EPA 906/9-75-003
ANALYTICAL REPORT
NEW ORLEANS AREA WATER SUPPLY STUDY
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
REGION VI
Dallas, Texas 75201
December 9, 1975
-------
EPA 906/9-75-003
ANALYTICAL REPORT
NEW ORLEANS AREA WATER SUPPLY STUDY
Submitted by
LOWER MISSISSIPPI RIVER BRANCH
SURVEILLANCE AND ANALYSIS DIVISION
REGION VI
U.S. ENVIRONMENTAL PROTECTION AGENCY
DALLAS, TEXAS
JUNE 1975
With Technical Assistance as Noted
-------
NOTICE
The mention of trade names or commercial products in
this publication does not constitute endorsement or
recommendation for use by the U.S. Environmental
Protection Agency.
-------
TABLE OF CONTENTS Page
LIST OF TABLES AND FIGURES ii
ACKNOWLEDGEMENTS „•}_ vi
INTRODUCTION 1
SUMMARY OF EXPERIMENTAL METHODS 4
Sampling Methods 4
Carbon Adsorption 4
XAD Resin Adsorption 6
Liquid-Liquid Extraction 6
Reverse Osmosis 6
Volatile Stripping (VOA) 7
Sample Preparation Methods 9
Mega Carbon Processing 9
CAM Carbon Processing 10
Mini Carbon Processing 11
XAD Resin Processing 11
Analytical Methods 12
Southeast Environmental Research Laboratory 12
Water Supply Research Laboratory 14
Processing of Blanks 16
ANALYTICAL RESULTS 20
PROJECT SUMMARY AND EVALUATION 36
APPENDIX: Technical Assistance Report Submitted by
Analytical Chemistry Branch, Southeast Environ-
mental Research Laboratory, Athens, Georgia 41
-------
LIST OF TABLES AND FIGURES
TABLE 1 DISTRIBUTION OF WORK OPERATIONS 18-19
TABLE 2 ORGANIC COMPOUND IDENTIFICATIONS 25-33
KEY TO SYMBOLS USED IN TABLE 2 34
11
-------
ACKNOWLEDGEMENTS
This project was assigned by the Region VI Administrator, U.S.
Environmental Protection Agency, Dallas, Texas, to the Lower Mississippi
River Branch Laboratory, Slidell, Louisiana, a facility of the Region VI
Surveillance and Analysis Division. The following individuals were
intimately involved in various phases of the project:
LMR Branch Chief: Thomas F. Beckers
Sampling Coordinator: Ernest Douglas
Analytical Coordinator and
Report Preparation: William D. Langley
Sampling Team: Burton 0. Pritchard
Albert L. Hebert
Mega-Carbon Sample
Processing: Luther L. Hunt
Clerical Assistance: Addye P. Smith
Ttie objectives of this project could not have been accomplished
without the extremely close cooperation', diversified abilities, and
high competence of other individuals and groups within the Environ-
mental Protection Agency who took-time away from other duties to
devote, themselves to this effort.
The Analytical Chemistry Branch, Southeast Environmental Research
Laboratory, Athens, Georgia, performed extensive analytical work on the
Mega-carbon, CAM carbon, and Mini-carbon chloroform extracts as well as
the XAD resin extracts and tetralin extracts from all three water plants
sampled. Their technical assistance, timely, exceptionally thorough,
and precise, is gratefully acknowledged by the Lower Mississippi River
Branch personnel. Individuals at SERL involved in this project were:
ACB Branch Chief: William T. Donaldson
Analytical: Arthur W. Garrison
Lawrence H. Keith
Mike H. Carter
Frank R. Allen
Terry L. Floyd
John D. Pope
Alfred D. Thruston, Jr.
m
-------
The technical assistance and close cooperation, freely given,
of the Water Supply Research Laboratory, Cincinnati, Ohio, is like-
wise gratefully acknowledged. Their application of the relatively
new volatile organics analysis technique to samples obtained from
the Carroll ton Water Plant enabled the detection and quantitation of
certain components which were not, nor could they have been, detected
by the carbon-chloroform extraction procedures. Individuals at WSRL
directly involved in one or more phases of this project were;
WSRL Director: Gordon Robeck
Analytical: Robert Melton
Frederick Kopfler
E. Coleman
R. Lingg
Ron Dressman
Don Mitchell
W. Kaylor
Not the least of importance in the successful accomplishment of
the analytical objectives was the job of processing the many samples
obtained by the carbon adsorption methods. This must be accomplished
in a well protected environment tinder closely controlled conditions
to minimize the possibility of contamination. For the CAM and Mini-
carbon samples, this job was performed by individuals within the
Subsurface Environment Branch, Robert S. Kerr Environmental Research
Center, Ada, Oklahoma, and the Houston Branch Laboratory of Region VI,
S & A Division. These individuals were:
RSKERC, CAM-Carbon Processing: William Dunlap
Roger Cosby
Houston Branch, Region VI Medardo (Mike) Garza
Mini-Carbon Processing
Assistance In transporting the exposed CAM samples to Ada was
provided by Jim MiHsap of the Region VI, S & A Division, Ada Branch
Laboratory.
Valuable assistance in the conduct of this project was also given
by certain individuals or groups outside the Agency, namely, Mr. Gregor
Junk and his associates, USAEC-Ames Laboratory, Iowa State University,
Ames, Iowa, and personnel of the Gulf South Research Institute, New
Orleans, Louisiana.
Finally, the entire staff of the Lower Mississippi River Branch
Laboratory wish to express their sincere appreciation for the assistance
and cooperation 1n the sampling operation extended by all plant personnel
at the Carrollton Water Plant (City df New Orleans), the Jefferson No. 1
Water Plant (Metairie) and the Jefferson No. 2 Water Plant (Marrero).
1v
-------
INTRODUCTION
In July, 1974, representatives of the State of Louisiana and the
City of New Orleans tendered a request to the Region VI Administrator,
United States Environmental Protection Agency (EPA), that a sampling and
analytical survey be undertaken by the agency to determine, to the
extent possible, the identities and quantitative concentrations of trace
organic compounds which might be present in the finished (i.e., treated)
water of the Carrollton Water Plant (City of New Orleans), Jefferson
Parish No. 1 Water Plant (Metairie), and the Jefferson Parish No. 2
Water Plant (Marrero). The request was accepted and agreed to by the
Regional Administrator. Immediately thereafter, a plan and schedule
were formulated for conducting the necessary sampling, and arrangements
were made for technical assistance in processing and analyzing the
samples obtained. The assignment for sampling and analytical coordina-
tion was given to the Lower Mississippi River Branch of the Region VI
Surveillance and Analysis Division. This branch facility was instructed
to have sampling operations completed by mid-August, 1974, and an analytical
report issued by the end of October. On November 8, 1974, the Draft
Analytical Report was released pursuant to these instructions. While
this report listed 66 organic compounds which had been identified and in
most cases quantitated in one or more of the tested waters, it was
stated that the report was not to be considered final as there were some
samples remaining to be analyzed and that this analytical work was
proceeding. All analytical work has now been completed, and the present
report can be considered the final report for this project.
1
-------
Subsequent to the release of the Draft Analytical Report by Region
VI, the National Administrator of the Environmental Protection Agency
directed that a nationwide survey for trace organics in drinking water
supplies be initiated, a study which is currently in progress at the
Federal level. It is, therefore, emphasized that this report is not a
part of that nationwide survey but is solely related to the New Orleans
area study previously committed. It is also emphasized that the study
reported herein was strictly an analytical survey and did not encompass
an evaluation of the public health significance of the compounds identified,
In accordance with directions from the Regional Administrator, a com-
prehensive sampling and analytical program was developed and placed in
operation. This program is illustrated in Table 1 and described in some
detail in the Summary of Experimental Methods. The carbon adsorption
methods sampling program was established on the assumption that an
individual might consume one liter of water per day, i.e., approximately
one quart. Thus the use in Table 1 and elsewhere throughout this report
of the terms 70-year equivalent, 10 year equivalent, etc. has reference
to the volume of water sampled equivalent to the amount a person might
consume in that period of time at the one liter per day rate. The other
methods were added to the project either by specific request or to
provide a means of detecting compounds of a type not amenable to detection
by the carbon adsorption-chloroform extraction methods. In some cases
it was hoped to provide some comparative evaluation of sampling methods
in an as yet experimental stage of development.
-------
To perform the necessary analtyical work for this project, the
analytical coordinator through the Regional Administrator requested and
was granted the technical assistance of several groups within the Environ-
mental Protection Agency's Research Centers having highly developed and
competent analytical expertise and the necessary instrumentation to per-
form the required analytical operations. While their assistance has
been acknowledged in a previous section, it should be stressed that the
results presented herein represent the combined analytical efforts of
personnel of the Analytical Chemistry Branch, Southeast Environmental
Research Laboratory (EPA), Athens, Georgia, and the Water Supply Research
Laboratory, EPA, Cincinnati, Ohio.
-------
SUMMARY OF EXPERIMENTAL METHODS
Shown in Table 1 is a distribution of work operations for the New
Orleans area water supply study. This table provides in short and
concise form information on the various sampling methods employed, the
specific water plants sampled, the inclusive dates on which sampling was
performed, the group performing the sampling operation, the water volume
sampled, the group preparing the sample for analysis, and finally the
group performing the analyses. Given below is a brief narrative descriptioi
of the various sampling, sample preparation, and analytical methods
used.
Sampling Methods
Carbon Adsorption. Three types of carbon adsorption units were
used in the sampling performed on this project.
(1) The Mega sampler is a trailer mounted unit fabricated by
personnel of the Region VI Lower Mississippi River Branch Laboratory for
a previous project (ca. 1968-70). It was retrieved for purposes of the
present project from NERC-Cincinnati. The Mega Sampler consists of four
12" X 30" cylindrical PVC columns which can be packed with activated
carbon and connected in series. It is also equipped with two Universal
Pool Products sand pre-fliters for use when turbidity or sediment must
be removed. For this project, however, the sand filters were unnecessary
and were therefore not used. In sampling the finished water at the
Carrollton Water Plant, the only location where the Mega sampler was
used, two of the carbon units were employed; the remaining two were
bypassed. Each of the two columns was packed with activated carbon in
the following manner: the lower strata was a 5" layer of coarse
-------
grade carbon (Pittsburgh Carbon Company Type F-300); the middle strata
was a 12" layer of fine grade carbon (Nuchar C 190 X 30 mesh); and the
top strata was another 5" layer of the coarse grade carbon. The total
weight of carbon in each unit was approximately 22 pounds.
(2) The CAM samplers used in this project were similar to the system
described in Standard Methods for The Examination Of Water And Wastewater.
13th Edition (1971) Method 139B, p.. 264. The carbon unit is a 3" X 18"
Pyrex cylinder which was packed .with approximately 12 ounces of Nuchar C
190 X 30 mesh. In collecting the 70-year equivalent samples, two packed
columns connected in series were used at each plant. In collecting the
10-year and 1-year equivalent samples, only one unit per sample was used.
(3) The Mini-sampler is a miniaturized version of the CAM sampler.
The carbon column is of PVC construction and of sufficient capacity to
contain 70 grams of 14 X 40 mesh activated carbon. As is the CAM sampler,
this unit is equipped with flow measuring and control devices. The Mini-
sampler unit is completely described in the publication The Determination
Of Organics-Carbon Adsorbable In Water, 3rd Edition, by J. K. Carswell,
R. W. Buelow, and J. M. Symons, U.S. EPA Water Supply Research Laboratory
(1972). The actual units used in this project were obtained on loan from
WSRL-Cincinnati.
In all of the carbon adsorption sampling operations described above
and in Table 1, the EPA personnel from the Lower Mississippi River Branch
were assisted by local operating personnel of the respective water plants.
-------
XAD Resin Adsorption. This sampling method, developed by the USAEC-Ames
Laboratory, Iowa, uses 5 cc of a macroreticular synthetic resin (Rohm
and Haas XAD-2) contained in a small glass column. Its dimensions allow
it to be quickly connected to any standard plumbing fixture using non-
contaminating connections. The unit consists solely of the extraction
column and is not equipped with flow control or automatic measuring
devices. Flow rate is obtained by collecting a measured volume in a
known period of time. The water sampling operations with the XAD resin
adsorption units were performed on this project by Mr. Junk of the
USAEC-Ames Laboratory assisted by EPA personnel. For reference to the
XAD resin adsorption method see Junk, G . A., et al., "Use of Macro-
reticular Resins in the Analysis of Water for Trace Organic Contaminants"
J. of Chromatog, 99, p. 745-762, 1974.
Liquid-Liquid Extraction. This sampling technique was adopted by
general agreement among the analysts prior to the initiation of the project
with the intention that it would facilitate the recovery and analysis of
highly volatile organics which it was feared might be lost in the sample
processing procedures associated with the adsorption techniques. At each
plant, triplicate one liter samples of finished water were extracted in
separatory funnels with 2 ml of high boiling solvent, tetralin. The
Immiscible liquid phases were allowed to separate, the water drained
and discarded, and the tetralin recovered into septum vials (Teflon-
lined septums), sealed and delivered to Southeast Environmental Research
Laboratory for analysis. This sampling operation was performed by personnel
of SERL and LMRB.
6
-------
Reverse Osmosis. This semi-permeable membrane water purification
method, while proven effective for certain purposes, is in an as yet
experimental stage of development for use as a solute concentration
method to facilitate trace organics analysis. Its application for this
purpose is undergoing evaluation at the EPA Water Supply Research Labora-
tory which requested its inclusion in the project with sampling performed
by Gulf South Research Institute, New Orleans, Louisiana. No analytical
data from this technique were derived for inclusion in this report.
It should be stated, however, that reverse osmosis is being used at
WSRL primarily as a method to generate samples of organics for toxicity
studies and not as a concentration step for trace organic analysis.
Volatile Stripping. (Volatile Organics Analysis, VOA; Bellar Techni-
que). This relatively direct sampling and analytical technique employs
helium gas stripping of the more volatile organics from a small water
sample with entrapment of organics on a Tenax or Chromosorb 103 column.
This column is then attached to the injection port of a gas chromatograph
or GC/MS instrument. At elevated temperature and with carrier gas flow
through the column, the components are desorbed directly into the analytical
instrument.
Under the direction of the Water Supply Research Laboratory, samples
by this technique were collected from a finished water tap in the CarrolHon
Water Plant on September 23, 1975, by personnel of the Gulf South Research
Institute. Several 50 ml vials specially prepared to eliminate any possible
organic contamination were provided by WSRL. Samples thus collected in a
carefully prescribed manner were pre-chilled in crushed ice and shipped by
-------
air freight to WSRL in Cincinnati for analysis. Details of this
procedure, developed prior to and independently of this project,
are found in the publication entitled "The Determination of
Volatile Organic Compounds at the yg/1 in Water by Gas Chromato-
graphy" by T. A. Bellar and J. J. Lichtenberg, Journal of the
American Water Works Association., December, 1974.
8
-------
Sample Preparation Methods
The description of the sample preparation methods given below is
devoted solely to the process operations which followed the carbon ad-
sorption and XAD resin adsorption methods of sample collection. Samples
collected by liquid-liquid extraction and for volatile stripping required
no further processing prior to analysis beyond that already described.
For reasons mentioned above, the reverse osmosis samples are not considered
further in this report.
Mega Carbon Processing. Immediately following its use in sample
collection at the Carroll ton Water Plant, the Mega sampler was transported
to the EPA-LMRB laboratories where the activated carbon was removed, dis-
tributed in trays in a forced draft convection oven equipped with an
activated carbon air intake filter to prevent contamination from labora-
tory air and dried for 10 days at a temperature of 40°C. After drying,
the carbon was sealed in new, pre-cleaned, five gallon cans and taken
on commerical aircraft to the Robert A. Taft Center in Cincinnati. Using
the large scale permanently installed extraction unit specifically
fabricated for Mega-sampler carbon extraction, the carbon was extracted
for 40 hours with 50 gallons of Analytical Reagent grade redistilled
chloroform. Following the 40 hour reflux extraction, the extract was
concentrated by conversion of the unit to a distillation mode and distill-
ing the excess solvent until a final volume of approximatley 1/2 gallon
remained in the kettle. The concentrated chloroform extract was then
recovered and transported in sealed Teflon bottles to the Southeast
Environmental Research Laboratory for analysis. The Mega carbon pro-
cessing was performed by Mr. Luther Hunt of the LMRB laboratory staff.
