EPA-R2-73-155
MAY 1973 Environmental Protection Technology Series
Environmental Applications of
Advanced Instrumental Analyses
Assistance Projects, FY 69-71
^ PRO^°
Office of Research and Monitoring
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
been grouped into five series. These five broad
categories were established to facilitate further
development and application of environmental
technology. Elimination of traditional grouping
was consciously planned to foster technology
transfer and a maximum interface in related
fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL
PROTECTION TECHNOLOGY series. This series
describes research performed to develop and
demonstrate instrumentation, equipment and
methodology to repair or prevent environmental
degradation from point and non-point sources of
pollution. This work provides the new or improved
technology required for the control and treatment
of pollution sources to meet environmental quality
standards.
-------
EPA-R2-73-155
May, 1973
ENVIRONMENTAL APPLICATIONS OF ADVANCED INSTRUMENTAL
ANALYSES: ASSISTANCE PROJECTS, FY 69-71
by
Lawrence H. Keith and Shirley H. Hercules
Southeast Environmental Research Laboratory
College Station Road
Athens, Georgia 30601
Project 16020-GHZ
Program Element 1B1027
NATIONAL ENVIRONMENTAL RESEARCH CENTER
OFFICE OF RESEARCH AND MONITORING
U. S. ENVIRONMENTAL PROTECTION AGENCY
CORVALLIS, OREGON 97330
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, B.C. 20402
Price $1.25 domestic postpaid or $1 GPO Bookstore
-------
ABSTRACT
A multitude of analyses involving the identification
and measurements of organic pollutants in water are
discussed under eleven project categories. In most
cases these analyses have helped to solve, or at least
understand more clearly, the related pollution incident
and in some cases provided evidence for enforcement of
regulatory legislation.
11
-------
CONTENTS
Page
Abstract ii
List of Figures v
List of Tables viii
Acknowledgments x
Sections
I Conclusions 1
II Recommendations 3
III Introduction 5
IV Projects
1 Tennessee River Contaminant (bis 12- 7
ethyIhexyl]phthaiate)
2 Fish Kill Contaminant Characteriza- 10
tions (pesticides)
3 Confirmation of PCB's in Environmental 20
Samples
4 Escambia Bay, Florida, Industrial 23
Pollution
5 Kansas City Landfill Contaminants 28
6 Identification of the Cause of Earthy- 29
Musty Taste and Odor Problems in Ohio
Water Supplies
7 Western Louisiana Industrial Waste 35
Survey
8 Color Body Characterization and 50
Related Studies from a Kraft Paper
Mill Effluent
9 Characterization of Foam from Kraft 59
Pulp Mills
111
-------
Page
10 Analysis of Oil Spills by NMR 63
Spectroscopy
11 Analysis of Oil Spills by Fluores- 72
cence Spectroscopy
V References 81
IV
-------
FIGURES
No. Page
1 Infrared Spectrum of the Tennessee River
Sample 8
2 Mass Spectrum (top) and NMR Spectrum (bottom)
of Tennessee River Sample 9
3 Comparison of the Infrared Spectrum of the
Tennessee River Sample with that of
bis(2-ethylhexyl)phthalate 11
4 Comparison of the NMR Spectrum of the
Tennessee River Sample with that of
bis (2-ethylhexyl)phthalate 12
5 Comparison of the Mass Spectrum of the
Tennessee River Sample with that of
bis(2-ethylhexyl)phthalate 13
6 Comparison of the Mass Spectrum of Boone Lake
Extract with that of Phenylmercuric Chloride 14
7 Comparison of the Mass Spectrum of Tombigbee
River Extract with that of Diazinon 17
8 Comparison of the Mass Spectra of the Locust
Fork Branch and Black Warrior River Extracts
with those of Malathion Obtained by GC-MS
and by Direct Probe Insertion 18
9 Comparison of the FID Gas Chromatogram of
Escambia Bay Sediment Extract with that of
Aroclor 1254 21
10 Comparison of Select Chlorine Isotope Clusters
from Escambia Bay Sediment Extract with that
from Corresponding PCB's in Aroclor 1254 22
11 Comparison of the FID Gas Chromatogram of
Great Miami River, Ohio Extract with that
of Aroclor 1248 24
12 Comparison of Select Chlorine Isotope Clusters
from Great Miami River, Ohio Extract with that
from Corresponding PCB's in Aroclor 1248 25
v
-------
No. page
13 The NMR Spectrum of "Celinol" 31
14 Comparison of the Mass Spectrum of "Celinol"
with that of Isoborneol 32
15 Comparison of the Infrared Spectrum of
"Celinol" with that of 2-Methylisoborneol 33
*
16 Comparison of the Mass Spectrum of Indian
Lake Extract with that of Geosmin 34
17 Chemically Characterized Gas Chromatogram
of Petrochemical Company "A" 37
18 Chemically Characterized Gas Chromatogram
of Petrochemical Company "B" 41
19 Chemically Characterized Gas Chromatogram
of Petrorefinery "A" 43
20 Chemically Characterized Gas Chromatogram
of Petrorefinery Company "B" 46
21 Chemically Characterized Gas Chromatogram
of Synthetic Rubber Company 48
22 Chemically Characterized Gas Chromatogram
of Chemical Company "A" 51
23 Chemically Characterized Gas Chromatogram
of Chemical Company "B" 52
24 Comparison of Kraft Pulp Mill Effluent and
Marsh Water Color Intensity as a Function of
pH 54
25 Chemically Characterized Gas Chromatogram of
Foam from a Kraft Papermill Bleach Stage
Caustic Clarifier 62
26 Comparison of NMR Spectra of Oil from the
Suspect Barge and from a Florida Beach 64
27 Comparison of NMR Spectra of Oil from
Duluth Harbor and the Primary Suspect Source 66
VI
-------
No.
28 Comparison of NMR Spectra of Oil from
Duluth Harbor and the Secondary Suspect
Source 67
VII
-------
TABLES
No. Page
1 Compounds Identified in the Escambia Bay,
Florida Industrial Waste Survey 26
2 Compounds Identified in the Kansas City,
Kansas Landfill Analysis 27
3 Chemical Characterization and Gross Pollution
Measurements of Petrochemical Company A 38
4 Gross Pollution Measurements and Chemical
Characterization of Petrochemical Company B 42
5 Gross Pollution Measurements and Chemical
Characterization of Petrorefinery Company
"A" 44
6 Gross Pollution Measurements and Chemical
Characterizations of Petrorefinery Company
"B" 47
7 Gross Pollution Measurements and Chemical
Characterization of Synthetic Rubber Company 49
8 Chemical Characterization of Chemical Company
A 53
9 Chemical Characterization of Chemical Company
B 53
10 Comparison of Kraft Pulp Mill Effluent and
Marsh Water Fluorescent Spectra 55
11 Color Leached from Lagoon Mud as a Function
of Calcium Hydroxide Concentration 56
12 Comparison of the Amounts of Acid Insoluble
Organics in the Lime Treated and the Untreated
Effluent Samples 57
13 Compounds Tentatively Identified in the
Neutral Volatiles Fraction of the Riceboro
Pulp Mill Effluent 58
Vlll
-------
No. Paqe
14 Compounds Tentatively Identified in the
Phenols and Acids Fraction of the Riceboro
Pulp Mill Effluent 58
15 Compounds Identified in Perdido Bay Kraft
Papermill Wastewater, Foam, and Scum Samples 60
16 Normalized Peak Areas of Proton Signals in
Samples from the St. Mark/ Florida, Oil
Spill 65
17 NMR Comparison of Oil from West Florida
Beaches with Suspect Crude Oil and Bunker C
Oil 70
18 Fluorescence Spectra of Duluth Harbor Samples 73
19 Fluorescence Spectra of Savannah River
Samples 74
20 Fluorescence Spectra of the Duluth Superior
Harbor Samples 76
21 Fluorescence Spectra of the Dundee Canal
Samples 77
22 Fluorescence Spectra of St. John's River
Samples 79
23 Other Analyses of St. John's River Samples 80
IX
-------
ACKNOWLEDGMENTS
The following NWCCRP staff were the principal investi-
gators of the projects listed in Section IV:
Project
1. L. H. Keith and A. W. Garrison
2. A. W. Garrison and A. D. Thruston, Jr.
3. A. W. Garrison
4. A. W. Garrison
5. R. G. Webb
6. A. W. Garrison and L. H. Keith
7. L. H. Keith
8. C. G. Gustafson
9. L. H. Keith
10. L. H. Keith
11. A. D. Thruston, Jr.
The assistance of A. L. Alford, T. L. Floyd, F. R.
Allen, A. C. McCall, and M. M. Walker in preparing the
samples for analysis, providing spectral data, inter-
pretation of spectra, and confirmation with standards
is gratefully acknowledged.
x
-------
SECTION I
CONCLUSIONS
The identification and quantification of specific
chemical compounds responsible for a wide variety of
pollution incidents have helped to solve, or at least
to understand more clearly, the problems associated
with the incidents. In some cases evidence was provided
for enforcement of regulatory legislation. The projects
have demonstrated the applicability of newly developed
laboratory techniques to environmental pollution
problems.
Because of the complexity and small size of the samples
available, gas chromatography-mass spectrometry has
been the most valuable technique for analysis. Nuclear
magnetic resonance and infrared spectroscopy were
helpful when a single compound was isolated in sufficient
quantity.
-------
SECTION II
RECOMMENDATIONS
Analyses of the type described will be necessary and
should be used for establishing and enforcing water
quality effluent standards for the industrial permit
program. The data provided will also be needed to
update these standards with respect to specific
organic compounds discharged into receiving waters.
Many more laboratories, including those at the state
level, should be equipped to conduct the type of
analyses described herein.
Newly developed techniques should be investigated for
their applicability to pollution analysis.
An annual report of unpublished contaminants charac-
terization work is recommended.
-------
SECTION III
INTRODUCTION
The National Water Contaminants Characterization
Research Program (NWCCRP) develops methods for the
identification and quantification of chemical pollu-
tants and identifies specific compounds associated with
various sources of pollution. Information thus
obtained may be used in the development of regulatory
legislation. In assessing the applicability of new
techniques to environmental samples and developing
information that the NWCCRP has a unique capability
for providing, a number of samples are analyzed that
are directly related to a wide range of environmental
problems.
Results of the aforementioned analyses are reported in
individual memoranda to the person requesting the
analysis and therefore receive limited distribution.
To acquaint other researchers and administrators with
the type of information that can be provided and to
help analytical chemists recognize the broad applica-
bility of the techniques used, a series of annual
reports is planned. Brief summaries will be presented
of the problems under study by the NWCCRP. This first
report, summarizing the problems studied during the
fiscal years 1968 through 1971, deals mostly with the
application of gas chromatography-mass spectrometry
since the NWCCRP was the only group in the Federal
Water Quality Administration with a GC-MS capability
during this period. Future reports will discuss
applications for a wider variety of techniques.
