not found in 67.5 percent

and 87.5 percent of the finished waters, respectively.  In about one-third

of the cases where these compounds were present in the finished water, they

were also present in the raw water, indicating they were environmental

contaminants and were not created during water treatment.  The cause for the

-------
                  - 38 -
 300
  0.1
    2  5 10 2030 50 70   90959899
PERCENT, EQUAL TO OR LESS THAN GIVEN
CONCENTRATION
FIGURE 2 FREQUENCY  DISTRIBUTION OF
         TRIHALOMETHANE DATA

-------
                                      39 -
appearance of these compounds in the finished water when they were not detected




in the raw water is not known at this time.  One possibility is that this may




have merely been an artifact caused by the varying limit of detection of the




analysis, see Table 2.




     Nonvolatile Total organic carbon.  In addition to studying the six




specific compounds discussed above, an attempt was made to investigate the




general organic level in finished drinking waters by measuring the




nonvolatile total organic carbon (MVTOC) concentration in all 80 locations.




The range of these data was from less than 0.05 mg/1 to 12.2 mg/1, but




again, the data were grouped toward the lower end of the range (See




Figure 3).  The median NVTOC concentration (one-half of the data above




and below) was 1.5 mg/1.

-------
    5.0
z _
< E
ฐZ
ซ/ ฃ,
00 2.0
^ z
5ฐ

-------
                                     - 41 -
     Influence of source type and treatment practice on trihalomethane




formation.   The second objective of the Survey was to determine, if possible,




the influence that the type of source water and the treatment practiced have




on the formation of chlorination byproducts.  An initial examination of the




data indicated that the dominant factor influencing the creation of  •




chlorination byproducts was the general organic level of the water, provided




sufficient chlorine was added to produce a chlorine residual at the time of




sampling.




     To test this idea, the total trihalomethane concentration for each




finished water was first calcualted by dividing each of the four concentrations




by the appropriate molecular weight and adding the quotients together.  This




yielded a total trihalomethane (TTHM) concentration in yMoles/liter*.  The




advantage of this procedure is that the concentrationsof all four compounds




are reflected in a single number.  Analytic techniques to determine this parameter




directly are presently being considered.




     These data were then plotted against the NVTOC concentration of the raw




 water.  The TTHM data were divided into NVTOC cells in ascending order, each




cell having a span of 0.5 mg/1 NVTOC.  The average TTHM concentration was




then calculated for each cell and plotted against the appropriate NVTOC




concentration.  This analysis is appropriate based on the assumption that




each cell is sufficiently large and heterogeneous, arbitrarily taken as 4 or




more, with respect to the other variables that their influence is damped out by




the averaging process.
Note:  1 yM/1 TTHM = 119 yg/1 chloroform if only chloroform is present.

-------
                                     - 42 -
     The data plotted in Figure 4 has a correlation coefficient of 0.98, in spite




of any error because of suspended solids in some of the raw water samples.




Because of the scatter of the individual data points, the correlation coefficient




is 0.75 when all of the data are considered.  This shows that because most




waters contain a chlorine residual (meaning an excess of one of the reactants




is present), the concentration of the product (TTHM) is related to the




concentrationof the other reactants (unknown precursors) and further that the




NVTOC concentration is a reasonable indication of their concentration.




     To examine the data another way, the chlorine demand (total chlorine




added minus total chlorine residual) was calculated for each location.  On the




basis that the formation of TTHM exerted some of the chlorine demand, chlorine




demand  was plotted verses average TTHM concentration (averaged over 1 mg/1




wide chlorine demand cells) in Figure 5. The correlation coefficient of




these data is 0.85, 0.61 when all of the individual data points are considered.




Although these parameters are related, the correlation is probably influenced




by the other forms of chlorine demand.  Therefore, raw water NVTOC concentration




was chosen as the dominant independent variable for the analyses that follow.




     After this conclusion was reached all of the data were  divided  into




six NVTOC concentration cells, 0-1 mg/1, 1-2 mg/1, 2-3 mg/1, 3-4 mg/1, 4-5 mg/1




and greater than 5 mg/1, to eliminate the influence of that variable and were




then sorted so that like source types and treatment practices were together.