-------
CAM Carbon Processing. On removal from the sampling location,
the CAM carbon cylinders were drained of excess water, sealed,
and shipped by commercial air carrier to Oklahoma City where they
were claimed by personnel of the Robert S. Kerr Environmental
Research Center, Ada, Oklahoma, and taken to the center by private
aircraft. The columns were stored at 4°C until carbon processing
could be initiated.
Columns were opened in a special carbon handling room designed
to minimize the potential for contamination. The carbon was trans-
ferred to Pyrex glass dishes and dried at 35-38°C for 48 hours under
a gentle flow of clean air in a mechanical convection oven. The
oven air inlet was equipped with a carbon filter to prevent atmos-
pheric contamination.
The dried carbon was transferred to 2200 ml soxhlet extractors
and extracted for 48 hours with chloroform. The chloroform extracts
were filtered through solvent extracted glass fiber filters to remove
carbon fines and then vacuum concentrated at temperatures not exceeding
27°C in rotary evaporators to final volumes of 30-60 ml. The con-
centrated extracts were transferred quantitatively to 10 ml ampules,
several ampules being required to accommodate each extract. The
ampules were purged with clean, dry nitrogen and sealed while the
contents were held at -50°C in a cold bath. The filled ampules were
maintained under regrigeration (4°C) until shipment by air mail to the
Southeast Environmental Research Laboratory for analysis. CAM carbon
processing was performed by Mr. Roger Cosby under the direction of
Dr. William Dunlap at the Robert S. Kerr Center.
10
-------
Mini Carbon Processing. The exposed Mini-sampler columns, drained
and sealed, were forwarded by air express to the EPA Region VI Houston
Laboratory. There the carbon was removed and dried at 39.5°C for a
period of 48 hours in a convection oven with activated carbon air intake
filter. The dried carbon was transferred to Soxhlet extractors equipped
with fritted glass thimbles and extracted for a period of 48 hours with
spectrophotometric quality chloroform.
Each of the two-month equivalent sample extracts were split in a 1:1
proportion with one portion being evaporated to dryness at 70°C for carbon-
chloroform extract (CCE) residue determination and the other portion reduced
in volume in Kuderna-Danish evaporative concentrators. The reduced volume
concentrates were quantitatively transferred to 25 ml volumetric flasks
and made up to volume with chloroform. The one-day equivalent sample
extracts were not split for residue determination, but the concentrative
evaporation procedures were followed.
The 25 ml extracts were later transferred into vials, sealed, and
shipped to the Southeast Environmental Research Laboratory for analysis.
The Mini-carbon processing was performed by Mr. Medardo Garza of the EPA
Houston Laboratory.
XAD Resin Processing. The XAD resin units, after exposure, were
hand carried or sent by air carrier to Mr. Gregor Junk at Ames, Iowa.
There the samples were extracted with redistilled ethyl ether by Mr.
Junk or his associates in accordance with his procedures. The ether
was dried and concentrated to 1 ml in a micro Kuderna-Danish evaporator
for GC and MS analysis at Ames. One fourth of each extract was carried
to Southeast Environmental Research Laboratory for analysis.
11
-------
Analytical Methods
With the exception of some analyses on resin extracts performed
at Mr. Gregor Junk's laboratory, all analyses performed for this project
were by the Analytical Chemistry Branch of the Southeast Environmental
Research Laboratory, .Athens, Georgia, and the Water Supply Research Labora-
tory in Cincinnati. The methods described below are intended only as a
brief summary of the predominance of analytical work on which the reported
results are based. A more complete description of the work performed at
SERL is given in the Appendix. The work performed at WSRL has been further
developed in the course of the ongoing nationwide drinking water survey and
will undoubtedly be reported in detail in future publications of that group
Southeast Environmental Research Laboratory
GC and GC/HS (Gas Chromatography-Mass Spectrometry)
Gas chromatography was performed using a Varian 1400 GC with a
flame ionization detector. Typical GC conditions were:
Column: 10' x 1/8" i.d. glass
Packing: 3% SP-2100 on 80/100 mesh Supelcon AW-DM
Program: 6 min. initial hold; then from 40 - 280°i
at 6°/m1n.
Carrier Gas: Helium at 20 ml/min.
Sample Size: 2 microliters
For the tetralin extracts, the temperature program was usually a
1 minute hold at 35°C (with the oven door open) followed by an increase
to 210°C at 10°/min.
12
-------
GC/MS instrumentation was a Finnigan 1015 system interfaced via a
Sholke separator to a modified Varian 1400 GC. This system was interfaced
to a Systems Industries System 150 computer for data aquisition, data
storage, data reduction and manipulation. Mass spectrometer electron
voltage was 70.
Some initial GC/MS work was done on a Varian MAT CH5/DF system
interfaced to a Varian 2740 GC via a Watson-Biemann separator and to a
Varian SS-100 Data System. This instrument was later used for confirmation
of the presence of Atrazine in the CarrolIton 70 year CCE by accurate mass
measurement at a resolution of about 5000 amu.
Gas chromatography on these GC/MS systems was performed using a similar
column and conditions to those employed in the GC runs described above.
Mass spectra stored on disks from the Finnigan GC/MS runs were com-
pared via acousticoupler connection with spectra in the EPA-Battelle
computer files at Battelle (Columbus).
Quantitative Analysis
The Perkin-Elmer PEP-1 Data System, interfaced to a Varian 1400 GC
containing a SP-2100 column and operated under the conditions described
above, was used for,computerized quantisation and retention time measure-
ments. Since Atrazine was present in all extracts of New Orleans samples,
it was chosen as an internal standard. A stock solution of 5 parts per
thousand of Atrazine (99.7% pure) in chloroform was the reference for
quantity of all identified compounds for which standards were obtained.
Standards, obtained from the laboratory supply or from commercial sources,
13
-------
were mixed in known concentrations with other standards and with a known
amount of the Atrazine reference stock solution. Mixtures we^e designed
so as to obtain good peak resolution. The Atrazine was assigned a flame
response of 1.000 and, since its concentration was known, the computer
system was able to calculate the flame response, as well as the relative
retention time, of ea^ch standard.
After tentative identification of compounds by GC/MS, a PEP-1 compute*
program was written for the GC-computer run of each extract, allowing the
computer to use the known flame responses to calculate concentrations. In
some cases, the flame response calculated for a standard was also used for
other compounds of the same chemical class. The relative retention times,
calculated for all compounds and printed out by the computer, were then
manually compared with those of the available standards. It was necessary
to dose the blanks with Atrazine as an internal standard, since it was
ascertained that Atrazine was not present in them.
In most cases, mass spectra of the standards was obtained on the
Finnigan GC/MS system for visual comparison with those of the identified
compounds.
Water Supply Research Laboratory
Volative Organics Analysis
In the various analyses performed at the Water Supply Research
Laboratory, the following instruments were used:
1. Perkin-Elmer Model 900 GC
2. Finnigan 1015D - System Industries 150 GC/MS
3. Phase Separations Limited TOC Apparatus
14
-------
The volatile organics were purged from aliquots of the water and
adsorbed on a small column containing either Tenax or Chromosorb 103 as
described by Bellar and Lichtenberg. Qualitative analysis was accomplish-
ed by GC/MS using a Chromosorb 101 column and operating the mass spectro-
meter in the electron impact ionization mode.
Quantitative analysis of the major components of the volatile organics
was accomplished by gas chromatography using the Perkin-Elmer gas chromato-
graph fitted with a 6 foot column of Chromosorb 101 and flame ionization
detectors.
Standards of chloroform and dichloroethane were prepared by intro-
ducing 5 yl and 2.5 yl respectively into one liter of distilled water
with a 5 yl syringe. This was thoroughly shaken until dissolution was
complete. This stock solution was then diluted 100 fold resulting in
concentrations (calculated from literaturs values of the densities) of
78 yg/1 chloroform and 31 yg/1 dichloroethane.
All Head Gas (HG) and VOA samples were injected onto a 6' x 2 mm
ID 50/60 mesh Chromosorb 101 GC column. Some confirmatory analyses
were performed on a hexaneiethyl ether extract (85:15 v/v). These
samples were run on a 8' x 2 mm ID 3% OV-17 GC column. TOC analyses
of the water were performed with the Phase Separations Limited apparatus.
The GC/MS was calibrated according to EPA standard procedures as
given by J. W. Eichelberger, L. E. Harris and W. L Budde, Anal. Chem.
45, 227 (1974).
15
-------
Processing of Blanks
The foregoing discussion of sampling methods, sample preparation,
and analytical procedures has been devoted to the processing of actual
samples. In order to assure that components identified were actually
derived from the original samples and were not artifacts, contaminants,
or inherent components deriving from the sampling method itself, or
from the sampling media, commercial solvents used for extraction, or
from any stage of the sample preparation, it was necessary to process
blank samples taken through all stages of the operations in parallel
with the actual samples, except that there had been no exposure to
the water from the water plant undergoing test.
An exception to the latter statement existed in the development of
a blank for the Mega-carbon sampling and extraction procedure. In the
case of the CAM and Mini-carbon procedures, blanks were developed by
passing distilled water through packed units at the Lower Mississippi
River Branch Laboratory. This carbon was thereafter dried and extracted
and otherwise handled in an identical manner to that of the samples.
However, the large scale of the Mega-carbon unit rendered this procedure
infeasible for the development of a Mega-carbon blank. In this instance,
two of the Mega-carbon adsorption cylinders were packed with carbon in the
t
same manner as previously described for the samples and the unit was
returned to the Carroll ton plant. For blank development, 100 gallons of
Carrollton water was passed through the two units connnected in series.
Following this short-term exposure, carbon from the first unit was
discarded without processing while the carbon from the second unit was
16
-------
dried and extracted in the same manner as the combined carbon from the
primary Mega-carbon sample. This procedure for blank development rested
on the assumptions that the first unit would serve as a pre-filter for
organics adsorption and that the volume of 100 gallons used in wetting
the carbon was insignificant in comparison to the 300,000 gallons used
in the sample development. In each instance of the extraction procedures,
additional solvent blanks were developed separately and independently of
the extraction blanks by taking a portion of the solvent through the
concentration procedure without using it for extracting carbon.
As a consequence of this processing and analysis of blanks, in the
interpretation of analytical results no components could be accepted
as deriving from the finished water samples unless these components were
not present at a signifiqant ieyel in the blanks relative to the samples
or unless they were identified independently in one or more of the other
methods in which they were absent from the corresponding blank.
17
-------
Sampling Method
Carbon Adsorption
with Chloroform
Extraction
Method Modification
Mega Sampler
CAM 70 yr. equiv.
CAM 10 yr. equiv.
CAM 1 yr. equiv.
Mini-Sampler
2 Mo, equiv.
Mini-Sampler
TABLE I
DISTRIBUTION OF WORK OPERATIONS
NEW ORLEANS AREA WATER SUPPLY STUDY
Plants Sampled
Carrol 1 ton
Carroll ton
Jefferson No. 1
Jefferson No. 2
Carroll ton
Jefferson No. 1
Jefferson No. 2
Carroll ton
Jefferson No. 1
Jefferson No. 2
Dates Samples
7/17-24/74
7/18-24/74
7/24-8/2/74
7/24-8/2/74
8/6-7/74
8/6-7/74
8/6-8/74
8/13/74
8/13/74
8/13/74
Carroll ton ,7/30r31/74
Jefferson No. 1 7/31-8/1/74
Jefferson No. 2 .8/1-2/74
Carroll ton (repeat) 8/6-8/74
Carroll ton 8/6/74
Jefferson No. 1 8/6/74
Jefferson No. 2 8/6/74
Sampled By_
LMF
LMF
LMF
LMF
LMF
LMF
LMF
LMF
LMF
LMF
LMF
LMF
LMF
LMF
LMF
LMF
LMF
Water Volume Sample Prepared By Analysis
300,000 Gals.
LMF
SERL
6,750 Gals.
6,759 Gals.
6,707 Gals.
' 963 Gals
965 GiH,
1,300 Gals.
/
74 Gals.
90 Gals.
97.5 Gals.
62 liters
65 liters
60 liters
58 liters
1 liter
1 liter
1 liter
RSKERL
RSKERL
RSKERL
RSKERL
RSKERL
R5KHL
RSKERL
RSKERL
RSKERL
HNF
HNF
HNF
WSRL
HNF
HNF
HNF
SERL
SERL
SERL
Stored in
sealed vials
'at RsKtRL-
Stored in
sealed vials
at RSKERL
HNF-SERL
HNF-SERL
HNF-SERL
WSRL
HNF-SERL
HNF-SERL
HNF-SERL
-------
Sampling Method
XAD Resin Adsorption
with Ethyl Ether
Extraction
Liquid-Liquid Contact
Extraction
Reverse Osmosis
Volatile Stripping
Method Modification
Developed by the
USAEC-Ames Laboratory
Tetralin Solvent
Cellulose Acetate
Membrane in Series
with Dupont Perniasep
Membrane
Bellar Technique for
Volatile Organics
Analysis (VOA)
TABLE I (CONTINUED)
DISTRIBUTION OF WORK OPERATIONS
NEW ORLEANS AREA WATER SUPPLY STUDY
Hants Sampled Dates Sampled Sampled By Water Volume Sample Prepared By Analysis By
Carroll ton
Jefferson No. 1 .
Jefferson No. 2-~
Car roll ton
Jefferson No. 1
Jefferson No. 2;
Carrol Hon
Carroll ton
7/30-8/1/74
7/30-31/74
7/31-8/1/74
7/31/74
7/30/74
8/7-9/74
•9/23/74
LMF-Junk
LMF-Junk
LMF
'i ' »
SERL-LMF
SERL-LMF
LMF .,
GSRI
GSRI
318 liters
365 liters
275 liters
Junk-SERL
Junk-SERL
Junk-SERL
3 ea. X 1 liter SERL
3 ea. X 1 liter SERL
3 ea. X 1 liter' SLRL
Approx. 400 Gals. WSRL
SO ml vials USRL
SERL
SERL
SERL
SERL
SERL
WSRL
WSRL
Key to abbreviations used in Table I
LMF: Lower Mississippi River Facility '(Region VI EPA) Slidell, Lou}siana _ _
SERL: Southeast Environmental Research laboratory (EPA; .JERC-Corvallis) Athens, Borgia
RSKERL- Robert S. Kerr Environmental Research Laboratory (EPA; NERC-Corvallis) Ada, Oklahoma
WSRL- Water Supply Research Laboratory (EPA; NERC-Cincinnati) Cincinnati Oh10
GSRI- Gulf South Research Institute, New Orleans, Louisiana
HNF: Houston Facility (Region VI, EPA) Houston, Texas
19
-------
ANALYTICAL RESULTS
The trace organic compounds or organic isomers of undetermined
specific structure which were identified by one or more methods in samples
derived from the finished water at the Carroll ton Water Plant, Jefferson
Parish No. 1 Water Plant, and Jefferson Parish No. 2 Water Plant are listed
in Table 2. Supporting data for these identifications exist at the Water
Supply Research Laboratory (Cincinnati, Ohio) or at the Southeast Environ-
mental Research Laboratory (Athens, Georgia). Compounds are listed in
Table 2 in the alphabetical order of their capitalized letter. Each compound
is numbered in the order of its listing. Any reference to a compound in
this discussion will be by its assigned number except where use of the name
may be necessary or desirable to clarify the selected nomenclature.