-------
SECTION IV
PROJECTS
1. Tennessee River Contaminant
(bis [2-ethylhexyl] phthalate)
When the Ohio River is at flood stage, the Tennessee
River dam above Calvert City, Kentucky, is closed,
releasing little or no flow down the Tennessee River.
During the spring of 1968 the dam was closed and the
industries on the lower Tennessee River complained
about the pooling of industrial wastes under the
stagnant river condition.
Samples of the contaminated water were taken by
personnel at the Robert A. Taft Sanitary Engineering
Center, Cincinnati, Ohio. Analysis by flame ionization
gas chromatography revealed the presence of one
compound whose concentration far exceeded that of all
others. IR and GC of chloroform extracts suggested
"the presence of a phthalate very similar to dinonyl
phthalate" (1). An extract was submitted to the NWCCRP
for molecular structural identification.
Both thin layer and gas chromatography indicated that
the extract contained only one major component in about
90% purity. The NMR, infrared, and mass spectra were
therefore recorded with no further purification. The
infrared spectrum (Figure 1) shows peaks characteristic
of a carbonyl, an ester, and an ortho substituted
phenyl (2). NMR and mass spectra (Figure 2) confirm
the presence of a phthalate and indicate it to be a
dioctyl phthalate ester.
Fifteen possible isomeric structures exist for dioctyl
phthalate, assuming the diester to be symmetrical.
Further analysis of the NMR peaks in conjunction with
the electronic integration yielded the structure
bis[2-ethylhexyl] phthalate (I).
-------
00
2.5
100
WAVELENGTH U
5 6
i....i .... i .... i .... i .
8
i.
10
i 11 .
12
15 20
...i .1.1.1.
80-
60 -
40--
20-
0--
TENNESSEE RIVER SAMPLE
4000
3500
3000
2500
2000 1800 1600 • 1400 1200 1000 800
FREQUENCY CfTl
.-I
600
Figure 1. Infrared Spectrum of the Tennessee River Sample
-------
10O
90-
80-
£ '
s
10-
0
TINNISSEE RIVER SAMPLE
49
,67
279
lio ' i4o lio i«o ' 2Jo
M*
3W
3^0 ' 2^0 ' 360 ' 3*0 ' 340 ' 3M ' 3^0
CH2
20 (o
T»nn»u«« River Sample
2o
Figure 2. Mass Spectrum (top) and NMR Spectrum (bottom) of Tennessee River
Sample
-------
Comparison of the NMR, IR, and mass spectra of the
sample with those of a standard confirmed the above
structure (Figures 3-5).
The results of the structure determination were
submitted to the Taft Center.
2. Fish Kill Contaminant Characterizations
(Pesticides)
Boone Lake
During the summer of 1968 a massive kill of approxi-
mately one million fish occurred in the Boone Lake
reservoir, Tennessee. After eliminating all of the
more common causes of fish kills, the contents of some
nearly empty 55-gallon drums found floating on the
reservoir were investigated. The metal drums were of
the kind commonly used to float docks and boathouses.
The manufacturer listed on one of the drum labels,
Buckman Laboratories, Inc., indicated that the drum may
have contained a mixture of phenylmercurie acetate and
2,4,6-trichlorophenol, a microorganism control product.
Subsequent analysis of the river water and drum residue
revealed the presence of the trichlorophenol and
diphenyl mercury, a decomposition product of the phenyl-
mercuric acetate. A residue sample analyzed by mass
spectrometry using the direct probe inlet was shown to
contain phenylmercurie chloride (Figure 6, top). How-
ever, recent work in the National Environmental Research
Center, Analytical Quality Control Laboratory of
Cincinnati, Ohio, indicates that the acetate is readily
converted to the chloride. Confirmation of the
structure was obtained by comparison of the mass
10
-------
WAVELENGTH
rH U
6 ~ 7
4OOO 3500 30OO 2500
-i i 1 1 1 1 1 r
2OOO I8OO 1600 , 1400 1200 IOOO 80O 60O
FREQUENCY CIT1
WAVELENGTH
4000 3500 30'00
zo'oo
1800 FREQUENCY *g°-
iooo s5o eAo
Figure 3. Comparison of the Infrared Spectrum of the Tennessee River Sample
with that of bis (2-ethylhexyl)phthalate
-------
NJ
n*M«« Kiv«r Sampl*
Aromatic
TIM
^^JC™
tfWjPy^AIVMWUWVfrK "WVl^tlV
I 1 1 1
hi.-(l-ETHYLHEXYL) PHTHALATE
-C-O-CHj-CM
TS.7S
Aromalk
r 120
'f
CH2
CH2
~CH
J
Figure 4. Comparison of the NMR Spectrum of the Tennessee River Sample with
that of bis(2-ethylhexyl)phthalate
-------
u>
IOO
90-
80-
70-
|so-
S 40-
|30-
20-
10-
O
100-
90-
60-
X70-
1 Sfr
J 40-
| 30-
3 20
1O-
0
TENNESSEE RIVER SAMPLE
, I, H
20
•
u
, . ti
1
* nlii .id li .,1
40 60 80 100
i 1 •"! 1 • 1
49 o
-^ J
OC^-H
0
ifirc~°H '°H (QC^H17
li^-C-OH ff^]"^"0" \^-~<^C>t8M17
167 \)
1 T 7
120 ' 140 160 180^200 ' 220 ' 240 ' 260 ' 280 ' 300 ' 320 ' 340 ' 360 ' 3^0 4OO
bii-(2-ETHYLHEXYL) PHTHALATE
AN
,1 ,
20
INDUSTRIAL POLLUTANT
,i
i
.1 1 1 , , 1,
T1 T 1 1 ~l J 1 1 *~
4O 60 80 IOO
o
(C3C/j~H
iWb
OH J- -,
o^t^H ^ oci8"17
167 ,^^-O-OH L "
K,)L— u
^^"^ ^tpNI 17 M*
, 279 390
1, . \
120 ' 140 160 ' 180 ' 200 ' 220 ' 240 260 2BO ' 300 32O 34O 360 ' 3*> ' 4OO
Figure 5. Comparison of the Mass Spectrum of the Tennessee River Sample with
that of bis(2-ethylhexyl)phthalate
-------
100
90
80
8 70
•f 60
I 50
> 40
I 30
at
20
10
0
EXTRACT OF SAMPLE FROM BOONE LAKE, TENN.
MT
314
202
l.i
279
40 60 80
1OO 120 140 160 180 200 220 240 260 280 300 320
m/9
100
90
80
8 70
| 60
| 50
S 40
I 30
ae
20
10
0
77
PHENYLMERCURIC CHLORIDE
-Hq-CI
202
I
279
M*
314
40 60 80 100 120 140 160 180 200 220 240 260 280 300 320
m/e
Figure 6. Comparison of the Mass Spectrum of Boone Lake Extract with that of
Phenylmercuric Chloride
-------
spectrum with that of a standard (Figure 6, bottom).
Bioassay conducted by TVA personnel confirmed the
toxicity of the phenylmercury compound to fish.
Analysis of water samples indicated the presence of
diphenylmercury at greater than acutely toxic levels,
establishing it as the primary cause of the fish kill.
Beginning October 7, 1968, all drums containing and
suspected of containing toxic materials were to be
removed from the area of the kill by personnel of the
Division of Reservoir Properties. The year 1972 was
set by the Tennessee Valley Authority as a deadline
for the complete removal from the reservoir of all
drums used as flotation devices. A subsequent fish
kill in 1969 prompted the suggestion that attempts be
made to remove the drums before the suggested deadline.
Tombigbee River, Alabama
In 1968, a fish kill involving about 30 thousand fish
occurred on the Tombigbee River in Alabama. Geigy
Chemical Corporation, a manufacturer of pesticides
among other products, was a suspected source of the
contamination that killed the fish.
Gas chromatography indicated the presence of the
organophosphorus pesticide, diazinon (II), in the
water samples taken from the area of the fish kill, but
more substantial evidence was required for,legal
prosecution.
CH,
/
CH
CH3CH2O
II
Accordingly, the sample was chromatographed by TLC and
the adsorbent containing the spot with an Rf corres-
ponding to that of diazinon was scraped from the plate,
15
-------
A similar area of adsorbant was scraped from a blank
plate. The residues were introduced into the MS by
direct probe; subtraction of the peaks present in the
spectrum of the blank from the peaks in the river
sample spectrum left a fragmentation pattern (Figure 7,
top) identical to that of a diazinon standard (Figure 1,
bottom) in the region above m/e 100.
A federal court awarded the state damages in the
amount of the value of the fish killed.
Black Warrior River and Locust Fork Branch Fish Kills
Fish kills in the Black Warrior River occurred at the
Locust Fork Branch, near Birmingport, Alabama in
October, 1969, and again near Demopolis, Alabama in
September, 1970. The 1969 kill involved 750 thousand
fish and the 1970 incident killed 8 thousand fish.
Both kills were suspected to have been caused by the
spraying of malathion (III) in conjunction with a U. S.
Corps of Engineers mosquito control program in this
area.
s
CH30-P-S-CH;
CH3O
ill
The presence of malathion was confirmed by GC-MS in
extracts of both the Locust Fork Branch (Figure 8-A)
and the Black Warrior River area near Demopolis
(Figure 8-B). The molecular ion (m/e 330) was not
observed in spectra of either of the samples or of the
standard when it was introduced from the gas chromato-
graphic column (Figure 8-C). Instead, the fragment
(m/e 173) resulting from cleavage of (0130)2PS2*
was the largest significant ion. A small parent ion
was found in a sample introduced by the direct probe
(Figure 8-D). The fragmentation pattern of 8-D is
16
-------
lOOj
90
80
S 70
e
"S 60
c
| SO
| 40
1 30
20
10
100
100
90
80
| 70
•f 60
3
3 50
> 40
1 30
QC
20
10
u
)0
1
ll
EXTRACT OF TOMBIGBEE RIVER SAMPLE
120 140 160 180
J 1
j iL
1 |
120 ' 140 160 U
io
(CORRECTED FOR BACKGROUND)
304
i
a
li II .LJ ll
200 220 240 ' 260 280 300
m/e
DIAZINON
,CH3
S __/Cv
II fj^A CH,
:H3CH2O-P-O-/QjN
CH3CH2O ^^CH
NT
304
bl
ll II 1 , I , .
