DOC
EPA
United States
Department of
Commerce
United States
Environmental Protection
Agency
National Oceanic and Atmospheric Administration NOAA TM
Environmental Research Laboratories
Boulder, Colorado 80302
Office of Research and Development
Office of Energy, Minerals and Industry
Washington, D.C. 20460
EPA-600/7-77-098
September 1977
PETROLEUM HYDROCARBONS
IN THE NORTHERN PUGET
SOUND AREA -- A Pilot
Design Study
Interagency
Energy-Environment
Research and Development
Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
NOAA Technical Memorandum ERL MESA-8
A PILOT STUDY ON THE DESIGN OF A PETROLEUM
HYDROCARBON BASELINE INVESTIGATION FOR NORTHERN
PUGET SOUND AND THE STRAIT OF JUAN DE FUCA
by
William D. MacLeod, Jr., Donald W. Brown, Rand G. Jenkins,
L. Scott Ramos, and Victor D. Henry
NOAA National Analytical Facility
Environmental Conservation Division
Northwest and Alaska Fisheries Center
National Marine Fisheries Service
2725 Montlake Boulevard East
Seattle, Washington 98112
Prepared for the MESA (Marine Ecosystems Analysis) Puget Sound
Project, Seattle, Washington in partial fulfillment of
EPA Interagency Agreement No. D6-E693-EN
Program Element No. EHE625-A
EPA Project Officer: Clinton W. Hall (EPA/Washington, D.C.)
NOAA Project Officer: Howard S. Harris (NOAA/Seattle, WA)
This study was conducted
as part of the Federal
Interagency Energy/Environment
Research and Development Program
Prepared for
OFFICE OF ENERGY, MINERALS, AND INDUSTRY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
Novenber 1976
UNITED STATES
DEPARTMENT OF COMMERCE
Elliot L. Richardson, Secretary
NATIONAL OCEANIC AND
ATMOSPHERIC ADMINISTRATION
Robert M. White. Administrator
Environmental Research
Laboratories
Wilmot N. Hess. Director
TMFNT<*
-------
Completion Report Submitted to
PUGET SOUND ENERGY-RELATED RESEARCH PROJECT
MARINE ECOSYSTEMS ANALYSIS PROGRAM
ENVIRONMENTAL RESEARCH LABORATORIES
by
NORTHWEST AND ALASKA FISHERIES CENTER
NATIONAL MARINE FISHERIES SERVICE
NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION
2725 MONTLAKE BOULEVARD EAST
SEATTLE, WASHINGTON 98112
This work is the result of research sponsored by the Environmental
Protection Agency and administered by the Environmental Research
Laboratories of the National Oceanic and Atmospheric Administration.
The Environmental Research Laboratories do no approve, recommend,
or endorse any proprietary product or proprietary material mentioned
in this publication. No reference shall be made to the Environmental
Research Laboratories or to this publication furnished by the
Environmental Research Laboratories in any advertising or sales
promotion which would indicate or imply that the Environmental
Research Laboratories approve, recommend, or endorse any proprietary
product or proprietary material mentioned herein, or which has as its
purpose an intent to cause directly or indirectly the advertised
product to be used or purchased because of this Environmental Research
Laboratories publication.