-------
                       - 43 -
     0.8
 u
 z
 o
 u
 LU
 z.
LU

O

x
OL
     0,7
     0.6
     0.5
0.4
     0,3
     0.2
     0.1
                                    (4)  H
                         (6)
                   9(7)  (NO.) - NUMBER OF
                         SUPPLIES IN NVTOC CELL
                     '(9)
              NOTE: I fl MOLE/LITER
              TTHM = 119 mg/l
              CHLOROFORM IF ONLY
              CHLOROFORM PRESENT
FIGURE 4,
   012345
   RAW WATER NON-VOLATILE TOTAL ORGANIC
   CARBON CONCENTRATION, mg/l
      CORRELATION OF TOTAL TRIHALOMETHANE
      AND NON-VOLATILE TOTALORGANIC
      CARBON CONCENTRATIONS

-------
                       - 44 -
z
o
   1.0
   0.9
   0.8
   0.7
   0.6
Z


8  0.5

LU
Z
<
X


   0.4
s  ฐ-3
O
<
Of
LU
   0.2
   0.1
              I        I        I        I        I


                                         (7)
              (20)
                                                 (7)
             (5)
(NO.) = NUMBER OF LOCATIONS

       IN EACH CL2 DEMAND CELL



 NOTE: IjuM/liter TTHM = 119 AI9/I

      CHLOROFORM IF IT WAS ALL

      CHLOROFORM
              I
  I
I
I
I
      0123456


    CHLORINE DEMAND (TOTAL DOSE - TOTAL RESIDUAL), mg/l



 FIGURE 5 CORRELATION OF TOTAL TRIHALOMETHANE

           CONCENTRATION WITH CHLORINE DEMAND

-------
                                     - 45 -
     Source Influence.  Except in the upper NVTOC cell, groundwater sources had




a lower average TTHM concentration than surface waters (See Table 8).   When




all NVTOC cells are considered, not much difference existed between the




various types of surface water, although river water sources had a higher




average TTHM concentration in _4 of the 6 NVTOC cells.




                                  Table 8




                             Source Influence






                          Raw NVTOC Range,  rag/1


Category
All Locations
Ground Water
River Water
Impounded Water
0-1
Avg.
TTHM
Cone.
n* yM/1
14 0.09
10 0.06
3 0.19
1 0.13
1-2
Avg.
TTHM
Cone.
n yM/1
16 0.27
2 0.21
6 0.27
8 0.29
2-3
Avg.
TTHM
Cone
n uM/1
13 0.34
1 0.11
6 0.46
6 0.27
3-4
Avg.
TTHM
Cone.
n vM/1
15 0.52
1 0.19
12 0.55
2 0.50
4-5
Avg.
TTHM
Cone.
n uM/1
9 0.67
Q
5 0.66
4 0.68
>5
Avg.
TTHM
Cone.
n yM/1
13 1.23
2 1.65
7 1.35
4 0.82
*Number of locations.

-------
                                    - 46 -
     Treatment influence.  Higher average TTHM concentrations were observed




at locations were raw water chlorination was practiced (see Table ง).   An attempt




was made to relate raw water chlorine dose to average TTHM production, but the




number of locations in each cell was too small to produce meaningful data.




The trend of average TTHM production was generally higher as raw water chlorine




dose increased, but the data were quite variable.




     The data on chlorine residual indicated that finished waters that did




not contain much free chlorine residual (Item 5, Table 9) had lower TTHM




concentrations than systems that had higher free chlorine residuals.




                                   Table 9




                          Chlorination Practice Influence





                              Raw NVTOC Range, mg/1
Category
All Locations
Raw Water Chlorination
No Raw Water
Chlorination
<0.4 mg/1 Combined Cl
Residual, >0.4 mg/1
Free Cl_ Residual
<0.4 mg/1 Free C10
0-1
Avg.
TTHM
Cone.
n* yM/1
14 0.09
4 0.18
10 0.06
5 0.11
7 0.05
1-2
Avg.
TTHM
Cone.
n yM/1
16 0.27
14 0.26
2 0.32
9 0.32
5 0.15
2-3
Avg.
TTHM
Cone.
n yM/1
13 0.34
9 0.36
4 0.31
8 0.40
3 0.13
3-4
Avg.
TTHM
Cone.
n yM/1
15 0.52
13 0.55
2 0.34
6 0.48
3 0.30
4-5
Avg.
TTHM
Cone.
n yM/1
9 0.67
7 0.73
2 0.47
2 0.51
3 0.47
>5
Avg.
TTHM
Cone.
n pM/1
13 1.23
11 1.27
2 1.06
4 1.56
6 0.63
  Residual
*Number of  locations.