200 220 240 260 280 300 320
m/e
Figure 7. Comparison of the Mass Spectrum of Tombigbee River Extract with that
of Diazinon
-------
93
III! 1 1
20 1 *> 1 60 1
1
i
1 SO
MAIATHION - LOCUST CREEK
173
08 ''
hi
WO 1 UO 1 180 I 200 '""220 l 240 < 260 ' 280 > 300 325 3
B
no
80
t
i
s
g 40
4
£
20
<
j
1
1
93
1 1
ill III III:
19
12
1
MAIATHION - HACK WARRIOR
CRECK
173
158
. 1 „ ..
l-ffl) -T4CT
no
80
j" 60
Z
1 40
5
20
d
no
80
s
£ 60
i
X
20
0
I
99
9"
j
1
ill
1 i 1,
J I
J
,
12
I
MALATHION STANDARD - GC/MS
173
158
„ Ll, ,1 ,
1 1
20 40 60 1 W 1 100 * 12O 140 ! 160 1 180 1 200 ! 220 I 240 > 260 ' 2% < 300 320 34O
CHjO»« MALATHION STANDARD - DIRECT PROBE
CHgO* " /CH3O. SS \t
MS \CH30- *SH/
\^ IKft
CH30»
CH3O*P
93-
.
i
- I
III ill
\
|
I , , 1
\
\
\
j
j
,
1 ,
!l> III
CH30-'"S^HCOOC»H9
o
+ II
O
173
M*
330
! i !
20 i «J i M 1 80 1 KX) 1 120 1 140 1 160 i 188 i 200 1 220 1 240 ! 260 290" ' 300 " 3gT ^ 340
Figure 8.
Comparison of the Mass Spectra of the Locust
Fork Branch and Black Warrior River Extracts
with those of Malathion Obtained by GC-MS
and by Direct Probe Insertion
18
-------
significantly different from 8-C (primarily due to
different sample pressures in the ion source),
illustrating the necessity of comparing mass spectra
of samples with standards under identical conditions.
The major fragmentations of malathion are indicated
in Figure 8-D.
The results of these analyses were used in a brief
prepared by the Alabama Department of Conservation and
submitted to the Alabama State Attorney.
Yalobusha River—Mississippi
A kill of an estimated 15 tons of fish occurred in
April, 1971, in the Yalobusha River, Grenada, Mississ-
ippi. The organic chemical responsible for the kill
had been tentatively identified by gas chromatography
as pentachlorophenol (IV) . Confirmation of the
structure was requested for legal action.
OH
IV
The samples and a standard of pentachlorophenol were
each methylated with diazomethane and analyzed by GC-MS.
Both retention time and mass spectral analysis confirmed
the unknown to be pentachlorophenol. The mass spectrum
indicated also the presence of tetrachlorophenol isomer
in some of the samples.
The results of the structure determination were
submitted to the Office of the Attorney General in
Mississippi for prosecution.
19
-------
3. Confirmation of PCB's in Environmental Samples
PCB's (polychlorobiphenyls), manufactured by the chlorina-
tion of biphenyl, are distillation fractions containing
20 or more isomers. They have been widely used in a
large variety of industrial and consumer products since
1929. Manufactured throughout the world, PCB's are sold
in the U. S. solely by Monsanto Company under the trade
name Aroclors(R). Because of their resistance to
chemical breakdown, PCB's have become ubiquitous pollu-
ters of the environment (4).
Only recently has the cumulative chronic toxicity of
the chlorinated biphenyls to animal life been observed.
It has therefore become important to develop methods
for their detection and identification in environmental
samples.
Mass spectrometry has been used to confirm the presence
of PCB's, tentatively identified by electron capture
GC, in water and mud samples. The flame detector
chromatograms of the extracts are invariably compli-
cated, but the specificity of the mass spectrometer
allows unequivocable identification of submicrogram
amounts of PCB's.
Escambia Bay, Florida
In one case an Escambia Bay, Florida, sediment extract
cleaned on a Florisil column was analyzed for Aroclor
1254. The GC flame detector pattern is shown in
Figure 9. Eleven of the peaks were shown by MS to be
due to PCB's. Their retention times and chlorine
numbers (number of chlorine atoms per PCB molecule, as
determined by MS) correspond with those of an Aroclor
1254 standard, also shown in Figure 9.
Mass spectra showing several chlorine isotope clusters
for some of the PCB's from the Escambia Bay sediment
are compared with corresponding spectra from an Aroclor
1254 standard in Figure 10. The parent ion was
observed in all the PCB's. The major fragmentation
path is loss of successive chlorine atoms from the
parent molecule. Each such loss gives a fairly
definitive cluster of isotope peaks whose intensity
20
-------
PCBs- SEDIMENT EXTRACT
CHLORINE
NUMBERS
1
•50
40
•30
20
• in *°
IU m
8
TO
O
Om
70
AROCLOR 1254
5
E K
665
I
30
20
10
14 12
10
8 6
- TIME, min.
4
Figure 9.
Comparison of the FID Gas Chromatogram of
EsGambia Bay Sediment Extract with that of
Aroclor 1254.
21
-------
PCB's-SEDIMENT EXTRACT
100
80
£
I"
> 40
<
S
20
M*4C1
2CI
3CI
(M*-CI)
I
100
80
£
j"
— 40
5
DC
20
4CI
2CI 3CI
200 22O 24O Z6O 290 30O 20O 2
100
M*.SCI go
20 240 260 28O XO 32O
§
|60
^40
i
20
34O 2-
M*6CI
I
I
4CI
SCI
SCI
1O 260 2UU 3OO 320 34O 360
to
to
PCB's-AROCLOR 1254
KM
80
i
« "
;
« *
i
20
M*,4CI
2CI
(M*-2CI>
SCI
(M*-CI)
|
100
80
|
5 60
g
5 *"
te
2C
4CI
2CI
SCI
|
100
M*SCI _ ^
1
|
»-
z
» 60
S
5 40
i
20
i
M*,6CI
!
4CI !
SCI ,
r
; L
,, ! i!!<
i s t>' pi
j 1 , 1 s i
1
)
Figure 10. Comparison of Select Chlorine Isotope clusters from Escambia Bay
Sediment Extract with that from Corresponding PCB's in Aroclor 1254
-------
ratios depend upon the number of chlorine atoms left
on the fragment. A comparison of the sample and
Aroclor standard spectra (Figure 10) shows similar
isotope ratios and fragmentation patterns.
Gre at Mi ami Rive r, Ohi o
Figure 11 shows the GC flame detector chromatogram of
a water extract from the Great Miami River in Ohio.
Mass spectra of the components eluting from the GC
showed 15 of them to be PCB's. They appear to come
from a mixture of two Aroclors. The retention time
and chlorine numbers of the first eleven of these
correspond to an Aroclor 1248 standard, also shown in
Figure 11. However, the last four PCB peaks with
chlorine numbers of 5, 6, and 7 appear to be from
Aroclor 1254 (cf. Figure 9). These four components
are in relatively large concentrations, as in the 1254
standard, whereas the standard Aroclor 1248 contains
only trace amounts, if any, of 6-chlorine and 7-chlorine
PCB's.
Mass spectra showing the clearly definable chlorine
isotope clusters of some of the PCB's from the river
sample are compared with corresponding spectra from
an Aroclor 1248 standard in Figure 12. Only 6 yg of
Aroclors were present in the total sample according to
electron capture GC analysis. Approximately 0.1 yg
of an individual PCB gave a good spectrum.
4. Esoambia Bay3 Florida Industrial Pollution
The polluted state of the Escambia River and Escambia
Bay near Pensacola, Florida, was suspected to be the
major cause of numerous fish kills during 1969. The
governor of Florida requested technical assistance in
evaluating pollution from sources on the river. The
NWCCRP was asked to characterize three effluent
samples by GC-MS.
The nine compounds identified and confirmed by GC-MS
are listed in Table 1 along with five that were
tentatively identified. In addition, diphenyl ether,
which may be implicated in odor problems and acrylo-
nitrile, were indicated in the Monsanto effluent.
23
-------
PCB's-OHIO RIVER WATER
•70
CHLORINE
NUMBERS
TIME,min.
Figure 11.
Comparison of the FID Gas Chromatogram of
Great Miami River, Ohio Extract with that
of Aroclor 1248
24
-------
RGB's-OHIO RIVER WATER
£40
1(1
M«3CI
2CI
It-
£60
Z
1»
s
*
TOD 5St JMT
3£l
|60
4CI
aaa 300 — aid 3*0
Ul
PCB's-AROCLOR 1248
1C)
M*,3tl '°°
a
w
t
* 60
i
£ 40
. S
s
**rt
JCI
no
M*,4CI
3CI
J _
80
£
i«o
5
i"
s
20
L
M*
JCI
1 •)
' 1 4CI !
' ' l'
"
•- -trim 4tm— — imn '-' wf~ 'vet^
Figure 12. Comparison of Select Chlorine Isotope Clusters from Great Miami
River, Ohio Extract with that from Corresponding PCB's in Aroclor
1248
-------
Table 1
Compounds Identified in the Escambia Bay, Florida
Industrial Waste Survey
Compound
Structure
Identi f ication
confirmed
a-Terpineol
Naphthalene
Diphenyl ether
Oibenzofuran
o-Nitrotoluene
p_-Nitrotoluene
Acrylonitrile
n-Hexadecane
n-Octadecane
n-Heptadecane
m-Nitrotoluene
2,4'-Dimethyldi-
phenylsulfone
Dibromopropene
2,6-Di-t-butylcresol
Yes
Yes
Yes
CH =CH-C=N
C16H34
C18H38
C17H36
(Isomer not determined)
(Isomer not determined)
Yes
No
No
No
No
26
-------
Table 2
Compounds Identified in the Kansas City. Kansas
Landfill Analysis
Compound
Phenol
£-Cresol
g-Cresol
Methylnaphthalene
DimethyInaphthalene
l-Methyl-4-ethylbenzene
3-Methylbiphenyl
1-Indanone
Diethyl phthalate
Di-n-butyl phthalate
Di-2-ethyl-n-butyl phthalate
A variety of n-alkanes
Structure
,OH
f"3- f«l
. "1
(6lJW^
^xS<0^Tsx'
O '*«—.
CH3{CH2)nCM3
27
-------
5. Kansas City Landfill Contaminants
Many complaints were registered in June, 1969, by
citizens of Kansas City, Missouri, concerning a medi-
cinal or "iodine-like" taste in their city water. An
investigation was undertaken to determine the source
of the contamination. The Kansas City Water Supply
Department, monitoring organics at their intake by gas
chromatography, showed a heavy organic load in the
water.
A suspected source of contamination was a landfill for
industrial wastes, recently reopened upstream of the
city. Colorimetric tests of extracts of earth and
fiberglass samples from the operation proved positive
for phenols (compounds that, when chlorinated during
water treatment, produce tastes similar to those
reported. Results indicated, therefore, that leaching
of soil and other materials being bulldozed into the
river was the source of the problem.
For corrective action, more positive identification of
the contaminants was required. Accordingly, samples
of fiberglass from the dump were leached for 30 minutes
in distilled water to simulate being pushed into the
river. An ether extract of the resulting solution was
examined by GC-MS. Positive identification was made
of phenol, p-cresol, and benzothiazole.
The Johnson County, Kansas, Water Treatment Plant
(Kansas City, Kansas) requested that a similar study
for potential pollutants be performed on yet another
landfill operation whose drainage could reach their water
intake. Should a similar problem occur in their city,
such information would permit immediate action.