ii
-------
TABLE OF CONTENTS
FOREWORD vi
INTRODUCTION 1
FIELD STUDIES 2
Site Selection 2
Site Description 3
Sampling Procedures 6
LABORATORY STUDIES 6
Analytical Overview 6
Specific Laboratory Studies 8
Tissue Digestion Studies 10
Sediment Extraction Studies 10
Extraction Solvent Studies 10
Extract Concentrating Studies 10
Eliminating Gels 11
Adsorption Chromatography Studies 11
Desulfurization Studies 11
Microgravimetric Studies 11
Solvent Displacement for GC Analysis 12
Gas Chromatography 12
ANALYTICAL RESULTS 13
RECOMMENDATIONS 30
General 30
First-Year Recommendations 30
Sample environment 30
Target samples 31
Site selection 31
Sampling 31
Field measurements and laboratory analysis 32
Special projects 32
ACKNOWLEDGEMENTS 33
REFERENCES 34
APPENDIX A*. ANALYTICAL PROCEDURES 35
APPENDIX B. INTERTIDAL SITES 45
iii
-------
FIGURES
Number Title Page
1. Peabody Creek intertidal sampling area at Port Angeles,
Washington. 4
2. Dungeness Bay intertidal sampling area at Dungeness, Washington. 5
3. Schematic of tissue and sediment analysis. 9
4. Gas chromatograms of saturated hydrocarbons extracted from
(a) Port Angeles sediments and (b) Dungeness sediments. 18
5. Gas chromatograms of unsaturated hydrocarbons extracted from
(a) Port Angeles sediments and (b) Dungeness sediments. 19
iv
-------
TABLES*
Number Title Page
1. Microgravimetric analysis of hydrocarbons extracted from
intertidal sediment, Port Angeles harbor (P) and Dungeness
Bay (D). 14
2. Alkanes extracted from intertidal sediment, Port Angeles
harbor (P) and Dungeness Bay (D). 15
3. Selected aromatic hydrocarbons extracted from intertidal
sediment, Port Angeles harbor (P) and Dungeness Bay (D). 16
4. Microgravimetric analysis of hydrocarbons extracted from
Mytilus edulis tissue, Port Angeles harbor (P) and Dungeness
Bay (D). 20
5. Alkanes extracted from Mytilus edulis, Port Angeles harbor
(P) and Dungeness Bay (D). 22
6. Selected aromatic hydrocarbons extracted from Mytilus edulis
tissue, Port Angeles harbor (P) and Dungeness Bay (D). 23
7. Microgravimetric analysis of hydrocarbons extracted from Thais
lamellosa tissue, Port Angeles harbor (P) and Dungeness Bay (D). 24
8. Alkanes extracted from Thais lamellosa, Port Angeles harbor (P)
and Dungeness Bay (D). 25
9. Selected aromatic hydrocarbons extracted from Thais lamellosa
10.
11.
12.
tissue, Port Angeles harbor (P) and Dungeness Bay (D) .
Reproducibility of replicate GC sample injections (N=5) .
Unsaturated compounds identified in Peabody Creek sediment.
Selected aromatics suggested to be reported in the baseline
study.
26
27
28
29
Lower case letter (e.g., a) associated with sample number indicates
duplicate/replicate aliquots of the composite sample; hyphen (-) denotes
"not detected."
-------
FOREWORD
Substantially increased petroleum tanker traffic, pipeline transport, and
refining operations are anticipated in the region of Northern Puget Sound and
Strait of Juan de Fuca when the Alaska pipeline comes into operation. To
assess the potential future environmental impact arising from these activi-
ties current hydrocarbon baseline levels must be measured. Under the Puget
Sound Energy-Related Project, the NOAA National Analytical Facility (NAF)
contracted to undertake a pilot study on the "Design of a Petroleum Hydro-
carbon Baseline Investigation for Northern Puget Sound and Strait of Juan de
Fuca." This pilot study was supported by U.S. Environmental Protection
Agency "pass-through" funds administered by the NOAA Marine Ecosystem
Analysis Program. This study was conducted in consultation with representa-
tives from NOAA (National Marine Fisheries Service, Environmental Research
Laboratories and Environmental Data Service), Washington State Department of
Ecology, University of Washington, U.S. Environmental Protection Agency, and
the Canadian Department of the Environment. This report presents the results
of the pilot study and offers recommendations for a first year Petroleum
Hydrocarbon Baseline Investigation.
vi
-------
INTRODUCTION
The greater Puget Sound region has accommodated the transportation and
refining of petroleum and its products for many years without serious diffi-
culties with massive oil spills or smaller scale chronic contamination. Until
recently, most of the crude oil requirements of the Pacific Northwest have
been supplied to the U.S. refineries in the Puget Sound region by pipeline
from Canada. Now, however, this supply has been greatly reduced and probably
will cease in the near future. To maintain refinery production, tanker
traffic has steadily risen, as has the risk of acute and chronic pollution in
the marine environment. This trend can be expected to continue with the open-
ing of the Trans-Alaskan pipeline. Furthermore, it is possible that the
greater Puget Sound region could become a petroleum transshipment point serv-
icing other parts of the country.