-------
     All of the locations that practice filtration were sorted into NVTOC

concentration cells and then resorted based on the use of polyelectrolytes,

either as a coagulant^or filter-aid.    This was to gain insight as to whether

or not polyelectrolytes could act as a precursor for TTHM formation. In the

study group, the polyelectrolyte dose varied from 0.02 mg/1 to 7.7 mg/1 on the

days of sampling.  At two locations the dose was unknown.  In Table 10, the

indication can be seen that using polyelectrolyte resulted in higher average TTHM

concentrations in all NVTOC cells.  Additional controlled experiments must

be done to definitely establish whether the polyelectrolyte is reacting directly

with the chlorine.

                                 Table 10

                            Filtration Practice Influence


                               Raw NVTOC Range, mg/1
Category
All Locations
All Filter Plants
w/Polyelectrolyte
w/o Polyelectrolyte
0-1
Avg.
TTHM
Cone.
n* yM/1
14 0.09
5 0.17
2 0.21
3 0.14
1-2
Avg.
TTHM
Cone.
n yM/1
16 0.27
14 0.28
4 0.32
10 0.27
2-3
Avg.
TTHM
Cone
n yM/1
13 0.34
9 0.36
2 0.53
7 0.31
3-4
Avg.
TTHM
Cone
n yM/1
15 0.52
13 0.56
3 0.68
10 0.52
4-5
Avg.
TTHM
Cone
n yM/1
9 0.67
9 0.67
3 0.94
6 0.54
>5
Avg.
TTHM
Cone.
N yM/1
13 1.23
13 1.23
2 2.56
11 1.01
 *Number of locations

Based on a comparison of NVTOC concentrations in the
in the 63 locations where filtration was practiced a
NVTOC occurred, on the average.
raw and finished waters
30 percent reduction in

-------
                                   - 48 -
     In the 17 treatment plants where precipitative softening is practiced,




all but 3 had a pH of 9.0 or more in the finished water.   Two plants produced




a finished water with a pH over 10.0.  The data in Table 11 shows a higher




TTHM concentration in 4 of the NVTOC cells.  Further, the average TTHM




concentration for all 17 precipitative softening plants was 0.84 yM/1, while




the average TTHM concentration for all 80 locations was 0.49 yM/1.  This




follows the expected trend for pH dependency of the classical haloform




reaction and indicates that chlorination at higher pH will produce higher




concentrations of trihalomethanes 4 other conditions being equal.







                                   Table 11




                       Influence of Precipitative Softening




                             Raw NVTOC Range, mg/1


Category
All Locations
Ppt Softening
0-1
Avg.
TTHM
Cone.
n* yM/1
14 0.09
2 0.13
1-2
Avg.
TTHM
Cone.
n yM/1
16 0.27
2 0.49
2-3
Avg.
TTHM
Cone.
n yM/1
13 0.34
0 -
3-4
Avg.
TTHM
Cone.
n yM/1
15 0.52
5 0.55
4-5
Avg.
TTHM
Cone.
n yM/1
9 0.67
2 0.35
>5
Avg.
TTHM
Cone.
n yM/1
13 1.23
6 1.61
*Number of locations.

-------
                                     - 49 -
     In the 19 treatment plants using powdered activated carbon (one plant




sampled twice) the dosage varied from 0.6 mg/1 to 17.5 mg/1 with a median of 2.3.mg/l,




All of these plants treated surface waters.  From Table 12 it can be seen that




(except for the highest NVTOC cell) locations where powdered activated




carbon (PAC) was used had average TTHM concentrations lower than those




locations not using PAC.  The PAC dosage for the 4 plants in the highest NVTOC




cell was not sufficiently different from the overall median dose of 2.3 mg/1




iftoted above to readily explain the apparent difference in treatment performance.




The NVTOC concentration in these waters may have been too high to be influenced




by the PAC or the number of locations in this cell may just be too small to




draw reliable conclusions.




                                  Table 12




                     Influence of Powdered Activated Carbon




                            Raw NVTOC Range, mg/1


Category
All Locations
All Filter Plants
With PAC
Without PAC
0-1
Avg.
TTHM
Cone.
n* yM/1
14 0.09
5 0.17
2 0.16
3 0.18
1-2
Avg.
TTHM
Cone.
n yM/1
16 0.27
14 0.28
3 0.20
12 0.28
2-3
Avg.
TTHM
Cone.
n yM/1
13 0.34
9 0.36
1 0.13
8 0.39
3-4
Avg.
TTHM
Cone,
n uM/1
15 0.52
13 0.56
6 0.46
7 0.64
4-5
Avg.
TTHM
Cone.
n yM/1
9 0.67
9 0.67
4 0.24
5 1.02
>5
Avg.
TTHM
Cone.
n yM/1
13 1.23
13 1.23
4 1.29
9 1.21
*Number of locations