Among the industries served by the disposal service
were a petroleum products plant, a paint company, and
several home appliance companies. Examination by GC-MS
of extracts of earth, water, and fiberglass indicated
the presence of the organic compounds in Table 2 in
addition to elemental sulfur.
The most interesting compounds from a taste and odor
standpoint are phenol, the cresols, and dimethylnaph-
thalene, which has an odor of burnt rubber and has been
28
-------
found in the effluent from a refinery's activated
sludge plant. Phenol and o- and p-cresol were
confirmed by comparison of mass spectra of standards
and the sample.
6. Identification of the Cause of Earthy., Musty
Taste and Odor Problems in Ohio Water Supplies
In March, 1968, the Advanced Waste Treatment Research
Laboratory in Cincinnati, Ohio, was requested to
conduct studies of an intermittent musty, earthy odor
in the Wabash River in Indiana. An investigation
showed that the odor was present constantly in Grand
Lake, Ohio, the principle headwater of the Wabash
River.
In 1929 an earthy odor in water was attributed to
actinomycetes (6) and recently a compound with a
similar odor, termed "geosmin", was isolated from
numerous actinomycetes cultures. Its structure was
identified as trans-1ylO^-dimethyl-trans-g-decalol (V) ,
(7,8). However, GC retention indices showed the river
pollutant to be different from "geosmin".
V
Rosen, Mashni, and Safferman of the Advanced Waste
Treatment Research Laboratory of the NERC, Cincinnati,
Ohio, isolated from a culture of streptomyces lavendulae
a compound that matched the odor and gas chromatography
retention indexes of the odorous compound from the lake
and river samples. It was shown to be different from
"geosmin" and was termed "celinol" after the city of
Celina, Ohio, which uses Grand Lake as a water supply.
29
-------
Infrared spectra indicated that the compound was an
aliphatic alcohol; high resolution mass spectra and
an element map showed a parent ion at mass 168 with
an indicated empirical formula of C-Q^QO. The sample,
believed to have a structure similar to "geosmin",
was submitted to NWCCRP for further structural
elucidation.
The NMR spectrum (Figure 13) considered in light of
the molecular weight and empirical formula, indicated
an aliphatic bicyclic structure. The mass spectrum of
"celinol" was compared with those of compounds
considered possible on the basis of the above evidence
—either a borane or a pinane. The fragmentation
pattern was similar to that of isoborneol, but the
major fragments were 14 mass units greater (Figure 14)
indicative of an added methyl group, which is consistent
with the NMR data. The extra methyl group, from NMR
evidence, was geminal to the hydroxyl, yielding the
structure 2-methylisoborneol (2-exo-hydroxy-2~methyl-
bornane).
2-Methylisoborneol had not been reported previously as
a naturally occurring compound and no commercial
standard was available. We therefore prepared the
compound by two different methods: 1) reacting camphor
(VI) with methyl magnesium bromide according to Zelinsky
(9) and 2) reacting camphor (VI) with methyl lithium
reagent.
I) CH3MgBr
° or 2) CHoLi
O
VI VII
Spectral comparisons (NMR, IR, and MS) confirmed the
structure of "celinol" to be identical to 2-methyl-
isoborneol (VII). Figure 15 shows the infrared compar-
ison.
30
-------
Figure 13. The NMR Spectrum of "Celinol"
-------
100
90-
80
r 70~
I 60-
- 50-
> 40-
.•
-H 30-
^20-
10-
"CELINOL" GC INLET
C3H5*
41
H20
1]
i '
'
0 ' 20
1 1
"T 1 i
,
,,l!
60
79 C^HQ4.
1
{
f T
93
|| [
T
ii
,
C8H11*
107
M-H20,CH3.
135
M-H20
C9H13* 150
121
1 i
0 ' 100 ' 120 140 160 '
m/e
100-
90
80
£70
M
g 60-
c
- 50-
|40J
D
^30-
at
20
10-
ISOBORNEOL
i
1
0 20 4
_^*
c
!-—•
MW:
1
, ll
•"_i — ' f"*1
0
A/
C
\
154
60
OH
YH
79
l .
l.
' 80
93
121
107
M-HoO
136
139
., ..
100 120 140 160
m/e
Figure 14.
Comparison of the Mass Spectrum of "Celinol"
with that of Isoborneol
32
-------
OJ
U)
"CELINOL"
2- METHYLISOBORNEOL
4OOO
350O
3OOO
20OO
WOO 1«00 1400
WAVENUMBER (CM"1)
"So"
1OOO
400
Figure 15. Comparison of the Infrared Spectrum of "Celinol" with that of
2-Methylisoborneol
-------
K>OT
80
149
M+182
Indian Lak* Samp)
60 80 100 120 140 160 1
80 100 120 140 160 180
M+182
Figure 16.
Comparison of the Mass Spectrum of Indian
Lake Extract with that of Geosmin
34
-------
A simultaneous, but independent, investigation of a
compound from three other species of actinomycetes
by Medsker, Jenkins, and Thomas (10) led to the same
structure as "celinol". Further samples from Grand
Lake were found to contain both "geosmin" and 2-methyl-
isoborneol. Samples from Indian Lake nearby contained
only "geosmin." Figure 16 illustrates the matching of
a known sample of "geosmin" with that of an extract
from Indian Lake.
The concentration of 2-methylisoborneol and geosmin
were estimated by Rosen to be 0.1 ppb and 0.03 ppb,
respectively in Grand Lake. The presence of 2-methyl-
isoborneol in a body of natural water had not previously
been reported.
Recently Piet, Zoeteman and Kraayeveld, Government
Institute for Drinking Water Supply, The Hague, Nether-
lands, detected "geosmin" and 2-methylisoborneol in
rivers of the Netherlands (11).
Having once been identified and associated with a
particular problem, a pollutant becomes much more
readily identified in future cases. Such an example
illustrates an important function of organic contami-
nants characterization.
7. Western Louisiana Industrial Waste Survey
In April, 1971, Dr. T. 0. Meiggs, EPA Division of Field
Investigations, Denver, Colorado, requested NWCCRP to
characterize chemically a number of industrial effluents
from an industrialized area in western Louisiana. The
information was for an enforcement conference. We
identified a total of 50 compounds in 7 different
industrial effluents; forty-nine of the identifications
were confirmed by comparison of GC-MS data and/or GC
retention times of standards with those of the
tentatively identified compounds. Some compounds were
common to two or more of the effluents.
Values for total organic carbon (TOG), chemical oxygen
demand (COD), specific conductivity, and concentrations
of total and suspended solids were supplied to the EPA
Division of Field Investigations from 24-hour composite
samples.
35
-------
One liter of each of the effluent samples was extrac-
ted with chloroform and concentrated. A 50-ft.
support coated open tubular (S.C.O.T.) column with
carbowax 20M-TPA provided adequate separation of all
but one of the mixtures for which a 50 ft. S.C.O.T.
column coated with SE-30 was used.
Chemical characterization of these seven industrial
effluents
• revealed the discharge of unexpected compounds
(not indicated from lists of products and raw
materials supplied by the manufacturers) along
with other data for enforcement of permit
regulations.
• furnished knowledge of treatment effectiveness
beyond that provided by gross pollution
measurements.
• gave another means of determining the signifi-
cance of an efiluent—whether the compounds
present are likely to cause oxygen depletion,
to contribute to taste and odor problems or to
cause chronic or acute toxic effects in biota.
Petrochemical Company A
Petrochemical Company A, producing olefins and oxygena-
ted hydrocarbons, discharges an estimated 4 million
gallons per day. Treatment consists of three ponds
connected in series, and covering a total of 5 acres
with a 5-6 day retention time. The first and largest
of the ponds contains aerators. The third has an over-
flow structure at the entrance of a bayou into which
the effluent is discharged; samples were gathered at
this point.
The effluent from Petrochemical Company A contained the
most complex mixture of organic compounds (Figure 17).
Because no standard was available for comparison, the
identification of 3-methylindene rests solely on mass
spectrometric data interpretation. All other compounds
were confirmed by comparison with standards. Phenol
and 2 ,6-dimethylnaphthalene were found to have the same
retention time under the gas chromatographic conditions
36
-------
50' S.C.O.T CARBOWAX 20M-TPA
70° ISO. / 2 MIN.; PROGRAM -> 200° AT 8V MINI.
i
8 10 12
MINUTES
14
16
18
20
Figure 17.
Chemically Characterized Gas Chromatogram
of Petrochemical Company "A"
37
-------
Table 3
Chemical Characterization and Gross Pollution
Measurements of Petrochemical Company A
GROSS POLLUTION MEASUREMENTS
Total Organic Carbon
Chemical Oxygen Demand
Total Solids
Suspended Solids
Specific Conductivity
180 mg/i
612 mg/i
868 mg/l
78 mg/J,
1,100 v mhos/cm
CO
CHEMICAL CHARACTERIZATION
Products
Propylene
Ethylene
Butadiene
Butane
Octane
Ethylene glycol
Ethylene oxide
Polyglycols
Ammonia
Raw Materials
Raw gas
Ethane
Refinery gases
Refinery C-^
stream
Refinery C3
stream
Propane
Butadiene
Nitrogen
Hydroformer gas
Platformer gas
Compounds Identified
m-Xylene
p_-Xylene
o-Xylene
Styrene
o-Methylstyrene
Indane
Indene
Methylindene
3 -Me thy 1 indene
Naphthalene
2-Methylnaphthalene
1-Methylnaphthalene
2 ,6-Dimethylnaphthalene
Phenol
Approximate
Concentr at ion
(mg/Jl)
0.008
0.002
0.006
0.031
0.001
0.007
0.026
0.002
0.003
0.053
0.030
0.025
0.015
0.060
Approximate
Discharge
(Ibs/day)
0.3
0.1
0.2
1.1
0.1
0.3
0.9
0.1
0.1
1.9
1.1
0.9
0.5
2.1
-------
employed, but the mass spectrometric fragmentation
patterns of each were readily discernable even though
they were present as a mixture. A mixture of two
standards confirmed the spectral interpretation.
Sulfur and several polycyclic aromatics such as
indenes and naphthalenes have been cited as the source
of odor in oil products (12) . They may therefore be
responsible for odor problems in drinking water
supplies derived from sources into which petroleum
by-products are discharged as waste.
Table 3 presents a summary of the gross pollution
parameters and a list of products and raw materials
supplied by the company, along with the identified
compounds, their approximate concentrations, and their
rates of discharge. A comparison of the list of
products and raw materials with that of the identified
compounds illustrates an important point—the lists
supplied by a company cannot be used exclusively as
a key for the identification of organic pollutants in
industrial discharge.
Petrochemical Company B
The second petrochemical plant, producing alcohols and
paraffins , discharges about 314 million gallons per
day. Wastes from a paraffin unit and an alcohol unit
enter a ditch at two points and pass through a "flow
structure" before emptying into a bayou. The effluent
from the paraffin unit passes through an American
Petroleum Institute (API) separator before entering
the ditch; the effluent from the alcohol unit, compris-
ing about 80% of the total effluent, is not treated.