In a study performed for the Washington State Legislature by the Oceanographic
Commission of Washington, it was estimated that refinery capacity could
triple and tanker transport of, crude oil could increase tenfold by the twenty-
first century (1). Major issues thus face the petroleum industry and various
levels of governments in this region—issues such as: deciding the ultimate
capacity of refining, pipeline and storage facilities; limitations on tanker
traffic and location of tanker terminals; and the appropriate response to
massive or chronic oil pollution. Decisions resolving issues such as these
require a more detailed knowledge of the marine environment of greater Puget
Sound than is presently available. Among the various physical, chemical, and
biological parameters which need to be better characterized are the hydro-
carbon baseline patterns in the environment prior to proj ected increased
petroleum operations.
Knowledge of the present distribution and concentration of hydrocarbons in
the environment, especially those found in petroleum, is necessary in order
to establish a baseline for measuring the future impact of petroleum pollu-
tion. This means that the current conditions need to be well-defined before
an effective monitoring program can be designed to determine (a) changes from
baseline levels, (b) impacts of pollution, and (c) trends in pollutant concen-
trations. Furthermore, unnaturally high baseline levels of petroleum compo-
nents may pinpoint areas exposed to current contamination before the Alaskan
oil traffic commences.
Previously, only one systematic study has been conducted in the Puget Sound
region to determine the extent of petroleum contamination (2). Currently,
Battelle Northwest, under a contract with the U.S. Energy Research and
Development Administration (ERDA), is carrying out both biological and
chemical studies in the Cherry Point and March Point refinery areas. ERDA
has also contracted the University of Washington to undertake chemical studies
in these areas. Within NOAA, the Energy Resources Program has a project with
the Pacific Marine Environmental Laboratory to analyze the water column,
including particulate matter, in the Northern Puget Sound area. The studies
recommended herein will supplement and interface with the above projects.
-------
The identification and quantitation of petroleum hydrocarbons in the marine
environment is extremely complex. Standardized field and laboratory tech-
niques have been devised only recently. The problem is complicated by bio-r
genie hydrocarbons in the environment, by the complex physical and chemical
nature of petroleum, and by uncertainties in analytical and statistical
procedures. The validity and utility of baseline data could be questioned
unless such problems are resolved. Therefore, it was appropriate to carry
out a pilot study of field and laboratory parameters pertinent to the design
of a baseline program. This involved workshop panel discussions, field
studies and laboratory studies.
A workshop panel was set up to identify important issues and recommend guide-
lines for the pilot study. The panel consisted of experts in the fields of
marine biology, oceanography, analytical chemistry and statistics. Repre-
sentatives from related programs were included. After the pilot study was
largely complete, the data and preliminary conclusions were presented to a
second workshop session which discussed aspects of the baseline design.
In the field studies, recommended biota and sediment were sampled at the
recommended intertidal sites. The organisms represented two intertidal
trophic levels. Methods were established to collect and preserve samples,
avoiding contamination. Samples were collected to establish the statistical
variability of the procedures.
In the laboratory, various sample extraction methods were assessed for effi-
ciency and precision. Techniques such as adsorption chromatography, micro-
gravimetry, gas chromatography, mass spectrometry, and automation were used
to process large numbers of samples. All these studies not only elucidated
methods for determining residual hydrocarbons in environmental samples, but
they also facilitated the development of specific strategies for the follow-
on baseline program.
FIELD STUDIES
Site Selection
The first workshop panel endorsed the recommendation that hydrocarbon resi-
dues in samples be compared from two physically similar sites that differ in
their known exposure to petroleum contamination. In response to the panel
discussions, we chose two areas:
1. Port Angeles Harbor - the possible site of a future supertanker
terminal (already exposed to petroleum hydrocarbons), and
2. Dungeness Bay - a wildlife refuge 10 miles east of Port Angeles
(presumably relatively uncontaminated).
It was agreed that the sampling should be confined to intertidal sediment and
biota (viz., Mytilus edulis and Thais lamellosa).
-------
Harry Tracy, Washington State Department of Ecology, suggested that the area
where Peabody Creek stream empties into Port Angeles harbor might show con-
tamination from a long-standing fuel tank leak. Sampling at a site known to
have received chronic exposure to petroleum hydrocarbons would permit a test
of the effectiveness of the sampling and analytical procedures when compared
with the results from a more "pristine" site.