No effort was made to employ a consistent nomenclature system in this
list of compounds. The name selected for a specific compound may be the
name most commonly used for it or by which it might be most readily
recognized. For example, compound 2 is called acetone although it might
also be correctly named dimethyl ketone or propanone. In some instances
the common name by which a compound is preferentially listed is followed
in parentheses or brackets by the name of the compound in the IUPAC (In-
ternational Union of Pure and Applied Chemistry) nomenclature system. This
system provides detailed information on molecular structure not given by
most common chemical names. Where this style of listing was employed, it
was done primarily to enable the reader to discern similarities or slight
differences in certain compounds appearing in the list, as for example,
compounds 3 and 14 or compounds 70 and 71.
20
-------
It is not intended that the name given for compound 4 (Alachlor,
one chlorine homolog) be taken as a legitimate compound name. Rather,
it is a designation used for an identified compound for which a satis-
factory common name was not found but which was determined by mass
spectral evidence to have a homologous relationship to Alachlor, com-
pound 3. It was determined that compound 4 was identical to compound 3
except for the inclusion of an additional chlorine atom probably bonded
to the same carbon atom as the single chlorine atom in compound 3. How-
ever, due to the lack of an available standard of compound 4 for absolute
confirmation of the identification, use of the probable IUPAC name for
compound 4 could not be justified.
Likewise, a close relationship exists between compound 8 (Atrazine)
and compound 9 (designated as Atrazine, deethyl). Compound 8 contains
an ethylamino group bonded to the fourth atom of the six membered
triazine ring. In compound 9 this position contains simply an ami no
group. The designation for compound 9 is thus intended to express its
similarity to Atrazine with the absence of the ethyl group. In this
instance, however, inclusion of the IUPAC name was justified as a
standard was obtained for confirmation.
Compound 82, now named as 1, 3, 5-trimethyl-2, 4, 6-trioxo-hexah.ydro-
triazine, was originally named in the draft report as trimethyl-trioxo-
hexahydrotriazine isomer. The present structure has been confirmed by
comparison with a standard. The compound having the structure as
presently given may also be called the trimethyl ester of isocyanuric
acid or the trimethyl ester of tricarbonimide.
21
-------
Where the name given for a compound is followed by the term isomer,
manual or computerized interpretation of the mass spectral data did not
permit the analysts to determine which one of more than one possible
molecular isomers bearing that name was present. Compounds 5, 6, and 7
are designated as alkylbenzene - Cg, 63, and C^ isomers, respectively.
There are four possible fy isomers of an alkylbenzene. Of these, two
are located elsewhere irNthe list as compound 51, ethylbenzene, and
compound 86, o-xylene. Compound 5 must then be m-xylene, p-xylene, or
a mixture of the two. There are three alkylbenzene - C3 isomers located
elsewhere in the list. Compound 52 is o-ethyltoluene; compound 53 is
m-ethyltoluene; and compound 81 is designated as trimethylbenzene isomer.
Compound 6 is thus most likely p-ethyltoluene or n-propylbenzene. There
are no alkylbenzene - C4 isomers other than compound 7 located elsewhere
in the list.
Given for most compounds in Table 2 is a quantitative value repre-
senting the "highest measured concentration" in micrograms per liter
(parts per billion). With the exceptions of compounds 22 and 34 (which
were determined by Volatile Organics Analysis directly from the water
medium) all concentration values were obtained through quantitative
GC/MS analyses of processed carbon-chloroform extracts or XAD resin-
ethyl ether extracts related back to the water medium from which the
samples were derived. This concentration relationship was possible
since the volumes of water passing through the carbon or resin units
were known. Therefore an expressed concentration value from any com-
ponent listed in Table 2 refers to concentration of that component in
the water medium.
22
-------
In order to express exact concentration values in the water,
however, it would be necessary to know the recovery efficiency values
for every stage of the sample collection, preparation, and analytical
process. That is, one would need to measure with standards for each
compound the efficiency of the carbon or resin adsorption from water,
component losses through volatility incurred in drying the carbon, the
efficiency of desorption by solvent extraction from the carbon, and losses
incurred in concentrating the solvent to a low volume for analytical pro-
cessing. For this project, determination of these efficiencies could be
considered an impossible, or at least infeasible, task. Consequently,
it is emphasized that the concentration values reported must not be inter-
preted as absolute concentration levels present in the water, but simply
represent the highest concentration values measured by the analysts. The
term "highest" is used because when values determined by two or more
different methods differed to any extent, the higher or highest of the
values was reported in the tabulation. All values are retained in the
analysts' records. The reader who is familiar with the Draft Analytical
Report for this project, previously released, will note that values for
a number of compounds are significantly higher than previously reported.
This is due to the fact that higher values for these compounds were found
on analysis of the 2-month mini-sample extracts which had not been performed
at the time of the earlier report.
It is also recognized that all of the specific organic compounds
identified and reported herein were, in the final stage of analysis,
23
-------
determined through some modification of gas chromatography. For a
compound to be analyzed by this technique, it must possess some degree
of volatility under the conditions of analysis. Consequently, non-
volatile organic substances would not have been detected under the
analytical conditions employed. The compounds listed in Table 2
represent those compounds which have been identified and quantitated
with a reasonable degree of certainty by the analysts from the samples
presented to them for analysis.
24
-------
TABLE 2
ORGANIC COMPOUND IDENTIFICATIONS
NEW ORLEANS AREA WATER SUPPLY STUDY
1
2
3
4
5
6
7
8
Compound
Acetaldehyde
Acetone
Alachlor *_
[2-chloro-2',6'-
diethyl-N- (methoxy-
methyl)acetanilide]
Alachlor, one chlorine
homolog
Alkylbenzene - Co
isotner
(m-Xylene or p-Xylene)
Alkylbenzene - C^
isomer
(p-Ethyltoluene or
n-Propylbenzene)
Alkylbenzene • C,
isomer
Atrazine *
(2-chloro-4-ethyl-
amino- 6- isopropy 1-
amino-js- triazine)
Highest Measured Concentration
yg/1 (ppb)
Carrollton
Water Plant
D-VOA
D-VOA
1.9
1.7
7.5
2.4
< 0.1
4.9
Jefferson #1
Water Plant
NE
NE
2.9
P
5.6
1.9
ND
5.2
Jefferson #2
Water Plant
NE
NE
2.1
0.05
6.2
2.2
ND
5.4
25
-------
TABLE 2 (Continued)
ORGANIC COMPOUND IDENTIFICATIONS
NEW ORLEANS AREA WATER SUPPLY STUDY
*
10
11
12
13
14
15
16
17
Compound
Afcrazine, 4^et%i*r :
(2-chloro-4-fcamino-
6- isopropy lamino-
£-triazine)
Benzaldehyde *
Benzyl butyl
phthalate *
Bromodichloromethane
Bromof orm *
Butachlor *
diethyl-N* (butoasy-
methyl) acetanilide]
Butanone
Butyl octyl
maleate *
Carbon disulfide
Highest Measured Concentration
V8/1 (ppb)
Carrol Iton
Water Plant
. \
0.78
0.03
1.4
<0.1
0.57
0.05
D-VOA
D-RE(J)
D-VOA
Jefferson #1
Water Plant
0.80
ND
1.8
<0.1
ND
0.06
NE
D-RE(J)
NE
Jefferson #2
Water Plant
0.75
ND
" 1.6
ND
ND
0.05
NE
ND
NE
26
-------
TABLE 2 (Continued)
ORGANIC COMPOUND IDENTIFICATIONS
NEW ORLEANS AREA WATER SUPPLY STUDY
18
19
20
21
22
23
24
25
26
Compound
,
Carbon tetrachloride
a-Chlordane
Chlordene *
bis (2-Chloroethyl) -
ether *
Chloroform *
bis (2-Chloroisopropyl]
ether * .
m-Chloronitrobenzene *
Cyariazine *
[2- (A-chlorO'6-ethyl-
amino-jS-triazin-2-
ylamino)-2-methyi-
propionitrile ]
DDE *
[ 2 ,2-bis (p-chloro-
phenyl)-l,l-dichloro-
ethylene]
Highest Measured Concentration
Carroll ton
Water Plant
D-VOA
<0.1 (T)
<0.1 (T)
O.04
133 (a)
0.18
D-.REXJ)
0.35
o.osGJ)
Jefferson #1
Water Plant
NE
ND
ND
0.16
NE
0.08
D-RE(J)
0.21
0.05(0)
Jefferson #2
Water Plant
NE
ND
ND
0.12
NE
0.03
ND
0.31
ND
.27
-------
TABLE 2 (Continued)
ORGANIC COMPOUND IDENTIFICATIONS
NEtf ORLEANS AREA WATER SUPPLY STUDY
w
28
*
29
30
31
32
33
34
35
36
37
Compound
r*t. '- -f ' 1 •'' - ' -"/-.
- fi-Decane *
Decane, branched
isomer
Dibromochloromethane *
Dlbromodlchloroethane
isomer
Dibutyl phthalate *
2,6-Di-t-butyl-p-
benzoqulnone *
tt-Dichlorobenzene *
1 , 2-Dichloroe thane
Dich loroiodomethane
Dichlorome thane
Dicyclopentadiene *
Highest Measured Concentration
yg/1 (ppb)
Carrol It on
Water Plant
2.4
5.8
1-1
0.33
0.10 -
0.22
<3
8 (a)
1.1
D-VOA
ND
Jefferson #1
Water Plant
iio
5.4
0.4
ND
0.36
\
0.21
<0.1
NE
1.3
NE
D-RE(J)
Jefferson #2
Water Plant
2.0
5.2
0.06
0.63
0.23
0.25
ND
NE
1.5
NE
ND
28
-------
TABLE 2 (Continued)
ORGANIC COMPOUND IDENTIFICATIONS
*NEW ORLEANS AREA WATER SUPPLY STUDY
38
39
40
41
42
43
44
t
45
46
47
48
Compound
Dieldrin *
Diethyl phthalate *
Di-(2-ethylhexyl)
adipate *
Di-(2-ethylliexyl)
phthalate *
Dihexyl phthalate
Dihydrocarvone
Diisobutyl phthalate *
Dirarthyl phthalate *
Dipropyl phthalate *
n-Dodecane *
Endrin *
Highest Measured Concentration
yg/1 (ppb)
Carrollton
Water Plant
0.04
0.24
0.10
11
0.05
0.14
0.59
0.60
0.07
0.10
0.004
(T)
Jefferson #1
Water Plant
0.07
0.10
ND
0.50
ND
0.06
ND
0.82
0.13
0.40
0.008
(T)
Jefferson #2
Water Plant
0.05
0.18
ND
1.2
0.16
0.07
ND
0.74
0.14
0.37
0.006
(T)
29
-------
TABLE 2 (Continued)
ORGANIC COMPOUND IDENTIFICATIONS
NEW ORLEANS AREA WATER SUPPLY STUDY
•
49
50
51'
52
53
54
55
56
57
58
59
Compound
Ethanol
Ethyl acetate
Ethylbenzene
o-E thy 1 toluene *
m-Ethyltoluene *
1,2,3,4,5,7,7-
Heptachloro-
norbornene *
Heptachloronorbornene
isomer
Hexachlor o-l,3-
butadiene *
Hexachloroethane *
Isophorone *
Limonene • *
Highest Measured Concentration
yg/1 (ppb)
Carroll ton
Water Plant
D-VOA
D-RE
2.3
ND
ND
0.06
0.06
0.70
4.3
2.9
0.03
Jefferson #1
Water Plant
NE
ND
1.6
0.04
0.05
0.07
0.04
0.27
0.19
2.8
ND
Jefferson #2
Water Plant
NE
D-RE
1.8
0.02
0.02
0.07
0.04
0.21
0*30
9.5
ND
30
-------
TABLE 2 (Continued)
ORGANIC COMPOUND IDENTIFICATIONS
NEW ORLEANS AREA WATER SUPPLY STUDY
60
61
62
63
64
65
66
67
68
69
1 !
Compound
Methanol
Methyl benzoate *
Methylnaphthalene *
Naphthalene *
. — _ — . — , — ,. ,
3-Methylbutanal
2-Methylpropanal
^ ....... .
n-Nonane *
Pentachlorpethane
Pentachlorophenyl
methyl ether
n-Pentadecane *
Highest Measured Concentration
yg/1 (ppb)
Carrollton
Water Plant
D-VOA
• ND
ND
D-RE(J)
D-VOA
D-VOA
2.4
<0.1
<0.1 (T)
0.03
Jefferson #1 Jefferson #2
Water Plant Water Plant
NE
<0.01
D-RE(J)
D-RE(J)
NE
NE
2.4
ND
ND
0.10
NE
ND
ND
ND
NE
NE
2.1
ND
ND
0.10
31
-------
TABLE 2 (Continued)
ORGANIC COMPOUND IDENTIFICATIONS
MEW ORLEANS AREA WATER SUPPLY STUDY
70
71
72
73
74
75
76
77
Compound
Propazine *
[2-chloro-4,6-bi8
(leopropy lamino) -.§-
triaiclnej
Simaxlne *
• [2»chloro-4;6-biB
(ethy lamino) '_|-
triazinej ~"
1, 1,1,2- Tetr«-
chloroethant*
Tetrachloroethylene *
n-Tetradecane *
Toluene *
l,l»2-Trlchloro-
ethane *
l»l»2-Trichloro-
cthylenfe
Highest Measured Concentration
yg/1 (ppb)
Carrollton
Water Plant
<0.1
<0.1
0.11
<1
0.10
11
6.2
D-VOA
Jefferson #1
Water Plant
<0.1
<0.1
ND
0.20
0.10
7.1
8.5
NE
Jefferson #2
Water Plant
<0.1
<0.1
ND
0.20
0.12
12
6.4
NE
32
-------
TABLE 2 (Continued)
ORGANIC COMPOUND IDENTIFICATIONS
NEW ORLEANS AREA WATER SUPPLY STUDY
78.
79
80
81
82
83.,
84
85
86
Compound
1,1,1-Trichloro-
propane
1,2,3-Trichloro-
propane
n-Tridecane *
Tritnathylbenzene
isomer
1,3,5-Trimethyl-
2,4,6-trioxo-
hexahydrotriazine *
Triphenyl phosphate *
n-Undecane *
Undecane, branched
isomer
o-Xylene *
Highest Measured Concentration
P9 PP
Carrollton
Water Plant
<0.1
<0.2
0.30
6.1
0.01
0.12
2.5
5.3
4.1
Jefferson #1
Water Plant
ND
ND
0.17
5.1
ND
ND
<10
ND
-2.3
Jefferson #2
Water Plant
ND
ND
0.20
5.3
ND
ND
2.1
ND
-3.4
33
-------
KEY TO SYMBOLS USED IN TABLE 2
* While all compounds listed in Table 2 were identified
by one or more methods, those marked with this symbol
gained added confirmation by gas chromatography re-
tention time match with an available standard of the
compound. Compounds 38 and 48 were further con-
firmed by gas chromatography retention time match on
two additional columns of varying polarity.
a The quantitative values for compounds 22 and 34 were
obtained on Volatile Organics Analysis by comparison
with standards of known concentration at the Water
Supply Research Laboratory. Compound 22 was detected
but not quantified in Tetralin extracts of CarrolHon
water at Southeast Environmental Laboratory, but not
in Tetralin extracts of Jefferson No. 1 or Jefferson
No. 2, waters which were not examined by the Water
Supply Laboratory. The Southeast Laboratory did not
detect compound 34 1n any of its processed samples.
0-VOA These compounds were detected by Volatile Organics
Analysis only as performed by the Water Supply
Research Laboratory. Quantitative values were not
obtained except for compounds 22 and 34 as indicated
above. Only the Carroll ton water was examined by this
method.
NE Not examined. This symbol is used for some compounds
reported by the Water Supply Research Laboratory
and not detected by the Southeast Environmental
Laboratory. It is used exclusively under the headings
for Jefferson No. 1 and No. 2, waters which were not
examined by the Water Supply Laboratory.