The sample from the ditch was taken after the effluent
passed through the "flow structure."
Flow Structure
39
-------
Only normal short chain alcohols with even carbon
numbers were found to any appreciable extent.
n-Hexanol was present in the largest concentration
along with lesser amounts of n-butanol and n-octanol
and a trace of n-decanol (Figure 18). The absence of
any traces of paraffinic hydrocarbons in the effluent
is conspicuous. Either the API separator did an
efficient job of removing them or their large molecu-
lar weight prevented their elution from the chromato-
graphic column. Dilution by the effluent from the
alcohol unit would also help mask their presence if
they were being discharged in small amounts.
The gross pollution measurements and chemical charac-
terization data are summarized in Table 4.
Petro re finery Company A
The first of the two refineries studied discharges an
estimated 3 million gallons per day and produces about
230,000 barrels per day. Treatment consists of a
simple 8-hour retention-time lagoon that discharges
through an overflow structure into a tidal pond of a
river. To avoid tidal influences, sampling was done at
the lagoon overflow structure.
Overflow
/Structure
£
Tidal
Pond
Refinery
River
The chloroform extractables of the effluent consisted
primarily of phenol, p-cresol, and normal long chain
aliphatic hydrocarbons (Figure 19). Undecane (Cii )
through nonadecane (C^g) were identified and confirmed.
Small amounts of unspecified branched isomers appear
between the peaks of these normal hydrocarbons.
Although both 1-methyl naphthalene and 2-methylnaphtha-
lene were present, no naphthalene was detected in
measurable concentration. The gross pollution measure-
ments and chemical characterizations are presented in
Table 5.
40
-------
50' S.C.O.T. CARBOWAX 20M-TPA
50° ISO./2 MIN.; PROGRAM ->200° AT 8VMIN.
o
B
a
o
e
a
i
10
12 14 16
MINUTES
18
20
22
24
26
28
Figure 18. Chemically Characterized Gas Chromatogram of Petrochemical Company
"B"
-------
Table 4
Gross Pollution Measurements and Chemical Characterization
of Petrochemical Company B
GROSS POLLUTION MEASUREMENTS
Total Organic Carbon 130 mg/i
Chemical Oxygen Demand Not available
Total Solids 2650 mg/X.
Suspended Solids 34 mg/8.
Specific Conductivity 4,000 u mhos/cm
CHEMICAL CHARACTERIZATION
Approximate Approximate
Products Raw Materials Compounds Identified Concentration Discharge
(mg/i) (Ibs/day)
Normal paraffin Ethylene 1-Butanol 16.0 90
Industrial alcohols Aluminum 1-Hexanol 65.0 375
Ethylene Hydrogen 1-Octanol 19.0 110
Methyl chloride Raffinate 1-Decanol 2.5 15
Ethoxylates Sulfuric acid
Ethylene oxide
Acetic acid
Caustic
Phosphoric acid
Kerosene
Ethane
Propane
Methanol
HC1
-------
50' S.C.QT CARBOWAX 20M-TPA
60° ISO. / 2 MIN.t PROGRAM -* 200° AT 8VMIN.
oo
i
i
10 12 14
MINUTES
16
18
20
22
24
Figure 19.
Chemically Characterized Gas Chromatogram
of Petrorefinery "A"
43
-------
Table 5
Gross Pollution Measurements and Chemical Characterization
of Petrorefinery Company "A"
GROSS POLLUTION MEASUREMENTS
Total Organic Carbon
Chemical Oxygen Demand
Total Solids
Suspended Solids
Specific Conductivity
15 mg/i
Not available
9,220 mg/i
38 mg/i
13,700 vi mhos/cm
CHEMICAL CHARACTERIZATION
Products
LPG propane
Propylene
Orthoxylene
Aromatics
Raw Materials Compounds Identified
Crude oil Undecane
Excess gas oil Dodecane
Extracts Tridecane
Tetradecane
Pentadecane
Hexadecane
Heptadecane
Octadecane
Nonadecane
2 -Me thy In aph tha lene
1-Methylnaphthalene
Phenol
o-Cresol
Approximate
Concentration
(mg/Jl)
0.027
0.031
0.042
0.039
0.030
0.026
0.022
0.017
0.013
0.013
0.005
0.200
0.120
Approximate
Discharge
(Ibs/day)
69
79
107
99
76
66
53
43
33
33
12
510
300
-------
Petrorefinery Company B
The treatment facilities of refinery B, which discharges
about 1 million gallons per day and produces about
70,000 barrels per day, are more extensive than those of
Company A. The effluent from the refinery flows through
an API separator and a corrugated plate interceptor
(CPI) into an activated sludge unit. It then passes
through a clarifier into a 24-hour retention-time
aerated lagoon from which it discharges into a bayou.
The chloroform extractables of the effluent produced a
chromatogram resembling that of a crude oil and mass
spectra characteristic of only aliphatic hydrocarbons.
The normal chain hydrocarbons were present in larger
amounts than the many branched isomers and thus stood
out as spikes superimposed upon a base of complex,
poorly resolved lumps in the gas chromatograph (Figure
20) . Undecane (C-, ^) through heineicosane (C^i) were
confirmed. No phenols or naphthalenes were identified
in this sample. The gross pollution measurements and
chemical characterizations are summarized in Table 6.
Synthetic Rubber Company
A synthetic rubber company discharges about 6 million
gallons per day with only primary treatment followed
by discharge via a drainage ditch to a bayou.
Styrene (one of the 3 raw materials listed), furfural,
and 1-methylnaphthalene were identified and confirmed
in the effluent. The compound producing the largest
peak in the gas chromatogram (Figure 21) and four
isomers of apparent molecular weight 156 could not be
identified from their mass spectra alone; lack of time
and manpower precluded further investigation. Table 7
summarizes the results of the analysis.
45
-------
50' S.C.QT CARBOWAX 20M-TPA
60° ISO./2 MIN.; PROGRAM -*200° AT 8VMIN.
10 12
14
16 18 20
MINUTES
22 24 26 28 30 32 34 36
Figure 20. Chemically Characterized Gas Chromatogram of Petrorefinery Company "B"
-------
Table 6
Gross Pollution Measurements and Chemical Characterizations
of Petrorefinery Company "B"
GROSS POLLUTION MEASUREMENTS
Total Organic Carbon
Chemical Oxygen Demand
Total Solids
Suspended Solids
Specific Conductivity
210 rag/Jl
676 mg/i
2430 mg/«.
182 mg/X,
3900 y mhos/cm
CHEMICAL CHARACTERIZATIONS
Products
LPG
Propane
Butane
Gasoline
Kerosene
Diesel fuel
Heating oil
#6 Fuel oil
Coke
Raw Materials Compounds Identified
Crude oil Undecane
Isobutane Dodecane
PVC Tridecane
Tetradecane
Pentadecane
Hexadecane
Heptadecane
Octadecane
Nonadecane
Eicosane
Heine icosane
Approximate
Concentration
(mg/fc)
0.05
0.22
0.39
0.58
0.49
0.42
0.34
0.33
0.31
0.30
0.19
Approximate
Discharge
(Ibs/day)
0.4
2.2
3.8
5.6
4.8
4.0
3.3
3.2
3.0
2.9
1.8
-------
50' S.C.O.T CARBOWAX 20M-TPA
70° ISO./2 MIN.; PROGRAM -»200°AT 6.5VMIN.
8 10 12
MINUTES
14
16
18
20
Figure 21.
Chemically Characterized Gas Chromatogram
of Synthetic Rubber Company
48
-------
Table 7
Gross Pollution Measurements and Chemical Characterization
of Synthetic Rubber Company
GROSS POLLUTION MEASUREMENTS
Total Organic Carbon 52 mg/l
Chemical Oxygen Demand 168 mg/i
Total Solids 3210 mg/t,
Suspended Solids 76 mg/i
Specific Conductivity 5000 \i mhos/cm
vo
CHEMICAL CHARACTERIZATION
Products
Synthetic rubber
Raw Materials
Butadiene
Styrene
Carbon black
Compounds Identified
Styrene
Furfural
1-Methylnaphthalene
Approximate
Concentration
(mg/i)
0.0026
0.0017
0.0017
Approximate
Discharge
(Ibs/day)
1.3
0.9
0.9
-------
Chemical Company A
Chemical Company A, a producer of chlorinated hydro-
carbons discharges about 4 million gallons of
effluent per day from their organics plant into a
ditch. Its treatment, after mixing with the effluent
from a caustic and chlorine unit, consists only of
dilution with cooling water before discharge into a
river.
1,1,2-Trichloroethane and 1,1,2,2-tetrachloroethane
were identified as the two main contaminants (Figure
22). The total discharge at the above mentioned rate
amounted to over 300 pounds of these chlorinated
hydrocarbons per day. Table 8 summarizes the findings
Chemical Company B
The second chemical company, a polymer producer,
discharges an estimated 2 million gallons of effluent
per day. The treatment consists of a series of six
lagoons, with a 4-5 day total retention, that discharge
into a river. The only two compounds of significant
amount, as determined by gas chromatography, were
identified as decane and undecane (Figure 23) . No
gross pollution measurements were available. Table 9
summarizes the findings.
8. Color Body Characterization and Related
Studies from a Kraft Paper Mill Effluent
The Interstate Paper Company at Riceboro, Georgia,
utilizes a recently developed lime treatment process
to remove "color bodies" from the effluent of Kraft
pulp mills. The lime treatment removes 80-90% of the
color from the effluent; however, during passage through
the stabilization lagoon, the water becomes colored
again. Because of the unusually long retention time of
the lagoon (3-6 months) the reappearance of the color
may result from one or both of two processes: 1) decom-
position of previously uncolored compounds that were
not removed by lime treatment and 2) leaching of natu-
rally occurring "color bodies" present in the soil at
the bottom of the lagoon. The waste treatment system
is presented below.
50
-------
50' S.C.O.T CARBOWAX 20M-TPA
70° ISO./2MIR; PROGRAM-^200° AT 8VMIN.
I I I
I I
0 2 4 6 8 10 12
MINUTES
Figure 22.
Chemically Characterized Gas Chromatogram
of Chemical Company "A"
51
-------
100' S.C.QI SE-30
60° ISO./2MIN.; PROGRAM -*200° AT I2VMIN.
c
•
"o
UJ
I
6 8
MINUTES
10
12
14
Figure 23.