-------
                                    - 50 -

     Only eight water treatment plants used granular activated carbon  (GAC)

as a combination filtration/adsorption media, and this number is  too small to

make an analysis as above. All treat surface water and chlorinate raw water,

and all but one had >0.4 mg/1 free residual in the finished water,  so soire of

the variables noted above were eliminated.  Because all but one of  the  location-.

sampled were using granular activated carbon that had been in place for at

least several months, the activated carbon was exhausted for NVTOC  removal in

these locations.  This is shown in Table 13.   The average NVTQC removal at

these locations was not much higher than the "greater Lh.iri '-> -, pe.\,en> '\ ;Tt)C
                                              (footnote, Table 10) .
removal" previously reported for all coagulation-filtration plants/
the TTHM concentration in these finished waters being higher  than  the TTHM

concentration in the theoretical "median" finished water for  the entire  survey

in 6 out of these 7 locations using GAC exhausted for NVTOC rejfiu.'al  is nut

surprising.  This is also true when the data are examined on  a  "TTHM production

per unit of NVTOC" basis.

                                  Table 13
                      Summary of Granular Activated Carbon Plants


                                                          TTHM   TTHH/
                                                        1  Cone.  Fin. NVTOC,
Plant
A
B
C
D
E
F
G
Avg. (A to G)
Fresh Coal-Base GAC'_ H
Fresh Lignite-base GAC
(Resample)-B
Finished Water
NVTOC
Cone.
mg/1
1.0
1.4
1.6
1.9
3.2
4.2
4.4

0.2

1.4
% Rem<
of NV'
55
30
56
53
41
30
32
42
95

79
                                                           0 . 31    0 . 31
                                                           0,14    0.10
                                                           0,82    0.51
                                                           0.60    0.31
                                                           1 , 36    0,4^
                                                           1.19    0,28
                                                           0,74    0,18
                                                                  0.30
                                                           0.08

Theoretical median water*   1.5                 -           0.22

*See Figures 2 and 3 for median concentrations.

-------
     Shortly after the Survey samples were taken at one of these  locations  (B),




the granular activated carbon was removed and replaced with virgin  lignite-




base material.  'Ihis location was resampled in an effort to evaluate the




performance of fresh granular activated carbon.  At another location fresh




coal base granular activated carbon had been in place only 2 weeks  at the




time of sampling,  The summary data for the fresh materials (Table  13)  show




a marked improvement in the parameters listed, indicating the effectiveness




of fresh grara'iac activated carbon for treatment.




     Another attempt was made to evaluate the performance of granular




activated carbon for treating a variety of waters by monitoring the




activate^ caibor- units installed in the five locations where additional




sampling WP.S being undertaken.  These samplers were 3-foot-long (90 cm)




colijnns ef COP 1-based granular activated carbon operated downflow at a




i:Iltrr'tion rate of 3.2 gallons per minute/square foot (8 meters/hour).




Finished waier v.as passed through them for 7 days.  The contact time




KOS 1 *o 4 Tinuul"1'.  Fresh granular activated carbon produced low NVTOC




.-••jTiceni ration at first in all locations (Table 14), except Miami, Florida,




where tlie organic loซd was so heavy that a different treatment mode



would be needed to produce a lower NVTOC concentration.

-------
                                 - 52 -

                                Table 14

Performance of Fresh Coal-Based Granular Activated Carbon Samplers Treating

                               Finished Water
Location
Day
      NVTOC Concentration - mg/1
Influent to      Effluent from   NVTOC
Sampler	Sampler         Removed
Miami,
Florida
Seattle,
Washington
Ottumwa,
Iowa
Philadelphia,
Pennsylvania
Cincinnati,
Ohio
0
7
0
7
0
7
0
7
0
7
8.1
7.1
1.9
0.8
3.6
3.4
2.0
1.9
1.2
1.6
1.3
3.5
1.9*
0.05
1.6*
0.9
0.3
0.5
0.1
0.1
84%
51%
0%*
94%
56%*
73%
85%
74%
92%
94%
*Data Suspect

-------
                                   - 53 -
     The two water utilities in the United States (at Strasburg, Pennsylvania




and Whiting, Indiana) currently using ozone as a treatment unit process were




included in this Survey.  At Strasburg ozonation is the only treatment.  The




source in this water system is 12 springs -- goundwater of high quality.




None of the six selected compounds were found in the source water, and it




had a very low NVTOC concentration.  The ozone residual immediately after




application was 0.4 mg/1.  The finished water was also of good quality.