Three other potential sampling sites in the Port Angeles area were consid-
ered: Morse Creek, about 3 miles east of Port Angeles; Francis St. in Port
Angeles; and Ediz Hook, also within Port Angeles harbor. However, a field
survey of these sites determined that all three sample substrates were found
only at the Peabody Creek location. Therefore, the beach at the mouth of
Peabody Creek was chosen as the relatively contaminated sampling site.
Three potential locations were also surveyed at Dungeness Bay: Dungeness
Spit, with access via Dungeness Beach State Park; Dungeness Beach, at the
Three Crabs Restaurant; and Dungeness Beach, ih miles west of the restaurant.
All three substrates (fine sediment, Mytilus edulis and Thais lamellosa) were
found at only the Three Crabs location so it was selected as the relatively
uncontaminated sampling site.
During the preliminary survey trip (May 11, 1976) Mytilus specimens were
collected in the most accessible locations at Port Angeles and Dungeness:
on rocks and pilings of old piers. Analysis showed that the Mytilus
attached to pilings at Dungeness contained significant amounts of
aromatic hydrocarbons known to be present in creosote. Since this
contamination was not directly petroleum related, samples from pilings
were henceforth avoided.
Due to the small number of Thais specimens available at Dungeness, another
supposedly pristine area, Freshwater Bay, was surveyed. A large population
of Thais organisms was found, and specimens of these as well as Mytilus were
collected. Examination of the organisms by Tony Roth, University of
Washington, revealed that these fauna were of a different species than those
at Port Angeles. They were, in fact, Mytilus californianus and Thais
emarginata. To avoid potential questions about inter-species variability, it
was decided that sampling would be confined to Port Angeles and Dungeness Bay.
Site Description
The outlet of Peabody Creek into Port Angeles harbor is situated at the foot
of Lincoln Street (lat. 48°07'14"N, long. 123025'42"W). A set of pilings in
the creek itself forms the reference point from which the sample points were
located. The exact sampling points are designated in Figure 1.
At Dungeness Beach, the remaining pilings of an old ferry landing form the
reference point for the sampling sites (lat. 48°9'11"N, long. 123°7'11"W)
shown on the map in Figure 2.
-------
PORT ANGELES HARBOR
SEDIMENT SITES
MYTILUS EDULIS SITES
20 meters
RAILROADS!.
LINCOLN ST.
CITY OF
PORT ANGELES
Figure 1. Peabody Creek intertidal sampling area at Port Angeles, Washington.
-------
DUNGENESS BAY
A SEDIMENT SITES
• MYTILUS EDULIS SITES
O THAIS LAMELOSA SITES
o PILINGS
D-3
D-l
D-2
O
40 meters
D-2
HIGH TIDE LIMIT
V
__ 3
BEACH **
A
D-l
A
D-2
A A A •
D-3 D-4 D-5
-^ (
— -~~^-o
! BOAT
RAMP
1
THREE CRABS
RESTAURANT
DUNGENESS-
SEQUIM ROAD
Figure 2. Dungeness Bay intertidal sampling area at Dungeness, Washington,
-------
Sampling Procedures
Composite sediment samples were obtained from multiple cores using 3 cm deep
by 7.5 cm diameter tin cans previously cleaned with solvents and 96% sulfuric
acid. These incremental samples were taken at the corners of a square meter
area and placed in a clean aluminum pail. The samples were composited by
thoroughly mixing with a hand trowel. A portion of each composite (an
aliquot) was transferred to either clean, wide-mouth glass jars or aluminum
foil. At both Port Angeles and Dungeness Bay, the entire sample of one sedi-
ment composite was retained to allow multiple replicate analyses.
For each composite sample, ten specimens of Mytilus edulis and of Thais
lamellosa were collected by hand and wrapped in foil for transport and
storage. Mytilus was relatively abundant at each location and presented no
collection problems; however, only enough Thais specimens could be found to
make one sample composite at Port Angeles and two sample composites at
Dungeness. All samples were stored in a cold chest for transport and placed
in a freezer (-20°C) within 6 hours.