NO This symbol means the compound was not detected in a
specific water by any of the methods employed. It
applies mostly under the Jefferson No. 1 and No. 2
samples since the Mega-sample, obtained only at the
Carroll ton Plant, contains some compounds not detected
by the other sampling techniques.
D-RE This symbol means the compounds were detected, but not
quantified, only in XAD resin extracts in samples
analyzed at the Southeast Environmental Reserach Labor-
atory.
D-RE(J) These compounds were detected but not quantified
(except for DDE) In XAD resin extracts analyzed by
Gregor Junk. Where the (J) symbol is used, the com-
pounds were not detected by resin extracts analyzed
at the Southeast Environmental Laboratory.
34
-------
P Present, but not quantitated.
(T) Detected only by Thruston at the Southeast
Environmental Laboratory by GC/MS after
fractionation by column chromatography.
35
-------
PROJECT SUMMARY AND EVALUATION
The sampling program as originally assigned to the Lower Mississippi
River Branch was completed on schedule 1n raid-August, 1974. Sampling for
volatile organics analysis, however, was performed in September by
personnel of the Gulf South Research Institute under direction of the
Mater Supply Research Laboratory, \6enera11y the time required in process
Ing samples for analysis prevented getting samples to the analysts before
*,
mid-August to early September. Thus to meet the initial schedule for
presentation of analytical results, the analysts had only eight to ten
weeks to perform extremely complex and demanding analyses while maintain-
ing precise control over sample integrity and adhering to scientifically
defensible techniques. It 1s a tribute to their abilities and dedication
to the established schedule that preliminary results were available for
presentation 1n the Draft Analytical Report released by Region VI on
November 8, 1974. Since that date, the analytical work for which committ-
ments were obtained has been completed and further work in the nature
of confirming, adding to, or revising, to some extent, previous results
have been accomplished. The present report represents the completion
of this project as assigned.
The previous Draft Analytical Report listed 66 organic compounds
identified with some degree of certainty in one or more of the three
waters sampled. The present report lists 86 such compounds including
64 of the original 66, although some compounds previously listed as
isomers of undetermined specific structure may have now been given
precise names following confirmation of structure. Of the original 66
36
-------
compounds listed, two were dropped from the list, these being one
alkylbenzene-C3 isomer and one undecane branched isomer. Other
alkylbenzene isomers which may appear to be missing in comparison
with the earlier list are among those compounds appearing elsewhere
under more precise names (see Analytical Results section). In the
updated list of Table 2, there are thus 22 new identifications. In
addition to the 86 compounds listed, mass spectra were obtained for
eight additional compounds, but the analysts were unable to interpret
the spectra with sufficient certainty for precise compound identifica-
tion.
There is no claim presented here, either stated or imp.lied, that
the compounds listed represent all or any known proportion of the
compounds which may be present in the water at any given time. While
the sampling and analytical methods used in this study are among the
most versatile available, their limitations are acknowledged without
apology. There are at least two relatively simple methods which are
designed to provide a gross measurement of organics in water at levels
which might occur in drinking water. These two methods were performed
at the Water Supply Research Laboratory on Carroll ton water samples
or extracts during the course of this study. One of these methods
is Total Organic Carbon (TOC), an instrumental measurement of elemental
f
carbon in organic form. While several commercial instruments based
on various techniques are available for accomplishing this measurement,
- • .••• • -'
using a Phase Separations Limited TOC apparatus, the WSRL reported to
this author a value of 2.5 mg/1 TOC in a water sample obtained at the
37
-------
Carrollton plant on August 9, 1974. Other values obtained for TOC
using an Oceanography International instrument at WSRL on 5 replicate
sealed ampules of Carroll ten Water Cdate sampled September 23, 1974)
were 2.4 +_0.2 mg/1 (personal correspondence with Dr. Robert Melton,
October 23, 1974). Another method for obtaining a gross measurement
for organics in drinking water 1s the carbon-chloroform extract residue
method. TRtS metHM ttfct beert referenced earlier under the description
of the CAM samplers 1n the experimental section. Using a Mini-sampler
modification of this method as developed at WSRL, that laboratory ob-
tained a value for CCE of 0.3 mg/1 in a Mini-sample extract sampled
8/6"8/74 at th« Carroll ton plant. On the Mini-sample extract from
Carroll ton on 7/30-31/74, Mr. Mike Garza of the Region VI Houston
laboratory obtained a value of 0.23 mg/1 for CCE; for the Jefferson
No. 1 Mini-sample extract of 7/31-8/1/74, he obtained a value of
0.31 mg/1; and fro* the Jefferson No. 2 Mini-sample extract of 8/1-
2/74 he obtained a CCE value of 0.46 mg/1. The CCE residue method
would reflect a gross mass measurement of organics which were relatively
non-volatile, adsorbable on carbon, and extractable Into chloroform.
There 1s generally no attempt at Identification of these organics.
Using the data presented In the paragraph above in comparison with
the concentration data presented in Table 2, the Interested reader may
attempt to satisfy for himself what proportion of organics have been
accounted for herein. However, recognizing the limitations of the
various methods and the fact they they are only superficially related,
38
-------
this author hesitates .to present such an estimate for fear that
such speculation would only be misleading.
Committments were not obtained for the analysis of all samples
collected and processed. Those samples on which no analytical work
was performed included the 10-year and 1-year equivalent CAM samples
and the 1-day equivalent Mini-samples. The CAM extracts were retained
under refrigeration at the Robert S. Kerr Environmental Research
Center while the 1-day Mini-sample extracts are located at SERL.
The predominance of the data presented herein has been derived
from the carbon adsorption sampling methods and the volatile organics
analysis (VGA) technique. These two methods appear to complement
each other very well. The latter technique appears to be best suited
to the types of compounds that would be lost in the sample processing
methods employed with carbon adsorption. Obviously, one of the major
components detected by the VOA method, chloroform, would have been
completely undetectable in the carbon methods where chloroform is
used as a carbon extractant. The XAD resins, while not as productive
of analytical data in this study, nevertheless deserve further com-
parison with carbon adsorption under more closely controlled conditions,
The reverse osmosis methods, not a source of data in this project,
are receiving further study at the research level as a potential
technique of organic concentration for analytical purposes. Direct
liquid-liquid extraction with tetralin was also a disappointing
source of data although some compound confirmations were obtained
from this source. While some comparisons of sampling and analytical
39
-------
methods naturally accrued from this study, it was not a primary
objective of the study to make such comparisons; therefore, it would
be a mistake to make firm conclusions regarding any of the methods
presented since strict sampling controls and precisely duplicated
conditions were not applied.
No judgement has been presented here regarding the possible
source of the identified compounds or their public health significance.
This project was initiated and conducted solely as an analtyical
study and the results have been presented accordingly. Questions
relating to the significance of the findings are now the subject of
further concern at the National level within EPA and other groups,
both public and private. The current status of these ongoing studies
is unknown to this author, and, in any event, would be beyond the
scope of this report.
40
-------
New Orleans Area Water Supply Study-
Analysis of Carbon and Resin Extracts
Prepared and Submitted to the
Lower Mississippi River Branch
Surveillance and Analysis Division
Region VI
U.S. Environmental Protection Agency
by the
Analytical Chemistry Branch
Southeast Environmental Research Laboratory
U.S. Environmental Protection Agency
Athens, Georgia 30601
June, 1975
Appendix A
-------
SUMMARY
Eighty-two organic compounds were tentatively
identified by gas chromatography-mass spectrometry in one or
more of the three New Orleans area finished drinking water
supplies. Fifty-five of these identifications have been
confirmed. Concentrations range from 0.004 pg/1 to 12 pg/1.
A-l
-------
ACKNOWLEDGEMENTS
Larry Keith, Mike Carter, Frank Allen, Terry Floyd,
John Pope, Al Thruston, and Wayne Garrison of the Analytical
Chemistry Branch, (ACB), SERL, spent about 20 man-months
($44,000) on this project between July 1, 1974 arid April 30,
1975. The work was conducted as technical a^3P.^stance to
Region VI, and regularly scheduled research tasks involving
the above personnel were curtailed as a result of their
participation.
The administrative assistance and technical advice of
the Branch Chief, William Donaldson, is appreciated. The
participation of Gregor Junk, USAEC—Ames Laboratory, Iowa
State University, Ames, Iowa, in the collection, extraction,
and analysis of the resin adsorption samples was very
helpful.
EXPERIMENTAL
SAMPLE HANDLING, EXTRACTION, AND FRACTIONATION
Descriptions of the New Orleans samples and
corresponding extracts analysed bv the ACB are contained in
Table 1,
Carbon Adsorption Samples
All carbon adsorption sampling was done by EPA Region
VI personnel. The carbon was processed at other
laboratories (Table 1); ACB personnel received only the
carbon chloroform extracts (CCE's), which were refrigerated
until analysis. Volumes of 70-year and mega-sanrple extracts
were adjusted by evaporation in Kuderna-Danish evaporators
so that 1 ul corresponded to 1 liter of water passed through
the carbon filter. Blank extract volumes were adjusted
accordingly. Mini-sample extracts were evaporated to 0.3 ml
so that 1 ul corresponded to 100 ml of water. Although a
number of fractionation and separation procedures were
applied to the carbon chloroform extracts, as described
below, most information was obtained by direct GC and GC-MS
(gas chromatography-mass spectrometry) without pre-
treatment.
A-2
-------
Tabla l. Description of Now Orleans Drinking Water Extracts>
Extract
Mega-sample CCE*
Mega-sample blank CCE
70-year filter fl, CCE
#2, CCE
#1, CCE
#2, CCE
fl/CCE
#2, CCE
blank CCE
filter *1, CAE***
blank CAE
Mini-sample CCE
Sample Site, Water Equivalent, and Dates Sampled
Carrollton WTP** 300,000 gal. . 7/17-24/74
Carrollton WTP
ti
Jefferson No. 2 WTP
H
Jefferson No. 1 WTP
it
Carrollton WTP
Carrollton WTP
n
Jefferson No. 2 WTP
n
Jefferson No. 1 WTP
25,500 liters
it
25,500 liters
1 liter
60 liters
1 liter
60 liters
1 liter
60 liters
7/18-24/74
7/24-8/2/74
8/6/74
7/30-31/74
8/6/74
8/1-2/74
8/6/74
7/31-8/1/74
Disposition
Extracted at AWTRL, Cinn. by Region
VI personnel. Extracts were hand-
carried to SERL in Teflon bottles.
"70-year" .samples correspond to the
amount of water a man drinks in 70
years at 1 liter/day. Extracted by
Bill Dunlap at the RSKERL in Ada, Okla.
and shipped to SERL in sealed glass
ampules. The f2 filters were in
series with the fl filters.
Mini-samples were extracted by Mike
Garza, Region VI, Houston, and shipped
to SERL in glass vials. Garza used
one-half of the 60 liter extracts.
The 1 liter extracts were not
examined.
blank CCE
XAD Resin, ether extract
, blank extract
Carrollton WTP
Jefferson No. 2 WTP
Jefferson No. 1 WTP
318 liters 7/30-8/1/74
275 liters 7/31-8/1/74
365 liters O/30-31/74
\7/31-8/1/7 4
Samples were collected and resin
extracted by SERL and USAEC, Ames
personnel.
Tetralin extract
, blank
Carrollton WTP 1 liter 7/31/74
Jefferson No. 2 WTP 1 liter 7/31/74
Jefferson No. 1 WTP 1 liter 7/30/74
Samples were extracted by SERL
personnel in New Orleans — see
Experimental Section.
*Carbon chloroform extract — obtained by chloroform extraction of carbon through which water sample had passed.
**Water treatment plant — samples were collected from a tap at; the plant after treatment.
***Carbon alcohol extract — obtained by extracting carbon filter #1 with ethanol after chloroform extraction.
-------
Aliquots of the mega-sample CCE and its blank were
fractionated by thin layer chromatography (TLC) to determine
whether an initial fractionation could improve GC separation
of components. This procedure is especially good for
isolating the polynuclear aromatic class of hydrocarbons.
Aliquots of the extracts were spotted on silica-gel TLC
plates and developed with benzene for 10 cm. Hydrocarbons
had previously been found to migrate to the 8-10 cm region
of these plates, polynuclear aromatics to the 6-8 cm region,
and more polar aromatic to the 0-6 cm region. These regions
of adsorbent were scraped off the plates and eluted with
chloroform. The eluates were evaporated to an appropriate
volume for analysis by GC and GC-MS.
Aliquots of the mega-sample CCE, the Carrollton 70-year
CCE, and their respective blanks were analyzed by the EPA
organochlorine pesticide method. l This procedure involves
fractionation by column chromatography with detection by
electron capture GC. The presence of dieldrin and endrin,
as determined by this technique,, was confirmed by matching
GC retention times with standards on two other GC columns of
varying polarity. Further confirmation, as well as
detection of chlordene, a-chlordane, and pentachlorop.»enyl
methyl ether, was obtained by GC-MS analysis of the
appropriate fractions.
The Jefferson #1 and Jefferson *2 CCE's were also
analyzed by electron capture GC for the detection and
measurement of endrin without prior fractionation.
Other attempts were made to improve GC peak separation
by various types of fractionation. Steam distillation
neatly separated the mega-sample CCE into volatile
distillables and non-distillables, but did little to improve
actual peak separation without major adjustment of GC
conditions. Solubility extraction of the CCE into strongly
and weakly acidic, basic, and neutral fractions served only
to show that most of the compounds are neutral.
The carbon alcohol extract (CAE) gave only a few, poorly
resolved peaks when examined directly by GC. Aliquots of
the extract and of the corresponding blank were evaporated
just to dryness and methylated with diazomethane by a
standard procedure 2, since most of the components were
expected to be highly polar materials, including carboxylic
acids and phenols. The methylated extracts were made to
appropriate volumes and analyzed by GC and GC-MS.
A-4
-------
Resin Adsorption Samples
Gregor Junk, USAEC—Ames Laboratory, Iowa State
University, Ames, Iowa, assisted ACB personnel in collecting
JTesin adsorption samples in New Orleans. Rhom and Haas XAD-
2 resin was cleaned and placed in glass columns according to
Junk's established technique. 3 The columns were connected
to finished water lines at the three New Orleans treatment
plants and samples were collected for about 24 hours
according to the schedule in Table 1. One Jefferson No. 1
resin sample was taken to Ames by Junk and the other samples
were removed from the water lines and mailed to Junk by
Region VI EPA personnel.
The resin samples and corresponding resin blank were
extracted with ethyl ether according to Junk's technique as
soon as they arrived in Ames. The blank consisted of an
equal amount of resin of the same batch used in the sample
columns. The ether extracts were dried and concentrated to
1 ml in a micro Kuderna-Danish evaporator for GC and GC-MS
analysis at Ames. One-fourth of each extract was taken to
SERL for GC and GC-MS analysis. Volumes were adjusted by
evaporation with a stream of dry nitrogen so that 1 yl
corresponded to 1 liter of water passed through the resin.
Tetralin Extractions
Triplicate 1-liter samples of finished water from each
of the New Orleans treatment plants were manually extracted
with tetralin (1, 2, 3, 4-tetrahydr©naphthalene). The first
two samples were extracted by SERL personnel in New Orleans.
The Jefferson No. 2 sample was extracted by Region VI EPA
personnel and mailed to SERL. A 2-ml portion of Matheson,
Coleman and Bell practical grade tetralin was used to
extract each sample, using the same tetralin-rinsed 1-liter
separatory funnel for all samples and the blank. The
tetralin had been checked by GC to assure presence of enough
"window" area to observe early-eluting volatile components
in the water extract. After standing for 15 minutes, the
water was drained slowly from the separatory funnel, leaving
1.0 to 1.2 ml of the tetralin extract. The blank consisted
of 2 ml of tetralin that had been combined with 2 ml of
distilled water and used to rinse the separatory funnel
prior to sample extractions. All extracts and the blank
were sealed in serum vials with teflon lined septum caps.
Septum caps were replaced after each syringe puncture. The
volume was not adjusted prior to GC or GC-MS analyses.