Only chloroform at a concentration of less than 0.1 yg/1 was found in the




finished water.  Although this concentration was the lowest found in the




Survey, it was a definite positive value.  The possible source of this




slight amount of chloroform could be the oxidation of the chloride in the




water to chlorine by the ozone, with the subsequent formation of some




chloroform.  The chloride level in this water was low, but sufficient




was present to account for the chloroform formation, if this reaction occurs.




     The situation was different in Whiting, Indiana.  Here, Lake Michigan




is the source.  The source water contained 0.1 yg/1 of chloroform and less




than 0.8 yg/1 of bromodichloromethane.  None of the other selected compounds




were found in the source water.  A nearby refinery takes water from Lake




Michigan, chlorinates it, uses most of the water itself and sends some to




the Whiting water treatment plant.  Therefore, when the sampled water arrived




at the water plant, it contained 16 yg/1 of chloroform, 11 yg/1 of




bromodichloromethane, and 3 yg/1 of dibromochloromethane.  The water was




then ozonated, with about 2 mg/1 being applied in two 21-foot (6.5m) deep




towers for a detention time of 6 to 8 minutes.  Following ozonation water




was then chlorinated again, with ammonia added to produce a combined chlorine

-------
                                       - 54 -
residual.  The water is then coagulated and filtered before distribution.




The finished water contained only 0.5 yg/l of chloroform and 0.3 yg/l of




bromodichloromethane.  At this time, the mechanism causing the reduction




in trihalomethane concentration is not known.  These organics could have




been stripped from the water during the process of ozonation; they could




have been oxidized by the ozone, or they could merely have been lost to




the atmosphere during the passage through the open settling basin and




filters.  Subsequent sampling at this plant has indicated that the latter




explanation is the most probable.




     The data shown in Tables 8 and 9 indicate that the use of surface




water as a source, raw water chlroination, and the presence of greater:than




0.4 mg/1 free chlorine residual enhances the formation of trihalomethanes.




To test this indication the data were sorted on these bases.  Out of the




entire survey, 39 locations met these three criteria.  Of these, 13




finished waters had an NVTOC concentration equal to or less than the Survey




median concentration of 1.5 mg/1; the remaining 26 had greater than the




median NVTOC concentration.  Of those 13 with a finished water NVTOC




concentration equal to or below the median concentration, 62% had a TTHM




concentration above the median TTHM concentration, indicating agreement with




the hypothesis.  Of those 26 with a finished water NVTOC concentration greater




than the median concentration, only 19% had TTHM concentrations below the




median TTHM concentration, in opposition to the proposed hypothesis.  This




indicates general agreement with the indications of Tables 8 and 9.




     The difficulty with using these data, and those of Tables 10, 11, 12,




and 13 is that although the averages in the various NVTOC cells show some trends,




the individual data are quite scattered, for reasons unknown.  Therefore,




while these analyses are useful to obtain an indication of source and

-------
                                    - 55 -
treatment influences, carefully controlled experiments are needed in the




future to understand these reactions more exactly.




     Alternate indicators of organic contaminant levels.  Because organic




contaminants vary in toxicity, specific organic compounds should be monitored




in finished waters.  This is the recommended procedure for monitoring




organochlorine pesticides, for example.    Except for a few specific




examples, this approach is beyond the capabilities of most water utilities




and to some degree even is beyond the capabilities of researchers, given




the current state of organic analysis.  For example, all individual organic




compounds present in water cannot now be identified and quantified.




     In the absence of measuring for specific organic compounds, the next




best alternative is to measure some organic parameter that includes a large




number of organic compounds and assume that the level of this parameter  is




proportional to the level of toxicity of the water.  On this basis, carbon-




chloroform extract (CCE-m)   was included in the proposed Interim Primary




Drinking Water Regulations.




     In the NORS nonvolatile total organic carbon was the parameter chosen




to represent the concentration of organics in the water.  In Figure 4 NVTOC




was shown to be generally proportional to trihalomethane formation, so a




measure such as this is probably useful, but little else is known about NVTOC.




     In an effort to find an easier analytic procedure for monitoring the




organic level in water, three other measurements were made on each raw and




finished water in addition to NVTOC concentration.  These were ultraviolet (UV)




absorption, emission fluorescence scan (EmFS) and the Rapid Fluorometric




Method (RFM).   An attempt was made to correlate those parameters, although




different organics absorb UV to differing degrees and different organics




fluoresce to differing degrees.  Therefore, although the a priori judgment

-------
                                    - 56 -
was that those three parameters might not correlate with NVTOC concentrations




because they would be heavily influenced by the types of organics present




in the water, the} hypothesis that different waters would be sufficiently




similar to make these procedures useful was tested.