Chemically Characterized Gas Chromatogram
of Chemical Company "B"
52
-------
Table 8
Chemical Characterization of Chemical Company A
Products
Chlorine
Caustic soda
Raw Materials
Sodium chloride
Ethylene
Compounds Identified
1,1, 2-Trichloroe thane
1,1,2, 2-Tetrachloroethane
Approximate
Concentration
(mg/*)
5.4
2.2
Approximate
Discharge
-------
Key: • • Lime Treatment Effluent
o o Lagoon Effluent
A A Marsh Water
0.090
0.080
0.070
O 0.060
c 0.050
0.040
0.030
9
PH
10
11
Figure 24. Comparison of Kraft Pulp Mill Effluent and
Marsh Water Color Intensity as a Function of
pH
54
-------
Lime
Flocculat jon
Stabilization
Pond
R
I
V
E
R
I
Some physical characteristics of the lagoon effluent
were compared with those of nearby marsh water and of
the effluent following lime treatment. Plots of color
intensity at 430 nm as a function of pH for the three
samples are shown in Figure 24. The curve for the
lagoon effluent parallels the curve for marsh water
but differs greatly from that of the effluent following
lime treatment. Fluorescence spectra were recorded for
each sample, both without any pH adjustment and after
acidification to pH 2 with HCl. The wavelengths of the
fluorescence maxima and their intensities relative to
that of the lagoon effluent are given in Table 10.
Table 10
Comparison of Kraft Pulp Mill Effluent and
Marsh Water Fluorescent Spectra
Sample
Fluorescence
Maxima
Intensity
Ratio
Lime treatment effluent
Lagoon effluent
Marsh water
After Acidification
Lime treatment effluent
Lagoon effluent
Marsh water
427 nm
450 nm
450 nm
390 nm
450 nm
450 nm
3.8
1.0
1.8
1.4
1.0
1.7
55
-------
The absorption and fluorescence data indicate that the
color characteristics of the lagoon sample closely
resemble those of the marsh water.
Two further experiments were performed to determine
the source of the color bodies. One liter of lime-
treated effluent was sparged with air for 120 days to
test for the presence of compounds capable of decom-
posing to colored products. During that time, however,
the color decreased probably because of a change in pH
caused by CO2 absorption.
In addition, mud from the lagoon bottom was dried,
powdered, and then leached with calcium hydroxide
solutions of two different concentrations. The 5-gram
mud samples were shaken with 100 ml of solution for 10
days. Absorption measurements at 420 nm were made
after 5 and 10 days. The data summarized in Table 11
indicate that the leaching of color from the soil is
pH dependent.
Table 11
Color Leached from Lagoon Mud as a Function
of Calcium Hydroxide Concentration
Solution
H20
10" *M Ca(OH)2
10 3M Ca(OH)2
Absorbance at 420 nm
5 days
0.117
0.124
0.181
10 days
0.142
0.136
0.182
As a result of the study, the color increase in the
lagoon effluent was attributed to the leaching action
of the alkaline waste from the lime treatment.
An effort was made to identify the compounds responsi-
ble for the color in the effluent. Freeze concentra-
tion followed by gel permeation chromatography of the
56
-------
lagoon effluent yielded molecular weight values for
the "color bodies" of 2,500 to 5,000. Further work on
the molecular weight characteristics of these samples
is being conducted in detail by the Institute for
Paper Chemistry under FWQA Contract 1240-DKO.
The amount of acid-insoluble organic matter is relevant
to the color body determination since lignin and other
color-producing large molecules can be precipitated
from pulp mill effluents by acidification. The
difference between the total organic carbon (TOC)
contents of a sample before and after acidification
gives an indication of the concentration of acid-
insoluble organics. Such measurements made on the
pulp mill effluent both before and after lime treat-
ment (refer to Table 12} show that most of the acid-
insoluble components are removed by the treatment.
Table 12
Comparison of the Amounts of Acid Insoluble
Organics in the Lime Treated and the
Untreated Effluent Samples
TOC Before TOC After Acidification
Sample Acidification Acidification Insolubles
CpH 1)
Lime
(Treated) 117 mg/1 111 mg/1 5%
Untreated 216 mg/1 112 mg/1 48%
Analyses of effluent samples from before and after lime
treatment and after lagooning were made to determine
the effect of the treatment on the organic wastes
discharged. (The technique is described in references
13 and 14.) Each sample was made basic to pH 11.5 and
extracted with chloroform to remove the neutral
volatiles. The extracts were concentrated and analyzed
by gas chromatography-mass spectrometry. A list of the
compounds identified is presented in Table 13.
57
-------
Table 13
Compounds Tentatively Identified in the
Neutral Volatiles Fraction of the
Riceboro Pulp Mill Effluent
2^-cymene
limonene
guaiacol
a menthene isomer
l-methoxy-4-(1-propenyl)-benzene
l-methoxy-4-pentylbenzene
The aqueous residue was methylated with dimethyl sulfate
and re-extracted with chloroform. GC-MS analysis of the
extract after concentration yielded the compounds listed
in Table 14.
Table 14
Compounds Tentatively Identified in the
Phenols and Acids Fraction of the
Riceboro Pulp Mill Effluent
dimethylsulfone
2,5-diethylthiophene
4-me thoxybenz aIdehyde
3,4-dimethoxyethylbenzene
^-methoxypropiophenone
3,4-dimethoxybenzaldehyde
3,4-dimethoxyacetophenone
3,4,5-trimethoxyacetophenone
methyl 3,4-dimethoxybenzoate
3,4-dimethoxypropiophenone
methyl pimarate
methyl isopimarate
methyl abietate
methyl dehydroabietate
58
-------
The results of this analysis indicated that
• most of the neutral volatiles were not
affected by lime treatment.
• most of the resin acids are significantly
diminished in concentration after lime
treatment.
9. Characterization of Foam from Kraft Pulp Mills
Source of an Environmental Contamination
Brown foam, floating to shore in masses, has presented
a persistent problem in Perdido Bay in Florida's
western panhandle. Tall oil in the discharge from a
kraft paper mill at the north of the bay was suspected
to be the cause of the foam.
Three samples were collected and analyzed. The first,
a water sample, was taken from below the second of 2
sedimentation ponds in a stream composed almost
totally of the|plant's effluent. A foam and scum
sample was collected in Perdido Bay about 17 miles
below the mill at a point where the foam was in
abundance. The last sample was of foam only and was
taken from a point about 24 miles below the papermill.
To determine the presence of resin acids and fatty
acids, the samples were acidified, extracted with
chloroform, esterified with diazomethane, and analyzed
by flame ionization gas chromatography. Structures
were confirmed by GC-MS. The compounds listed in
Table 15 were identified (the acids as the methyl esters)
for the three samples.
The identification of the same and similar compounds in
all three samples indicates that the paper mill is the
source of the foam contamination in the bay. The mill
has begun installing secondary treatment. In-plant
changes have reduced the organic content of the
effluent to one-third of its former load.
59
-------
Table 15
Compounds Identified in Perdido Bay Kraft Papermill
Wastewater, Foam, and Scum Samples
Mill
Wastewater
Foam & Scum (17 miles
below mill)
Foam (24 miles
below mill)
Pimaric acid
Isopimaric acid
Dehydroabietic acid
6,8,11,13-Abietatetraen-
18-oate
Unidentified resin acid
Palmitic acid
Stearic acid
Lignoceric acid
Pimaric acid
Isopimaric acid
Dehydroabietic acid
6,8,11,13-Abietatetraen-
18-oate
Unidentified resin acid
Unidentified resin acid
Stearic acid
Eicosanoic acid
Lignoceric acid
Pimaric acid
Isopimaric acid
Dehydroabietic acid
6,8,11,13-Abietatetrean-
18-oate
Unidentified resin acid
Unidentified resin acid
Palmitic acid
Stearic acid
Eicosanoic acid
-------
Constituents from a Bleach Stage Caustic Clarifier
The International Paper Company at Springhill,
Louisiana, was involved in developing a "massive lime
process" to remove color from the effluent of the
"bleach plant caustic stage." After the lime was
added to the effluent and the mixture was passed to a
"caustic clarifier", however, excessive foaming
developed that caused flow stoppage and solids carry-
over. Identification of the foam constituents was
desired in order to take action to reduce the foaming.
The analytical procedure was similar to previous
characterization of paper mill effluents (13,14). The
foam was mixed with distilled water in a separatory
funnel and made strongly alkaline with sodium hydroxide.
The chloroform extracts of the basic solution contained
the neutral volatile constituents, which were only
minor components of the mixture and were not examined
further.
The acidic fraction in the aqueous phase, comprising
the majority of the organic componnents, was methylated
by two procedures to produce derivatives that could be
separated by gas chromatography. Methylation of the
chloroform extract of the acidified aqueous phase by
diazomethane proved superior to direct methylation
with dimethyl sulfate; the amounts of methyl esters
were greater using the former method.
A GC-MS analysis showed no resin acid or phenolic
compounds in detectable amounts, in contrast to previous
studies involving total effluents. Instead, the foam
consisted primarily of long-chain carboxylic ("fatty")
acids. Palmitic and stearic acids were in predominance
(Figure 25). Also present in significant amounts were
identified isomers of the di- and triunsaturated
analogs of octadecanoic acid. The identifications of
the numbered peaks in Figure 25 are listed below:
Peak Number Methyl Ester of
1 heptanoic acid (heptylic acid)
2 octanoic acid (caprylic acid)
3 nonanoic acid (pelargonic acid)
4 decanoic acid (capric acid)
5 undecanoic acid
6 dodecanoic acid (lauric acid)
61
-------
DIAZOMETHANE METHYLATION
en
to
<4 42 40 3a 36 34 32 30 28 26 24 22 20 |g u 14 12 10
FOAM
if
Ji,
6420
till
Figure 25. Chemically Characterized Gas Chromatogram of Foam from a Kraft
Papermill Bleach Stage Caustic Clarifier
-------
Peak Number Methyl Ester of
7 tetradecanoic acid (myristic acid)
8 hexadecanoic acid (palmitic acid)
9 octadecanoic acid (stearic acid)
10 diunsaturated analog of octadecanoic
acid
11 unidentified ester; molecular weight
262
12 unidentified ester; molecular weight
280
A mechanical means of foam control was tried; however,
the plant was closed down for lack of funds and no
chemical solution to the problem was attempted.
10. Analysis of Oil Spills l>y NMR Speatrosoopy
Nuclear magnetic resonance (NMR) spectroscopy is a
useful method for helping to determine the source of
an oil spill. NMR can rapidly distinguish between
different types of oils and sometimes can identify
the type of oil in a spill without reference to a
standard. Its greatest disadvantage is that it cannot
distinguish between different oils of the same general
type.
Shell Point Beach Oil Spill
An oil spill occurred in February, 1969, off the coast
of St. Mark, Florida, on Shell Point Beach. The U. S.
Coast Guard collected samples from a nearby punctured
barge and from Shell Point Beach 3, 5, and 7 days
after the spill was noticed. An analysis was performed
in conjunction with the Chemical Services Branch to
determine whether the barge was the source of the
spill.
A small amount of each sample was dissolved in carbon
tetrachloride containing 2-3% tetramethylsilane as a
reference standard. The NMR spectra were scanned and
electronic integrals of the areas under each signal
in the spectra were made in triplicate (Figure 26).
Since each sample was run at least three times, the
mean signal areas are averages of at least nine
integrals. The average areas under the signals were
normalized (Table 16).