Extracts were stored in a refrigerator.
A-5
-------
IDENTIFICATION METHODS
Gas chromatography was performed using a Varian 1400 GC
with a flame ionization detector. Typical GC conditions
were as follows:
Column: 10 ft x 1/8 in. i.d. (1/4 in. o.d.) glass
Packing: 3% SP-2100 on 80-100 mesh Supelcon AW/DMCS
Program: 6 rain initial hold? then from 40°—280°
at 6 °/min.
Carrier gas: helium at $0 ntl/min.
Sample size: 2 yl
For the tetralin extracts, the temperature program was
usually a 1 rain hold at 35 ° (with the oven door open)
followed by an increase to 210 ° at 10 Vmin.
A Finnigan 1015 mass spectrometer interfaced via a
Gohlke separator to a modified Varian 1400 GC was used for
GC-MS analysis. The GC-MS was interfaced to a System
Industries System 150 computer for data acquisition, data
storage, and data reduction and manipulation. Ionizing
voltage was 70 eV.
Some initial GC-MS work was done on a Varian MAT CH5/DF
system interfaced to a Varian 2740 GC via a Watson-Biemann
separator and to a Varian SS-100 Data System. This
instrument was later used to confirm the presence of
atrazine in the Carrollton 70-year CCE by accurate mass
measurement at a resolution of about 5000. It was also used
for determination of empirical formulae of the major mass
spectral peaks of alachlor and the alachlor homolog.
The GC conditions for GC-MS analyses were similar to
those used for GC analyses described previously.
Mass spectra stored on disks from the Finnigan GC-MS
runs were compared via acousticoupler connection with
spectra in the EPA-Battelle computer files at Battelle
(Columbus). Mass spectra were also compared through a
computer terminal and acousticoupler with spectra in the NIH
Mass Spectral Search System in Bethesda, Maryland.
Chemical ionization mass spectrometry was performed on
selected components in the Carrollton 70-year CCE with a
separate computerized Finnigan 1015 mass spectrometer
interfaced to a Finnigan 9500 GC using methane as a
carrier/reactant gas.
A-6
-------
Combination gas chromatography-infrared spectroscopy
instrumentation "* was a computerized Digilab FTS-14D/IR
Fourier transform spectrophotometer equipped with the
Digilab GC/IR accessory and interfaced to a Perkin-Elmer 990
GC. This instrument was used to confirm the presence of
alachlor and atrazine in the Carrollton mega-sample CCE.
QUANTITATIVE ANALYSIS
A Perkin-Elmer PEP-1 Data System, interfaced to a
Varian 1400 GC operated under the conditions described
above, was used for computerized guantitation and retention
time measurements. Since atrazine was present in all
extracts of New Orleans samples, it was chosen as an
internal standard. A stock solution of 5 parts-per-thousand
of atrazine (99.7% pure) in chloroform was the reference for
quantitation of all identified pollutants for which
standards were obtained. Solutions of known amounts of pure
reference compounds were prepared and mixed with a known
amount of the atrazine reference stock solution. Mixtures
were designed so as to obtain good GC peak resolution. The
atrazine was assigned a flame response of 1.000 and, since
its concentration was known, the computer system was able to
calculate the flame response, as well as the relative (to
atrazine) retention time, of each standard.
After tentative identification of pollutants by GC-MS,
a PEP-1 computer program was written for the GC-computer
analysis of each extract, allowing the computer to use the
known flame responses to calculate pollutant concentrations
(Figure 1). In some cases, the flame response calculated
for one standard was also used for other compounds of the
same chemical class. The relative retention times,
calculated for all pollutants and printed out by the
computer, were then manually compared with those of the
available standards. The blanks had to be dosed with
atrazine as an internal standard, since atrazine was not
present in them.
In most cases, mass spectra of the standards were
obtained on the Finnigan GC-MS system for visual comparison
with those of the identified compounds.
STANDARDS FOR CONFIRMATION
Standard reference compounds were obtained from
commercial chemical suppliers (e.g., Aldrich Chemical Co.
and Columbia Organics Chemical Co., Columbia, S.C.), or, in
A-7
-------
FUN
121 37.9 9/ 9/ 74 CCE 1 CARR9LTBN
INST
, GC METH0D 42 , OC FILE
TIME
3.48
4.26
3.40
5.95
6.59
6.89
T.38
7.99
8.49
9.01
9.75
9.96
10.35
10.64
11.03
11.37
12. 42
13.10
13.58
13.83
14.06
14.38
14.85
15.49
16.13
16.39
16.69
17.18
17.62
17.95
18.65
20.32
21.16
21.59
21.95
28.54
22.80
23.09
23.42
23.75
24.36
24.84
25.72
26.23
26*90
27.36
27.85
28.52
28.69
29.11
30.12
30.65
31.65
32.25
32.66
33.07
33.64
34.23
34.54
35.02
35.76
35.91
36.34
36.84
37.08
38.31
38.70
39.37
39.85
40.25
40.56
41.03
41.37
42.48
44.22
AP.EA
1.4457
1.0659
,0066
.7193
.2873
1
1
1
,
4
4
•
7331
• 4588
0368
5926
0438
0554
0464
1933
3883
0783
.6423
.
I.
iiao
1664
.0677
.1578
*
4
«
*
.
•
*
0668
1710
2593
3321
1760
5249
5580
1.5324
.
.
0975
0707
8.2608
.
•
.
.
.
.
.
.
•
•
,
.
1.
1.
.
5639
3025
0696
0418
0330
0557
0980
0692
1090
0017
0895
2123
03E7
1325
3777
.2404
.
.
.
.
.
.
7.
.
.
.
.
.
U
.
0273
1072
1895
2243
7336
2206
7993
3798
3955
0662
0254
1663
7922
4150
.5095
.
.
.
.
«
.
.
.
.
.
3.
6254
0641
1093
0644
1261
I1E2
1392
1443
J976
0937
1139
.7768
.
4542
RRT
.107,
.132.
. 167,
. 184,
.204,
.213,
• 228,
.247,
.261.
.279,
.302,
.308,
.320.
.329,
• 342,
.352,
.378,
.406,
.421,
• 428,
.435,
.445,
.460,
.430.
.500,
• SOS,
.517,
.532.
.546,
• 536,
.578,
.630,
.656,
.669,
• 680,
.698,
.706,
.715,
.726,
.736,
.755,
.770,
.797,
.814,
.834,
• 848,
.863,
.834,
.395,
.902,
.933,
.950,
.981,
• OOO,
• 012,
.025,
.043,
.061,
.071,
.085,
.108,
.113,
.126,
. 142,
.149,
.187,
.200,
• 220,
.235,
.248,
.257,
.372,
.282,
.317,
.371,
RF
1.000,
1.000,
1 . 000.
1.000.
1.319,
.214,
2.380,
1 . 000,
2.500.
1.000,
1.000,
LOCO,
4.733,
.208,
1 . 000,
1.642,
.184,
1.000,
1 . 000,
.217,
.957,
• 184,
.240,
.184,
.316,
.557,
.184,
1.635,
1.000,
I. 000,
.298,
.278.
.866,
.184,
1.000.
I. 000,
1.000,
.184,
1.000,
1.000,
1.000,
1.000,
.184,
.412,
• 310,
1.000,
.134,
1.000,
I. 000,
.314,
• 416,
l.QOO,
.416,
1.000,
.314,
1 . 000,
1.000,
1.000,
1.000,
.715,
.314,
1.112,
.715,
1 . 000,
1.000,
1 * 000,
.700,
.613,
.350,
1.000,
1.000,
1.000,
.329,
.243,
.352,
C
.8987,
.6626,
.0041,
.4474,
.2356.
.0975,
.6788,
.0229,
.9209,
.0273,
.0345,
\ .0288.
\.5694.
.0502.
.0487,
.6556.
.0128,
.7250,
.0421,
.0213,
.0398,
.0196,
.0387,
.0380,
.0346,
.1817,
. 0638.
1.5575.
. 0607,
.0439.
1.5302,
.0983,
.1629,
.0080.
. 0260,
• 0237,
.0346,
• 0112,
.0431,
.0678,
.0011,
.0556,
.0243,
.2653,
.2182,
. 1726,
. 0275,
.0170,
.0667,
.0370,
.0580,
.4871,
. 057 1 .
4.8480,
.0741,
.2459,
• 0412,
.0158,
• 1034,
.7965,
.0810.
.3522,
.2780.
.0398,
.0630,
.0400,
.0549,
.0423',
.0303,
.0897,
.1228,
.0533,
.6368,
.1198,
.0994,
NAME
1 1 2-7R! CHLeFUETHANE t
T0LUEN£i
CHL0n0t>iBTt0M EM ETHANE:
i
DICHLOKOTODOMETH ANF
I
H- 0R P-XYLEMEl
.'
C2H2CL2BR2, 3-XYLEMEj
K0NANEI
WJK MV 145:
1
P-ET TBL «H PR3P BZj
BI5-2-CL-ET ETHEP.t
BRANCHED DEC AWE:
T3IMETHYU BZ ISSJIERi
DECANE, UNK ARBMs
LIM0NENEI
BIS-2-CL-1PR ETHER!
BRANCHED UN DECANE I
HEXACHL0R0 ETHANEt
1
!
IS0PH0ReWE:
DIHYCR0CARV0NE1
HEXACHL0R0BUTADIENE«
D0 DECANE:
-RIDECANEi
3-M-3-B-6-H-TRIAZINE:
!
TETRADECANEi
DIMETHYL PHTH ALATE :
DI-T-SU-BEWZ8«UIUeNEs
I
FJNTADECANEl
1
I
DI ETHYL PHTHALATEt
CL7 N0RB0RNENE:
DE-ETHYL ATRAZI«J£:
CLTNBRUamJENE IS3MER!
ATRAZlNEt
DIPRBPYI. PHTHALATEt
1
LASS8 -- ALACHL0R:
D1BUTYU PHTHALATEi
ELAOEXt
UASSB CL H0M9l.aGt
I
!
I
MACHETE*
DIELDP.INt
D1HEXYI. PMTHALATEJ
I
!
I
EBNZY1. BUTYL PKTKALAi
TRIPHENYL PHBSPlIATEi
DI-2-ET HEXYL PHTHALs
w
u
D
H
I M
o o
r- 4J
O
c m
o M-I
4J
iH OJ
•H W
o c
(0 CO
U 0)
M
n
O II
C-H tr>
•H -P a
ace
O -H
0) -P C
•P C O
3 tt)-H
Q» -P -P
g (U co
O U M
O -P
-------
the case of some herbicides, from the EPA Registration
Division, Chemistry Branch, Pesticides Reference Standard
Section in Washington. 1,2,3,4,5,7,7-Heptachloronorbornene
was obtained from John Roach, Industrial Chemicals
Contaminants Branch, Bureau of Foods, FDA, Washington.
Deethylatrazine was obtained from the Ciba-Geigy Corp.
1,3,5-Trimethy1-2,4,6-trioxohexahydrotriazine (trimethy1
isocyanurate) was synthesized using a two-step procedure
starting with cyanuric chloride5 by R. G. Webb of the ACB.
The melting point and the IR and mass spectra of the product
matched reported values.
RECOVERY STUDIES
The recovery of atrazine from water was studied to
determine the approximate recoveries that might be expected
for the atrazine-type herbicides. Four extraction methods
were tested with spiked samples:
(a) Liquid-liquid extraction—Two 1-liter tap water
samples were spiked with 100 and 200 yg, respectively, of
atrazine, allowed to stand 5 days in laboratory light, and
extracted 3 times with 100 ml of methylene chloride. The
extracts were evaporated to 1 ml using a Kuderna-Danish
evaporator followed by a slow stream of dry nitrogen, and
analyzed by GC.
(b) Activated cocoanut charcoal (an incidental
laboratory supply)—Five liters of tap water was spiked with
100 yg of atrazine and passed by gravity flow through a
column of 11 grams (20-ml volume) of the charcoal (6-14
mesh). A control consisted of a similar charcoal column and
5 liters of unspiked water. For a direct recovery test, 100
yg of atrazine in 2 ml of chloroform was added directly to
11 grams of the same charcoal. The 3 charcoal samples were
air-dried at room temperature overnight, then extracted with
methylene chloride in a Soxhlet extractor for 6 hours. The
extracts were evaporated to 1 ml as in experiment (a), and
analyzed by GC.
(c) "Fine" mesh carbon (used at New Orleans)—
Experiment (b) was repeated with 3.5 grams (20-ml volume) of
the "fine" mesh Nuchar 190 carbon used in the New Orleans
study.
(d) "Coarse" mesh carbon (used at New Orleans)—
Experiment (b) was repeated with 10 grams (20-ml volume) of
the "coarse" mesh carbon used for the New Orleans study.
A-9
-------
Duplicates were run beginning with the atrazine spiked
water.
RESULTS
The results of this New Orleans area drinking water
supply analysis are summarized in Table 2 (at the end of the
report). A total of 82 organic compounds were tentatively
identified by GC-MS in one or more of the three finished
drinking waters. Fifty-five of^ these identifications were
confirmed. Nine compounds were not identified as to
specific isomer, and eight others were identified only
partially by certain structural features. Concentrations
ranged from about 0.004 ug/l(the detection limit for most
compounds except endrin) to 12 yg/1. Table 3 is a
classification of the 82 compounds by chemical family or by
use.
Table 3. Classification of Organics in New Orleans Drinking Water
Chemical Family or Us-a Compounds Present
Halogenated aliphatics 18
Chlorinated aromatics 5
Chlorinated aliphatic ethers 2 «
Herbicides & associated compounds 8 b
Pesticides 5 c
Aromatic hydrocarbons 11
Aliphatic hydrocarbons 9
Phthalates 11
Miscellaneous 13
Total 82
a bis-2-Chloroethyl ether and bis-2-chloroisopropyl
ether
Alachlor, alachlor homolog, atrazine,
deethylatrazine, butachlor, cyanazine, propazine, and
simazine.
f*
ot-chlordane, chlordene, DDE, dieldrin, and endrin
The 8 compounds only partially identified are listed as
unknowns at the bottom of Table 2 and consist of 2
dichlorinated compounds (probably aromatics), 3 phthalates,
A-10
-------
1 compound of apparent molecular weight 145, and 2
chlorinated-fluorinated hydrocarbons. In addition, 1 each
C2, C3, and CH alkylbenzene compounds, 2 branched
hydrocarbons (decane and undecane), a trimethylbenzene, an
alachlor homolog, a dibromodichloroethane, and a
heptachloronorbornene compound were not identified as to
specific isomer.
No polynuclear aromatic hydrocarbons were identified in
the carbon chloroform extracts, even after fractionation by
TLC (detection limit is estimated at 0.1 yg/1 after TLC
separation), but naphthalene and 1- or 2-methylnaphthalene
were identified in one XAD-resin extract.
No carboxylic acids or phenols were identified in any
of the extracts, including the carbon alcohol extract, which
was examined before and after methylation by diazomethane.
An independent investigator examined New Orleans
drinking water and the same CCE extracts used for this study
for volatile N-nitrosamines, but none were detected at 1
ng/1 or above.
a-Chlordane, chlordene, and pentachlorophenyl methyl
ether were identified only qualitatively in the Carrollton
mega-sample by GC-MS of fractions separated by column
chromotography (see Experimental section). Endrin was
detected and quantitated by electron capture GC of the
appropriate fraction. None of these 4 compounds was
detected by GC-MS analysis of the total extracts, in
contrast to most of the compounds in Table 2.
Butyl octyl maleate, m-chlor©nitrobenzene, DDE,
dicyclopentadiene, methylnaphthalene, and naphthalene were
detected and confirmed by Junk at Ames, Iowa, during his
independent GC-MS analysis of the Carrollton and Jefferson
No. 1 resin extracts. They were not quantitated, and were
not detected in the CCE's analyzed at SERL.