     Just as particulates in some raw waters caused some error in the NVTOC




measurement, the resultant turbidity interfered with the UV, EmFS, and RFM




measurements; therefore these data were not  (analyzed.  Plots of NVTOC




concentration versus each parameter for finished water (Figures 6,7, and 8)




show a wide scatter of data.  In Figure 6, 39 data points are included in a




one-milligram-per-liter-wide band and only 28 data points (up to an NVTOC




concentration of 3.5 mg/lj are excluded.  The overall correlation, however,




is not very good.  The two fluorescence techniques correlated reasonably well




with each other, but not with NVTOC concentration  (see Figure 9).




     At 11 Survey locations, concentrations of nonvolatile total organic




carbon and carbon chloroform extract (CCE-m) were measured simultaneously.




Figure 10 indicates that some correlation existed between these two




parameters, but the data are rather scattered.


                                12
     Sylvia, Bancroft and Miller   have proposed the RFM measurement as a




rapid indicator of CCE-m concentrations as they found good correlations




between these two parameters in the water they were studying. T.o test their




proposal on the NORS data, the 11 data pairs were plotted in Figure 11.




For these data, the RFM measurement was poorly correlated with CCE-m




concentration.

-------
                     - 57 -
   .18
   .16
   .14
c
3
LU
u
Z  .12

CO

O
to
OQ  . ..
<  .10
g
>
  .08
m 0.6
Q
LU
I .04
CO
  .02
   Ok
  •.:••* •.  ฎ
        ••*    '
•
               .
      *
      •••
     01       2345

                FINISHED WATER NVTOC, mg/l

FIGURE 6. CORRELATION BETWEEN ULTRA-VIOLET

          ABSORBANCE AND NON-VOLATILE TOTAL
          ORGANIC CARBON IN  FINISHED WATER

-------
                     - 58  -
    21
   19
.r   17
c
3

O
O
X   15
u
of   13
LU

O
on
O   1
<   9
c*
                  ••
     0123456
                FINISHED WATER  NVTOC, mg/l

FIGURE 7. CORRELATION OF RAPID  FLUOROMETRIC METHOD
          AND NON-VOLATILE TOTAL ORGANIC COMPOUND
          CARBON IN FINISHED WATER

-------
                     - 59 -
  1900
  1700
  1500
Z

<
u
U
O
z

O
  1100
   900
   700
   500
   300
    i     ซ•  • •
•  .  •    •
     0
            2345

          FINISHED WATER NVTOC, mg/l
 IGURE 8.  CORRELATION OF EMISSION FLUORESCENCE

          SCAN AND NON-VOLATILE TOTAL CARBON

          IN FINISHED WATER

-------
                            - 60 -
  2000
   1800
   1600
   1400
S  1200

Z
LU
U
CO
Ul


O  1000
z
o
./>  800

I
LU
ee.
LU
S  600
o
UJ
I
^  400
Z
   200
                              I    I
I    I
          I    I    I    L   1    I    I   I    I    I    I    I    I
      0   2   4   6   8   10  12  14   16   18  20  22  24  26  28

           FINISHED WATER RAPID FLUOROMETRIC METHOD, UNITS


   FIGURE 9  CORRELATION OF THE TWO FLUORESCENCE

             MEASUREMENTS PERFORMED  ON FINISHED WATER

-------
                            - 61 -
O>
   1.2
u
—  1.0
ป—
u
  0.8
2
o
X
u
O
03
CX.
  0.4
  0.2
     01       23456
            NON-VOLATILE TOTAL ORGANIC CARBON, mg/l

  FIGURE  10. CORRELATION OF NON-VOLATILE TOTAL
             ORGANIC CARBON WITH CARBON
             CHLOROFORM EXTRACT

-------
               - 62 -
  20,
   18
   16
   14
   12
ง10
o
2  6
            I	|	|	|	I
   "0      0.2     0.4      0.6     0.8      1.0     1.2
        FINISHED WATER CCE-m CONCENTRATION, mg/l
FIGURE 11 CORRELATION BETWEEN CARBON CHLOROFORM
         EXTRACT CONCENTRATION AND  THE RAPID
         FLUOROMETRIC METHOD IN FINISHED WATER