LABORATORY STUDIES
Analytical Overview
For more than a decade, petroleum chemists, organic geochemists and chemical
oceanographers have been applying modern analytical organic methods to
analyze marine sediments and biota for traces of hydrocarbons. These develop-
ments are reviewed biennially (odd years) in Analytical Chemistry under
"Petroleum" (3). Recently, the need to simplify and harmonize the numerous,
lengthy, painstaking, and individualized procedures has led to a series of
conferences and studies focusing on these problems. In the workshop of 26
scientists, funded by NOAA in 1972 at the Santa Catalina Marine Biological
Laboratory, entitled: "Marine Pollution Monitoring: Strategies for a
National Program," Farrington, Giam, Harvey, Parker, and Teal (4) examined
the analytical situation for petroleum contamination and recommended specific
monitoring procedures. The following year the National Academy of Sciences
sponsored a larger workshop (62 scientists) devoted to "Petroleum in the
Marine Environment" (5). Chapter 2, "Analytical Methods," contains numerous
brief yet explicit descriptions of useful techniques. Most analytical
methods for hydrocarbons in the marine environment follow a basic scheme:
substrate digestion, where necessary, followed by solvent extraction, then
adsorption chromatography for sample cleanup or class separation, and gas
chromatography (GC) to determine the hydrocarbon constituents. A somewhat
different approach to hydrocarbon baseline analysis was taken by the National
Bureau of Standards in a study on the Gulf of Alaska (6). Their procedure
employs gas stripping of the volatile compounds which are trapped on Tenax-GC
polymer and analyzed by capillary GC. Residual compounds are analyzed, in
part, by high-performance liquid chromatography (HPLC) with ultraviolet (UV)
detection. The "Pilot Study of the Buccaneer Oil Field" (7) in the Gulf of
Mexico follows aspects of several analytical protocols, but mainly that of
Farrington et al. (4).
-------
Despite the above efforts (or perhaps because of them), it has since become
evident that variations in methodology should be intercompared to determine
which options are best for a "standard method." In their report to the
Bureau of Land Management (BLM), "Evaluation of Extraction Techniques for
Hydrocarbons in Marine Sediments," Rohrback and Reed (8) describe such ex-
periments. Their recommendations closely resemble the current "official" BLM
procedures for hydrocarbon baseline analyses (9).
In our efforts to evaluate and recommend analytical procedures for the hydro-
carbon baseline studies described in this report, the Rohrback and Reed re-
port (8) provided valuable insight. Specifically, their experiments showed
Soxhlet extraction to be only slightly more efficient (4%) than simple solvent
extraction with agitation. Furthermore, the Soxhlet extraction required time-
consuming, freeze-drying of the sediment for best results. In contrast,
Warner (10) proposed a simpler procedure in which an aqueous slurry of sedi-
ment was extracted when tumbled in contact with diethyl ether. Admittedly,
Soxhlet extraction is important with difficult-to-extract, compacted, fossil
sediment (i.e., rocks), but it offers no evident advantage here with uncon-
solidated sediments.
After sample extraction, both the Warner procedure (10) and the BLM procedure
(9) recommend adsorption chromatography for sample extract cleanup and hydro-
carbon group classification, followed by GC determination of hydrocarbon con-
stituents. From time to time, shortcuts to these procedures are proposed
for assessment of petroleum pollution such as gross analysis of the extracts
by infrared, UV, or fluorescence spectrometry. However, Gordon, Keizer, and
Dale (11) point out that if the chemical composition of the substrate is un-
known and free to vary, as is usually the case, quantitative results have no
meaning because the reference calibration cannot be defined. This problem is
not serious with capillary (high-resolution) gas chromatography because of
its capability to separate complex hydrocarbon mixtures into hundreds of con-
stituents and measure many of them at 10~^ g levels or lower. These separated
compounds may be provisionally identified by comparing their GC retention
times with that of the corresponding known standard. This standard also
serves to calibrate the quantitative response factor. Confirmation of identi-
ty may be obtained by an equally sensitive and useful technique, gas chroma-
tography /mass spectrometry (GC/MS). The popularity of GC for hydrocarbon
analysis can be further attributed to its comparatively low cost and broad
analytical range which covers hydrocarbons from C^ to over C^Q. Although GC/
MS is more expensive, it is needed to identify unknown compounds and to con-
firm provisional GC identifications.