Using the VOA technique7, which has proven excellent
for very volatile organics in water, the 12 compounds listed
below8, not contained in Table 2, were identified only at
the EPA's Water Supply Research Laboratory (WSRL),
Cincinnati.
acetaldehyde dichloromethane
acetone ethanol
2-butanone methanol
carbon disulfide 3-methyl butanal
carbon tetrachloride 2-methyl propanal
A-11
-------
l,2~dichloroethane (8 yg/1) 1,1,2-trichloroethy.lene
Chloroformr identified at SERL by the tetralin extraction
method, was detected at 133 yg/1 by the HSRL. Five other
compounds identified at SERL were also identified by the VGA
technique at the WSRL {see footnotes to Table 2).
DISCUSSION
SEPARATION EX" .1RIMENTS
Organic analyses at SERL were performed using
concentrated extracts from carbon and XAD-2 resin adsorption
columns. Various attempts were made to fractionate the
extracts to improve subsequent GC peak separation—most of
these were unsuccessful (see Experimental section) .
The EPA r-olumn chromatographic method for organo-
^hlorine pesticides1 was useful for the detection and
quantitation of -andrin and dieldrin, and for the detection
of chlordene, a-chlordane, and pentachlorophenyl methyl
ether. With the exception of dieldrin, these compounds were
present at concentrations too low for GO-MS detection using
the total extracts. Dieldrin was detected in the total
extracts only after searching for characteristic peaks by
limited mass range mass spectral techniques.
Thin layer chromatography was successful in
fractionating the mega-sample CCE. Eighteen of the
compounds identified by GC-MS analysis of the total extracts
veire also identified by GC-MS analysis of the eluates from
the three regions of the developed TLC plots. However,
little was gained by the additional TLC fractionation.
Fortunately, extract fractionation was not necessary
for qualitative or even quantitative analysis. Computer
controlled GC :4S analysis, with computer manipulation of
acquired data,- allowed identification of compounds by direct
injection of CCE's. Figure 2 includes a FID chromatogram of
the unfractionated Carrollton 70-year CCE. Chromatograms of
the Jefferson No. 1 and Jefferson No. 2 CCE's were similar.
At this stage of separation, extracts were considered to be
ready for GC-MS analysis.
A-12
-------
n 03 *d
o QJ H-
H W 03
ro n n
w 3- (D
'O "I
O O NJ
3 3 •
d» 0*
ft
rt o »
O O3 (D
^ O
O &> O
033
73 l-h ft
O O it
c n c
3 O
O. O ft
W &> 0)
h{ DJ
p- n
3 O iQ
M 0)
i-3 I-1 en
0) ft
cr o n
M 3 3"
ro n
M^i
• I QJ
^< ft
n> o
&i 03
pi
n 3
n
w —
n
n
3 0)
tr 3
a> a
n
M '-fl
M
a
i
k
<
n^
t
IK
(>
R
8
CARROLLTON 70yr.-CCE
RGC
v IBB 110 in i
i« 10 iw ITB in t» at 211 a» zm
spectrum number
CARROLLTON 70yr.-CCE
27!
63
• -
68
,:. ;
54 ,1V1
U*
; • '\ i
1 i\/ ^
t_jgW
mins.
-------
IDENTIFICATION OF COMPOUNDS
The initial output of the GC-MS-Computer system is a
computer reconstructed gas chromatogram (RGC), which is a
recording of the MS operating as a GC detector, and must
ultimately be correlated with the FID chromatogram of the
same sample. The RGC and FID chromatograms for the
Carrollton 70-year CCE are shown in Figure 2. Peaks are
numbered to correspond with the alphabetical listing of
compounds in Table 2. High quality mass spectra, necessary
to identify the peaks, were^obtained using a separate PDP-8
computer with CRT and hard copier to obtain copies of the
disk-stored mass spectra of sample components. In so doing,
spectrum numbers corresponding to RGC peaks and a background
subtraction routine were used to obtain the best possible
spectra for all RGC peaks and shoulders.9
The first step in compound identification was an
attempt to match each disk-stored sample spectrum with a
spectrum in the EPA-Battelle (Columbus) computer file of
9500 mass spectra by direct acousticoupler-computer
connection.9 This was followed by a manual search of the
Aldermaston Eight Peak Index of Mass Spectra,10 regardless
of the degree of computer match, or "Similarity Index".
Compounds having a high Similarity Index (0.4 or above) with
the computer selection and matching the peak listing in the
eight-peak index or some other literature source, were
considered "tentatively identified" with a high level of
confidence. Compounds whose spectra matched the eight-peak
index or other literature source, but whose spectra were not
in the EPA-Battelle file, were also considered as
"tentatively identified", but with a lower level of
confidence.
Alachlor, cyanazine, and butachlor were identified by
searching a herbicide handbook11 for compounds of the same
molecular weight and chemical family as indicated by
examination of the mass spectra. These spectra were then
matched with those of the pure compounds, samples of which
were obtained from the EPA Pesticides Reference Standards
Section.
Several compounds were identified both in a sample
extract and in its blank. These were usually not considered
as being in New Orleans water and are therefore not listed
in Table 2. However, some compounds present in the blank,
especially the mega-sarople blank, at trace levels were
present in the sample extract at more than five times the
concentration in the blank. In such cases, the blank
A-14
-------
concentrations were subtracted from the sample
concentrations to give the quantitative values in Table 2.
Spectral Matching by Computer
About 75 apparently usable mass spectra were obtained
from the Carrollton 70-year CCE. From these spectra 44
compounds were subsequently identified (not counting
partially identified "unknowns"). The spectra of 17
compounds in the EPA-Battelle file were listed as "first
hits" (best matches) in spectral searches with the usable
sample spectra and later were determined to be correct
identifications by comparison with standards. The
similarity indices of 14 of the "first hits" were in the
0.4-0.8 range. Two were about 0.3, and one, dieldrin, was
0.216. Dieldrin, present at only 0.04 pg/1, gave a spectrum
with many extraneous peaks, resulting in the low similarity
index.
After spectra were checked against the EPA-Battelle
file and Eight Peak Index of Mass Spectra1°, remaining
unknown spectra were compared with those in the NIH Mass
Spectral Search System computer file in Washington. Three
or four compounds were identified by this comparison.
The EPA and NIH search systems have recently been
combined into the Mass Spectral Search System file, handled
by the Cyphernetics Corp., Ann Arbor, Michigan, and updated
to include a total of 35,000 reference mass spectra.9 This
combination and expansion has resulted in a much more useful
spectral matching system.
Limited Mass Range Searches
Simazine and propazine have almost the same GC
retention time as atrazine, and since atrazine was present
at much greater concentrations in most samples, its GC peak
obscured the smaller simazine and propazine peaks.
Indications of their presence were obtained by observations
of their molecular ion peaks and molecular ion-minus-methyl
peaks in the atrazine mass spectrum. Limited mass range
mass spectrometry, in which the disk-stored mass spectral
data are searched for ions that are definitive for a certain
compound, produced evidence that simazine was present in the
leading edge of the atrazine peak, and propazine in the
trailing edge.
The limited mass range technique was also useful in
searching the stored data for the spectrum of dieldrin,
which did not produce a discernible peak on the computer
A-15
-------
reconstructed gas chromatogram. Other compounds of special
interest—bis-chloromethyl ether, geosmin, 2-roethyl
isoborneol, and deisopropylatrazine—were searched for by
this technique in the Carrollton 70-year CCE, but were not
detected.
Identification and Confirmation of Alachlor by Advanced
Instrumentation
Alachlor, MW 269, does not give an easily discernible
molecular ion peak by electron impact mass spectrometry, and
its spectrum was not in the \computer data bank, which made
its identification difficult;^ The inability to identify the
compound was particularly frustrating since its GC peak
indicated that it was present in a relatively high
concentration. Also, an associated peak in the gas
chromatogram had a similar mass spectrum. Chemical
ionization mass spectrometry, however, gave a characteristic
molecular-ion-plus-one peak (m/e 270) for the compound
(Figure 3). High resolution mass spectrometry, using the GC
effluent, produced possible empirical formulae of the
important mass fragments, including the molecular ion.
These empirical formulae were correlated with reasonable
fragmentation modes of the parent molecule, so that the most
probable parent ion formula was deduced (Figure 3). Since
other herbicides had been identified in the samples, the
herbicide handbook1 l was searched for compounds having the
appropriate molecular formula and structural
characteristics. Alachlor was the only possibility. A
standard of alachlor produced the same low resolution mass
spectrum and GC retention time, confirming the
identification.
The identity of alachlor was also confirmed by Fourier
transform gas chromatography-infrared spectroscopy (see
Figure 3) , with a spectrum obtained on the fly from the
Carrollton mega-sample extract injected into the GC. This
spectrum matched that of the standard obtained in the same
mode.
The electron impact spectrum of the alachlor homolog
(compound 2, Table 2) did not give a definitive molecular
ion. Chemical ionization MS showed the molecular weight to
be 303 and the chlorine isotope pattern indicated the
presence of two chlorine atoms instead of one. High
resolution mass spectrometry supported this evidence,
yielding an empirical formula of C1IfH19 NO2C1? Mass
spectroraetry indicated the molecular structure to be the
same as that of alachlor except for the presence of two
chlorine atoms instead of one on the carbon atom alpha to
A-16
-------
INFRARED SPECTRUM
Alachlor
C-CI
3000
2500
2000
1500
1000 cnr
LOW RESOLUTION CHEMICAL IONIZATION SPECTRUM
100-
80 :
60-
20 :
0
2
M + 41
i,
i l l i i " "i
10 230 250 2
¥ l M + 29
I,
l , uj
70 290 3
J 1
10
HIGH RESOLUTION MASS SPECTRAL INFORMATION
Measured mass Possible Measured mass Possible
(amu) formulas (amu) formulas
100;
50-
0-
160.1122 *CnH14N
C8H160}
188.1066 *C]2Hi4NO
C7HnN303 -,
C,H17N2CI
\
LOW RESOLUTION ELECTRON
IMP ACT SPECTRUM
\
Carrollton 70-yr CCE 9/16/74
Spectrum 341-339 N.
X
•'•I"' J ' l». J. .. 0. ,.4 -L
50 100 150
238.1020 CI6Hi402
*Ci3HI7NOCI
r CIOH20N2CI2
Ci3H17N2CI
C10Hi,N03CI
269.1175 C20H15N
Ci5Hi5N302
CnHir03
'CuHzoNOjCI
CnH23N2OCI2
C]2H23N2CI2
^N 1
, ... L .. , .
" ii 1 i1 "r i ''I'T i i \
200 250
Figure 3. Spectra of alachlor in Carrollton 70-year CCE.
Starred (*) formulas correspond to correct alachlor ion
formulas.
-------
the carbonyl group. The Fourier transform GC-IR spectrum
was very much like that of alachlor. On the basis of the
above structural information the compound was tentatively
identify ed as 2 12-dichloro-2 ' , 6' -diethyl-N- (methoxympthyl)
acetanilide.
CONFIRMATION OF COMPOUNDS
After making tentative identifications (see under
Identification of Compounds } attempts were made to confirm
the identity of compounds in the drinking water by obtaining
standards, matching GC retention times and mass spectra, and
synthesizing compounds^ Fifty-five compounds were confirmed
in this way (Table 2). The mass spectra of selected
compounds are shown in Figures 4 through 13. The degree of
reliability of the "unconfirmed" identifications varies.
Some compounds were identified by more than one method or in
more than one laboratory. Also, most compounds were found
in more than one of the drinking water supplies, and at the
same sample site by more than one adsorption or extraction
technique. Repeat findings indicate that such compounds are
not analytical artifacts.
Alachlor, One Additional Chlorine Homolog
To confirm the identification of the alachlor homolog
and to determine whether it could have been produced during
cbJorination treatment of Hew Orleans drinking water,
solutions of 5 mg/1 alachlor in distilled water were treated
with up to 138 mg/1 of chlorine gas for 22 hours at room
temperature. Two chlorinated compounds were found, the
concentrations of which were dependent upon the
concentration of chlorine. The concentration of alachlor
diminished correspondingly. These components were extracted
with methylene chloride, which was then concentrated and
analyze^ *>Y riC~MS. The spectra of these compounds did not
natch that of the alachlor homolog found in the I.'ew Orleans
water; they appeared to correspond to 1- and 2-chlorine
additions to the aromatic ring of alachlor.
.therefore the homolog is probably not formed by
chlorination of alachlor in the i-Jew Orleans water supply,
but is possibly a by-product of alachlor manufacture.
Dee_tny 1 at raz in e
Deethylatrazine (desethylatrazine) was found in New
Orleans water only in samples collected by carbon
adsorption, not in resin adsorption samples, Wolfe and
A-IS
-------
SPECTRfi IO8ER 1
CflHflOJLTIU HEGW 9 27 7-t
8.
SsS.
28 30 •» SO 60 70 BO 30 193 110 120 138 IW ISO ICC 170 180 130 2BO 210 229 230 2« 2SO 260 770 280 SO 308 310 3ffl 338
MX E
g
R.
bPECtllH KU«ER I
CfflHQU-TO) HCSIN 3 77 7-t
Figure 4. Mass spectra of alachlor (top) and alachlor
homolog, both from the Carrollton resin extract.
A-19
-------
UK \ 1 23 7S
s.
fcR.
I It
li ilJ li, nit
1
30 18 99 GO TB M 90 108 118 1Z3 -.30 118 125 ICO 178 1M I9B MB ZIB Z35 Z38 218
It./ E
2BB 279 28B 290 300 31S 32O
zsc
W TWI OE 9 It 71
B.
1
3B 3B 18 » . S0 TB
ns c
jo so loa ti0 ta ias no ISB ica iw iw i» zoo no as SB zio ZBB ve m aa taa 300 3
Figure 5. Mass spectra of atrazine standard (top) and
deethylatrazine from Carrollton 70-year CCE.
A-20
-------
SPCCTlUt MtCER 1
MlS-1 RUM 1 23 7S
B.
D
30 10 SO CO 70 90 30 100 110 120 130 110 ISO 160 170 180 ISO 200 210 220 230 210
M/ £
SO 2GO 270 200 2^1 300
B
R-l
8
l«-1 IW I 11975
B 18 SB SB 78
H/E
38 103 118 Iffl 130 118 ISO ItS ITS 1W 130 JOB 110 Z28 ZB Z18 Z2I 260 2T8 2B8 290 300 310 32B
Figure 6. Mass spectra of bromodichloromethane (top)
and bromoform, both standards.
A-21
-------
SFECTUtlUKR 1
ML
8.
jiUUJJU
i
D 3D « SB CD ID IB 30 JOB 118 «B 130 !« ISO ISO 110 IM 19 109 tit Ot Z30 JIB 29 SB CTB
nx t
znznanaiaanaiawaB
e
8.
8.
r-
p
k£.
in - in
.nx •mw ox a ic n
ai 30 10 a> «o TB
HXC
«0 30 i(H 119 1ZO 1JO l« ISO 1CB 1T8 1W 130 280 Z18 ZZD Z30 ?W SI ZB8 ITO 2H 29D an 31B Jffl 330 318 3
Figure 7. Mass spectra of butachlor (Machete ^ standard
(top) and bis-2-chloroethyl ether from the Carrollton
70-year CCE.
A-22
-------
138 - 117
amu-KM -ram or a a «
8.
B 3B
H/'C
•n in sa 'ion 'no iffl 'aa i« ISB \se Vie in i» ioo iio ia ao iio iso » ITO ae oo MJ iio aao iao an) kr
Figure 8. Mass spectra of bis-2-chloroisopropyl ether from
the Carrollton 70-year CCE (top) and cyanazine (Bladex® )
standard.
A-23
-------
nteen i
D 1 1 2S 7S
1 i
SU
8.
P-
ML
R-
o
Lll
, iiL, ill
ae 10 so €0 TO 00 9
-------
srecTHm
QfflLLTON RESIN 9 27 71
jia.illljl.lllJ.lll
110 I2B 130 110 IB ICO |TO 'l»0 ISS ZOO ZI8
ZBO 270 280 230 30D
feR.