-------
     Significance of findings.  Most water treatment plants are not designed



to remove soluble organic compounds from raw water, and disinfection creates



some compounds that were not originally present in the raw water.   Therefore,



the finding that all finished waters in the Survey contained one type of



organic compounds or another was not surprising.   The presence of an



organic compound in a finished water is not significant, however,  unless


                                                       13 14
its concentration is such that it poses a health hazard  '  .   These



data, therefore, must be combined with health effects data before any



significance can be attached to the findings.  If a health hazard is found



to exist with any contaminant, then the treatment information currently



being developed by the research program of the Water Supply Research



Laboratory  '   must be applied to remove that contaminant,

-------
                                     - 64 -
Summary and Conclusions



     1,  The four trihalomethanes:  chloroform (trichloromethane),




bromodichloromethane, dibromochloromethane, and bromoform (tribromomethane),



are widespread in chlorinated drinking waters in the United States and result



from the chlorination treatment process.



     2.  The four trihalomethanes were not found or were present in low



concentrations in the raw waters tested.  Carbon tetrachloride was not found



in the raw water of 95% of the locations surveyed.  1,2-Dichloroethane



was not found in the raw water of 86% of the locations investigated.



     3,  The median concentration of three of these compounds found in



finished water was:  chloroform, 21 yg/1; bromodichloromethane, 6 yg/1; and



dibromochloromethane, 1.2 yg/1.  Bromoform was not found in finished water in



68.8% of the supplies surveyed.  The range of concentration of all four



trihalomethanes was:  chloroform, less than 0.1 yg/1 to 311 yg/1;



bromodichloromethane, none found to 116 yg/1; dibromochloromethane, none



found to 100 yg/1; and bromoform, none found to 92 yg/1.



     4.  Carbon tetrachloride and 1,2-dichloroethane were not detected very



frequently nor found in high concentrations in finished water in the



Icoations studied.  Specifically, carbon tetrachloride was not found in 87.5%



of the finished waters waters tested and the highest concentration found was



3 yg/1.  1,2-Dichloroethane was not found in 67.5% of the finished waters



tested, and the highest concentration found was 6 yg/1.



     5.  In general, total trihalomethane concentrations were related



to the organic content of the water, as measured by the nonvolatile total



organic carbon test, when sufficient chlorine was added to create a



chlorine residual.

-------
                                    - 65 -
     6.  In general, when the following conditions occurred, higher




concentrations of total trihalomethanes were found:  surface water as the




source water, raw water chlorination practiced, and more than 0.4 mg/1 free




chlorine residual present.




     7.  In general, where precipitative softening was practiced and the




finished water had a high pH, higher total  trihalomethane concentrations were




found.




     8.  At 7 out of the 8 locations, based on the removal of nonvolatile




total organic carbon, the granular activated carbon being used as a combination




filtration/adsorption media was exhausted.




     9.  When fresh granular activated carbon  (coal-base  and lignite-base)




was monitored at two locations, removal of nonvolatile total organic carbon




was higher than average and the total trihalomethane concentration was




lower than the average values obtained from plants where the granular




activated carbon had not been replaced for some time.




     10.  Nonvolatile total organic carbon concentrations did not correlate




well with ultraviolet absorption, fluorescence, and carbon chloroform




extract concentration data and the Rapid Fluorometric Method did not correlate




well with carbon chloroform extract concentration data.




     11.  One location treated a high quality groundwater only with ozonation




for disinfection.  The finished water at this location had the lowest total




trihalomethane concentration of any location surveyed.  At another location,




water containing the trihalomethanes was ozonated.  Through the treatment




plant the concentration of total trihalomethanes was reduced, but follow-up




sampling indicated the mechanism causing this reduction is most likely lost




to the atmosphere through the settling basins and filters.

-------
                                      - 66 -
 Acknowledgments




      The authors wish to thank the many others who contributed to this paper:




The Regional Water Supply Engineers who collected all of the samples;  the members




of the Methods Development and Quality Assurance Research Laboratory who made




most of the qualitative confirmation analyses; the members of the Criteria




Development Branch, Water Supply Research Laboratory who made the remainder




of the qualitative confirmation analyses; and to Mrs. Maura M. Lilly who




typed the manuscript.

-------
                                     - 67 -

 References

 1.  Coleman, W.E., Lingg, R.D., Melton,  R.G. ง Kopfler, F,K., GC/MS
     Techniques for the Identification of Volatile Organics in Tap Water.
     In preparation (June 1975).