63
-------
St. Mar*, Fia. OiI Spi
S-nple No. 5ST4
Crude 0>I from Barqe
r o
T o
10
Figure 26.
Comparison of NMR Spectra of Oil from the
Suspect Barge and from a Florida Beach
64
-------
Table 16
Normalized Peak Areas of Proton Signals in Samples from the
St. Mark, Florida, Oil Spill
Sample T9.05-9.60 T 8.00-9.05 T6.80-3.00 f 2.00-4.00
Crude oil from barge 0.27 0.60 0.08 0.04
(Figure 26, top)
Sample from Shell 0.23 0.62 0.09 0.06
Point Beach—
^ Weathered 3 days
Ul
Sample from Shell 0.23 0.62 0.11 0.05
Point Beach Weeds—
Weathered 5 days
(Figure 26, bottom)
Sample from Shell 0.21 0.65 0.11 0.04
Point Beach Weeds—
Weathered 7 days
-------
•' I • I
Harbor
',', Source A
1 • i • 1 • i - 1
Figure 27.
Comparison of NMR Spectra of Oil from
Duluth Harbor and the Primary Suspect Source
66
-------
Harbor
Source B
Figure 28. Comparison of NMR Spectra of Oil from Duluth
Harbor and the Secondary Suspect Source
67
-------
The similarity between the chemical shifts and the
relative peak areas of each signal indicates that the
samples are all of the same oil type and that the barge
could have been the source of the spill. A comparison
of infrared spectra and the sulfur and vanadium content
of pollution samples with those of oil from the suspect
barge also indicated that they were the same.
Small but distinct differences appeared between the
barge sample and the spill samples and also among the
spill samples themselves. The weathered samples showed
a decrease in the relative CH3 signal area and an
increase in the CH2 signal area as weathering time
increased. A mixture of hydrocarbons containing both
volatiles and nonvolatiles would be expected to show
such relative change in signal areas. The volatile
materials, having a larger ratio of 013 to CH2 groups,
evaporate faster, leaving the long chain hydrocarbons
and other larger molecules behind. The result is a
decrease in the 013 to CH2 ratio.
Duluth Harbor Oil Pollution
In 1969, the harbormaster in Duluth, Minnesota,
submitted samples of oil from the persistently polluted
harbor to the National Water Quality Laboratory in
Duluth. These samples and samples of a suspected
source (a railway company) were sent to NWCCRP for NMR
analyses. Also sent later was a sample from a second
possible source, an oil company.
The NMR spectrum of the harbor sample was very different
from that of the railway company sample (Figure 27) .
It closely resembled, however, the spectrum of the oil
company sample (Figure 28) implicating the oil company
as the probable source of pollution. Prosecution of
the oil company resulted in conviction and a fine.
Tampa Bay Oil Spill
The A.'iooa Trader was suspected to be the source of an
oil slick on Tampa Bay, Florida, in 1969. However, NMR
spectra of samples from the slick and from the ship's
oil cargo displayed significant differences. The
proportions of the various types of protons in the
68
-------
samples as calculated from the integrated intensities
are presented below:
Methyl Methylehe Misc. Aromatic
Tampa Bay Sample 0.20 0.55 0.16 0.10
Alooa Trader Sample 0.28 0.61 0.07 0.04
The differences in the spectra are not believed to be
due merely to weathering effects on the slick samples
since the changes in some proton proportions do not
correspond to those predicted from weathering, in
studies of the weathering on other oil samples (crude
oil and Bunker C) three changes were noted:
• weathering lowers the proportion of methyl
proton,
• it increases the methylene proton proportion,
and
• it has a small or no effect on the aromatic
proton percentage.
On the bases of these data, therefore, the oil from the
slick was judged to be from a source other than the
Alcoa Trader or was highly contaminated with oil from
another source.
Florida West Coast Oil Contamination
In the spring of 1970, oil pollution was reported along
an 80-100 mile stretch of Florida's northwest coast.
Two possible sources were the Greek tanker, Delian
Apollon, wrecked off Tampa in February, 1970, carrying
Bunker C oil, and a large Chevron crude oil fire from
a well off the coast of Louisiana. Because the two
sources represented two different types of oil, the
analysis was performed by NMR.
Spectra were scanned of samples from the two suspected
sources and from the polluted areas (Pensacola Beach,
Panama City .Beach, St. Andrews Park, Indian Pass,
Destin Beach, and submerged oil out from Destin Beach).
Data are summarized in Table 17 along with data from
previous weathering studies on Bunker C and crude oil.
69
-------
Table 17
NMR Comparison of Oil from West Florida Beaches with Suspect
Crude Oil and Bunker C Oil
Sample
Environmental samples
Delian Appollon
Chevron oil well and
oil slick
Weathered Bunker C
Weathered crude
Methyl
0.21-0.22
0.22
0.26-0.27
0.17-0.21
0.21-0.24
Normalized Proton
Methylene
0.68-0.71
0.61
0.65-0.66
0.56-0.59
0.69-0.71
Signal Areas
Misc.
0.05-0.07
0.10
0.05
0.10-0.15
0.04-0.06
Aromatic
0.02-0.04
0.07
0.02-0.03
0.10-0.13
0.02-0.03
-------
The spectra of the spill samples were all similar and
matched those of weathered crude oil but not those of
Bunker C or weathered Bunker C oil.
The Chevron oil well crude oil was therefore indicated
as the more probable source of the pollution.
Miscellaneous Oil Analyses
In the fall of 1970 an oil slick was discovered by the
U. S. Coast Guard around the Dutch Cargo carrier Maya.
Samples of the oil slick, of water being discharged from
the Maya and of the oil in the bilge of the Maya were
taken for analysis by EPA Region IV Surveillance and
Analysis Division at SEEL. An NMR analysis was also
requested since prosecution was expected and confirma-
tory data were desired to show that the ship was the
source of the slick.
The NMR spectra of the samples, were all similar,
indicating a partially refined crude oil. Infrared
analysis likewise confirmed that the samples were all
very similar and probably from the same source. The
data were entered as evidence in court by the State of
Florida. The case, however, has been postponed by a
Federal injunction charging that Florida's coastal
pollution laws are unconstitutional.
In the winter of 1970 Mr. Warren T. McFall, EPA Alaska
operations Office, requested an analysis of oil on the
feathers of a bird that had been killed by oil pollu-
tion and crude oil from a suspect source near Cook
Inlet. The NMR spectra of the two samples were quite
different—more than would be expected from weathering
and contamination from the bird's natural oils.
Fluorescence data also indicated that the two were not
from the same source. We concluded that the Cook
Inlet crude oil was not the source of the oil that
killed the bird. No further work was done concerning
this problem.
Also in the winter of 1970, the Chemical Services Branch
received an oil sample taken by the National Marine
Fisheries Service during exploratory fishing at 330
fathoms northwest of the Dutch West Indies island of
Aruba. From NMR analysis we concluded that the oil
was unlike any crude, Bunker C or refined oils that we
examined previously. The source remains unknown.
71
-------
11. Analysis of Oil Spills
by Fluorescence Spectroscopy
Fluorescence spectroscopy represents yet another method
of analysis that may be used to characterize an oil
spill and determine its source. Whereas NMR can differ-
entiate only between oil types, different oils of the
same type have sufficiently distinctive fluorescence
characteristics to distinguish between them (15).
Several samples are cited comparing fluorescence data
from oil spills with data from other techniques.
Duluth Harbor Oil Pollution
In a Duluth Harbor oil spill mentioned previously the
NMR spectra indicated that the oil of the slick
matched that of the suspect oil company and did not
match any of three oil samples from the suspect railway
company. Fluorescence analysis showed that the railway
samples were radically different from the harbor oil,
corroborating NMR conclusions.
NMR spectra of both oil slick samples and the samples
from the oil company, however, were identical. Also,
the relative fluorescent intensities of the oil company
samples and the intensity ratios closely matched those
of the first of the two harbor samples (Table 18).
Harbor Sample #2 did not match the oil company samples
as well. But a mixture of 1 part oil from the railway
company and 14 parts oil from the oil company produced
a fluorescence spectrum similar to that of the second
harbor sample. Thus, although the oil company was the
main source of the pollution there was evidence of a
measurable contribution from the railway company.
Savannah River Spills
In April of 1971, two different oil spills occurred on
the Savannah River. Fluorescence analyses were
performed on the spill samples and on samples from
suspect sources. The data obtained were compared with
that from gas chromatographic analysis using flame
ionization and photometric detectors.
72
-------
Table 18
Fluorescence Spectra of Duluth Harbor Samples
Samples
Harbor
Harbor
Oil Co
Oil Co
Oil Co
Sample #1
Sample #2
. Sample #1
. Sample #2
. Sample #3
Fluorescence Spectra (Excitation at 340 mji)
Intensity ratio at 400 mja of
harbor sample/oil co.
Dilution
1:1000
0.97
1.98
1.14
1.00
1.00
Dilution
1:10,000
1.00
3.67
0.97
1.00
0.90
Intensity ratio, shoulder
at 440 mu/max at 400 mju
Dilution
1:1000
0.39
0.58
0.40
0.41
0.39
Dilution
1:10,000
0.30
0.48
0.31
0.30
0.37
-------
Table 19
Fluorescence Spectra of Savannah River Samples
Spill
Spill A
Spill B
Sample
Fuel line
Drip pan
Bilge
River sample 11*
River sample #2*
Suspect
River sample
10,000 Dilution
Maximum
intensity
100
17
72
92
95
100
46
440 nm
368 nm
.40
.15
.35
.40
.40
.80
.55
50,000 Dilution
Maximum
intensity
100
21
75
93
94
100
42
440 nm
385 nm
.40
.22
.32
.40
.39
.66
.49
*The 440nm/368nm ratios of the spill and suspect match very well. The
intensity readings are slightly low for the two spill samples but this is not
uncommon for oils that have been weathered.
-------
Spill "A"
Two samples from spill "A" were compared by fluorescence
analysis with three oils taken from the fuel line, a
drip pan, and the bilge of a nearby suspect ship.
Fluorescence data are presented in Table 19. The
relative intensity measurements from the spill samples
matched those of the fuel line from the ship, confirming
results of GC analysis.
Spill "B"
In the second spill, which appeared to be a Bunker fuel,
the suspected source was the fuel tank of a nearby
vessel. The fluorescence data (Table 19), however,
indicated that the two oils were not similar, again
confirming GC data.
Duluth Superior Harbor Spill
In an oil spill on Duluth Superior Harbor in 1970 a
British and an American ship were suspects. Spill
samples were collected from the bay and from the slip
in which both ships had been docked at different times.
Comparisons were made with oils obtained from the ships.
Each sample was diluted with oils obtained from the
ships. Each sample was diluted wt/vol 1:7500, 1:10,000,
and 1:100,000 with cyclohexane. The relative fluores-
cence intensity data for the spill and suspect samples
are presented in Table 20.
All parameters of the sample from the British freighter
match closely those of the spill sample, implicating
that ship as the source of the pollution. These results
were turned over the Coast Guard for legal action.