SFEnHK HJCER 138 - 190
OWOJ-TCH TOnt HE 3 It 71
ZO 39 W 9) CO 70
ny E
* " 100 no' "liji i3o \ia ^isa~~wo~Tn in"Tsi"ia zio 'zai z» !MO aai jca"in
Figure 10. Mass spectra of 1,2,3,4,5,7,7-heptachloro-
norbornene from the Carrollton resin extract (top) and
hexachloro-l,3-butadiene from the Carrollton 70-year CCE.
A-25
-------
s.
f-
I8-
1
K.
e>
-~J*
,
il
ii Li..ilJL.
,
I™,
ilu
Jl
1 1 ,,,
(
ia
MX E
3» 310 3JO 3QO XO 3
omuia* ion ox s ic 71
B.
ZB JO 18
ny t
CO TO H 99 IBB 118 !ZO 130 1« ISO 1(B 110 1H l» IBB ZIB Z» ZB Z« SB IBB ITB MO 230 3OO 318
Figure 11. Mass spectra of hexachloroethane standard (top)
and isophorone from the CarrolIton 70-year CCE.
A-26
-------
COL-3 PAH ia-iS-7-1
SPECTRUn 331 - 3H8
T
30
50
70 90
110
130 150 170 1S0 £19 230 8S0 270 29<
11 -1
aWaiTO! TOW O£ 3 1C TI
R.
8.
bfi.
isa i«o ITS UB igg 200 iio
bo
Figure 12. Mass spectra of pentachlorophenyl methyl ether
from the Carrollton Mega Sample CCE (top) and 1,1,2-
trichloroethane from the Carrollton 70-year CCE.
-------
hR_
9PETTHJ1 M-TBEB 1
aw i i & is
L
30 1B £8 B> TO
lag no ize ias lie iso its iTO iw lae no ?IB no oe no se so rra 38 SB 300 Jia azo 3»
8
8
8.
f
8s
hS_
SPECTHJI KlflER S - a
amu.i
-------
Zepp12 reported that deethylatrazine was the major
photochemical degradation product of N-nitrosoatrazine in
aqueous and chloroform solutions, but not in ethyl ether
(which is used for eluting organics from the XAD-resin).
Therefore, experiments were conducted to determine whether
deethylatrazine could be produced by degradation of atrazine
or of tT-nitrosoatrazine on carbon or on the resin when
exposed to varying intensities of light. Only atrazine was
recovered from any of the experiments. Unfortunately, N-
nitrosoatrazine decomposes to atrazine in a GC, the
separation method used in these analyses. Repeated MS data,
however, indicate that deethylatrazine is present in the Hew
Orleans finished water, and is not an analytical artifact.
It simply may not be adsorbed by XAD-resin, or may not elute
from the resin.
1,3,5-Trimethyl-2,4,6-trioxohexahydrotriazine
1,3,5-Trimethyl-2,4,6-trioxohexahydrotriazine
(trimethyl isocyanurate), was tentatively identified by the
Mass Spectral Search System, using the EPA-Biemann search
mode (Similarity Index = 0.509). It is not available
commercially and had to be prepared for confirmation
(Experimental section). The mass spectra and GC retention
time of the product matched that of the sample component,
confirming the identification.
UNIDENTIFIED COMPOUNDS
Fifteen compounds at low concentrations (between 0.1
and 0.3 ug/1) yielded usable mass spectra, but could not be
identified. Many other compounds in even lower
concentrations were detected by GC, but did not give usable
mass spectra. About 100 peaks can be counted in the gas
chromatogram of the Carrollton 70-year CCE #1, but only 47
were identified.
Mass spectra of selected unidentified compounds in the
New Orleans drinking water are shown in Figures 14 and 15.
In the New Orleans water samples, as in many other
samples we have analyzed, the number of detected compounds
increases almost exponentially as the detection limit
decreases. This trend is summarized in Table 4.
A-29
-------
CAPROLLTON 70YR CCE 3/IS/74
SPECTRUM 238 - 236
OJ
o
90-
78-
58-
38-
10-
3
, ,'. 1 I'j t>
i • • • • i *
€>
Mr I4-.1 i
...llLl.U
i. ,, 1.1. iii .1.1. i, . ullii i.ll.i h h i
58 70 98 110 130 150 170 190 210
CaPPCLLTON 70YR CCE 9/16/74
SPECTRUn 94 - 89
go-
70-
50-
38-
10-
3
T~
0
l , 1.
50 70 90 119 139 153
Figure 14. Mass spectra of unknown chlorinated fluorinated
hydrocarbon (top) and unknown compound, apparent MW 145,
both from the CarrolIton 70-year CCE.
-------
MEGASAPIPLE (1*107) 9/19/74
SPECTRUM E74 - S71
90-
70-
50-
30-
10-
3
4|lJUklt/j;Uff .Ji n
6 50 70 90 110 130 150 178 190 210 230 250
srccnui m«ER »7 - am
g omLLTW mm oca an
R.
8.
p
hR.
Is.
B.
O
1
j,U,U...,
'
.. \ it
IB 30 10 a eo TO
MXE
Figure 15. Mass spectra of unknown dichlorinated compound,
MW 200, from the Carrollton Mega Sample CCE (top) and un-
known dichlorinated compound, MW 249, from the Carrollton
70-year CCE.
A-31
-------
Table 4. Classification of Organics in Carrollton CCE's by
Concentration
Concentration Number of Compounds Present
Range, pg/1 Within Concentration Range
70-year CCE Mega-sample CCE
0.01 - 0.09 74*
0.10 - 0.99 19
1.0 - 10 __4
Total detected 97 103
*Includes compounds detected by FID/GC (manual peak
count) but not identified.
Following this study, one FID gas chromatogram of the
Carrollton 70-year CCE was obtained with a SERL-prepared
glass capillary column-50 ft x 0.02 in. coated with SP-2100
(the packed column used 3% SP-2100). A total of 182
components were detected, about twice as many as were
detected using the packed column, as a result of the lower
detection limit caused by improved separation.
RECOVERIES
Several methods of extraction were used in these
analyses—chloroform extraction of carbon filters in series,
chloroform extraction of different sizes of carbon filters,
alcohol extraction of carbon filters, liquid-liquid
extraction with tetralin, and ether extraction of XAD-resin
filters. The carbon filters in series gave better
recoveries for the largest number of compounds.
In general, the concentrations for each of the three
sampling sites are in good agreement for samples collected
by the same techniques. For example, atrazine values for
the 70-year CCE's are 4.9, 5.2, and 5.4 yg/1, and butachlor
values are 0.05, 0.06, and 0.05 yg/1.
Recoveries from Carbon Filters in Series
Quantitative results in Table 2 for the 70-year CCE
compounds are sums of the amounts recovered from two carbon
A-32
-------
filters in series—filter #1 and filter #2. Table 5 gives
for each compound at a concentration >• 0.05 yg/1 the percent
of the total concentration that was recovered in each of the
two filters for the Carrollton 70-year sample.
Of the 29 compounds in Table 5, 24 were collected on
filter SI to the extent of 96% or more, indicating
quantitative adsorption on carbon (but not quantitative
recovery). These 24 include a wide variety of compounds,
indicating a generally high degree of adsorption. However,
there are exceptions. Three trihalogenated methanes or
ethanes were adsorbed to the extent of only 64 to 85% on
filter #1, and more hexachloroethane was found on filter #2
than *1.
Recoveries From Carbon Filters of: Different Sizes
The two-month equivalent (mini-sample) CCE's contained
much higher concentrations—up to 10 fold—of lower boiling
compounds than did the 70-year and mega-sample CCE's, but
about the same concentrations of the less volatile
compounds. This may be seen by comparing the FID
chromatogram of the Carrollton mini-sample CCE with that of
the mega-sample, both in Figure 16, and with that of the 70-
year sample in Figure 2. Although the 70-year sample was
collected about ten days before the mini-sample, during
which time the compound quantities in the river may have
changed, the concentrations should not have changed as a
function of volatility. These results may be an indication
of organic overloading on the 70-year carbon filter,
although this explanation is contrary to most of the results
shown in Table 5. Seventy grams of carbon were used to
extract 60 liters of water in the mini-sample filters (0.9
1/g)/ while 340 grams were used for 25,500 liters in the 70-
year filters (75 1/g). This is an 83-fold difference in the
water/carbon ratio.
Concentrations in the mega-sample CCE's were low
relative to the 70-year CCE's. Of 30 compounds quantatated
in the Carrollton drinking water by both methods, the
concentrations of 21 in the mega-sample were between 25 and
76% of that found in the 70-year sample. The water/carbon
ratio for the mega-sample was 57 1/g.
Several operational factors were biased against the
mega-sample quantitations relative to those for the 70-year
sample. The mega-sample carbon was dried for 10 days before
extraction, while the 70-year sample carbon was dried for
only 2 days. The mega-sample CCE was concentrated three
times more than the 70-year CCE, and concentration was not
A-33
-------
Table 5. Percent of Compounds Recovered from Carrollton
70-year Filters #1 and #2 in Series*
Total Concentration % From % From
Compound pg/1 Filter #1 Filter
Alachlor 0.82 97 3
Alachlor homolog 0.28 100 0
Atrazine 4.9 98 2
Atrazine, deethyl 0.51 96 4
Benzyl butyl phthalate 0.64 100 0
Bromoform 0.57 100 0
Butachlor 0.05 100 0
Cyanazine 0.35 100 0
Chlorodibromomethane 1.1 64 36
bis-2-Chloroethvl ether 0.04 100 0
bis-2-Chloroisopropyl ether 0.18 100 0
Dibromodichloroethane isomer 0.33 100 0
Dibutyl phthalate 0.10 80 20
2,6-Di-t-butyl-p-benzoquinone 0.22 100 0
Dichloroiodomethane 1.1 85 15
Dieldrin 0.04 100 0
Di-(2-ethylhexyl) phthalate 0.10 100 0
Dimethyl phthalate 0.27 100 0
Dipropyl phthalate 0.07 100 0
1,2,3,4,5,7,7-Heptachloronorbornene 0.06 100 0
Heptachloronorbornene isomer 0.06 100 0
Hexachloro-1,3-butadiene 0.16 100 0
Hexachloroethane 4.3 36 64
Isophorone 1.6 98 2
lfl/2-Trichloroethane 0.35 69 31
Trimethylbenzene isomer 0.04 100 0
Triphenyl phosphate 0.12 100 0
Undecane, branched 0.06 100 0
o-Xylene 0.33 100 0
*Including only compounds present in total concentration of
0.05 yg/1 or greater.
A-34
-------
3 H-
aiQ
3 M
n> ro
i£)
0) H-1
I C~\
O •
o
M
Z 0)
c w
la
CD IT
H H
W O
O P
O rt
H O
II iQ
CD n
en CD
'O 3
O W
DJ HI
O
rt if
O
rt
o y
o CD
3
T3 n
o &
c n
3 ^
a o
H- rt
3 O
tr p-
M 3
CD H-
to
CARROLLTON MINI-CCE
3,. 7 tj^JjJi
UlA^
25 30
mins.
CARROLLTON MEGA-CCE
33
mins.
-------
as carefully controlled, in that direct distillation of the
chloroform was used for the mega-sample whereas vacuum
evaporation at room temperature was used for the 70-year
CCE's.
Carbon alcohol extract (CAE)
The methylated carbon-alcohol extract {Carrollton 70-
year) was expected to contain carboxylic acids and phenols,
but none were found. Only compounds 1, 6, 1, 37, 49, and 69
(Table 2), which were identified in the 70-year CCE, were
found in the CAE. Of these, the one present in highest
concentration was atrazine atr\J}.3 wg/1. (This is included
in the value of 4.9 yg/1 given for the CCE in Table 2.)
Recovery from Resin
Not enough data are available from the New Orleans
study to make a good comparison of XAD-resin vs. carbon
extraction efficiencies. More compounds and higher
concentrations were obtained from the carbon, but sampling
circumstances were biased against the resin technique. I\
well-designed comparison of these two concentration
techniques is needed.
Recoveries Using Tetralin Extraction
Chloroform was the only compound identified by the
tetralin extraction technique. Only 2 or 3 GC peaks were
observed that were not also in the tetralin blank. This
technique is good for volatile compounds, since the high
boiling solvent leaves a window for early eluting GC peaks,
but its sensitivity is limited since tetralin cannot be
evaporated to concentrate extracted pollutants. This
especially limits the usefulness of tetralin in drinking
water analysis. However, tetralin has now been purified at
SERL to eliminate interfering peaks so that 1 ml can be
used to extract 2 liters of water, thus increasing the
sensitivity. Although more complicated, the gas stripping
VOA technique is far more productive for isolating these
volatile compounds.
RECOVERY STUDIES WITH ATRAZINE
Results of the atrazine recovery experiment are
summarized below:
Extraction with methylene chloride, 100 yg/1 spike 98
Extraction with methylene chloride, 200 Wg/1 spike 95
A-36
-------
Cocoanut charcoal +100 yg atrazine in 5 liters of water 29?»
Cocoanut charcoal +100 yg atrazine added directly 283
New Orleans fine carbon +100 yg atrazine in 5 liters of water 922
New Orleans fine carbon +100 yg atrazine added directly 103?
New Orleans coarse carbon +100 yg atrazine in 5 liters of
water 59%
New Orleans coarse carbon +100 yg atrazine in 5 liters of
water (Duplicate) 71%
New Orleans coarse carbon +100 yg atrazine added directly 102
Liquid-liquid extraction with methylene chloride is a
good method for recovery of atrazine from water, as is
carbon adsorption with the fine grade carbon. In the New
Orleans 70-year carbon samplers, the coarse carbon was used
on each end of the column as a filter and holder for the
fine grade carbon, which was assumed to be the main
adsorbing material.
o.
A-37
-------
REFERENCES
1. National Pollutant Discharge Elimination System,
Appendix A. Federal Register 38, £75, Part II,
November 28, 1973.
2. Webb, R. G., A. W. Garrison, L. H. Keith, and J. M.
McGuire. Current Practice in GC-MS Analysis of
Organics in Water. Southeast Environmental Research
Laboratory, Athens, GA. EPA Report Ho. EPA-R2-73-277,
August 1973. 91 p.
3. Junk, G. A., J. J. Richard, M. D. Grieser, D. witiak,
J. L. Witiak, M. D. Arguillo, R. Vick, H. J. Svec, J.
S. Fritz, and G. V. Calder. Use of Macroreticular
Resins in the Analysis of Water for Trace Organic
Contaminants. J. Chromatogr. 99, 745-762, 1974.
4. ?izarraga, L. V. and A. C. McCall. Infrared Fourier
Transform Spectrometry of Gas Chromatography Effluents.
Southeast Environmental Research Laboratory, Athens,
GA. EPA Report Uo. EPA-660/2-73-034, January 1074. 61
P-
5. Cignitti, M. and L. Paoloni. Tautomeric Forms of Oxy
Derivatives of 1,3,5-Triazine. I. Infrared Spectra.
Rend. 1st. Super. Sanita, 23, 1037-1047, 1960; Chem.
Abstr., 56, 3037, 1962.
6. Fine, D. H., D. P. Rounbehler, F. Huffman, A. W.
Garrison, N. L. Wolfe, and S. S. Epstein. Analysis of
Volatile IJ-Nitroso-Compounds in Drinking Water at the
ppt Level. Bull. Environ. Contam. Toxicol. In press,
14_, 1975.
7. Bellar, T. A. and J. J. Lichtenberg. The Determination
of Volatile Organic Compounds at the pg/1 Level in
Water by Gas Chromatography. Methods Development and
Quality Assurance Research Laboratory, Cincinnati,
Ohio. F.PA Report 'Jo. EPA-670/4-74-009, November 1974.
27 p.
3. Melton, 3v., Water Supply Research Laboratory,
Cincinnati, Ohio. Personal Communication, October
1974.