 2.  Lingg, R.D., Melton, R.G., Kopfler,  R.K. ง Coleman, W.E., GC/MS
     Techniques for the Quantitation of Volatile Organics in Tap Water.
     In preparation (June 1975).                           ,

 3.  Tardiff, R.G., Budde, W.L., Coleman, W.E., DeMarco, J., Dressman, R.C.,
     Eichelberger, J.W., Kaylor, W.H.,  Keith, L.H., Kopfler, F.K.,
     Lingg, R.D., McCabe, L.J., Melton, R.G., ง Mullaney, J.L., Organic
     Compounds in Drinking Water: A5-City Study.  In preparation (June 1975).

 4.  Bellar, T.A. ง Lichtenberg, J.J.,  The Determination of Volatile
     Organic Compounds at the yg/1 Level  in Water by Gas Chromatography.
     USEPA, National Environmental Research Center, Cincinnati, Ohio,
     EPA-670/4-74-009, Nov. 1974.  See also, Bellar, T.A. ง Lichtenberg,  J.J.,
     Determining Volatile Organics at the yg/1 Level in Water by Gas
     Chromatography.  Jour. AWWA, 66:739 (Dec. 1974).

 5.  Stevens, A.A. ง Symons, J.M., Analytical Considerations for Halogenated
     Organic Removal Studies.  In:  Proceedings AWWA Water Quality
     Technology Conference, December 2 $  3, 1975, Dallas, Texas, pp. XXVI-1
     (1975).

 6.  Dobbs, R.A., Wise, R.H. ง Dean, R.B., The Use of Ultraviolet Absorbance
     for Monitoring the Total Organic Carbon Content of Water and Wastewater.
     Water Research, 6:1173 (Oct. 1972).

 7.  Sylvia, A.E., Detection and Measurement of Microorganics in Drinking
     Water.  Jour. NEWWA, 87: No. 2 (Jun. 1973).

 8.  Rook, J.J., Formation of Haloforms During Chlorination of Natural
     Waters.  Water Treatment and Examination, 23: Part 2, 234 (1974).

 9.  Bellar, T.A., Lichtenberg, J.J., ง Kroner, R.C., The Occurrence of
     Organohalides in Chlorinated Drinking Water, Jour. AWWA, 66:703
     (Dec. 1974).

10.  Interim Primary Drinking Water Standards.  Federal Register, 40:No.  51,
     Part II, 11190 (Mar. 14, 1975).

11.  Buelow, R.W., Carswell, J.K. ง Symons,  J.M., An Improved Method for
     Determining Organics by Activated Carbon Adsorption and Solvent
     Extraction  (Parts I and II).  Jour.  AWWA, 65:57, 195 (Jan.-Marh. 1973).

-------
                                       - 68 -
12.    Sylvia, A.E., Bancroft, D.A. ฃ Miller, J.D., Detection and Measurement
      of Microorganics in Drinking Water by Fluorescence.  In:  Proceedings
      AWWA Water Quality Technology Conference,  December 2 & 3, 1974,
      Dallas, Texas, pp. XXVII-1  (1975).

13.    Murphy, S.D., A Report - Assessment of Health Risk from Organics in
      Drinking Water by an Ad Hoc Study Group to the Hazardous Materials
      Advisory Committee - U.S. Environmental Protection Agency, Washington, D.C.,
      (April 30, 1975).  Mimeo, 59 pp, plus Attachments.

14.    Tardiff, R.G., Craun, G.F., McCabe, L.J. and Bertozzi, P.E. , Preliminary
      Assessment of Suspected Carcinogens in Drinking Water - Interim Report
      to Congress, Appendix VII, Health Effects Caused by Exposure to
      Contaminants.  U.S. Environmental Protection Agency, Washington, D.C.,
      (Jun. 1975).  In Press.

15.    Love, O.T., Jr., Carswell, J.K., Stevens, A.A., Sorg, T.J., Logsdon, G.S.,
      and Symons, J.M., Preliminary Assessment of Suspected Carcinogens in
      Drinking Water - Interim Report to Congress, Appendix VI, Preliminary
      Results of Pilot Plants to Remove Water Contaminants, U.S. Environmental
      Protection Agency, Washington, D.C. (Jun. 1975).  In Press.

16.    Love, O.T., Jr., Carswell, J.K., Stevens, A.A. and Symons, J.M.,
      Treatment of Drinking Water for Prevention and Removal of Halogenated
      Organic Compounds  (An EPA Progress Report).  Presented at the 95th
      Annual Conference of the American Water Works Association, June 8-13,
      1975, Minneapolis, Minn.
                                                             U S GOVERNMENT PRINTING OFFICE 1975— 657-641/1009

-------