A more reqent development, high-performance liquid chromatography (HPLC) with
UV absorbance or fluorescence detection, is becoming increasingly useful for
analysis of aromatic hydrocarbons in marine substrates (6, 10), especially
with compounds which are not sufficiently heat-stable or volatile for GC.
One promising HPLC application for aromatic hydrocarbons is a group-analysis
approach which totals the compounds having a like number of benzene rings
(one, two, etc.).
-------
Specific Laboratory Studies
We have evaluated the latest analytical procedures, techniques, and equipment
to determine the best means of obtaining convenient, reliable, and useful
quantitative measurements of petroleum hydrocarbons in marine sediments and
tissues. Historically, this has not been an easy task, especially for large
numbers of environmental samples. Rohrback and Reed (8) provided valuable
information on the extraction of sediments and other analytical techniques.
We made detailed comparisons of the techniques of Warner (10), BLM (9), NBS
(6), and Rohrback and Reed (8). After careful consideration, Warner's pro-
cedures (10) were adopted as the framework for further development. With
several important modifications, they constitute the procedures currently re-
commended by this laboratory (Appendix A).
In our procedures (Fig. 3), acidified sediment or alkali-digested, homogenized
tissue is extracted with ether and chromatographed on silica gel. The latter
separates the ether soluble extract into two fractions: the saturated hydro-
carbons and the unsaturated hydrocarbons. The fractions are concentrated
separately; those from sediment are desulfurized with activated copper prior
to further analysis. An aliquot of each fraction is weighed on a microbal-
ance, and the weight is compared to the dry weight of the sample. Each frac-
tion is then analyzed by automated gas chromatography (GC), using high reso-
lution glass capillary columns for quantitation of specific compounds. The
identity of all hydrocarbons reported is periodically verified by gas chroma-
tography/mass spectrometry (GC/MS). Details of the analytical procedures ap-
pear in Appendix A.
Our method differs from Warner's (10) in the following respects:
1. Tissue digestion. Tissue samples of marine organisms were di-
gested in alkali overnight at 30°C in Teflon-lined, screw-
capped centrifuge tubes. In the Warner procedure (10) similar
digestion for 2 hours at 90°C showed frequent losses of moder-
ately volatile hydrocarbons (e.g., substituted naphthalenes),
due to imperfectly sealing caps.
2. Silica gel. A coarser grade (100-200 mesh) of silica gel (MCB
SX0144-06) gave satisfactory class separation of the saturated
and unsaturated hydrocarbons. Warner's procedure (10) required
pneumatic pressurization with nominal 200 mesh and finer silica
gel.
3. GC columns. High-resolution glass capillary columns were used
instead of packed columns. The capillaries gave much better
hydrocarbon separations than packed columns, and they also al-
lowed the aromatic hydrocarbon class to be analysed as a single
GC sample rather than as two samples.
Depending on the size and nature of the sample, selected individual hydrocar-
bons can be detected and measured from parts-per-million (ppm) down to parts-
per-billion (ppb) levels, based on dry weight. For tissue samples the
8
-------
SAMPLE
Saponification or Acidification
(tissue)
(sediment)
Ether extraction
Extractables
i
Adsorption chromatography
Saturated hydrocarbons
Unsaturated hydrocarbons
Non-extractables
I
discard
Microgravimetry
Automated
GC analysis
I
GC/MS confirmation
(10% of samples)
Microgravimetry
Automated
GC analysis
I
GC/MS confirmation
(10% of samples)
Figure 3. Schematic of tissue and sediment analysis.
-------
sensitivity is currently about 20 ppb, for sediment it is about 1 ppb. When
available, the GC manufacturer's glass capillary and pre-column assembly
should improve this sensitivity 5- to 100-fold.
Tissue Digestion Studies. Tissue samples spiked with moderately volatile
hydrocarbons were digested in alkali (NaOH) at 90°C for 2 hours in 40 ml cen-
trifuge tubes sealed with Teflon-lined screw caps according to the Warner pro-
cedure (10). About 90% of the added compounds were lost in these digestions
probably due to poor sealing of the tube caps and the elevated temperature.