A-38
-------
9. Heller, S. R., J. il. McGuire, and W. L. Budde. Trace
Organ!cs by GC-MS. Environ. Sci. Technol. 9_, 210-213,
1975.
10. Mass Spectrometry Data Centre. Eight Peak Index of
Mass Spectra, 1st, Ed. AWRE, Aldermaston, Reading,
England, 1970. T493 p.
11. Weed Science Society of America. Herbicide Handbook,
3rd Ed. Champaign, Illinois, 1974. 430 p.
12. Wolfe, N. L., R. G. Zepp, J. A. Gordon, and R. C.
Fincher. W-IJitrosamine Formation from Atrazine. Bull,
Environ. Contain. Toxicol. In press 14, 1975.
A-39
-------
TABLE 2
ORGANIC COMPOUNDS IN NEW ORLEANS DRINKING WATER—CONCENTRATIONS IN yg/1**
COMPOUND
1 Alachlor* (Lasso"?
(2-chloro-2' ,6'-
diethyl-N- (mathoxy-
methyDacetanilide]
Z Alachlor, one addi-
tional chlorine
homolog
3 Alkylbenzene-C- (nv
xylene or p-xylene)
4 Alkylbenzene-C, (p-
ethyltoluene or
n-propylbenzene)
5 Alkylbenzene-C,
6 Atrazine* (2-
chloro-4-ethyl-
amino-6-isopropy-
laraino-s-triazine)
~
7 Atrazine, deethyl*
(2-chloro-4-amino-
6-isopropylamino-3_-
triazine)
(desethylatrazine)
RRT+
•
1.09
1.14
0.35
0.42
•vO.52
..
1.00
0.96
CARROLLTON WATER PLANT
CCS
0.82
0.28
0.03
<0.1
4.9
0.51
2-Month
CCE
1.9
1.7
7.5
2.4
3.7
0.78
Mega-
Sample
CCE
0.44
0.18
0.05
0.01
2.7
0.22
Resin
Extract
0.67
0.21
J
1.0
JEFFERSON 11 WATER PLANT
CCE
2.3
' • '
.
5.2
0.27
2 -Month
CCE
2.9 .
T'
P
5.6
. •
1.9
4.8
0.80
Resin
Extract
0.35
/
0.64
JEFFERSON #2 WATER PLANT
70-Year++
CCE
2.1
0.05
0.03
5.4
0.27
2 -Month
CCE
1.4
P
6.2
2.2
3.2
0.75
Resin
Extract
0.17
0.18
Notes
c,d,f,
-------
TABLE 2 (Continued)
ORGANIC COMPOUNDS IN NEW ORLEANS DRINKING WATER—CONCENTRATION IN pg/1**
COMPOUND
8 Benzaldehyda*
9 Benzyl butyl
phthalata*
10 Bromodlchloro-
jna thane .
11 Bromoform*
12 . Butachlor*
. (Machete^*) [2-chloro-
2' ,6'-diethyl-N-
(butoxymethyl) -
3=» acetanilide]
-i
£ 13 Butyl octyl
malaate*
14 a-Chlordane
15 Chlordene*
16 bis-2-Chloroethyl
ether*
17 Chloroform*
18 bis-2-Chloroisopro-
pyl ether*
19 m-Chloronitroben-
zena* .
RRT+
0.44
1.29
0.18
0,32
1.21
0.44
0.51
CARROLLTON WATER PLANT
70-Year++
CCB
0.64
P
0.57
0.05
0.04
0.18
2-Month
CCE
1.4
Maga-
S ample
CCE
0.03 .
0.24
'
0.02
'' '
<0.1(T)
<0.1(T)
P
Resin
Extract
<0.1
<0.1
J
J
JEFFERSON *1 WATER PLANT
70-Yeart+
CCE
0.83
0.06
0.16
0.08
2-Month
CCE
1.8
Resin
Extract
<0.l
J
j
JEFFERSON 12 WATER PLANT
70-Year"1"1"
CCE
0.75
0.05
0.12
0.03
2 -Month
CCE
1.6
Resin
Extract
o.oe
Notoa
h
g,h
b
b,g
\
g,h
h
h
g
a,b
g/h
-------
TABLE 2 (Continued)
COMPOUND
20 Cyanazine* (Bladetf^J
[2-(4-chloro-6-
ethylamino-s-tri-
azin-2-ylamTno) -2-
methylpropionitrile ]
21 DDS*[2,2-bis-{p-
chlorophenyl) -1,1-
dichloroethylene]
22 n-Deoane*
23 Decane, branched
24 Dibromochloro-
me thane* .
25 Dibromodichloro-
J* ethane isomer
^
N>26 Dibutyl phthalate*
27 2,6-Di-t-butyl-p-
benzoquinone*
28 m-Dichlorobenzene*
29 Dicyclopentadiene*
30 Dieldrin*
31 Diethyl phthalate*
RRT*
1.13
0.48
0.44
0.23
0.35
1.12
0.84
0;46
1.23
0.91
ORGANIC COMPOUNDS IN NEW ORLEANS DRI
CARRQLLTON WATER PLANT
70-Year*+
CCE
0.35
0.02
1.1
0.33
0.10
0.22
-
0.04
0.03
2-Month
CCE
2.4
5.8
<10
<3
0.24
Mega-
Sample
CCE
<0.01
0.06
0.03
0.6
0.16
0.09
•
0.01
0.02
Resin
Extract
0.05(J)
<0.1
0.05
-
O.Ol(J)
J
NKINQ WATER—CONCENTRATION IN
JEFFERSON 11 WATER PLANT
70-Year'H'
CCE
0.21
0.4
0.36
0.21
— ^
0.07
0.03
2-Month
CCE
2.0
5.4
0.10
Resin :
Extract
0.05(J)
<0.1
0.01
•
-------
TABLE 2 (Continued)
ORGANIC COMPOUNDS IN NEW ORLEANS DRINKING WATER—CONCENTRATION IN Jjg/l**
COMPOUND
32 Di-(2-ethylhexyl)
adipate*
33 Di-(2-ethylhexyl)
phthalate*
34 Dihexyl phthalate
35 Dihydrocarvone
36 Diisobutyl phtha-
late*
37 Dimethyl phthalate*
38 Dipropyl phthalate*
39 n-Dodecane*
40 Endrin*
41 Ethyl acetate
42 Ethylbenzene
43 o-Ethyltoluene*
44 m-Ethyltoluene*
RRT+
1.32
1.38
1.24
6.63
1.06
0.83
1.02
0.67
1.27
(1st
>eak)
0.32
0.45
0.42
CARROLLTON WATER PLANT
70-Year++
CCE
0.10
0.03
0.14
0.27
0.07
0.01
2 -Month
CCE
0.40
<1
0.60
0.10
2.3
Mega-
Sample
CCE
0.10
11
0.59
P
0.01
0.004
(T)
0.02
Resin
Extract
0.05
0.05
<0.05
P
JEFFERSON 11 WATER PLANT
70-Year"1"*"
CCE
0.46
0.06
0.13
0.13
0.008
(Tl
0.04
0.05
2-Month
CCE
0.50
0.82
•
0.40
1.6
Resin
Extract
J
JEFFERSON #2 WATER PLANT
70 -Year"1"1"
CCE
0.27
0.07
0.18
0.14
0.006
(T)
0.02
0.02
2 -Month
CCE •
1.2
0.74
0.37
1.8
Resin
Extract
0.16
0.16
P
Notes
d,f,g
g
f
c,f,g,h
g
-------
TABLE 2 (Continued)
ORGANIC COMPOUNDS IN NEW ORLEANS DRINKING WATER—CONCENTRATIONS IN vq/1**
COMPOUND
45 1,2,3,4,5,7,7-
Heptachloronorbor-
ncne*
46 Heptachloronorbor-
nene isomer
47 Hexachloro-1,3-
butadiene*
48 Hexachloroethane*
49 Isophorone*
50 Limonene*
1,51 Methyl benzoate*
•C-
52 Methylnaphthalene*
53 Naphthalene*
54 n-Nonane*
55 Pentachloroethane*
56 Pentachlorophenyl
methyl ether
57 n-Pentadecane*
RRT+
0.94
0.98
0.65
0.53
0.60
0.50
0.58
0.38
•\.0.38
0.86
CARROLLTON WATER PLANT
70-Year"*"1
CCE
0.06
0.06
0.16
4.3
1.6
0.03
0.03
0.03
2-Month
CCE
0.70
2.9
2.4
Mega-
Sample
CCE
0.03
0.02
0.07
0.39
(T)
Resin
Extract
0.02
0.04
0.03
J
<0.1
JEFFERSON 11 WATER PLANT
70-Year»*
CCE
0.07
0.04
0.27
0.19
2.8
2-Month
CCE
<11 '
2.4
0.10
Resin
Extract
0.04
0.12
<0.1
<0.01
J
J
0.01
JEFFERSON #2 V7AT2R PLANT
70-Year"1"1"
CCE
0.07
0.04
0.21
0.30
2. 9
2-Month
CCE
9.5'
2.1
0.10
Resin
Extract
0.01
<0.01
<0.01
Notes
d,f,
g,h
f ,g,h
f,g,h
c,g
e
-------
ORGANIC COMPOUNDS IN NEW ORLEANS DRINKING WATER—CONCENTRATIONS IN pg/1**
COMPOUND
58 Propazine* [2-
chloro-4,6-bis-
(isopropylamino) -
s_-triazine]
59 Simazine* [2-
chloro-4 ,6-bis-
(ethylamino) -s-
triazine]
60 1,1,1,2-Tetra-r
chloroe thane *
61 Tetrachloroethy-
lene*
62 n-Tetradecane*
4> 63 Toluene*
Ut
64 1 , 1 , 2 -Trichloro-
ei jane*
65 1,1,1-Trichloro-
propane
66 1,2, 3-Trichloro-
propane
67 n-Tridecane*
68 Trimethylbenzene
isomer
RRT+
-H 0
•vl 0
0.30
0.26
0.80
0.22
0.21
M>.07
0.36
0.72
0.46
CARROLLTON WATER PLANT
TO-Year*4
CCE
<0 1
0.04
0.02
0.35
0.01
0.04
2-Month
CCE
<1
0.10
11.
6.2
0.30
6.1
Mega-
Sample
CCE
<0 1
0.11
0.08
<0.2
<0.2
0.02
Resin
Extract
• J
<0.1
<0 . 1
<0. 1
0.01^
JEFFERSON #1 WATER PLANT
70-Year'1"'"
CCE
<0. 1
-------
TABLE 2 (Continued)
ORGANIC COMPOUNDS IN NF.W ORLRANS DRINKING WATER— CONCENTRATIONS
COMPOUND
69 1, 3, 5 -Trine thy 1-
2,4,6-trioxo-
hexahydrotriazina*
70 Triphenyl phos-
phate*
71 n-Undecane*
72 Undecane, branched
73 o-Xylene*
74 Unknown chlori-
nated fluorinated
hydrocarbon
75 Unknown chlori-
nated fluorinated
hydrocarbon
76 Unknown compound.
apparent MW 145
77 Unknown dichlori-
nated compound
MW 200
78 Dichloroiodo- .
methane
79 Unknown dichlori-
nated compound
MW 249
RRT*
0-.73
1.31
0..58
0.52
0.33
0.78
0.82
-0.40
0.82
0.26
H.O
CARROLLTON WATER PLANT
70-YearH>
CCE
0.01
0.12'
0.06
0.33
.
<0.01
<0.01
<0.7
1.1
<0.04
2-Month
CCE
2.5
5.3
4.1
Mega-
Sample
CCE
0.01
0.03
0.03
0.04
0.12
<0.3
0.05
0.84
<0.02
Resin
Extract
<0.1
<0.1
JEFFERSON il WATER PLANT
70-Year'*"''
CCE
<0.9
1.3
P
2-Month
CCE
<10
T
2.8
I
*
*
Resin
Extract
-------
TABLE 2 (Continued)
ORGANIC COMPOUNDS IN NEW ORLEANS DRINKING WATER—CONCENTRATIONS IN pg/1**
COMPOUND
80 Unknown phthalate
81 Unknown phthalate
S2 Unknown phthalate
RRT+
1.22
1.24
1.36
CARROLLTON WATER PLANT
ZO-Year"1"1"
CCE
2 -Month
CCE
Mega-
Sample
CCE
0.01
0.01
0.12
Resin
Extract
JEFFERSON #1 WATER PLANT
70-Year+4-
CCE
2-Month
CCE
Resin
Extract
JEFFERSON #2 WATER PLANT
70-Year++
CCE
2-Month
CCE
Resin
Extract
Notes
* Confirmed by matching GC retention time and, in most cases, mass spectrum with that of a standard.
** Blank spaces indicate that compound was not detected in that specific sample. .,
+ Relative retention time (to atrazine); mostly determined by the GC computer system.
f+ Concentration values for 70-year CCE compounds are sums of concentrations in CCE #1 (filter #1) and CCE #2 (filter #2).
J Detected only by Junk, USAEC-Ames Laboratory. Identified in the Carrollton or Jefferson No. 1 resin extracts by GC-MS,
but not quantitated except for DDE and dieldrin. These compounds were not detected at SERL in the resin extracts.
P Present, usually in very low concentrations (<0.05 yg/1), but not quantitated due to interferences with other GC peaks or
lack of a discernable GC peak (detected only by mass spectrometry).
T Detected only by Thruston at SERL by GC-MS after fractionation by column chromatography.
NOTES:
a Chloroform was identified at SERL only in the tetralin extract of the CarrolVton Water.
b These compounds were also identified at the Water Supply Research Laboratory/'EPA, Cincinnati, by VOA analysis of the
Carrollton water.
c Also identified in the Carrollton CAE.
d Also identified by Junk, USAEC-Ames Laboratory, in the Carrollton resin extract.
-------
>
00
Notes (Continued)
Also identified by Junk, USAEC-Ames Laboratory, in the Jefferson II resin extract.
A1SO lUBIll.iJtAOV* JJJ w Ulii» , KW.^.— .-.—— * , ( _
Also identified by Melton at the Water Supply Research Laboratory, EPA, Cincinnati, by GC-MS analysis of a solves ox;,act
of a 1 liter grab sample of the Carrollton water.
g Also identified by duplicate analysis of the Carrollton 70-year CCE using the Varian CH5/DF system.
h Also identified by GC-MS (Finnigan) analysis of the Carrollton mega-sample CCE after fractionation by TLC.
i Confirmed by high resolution mass spectroroatry {only a structural indication in the casa of compound 12).
j Confirmed by GC-Ft-IR.
k Identification based on a report received after this table was prepared (EPA Mass Spectrometer User's Group Newsletter
No. 15, April 1975)—therefore not in alphabetical order.
® Registered trade names are given only as an aid to compound recognition, and,do not indicate the source of the compound.
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing!
1. REPORT NO.
EPA-906/9-75-003
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Analytical Report
Supply Study
New Orleans Area Water
5. REPORT DATE
June 1975
6. PERFORMING ORGANIZATION CODE
906
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
Surveillance ft Analysis Division, Region VI
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Surveillance ft Analysis Division, Region VI
U. S. Environmental Protection Agencv
1600 Patterson, Suite 1100
Dallas, TX 75231
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agencv, Region
1600 Patterson, Suite 1100
Dallas, TX 75231
13. TYPE OF REPORT AND PERIOD COVERED
n Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Detailed descriptions of sampling and analytical methods used to identify and
quantify trace organic compounds present in the finished (i.e. treated) water
of the New Orleans, Louisiana area water supply, specifically the Carrollton
Water Plant (City of New Orleans), Jefferson Parish No. 1 Water Plant (Metairie)
and the Jefferson Parish No. 2 Water Plant (Marrero).
Organic compound identifications and concentrations are presented in Table 2.
Methods of analysis of carbon and resin extracts are presented in the Appendix.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS C. COSATI I icld, (.,.. i,
Water supply
trace organics
Gas Chromatography/Mass Spectroscopy
carbon adsorption
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
54
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
A-49
-------
|