Dundee Canal, Savannah River
An oil spill covering a two to three mile length of the
Dundee Canal occurred in February, 1971. A tank trailer
carrying Bunker fuel was the suspect source. Five spill
samples and a sample from the trailer were collected.
The broad range of intensity values (Table 21) at
1:50,000 dilution indicates weathering of the spill
75
-------
Table 20
Fluorescence Spectra of the Duluth Superior Harbor Samples
Sample
British freighter
American freighter
Spill sample
Intensity @
Dilution
1:7,500
94
30
93
Dilution
1:10,000
87
20
81
Intensity ratio
386nm 440nm/386nm
Dilution
1:100,000
65
6 >-:.
52
Dilution Dilution
1:7,500 1:10,000
0.67 0.64
0.24 0.25
0.63 0.60
Dilution
1:100,000
0.57
0.28
0.55
-------
Table 21
Fluorescence Spectra of the Dundee Canal Samples
Samples
Suspect tank trailer #82
#1 — Puddle on ground
behind trailer
#2 — Ditch along Hwy 80
under highway
#3 — Ditch along Hwy 80
#4 — Dundee Canal
#5 — Dundee Canal at
Savannah River
1:10,000 Dilution
Ratio
Max. 440 nm
Intensity 386 nm
100 .66
91 .60
94 .64
97 .65
100 .66
77 .59
1:50,000 Dilution
Ratio
Max. 440 nm
Intensity 386 nm
100 .56
69 .53
73 .56
84 .56
94 .57
54 .50
Ni
(ppm)
44
28
32
33
43
25
V
(ppm)
365
198
285
308
348
210
Ni/V
.12
.14
.11
.11
.12
.12
-------
samples. A decrease in fluorescence intensity is common
in weathered oil.
Fluorescence data confirmed the metal ratio (Ni/V) analy-
ses and GC data indicating that the spilled oil was non-
homogeneous. Spill sample #4 (Dundee Canal), however,
matched the suspect oil in all parameters.
St. Johns River, Jacksonville/ Florida
An oil spill occurring in the St. Johns River covered a
one square mile area. Four samples (numbered 1-4) were
taken from the slick and six suspect samples (numbered
5-10) were collected from oil depots, ships, and indus-
tries surrounding the spill area. All samples were
cleaned, dried, and analyzed by the Region IV Chemical
Services Branch for four parameters: 1) percent
sulphur, 2) nickel-vanadium ratios, 3) infrared ratios,
and 4) color comparison under UV light. The NWCCRP was
requested to analyze the samples by the fluorescence
technique.
Fluorescence spectra were taken at various dilutions of
the oils with cyclohexane (1:10,000, 1:50,000 and
1:100,000). The ratio of the maximum intensity (386 nm)
to that of a shoulder (440 nm) was calculated for each
spectrum (15). Table 22 summarizes the data for
fluorescence analyses and Table 23 summarizes the
results of the other analyses. Both sets of data
indicate that the four spill samples came from the same
source.
Of the six suspect samples, only one matched the spill
sample in all fluorescence parameters. The same
suspect sample matched the spill with all other
analytical techniques employed, confirming the
identification.
78
-------
10
Table 22
Fluorescence Spectra of St. John's River Samples
Sample
no.
Maximum
Dilution
1:10,000
intensity
Dilution
1:50,000
at 386
nm
Dilution
1:100,000
shoulder
Dilution
1:10,000
Intensity ratio
at 440 nm/386 nm
Dilution
1:50,000
max.
Dilution
1:100,000
Parameters
that agree
Spill samples
1
2
3
4
Range of
spill
samples
70
65
66
72
65-72
77
81
72
72
72-81
0.67
0.70
0.71
0.67
0.60
0.61
0.63
0.61
0.54
0.55
0.55
0.53
72
90
76
75
72-90 0.67-0.71 0.60-0.63 0.53-0.55
Suspect samples
5
6
7
8
9
10
67
50
73
78
50a
100
81
60
68
76
73
100
90
64
65
71
99
100
0.71
0.62
0.57
0.58
0.82a
0.56
0.63
0.55
0.53
0.52
0.72
0.51
0.55
0.47
0.45
0.45
0.59
0.44
6 of 6
0 of 6
0 of 6
1 of 6
1 of 6
0 of 6
aMaximum is at 405 nm instead of 386 nm.
-------
00
o
Table 23
Other Analyses of St. John's Diver Samples
Infrared ratios
Sample
no.
1
2
3
4
Range of
spill
s
6
7
8
9
10
%
Sulfur
a
2.59
2. 78
2.62
2.59-2.78
2.65
2.30
2.56
2.02
2.08
1.68
Ni/V
a
0.31
0.38
0.29
0.29-0.38
0.31
0.19
0.27
0.21
0.61
0.24
720 cm~l
1375 cm"1
0.28
0.29
0.34
0.31
0.28-0.34
0.29
0.19
0.33
0.26
0.35
0.16
3050 cm"1
2925 cm"1
0.023
0.020
0.034
0.026
0.020-0.034
0.023
0.023
0.023
0.017
0.041
0.034
810 on"1
1375 CUT1
0.24
0.27
0.36
0.26
0.24-0.36
0.27
0.25
0.20
0.21
0.47
0.26
810 cm"1
720 am'1
0.88
0.94
1.09
0.86
0.86-1.09
0.95
1.34
0.60
0.82
1.36
1.66
1600 cm"1
1375 cm"1
0.13
0.15
0.18
0.16
0.13-0.18
0.16
0.15
0.09
0.11
0.27
0.14
Parameters
uv Colors
yellow
yellow
yellow
yellow
yellow
same yellow
darker (brown)
lighter (yellow)
lighter (yellow)
darker (brown)
lighter (yellow)
that agree
8 of 8
3 of 8
2 of 8
0 of 8
0 of 8
•Sample no. 1 did not contain enough oil for sulfur, nickel, or vanadium analysis.
-------
SECTION V
REFERENCES
1. Lichtenberg, J. J. , personal communication, July 16,
1968.
2. Nakanishi, K. , "Infrared Absorption Spectroscopy,"
Holden-Dayf Inc./ San Francisco, 1964, p. 182.
3. Beynon, J. H. , "Mass Spectrometry and Its Applica-
tions to Organic Chemistry," Elsevier Publishing
Co., New York, 1960, pp. 297-301 and 554-563.
4. Gustafson, C. G., "PCB's—Prevalent and Persistent,"
Environ. Sci. Techno 1. , 4_, 814 (1970) .
5. "Effects of Pollution on Water Quality, Escambia
River and Bay, Florida," U. S. Department of
Interior, F.W.P.C.A., Southeast Water Laboratory,
pp. 31-32 (January, 1970).
6. Adams, B. A., "The Cladothrix Dichotoms and Allied
Organisms as a Cause of 'Intermediate1 Taste in
Chlorinated Water," Water and Water Eng., 31, 327
(1929).
7. Gerber, N. N. , and Lechevalier, H. A., "Geosmin, an
Earthy-Smelling Substance Isolated from Actinomy-
cetes," Appl. Microbiol., 13, 935 (1965).
8. Gerber, N. N. , "Geosmin, from Microorganisms in
Trans-l,10-dimethyl-trans-9-decanol," Tetrahedron
Lett. , 25_, 2971 (1968) .
9. Zelinsky, N., "Uber eine Synthese der Cyclischen
Tertiaren Alkohole mit Hilfe von Magnesiumhalogen-
alkylen," Ber. Dtsch. Chem. Ges., 34, 2877 (1901).
10. Medsker, L. L. , Jenkins, D., and Thomas, J. F.,
"Odorous Compounds in Natural Waters. 2-Exo-
hydroxy-2-methylbornane, the Major Odorous Compound
Produced by Several Actinomycetes," Environ. Sci.
Technol., 3, 476 (1969).
81
-------
11. Piet, G. J., Zoeteman, B. C. J. , and Kraayeveld,
A. J. A., "Earthy-Smelling Substances in Surface
Waters of the Netherlands/" Water Treatment and
Examination, 21 (4), 281 (1972).
12. Zoeteman, B. C. J., and Kraayeveld, A. J. A., and
Piet, G. J., "Oil Pollution and Drinking Water
Odor," H20, 4., 367 (1971).
13. Keith, L. H., "Identification of Organic Contami-
nants Remaining in a Treated Kraft Paper Mill
Effluent," American Chemical Society, Division of
Water, Air, and Waste Chemistry Preprints, 76 (1969),
Minneapolis, Minnesota.
14. Garrison, A. W., Keith, L. H., and Walker, M. M.,
"The Use of Mass Spectrometry in the Identification
of Organic Contaminants in Water from the Kraft
Paper Mill Industry," American Society for Mass
Spectrometry and Allied Topics Preprints, B205
(1970), San Francisco, California.
15. Thruston, Jr., A. D., and Knight, R. W., "Charac-
terization of Crude and Residual-Type Oils by
Fluorescence Spectroscopy," Environ. Sci. Techno!.,
5, 65 (1971).
U. S. GOVERNMENT PRINTING OFFICE : 1973—514-156/364
82
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
/. RefoftNo.
2.
w
4. Title 5," heport Date
ENVIRONMENTAL APPLICATIONS OP ADVANCED INSTRUMEN-
TAL ANALYSES: ASSISTANCE PROJECTS, FY 69-71 ,
8.: Fig/or/a * af Org* g» . -.atioa
. .. . .
7. Authttr(s)
Lawrence H. Keith and Shirley H. Hercules
9. Orgnaizatioa
National Water Contaminants Characterization
Research Program, Southeast Environmental
Research Laboratory
10. PtojM ,Vo.
16020 GHZ
ftro
n.
Environmenal Protection Agency
|-f. Type
Period Covered
IS. Supplementary
Environmental Protection Agency report number EPA-R2-73-155 ,
May, 1973
IS. Abstract
A multitude of analyses using gas chromatography-mass spectrometry
(GC-MS), nuclear magnetic resonance (NMR), infrared, and fluorescence
spectroscopy to identify and measure organic pollutants in water are
discussed under eleven project categories. In most cases these
analyses have helped to solve, or at least understand more clearly,
the related pollution incident and in some cases provided for
enforcement of regulatory legislation. Projects included identifi-
cation of pesticides and PCB's from natural waters, organics from
industrial wastewaters and landfill runoffs, organics from paper
mill wastewaters and foam, and analyses of oils from oil spills
and suspect sources.
Ha. Descriptors *Water pollution, *Chemical wastes, *Chemical analyses,
*Mass spectrometry, *0il spills, *Pesticides
i~b. identifiers *Fishkill, *Polychlorinated biphenyls, *0rganoleptic
properties, *Nuclear magnetic resonance, *Fluorometry, Taste,
Odor, Oil pollution, Paper mill pollution, Phthalates, Fatty acids
Resin acids
171. COWRRField & Group 05A
,w- ' m
18 Availability ' tf- St<.uri"sCfa8. *
------- |