Alternatives that were tested include:
1. Enzyme digestion (Papain), 25°C;
2. Combination of enzyme and alkali (NaOH) digestion, 25°C; and
3. Alkali only, varying the NaOH concentration and the digestion
temperatures and times.
Enzyme digestion with 0.4-1% purified papain was unacceptable because a stable
emulsion formed during solvent extraction of the digestion mixture. Among
several other experiments,we found that tissue was satisfactorily digested in
4N NaOH at 30°C in 16-18 hours (overnight). This gave a 90% recovery of
spiked hydrocarbons (C,,and up).
Sediment Extraction Studies. Several techniques for sediment extraction
were considered including: (a) exhaustive Soxhlet extraction, (b) agitation
on a shaker table, (c) refluxing, and (d) rolling on a ball-mill tumbler. A
study comparing the first three of these methods was made by Rohrback and
Reed (8) for the BLM. They favored Soxhlet extraction; however, Soxhlet ex-
traction for 50 hr was only 4% more efficient than shaker table extraction
for 20 hr. Refluxing was the least preferred of these methods. A fourth
method, developed by Warner (10), consisted of extracting an aqueous slurry of
the diethyl ether by rolling the mixture in a sealed bottle on a ball-mill tum-
bler. This method, similar to the shaker table technique, was adopted because
it is both efficient and convenient.
Extraction Solvent Studies. Diethyl ether with 2% ethanol preservative
is not acceptable for this procedure. Up to 0.5 ml of ethanol remains in the
extract and upon concentration ethanol separates as a second phase. This etha-
nolic phase may cause bumping and loss of sample. Furthermore, the alcohol
may deactivate the silica gel column and nullify the separation of saturated
and unsaturated hydrocarbons. Freshly-opened, unpreserved ethyl ether must be
used to avert these problems. The level of the diethyl peroxide should be
monitored with peroxide test paper. If the peroxide concentration exceeds
15 ppm, the ether should be purified or discarded.
Extract Concentrating Studies. Several solvent concentrating techniques
were evaluated including: (a) evaporation under a stream of dry nitrogen,
(b) evaporation on a Kontes tube heater, and (c) a combination of the two.
Solvent evaporation under nitrogen alone was too slow to be useful. Evapora-
tion with the tube heater was faster when combined with the nitrogen stream
10
-------
technique. Either choice gave comparable hydrocarbon recoveries. Solvent
boiling techniques employing heat required an ebullator (boiling tube) to pre-
vent bumping and resultant loss of sample. Ultimately, an aluminum-foil
shroud used around the Kontes heating apparatus and tubes facilitated evapora-
tion by heat. This obviated the need for a nitrogen stream.
Eliminating Gels. Some tissues extracted with diethyl ether will form a
gel upon solvent evaporation. This highly viscous material can plug an ad-
sorption chromatography column, preventing solvent flow. However, the sample
gel can be dissolved in methylene chloride and the causative agents can be
removed from the extract solution by filtering the solution through a bed of
chromatographic silica gel prewashed with methylene chloride. The methylene
chloride solvent is then displaced by hexane for adsorption chromatography.
Adsorption Chromatography Studies. We modifed the silica gel chromato-
graphy technique to attain the highest possible flow rate at ambient pressure,
while resolving (separating) the saturated hydrocarbons from the unsaturated
hydrocarbons. The MCB silica gel (nominal 200 mesh and finer) normally used
gave very slow flow rates (<0.5 ml/min) and were easily plugged by moderately
viscous extracts. Faster flow rates were obtained using less silica gel but
the class separation was incomplete. Several chromatographic adsorbents
(silica gels, aluminas, magnesium silicate) in the 100-200 mesh range were
evaluated and found to differ widely in their hydrocarbon resolving charac-
teristics. MCB SX0144-06 (100-200 mesh) silica gel gave a desirable flow
rate, and it completely resolved the saturated hydrocarbons from the alkyl-
substituted, aromatic hydrocarbons.
Detailed column packing techniques were developed for preparing uniform sili-
ca gel columns (see Appendix A). Solvent ratios and fraction volumes were
evaluated to optimize the separation of saturated hydrocarbons from polyun-
saturated and aromatic hydrocarbons to give two fractions of minimum volume.
Desulfurization Studies. GC/MS analysis of sediment extracts showed
sulfur (Sg) in virtually all instances. Although Sg does not interfere with
GC analysis, it does interfere with GC/MS analysis and microgravimetry. Sul-
fur is removed by contact with activated copper (see Appendix A). Initially,
fine granular copper was cleaned with concentrated HC1, then washed with ace-
tone, oven dried at 80°C and stored under petroleum ether. However, when
used in the procedure, GC/MS analyses proved that Sg removal was incomplete.
Oven drying apparently reverses activation of the copper because when this
step was omitted, the copper effectively desulfurized sediment extracts.
This procedure will have to be reinvestigated for analyses of organosulfur
compounds, in which case sulfur levels will have to be reduced effectively
without disturbing the organosulfur compounds.
Microgravimetric Studies. Microgram weighing procedures were evaluated
using a Cahn microbalance. When aliquots of solutions of pure compounds, such
as pentadecane, were air-dried and weighed, the residues indicated that about
10% was recovered. Extracts of environmental samples in solvent generally
contain non-volatile material which acts as a "keeper" to minimize losses of
11
-------
moderately volatile hydrocarbons. In an analogous weighing experiment micro-
gram amounts of a light machine oil dissolved in solvent lost only about 10%
due to evaporation.
Solvent Displacement for GC Analysis. The solvents used in silica gel
chromatography were displaced by carbon disulfide to minimize their inter-
ference with the GC detector. In the first experiments, the fractions from
silica gel chromatography were evaporated in the heater block to 0.5 ml, then
placed in uncapped GC vials, and allowed to evaporate to dryness at room tem-
perature. The residues were taken up in 0.5 ml carbon disulfide (CS^) and
analyzed by GC. Analysis showed that only about 10% of pentadecane and
naphthalene was recovered from spiked samples. Subsequently, the expected
analytical recovery efficiency (70-95%) was attained by avoiding complete
evaporation. In this procedure, the petroleum ether and/or methylene chlor-
ide solvents were displaced with CS2 by adding 1 ml C$2 to the 0.5 ml chroma-
tographic concentrate. Then the mixture was reconcentrated to 0.5 ml in the
heater block. Conveniently, an internal standard, such as hexamethylbenzene,
can be added in the 1 ml CS2^
Gas Chromatography. The tissue and sediment extracts are extremely com-
plex organic mixtures. A GC system capable of the highest possible resolu-
tion is needed to obtain useful analyses from such samples. Since packed
columns are limited in separation capability compared with capillary GC col-
umns, capillaries were chosen for this study. Many instrumental modifica-
tions of both the GC and GC/MS systems were necessary to achieve optimum con-
figurations for capillary column operation (see illustrated details in
Appendix A).
12
-------
ANALYTICAL RESULTS
Over 60 composite marine intertidal samples of sediment, mussel (Mytilus) and
snail (Thais)>were extracted for residual hydrocarbons and analyzed according
to the recommended procedure. The saturated and unsaturated hydrocarbon
classes separated by adsorption chromatography were determined by micro-
gravimetry. Selected individual hydrocarbon compounds were determined by GC.
Each composite was analyzed in replicate (duplicate or more) as denoted by
the sample code (P-la, P-lb, etc.). The results of these analyses appear in
Tables 1-9.
In many instances the agreement of the amounts of individual compounds
(Tables 2,3,5,6,8, and 9) from duplicate analyses is excellent. Other
instances show varied comparisons, i.e., one compound may be high in sample a,
low in sample b, whereas the reverse is true for a different compound in the
paired samples. Occasionally the difference is biased one way throughout the
range of compounds.
Table 10 shows the GC reproducibility of a single extracted sample injected
in replicate (5 times). Although further improvement in these data can be
virtually assured at this time, it can be seen that the GC precision shown in
Table 10 is good-to-excellent except for the n-C3Q and n-C3i alkanes.
As far as the overall analytical procedure is concerned, the relative stand-
ard deviation (or coefficient of variation) at the 95% confidence level
averaged less than 20% for individual hydrocarbon compounds at the practical
limit of sensitivity. For 100 g sediment samples (wet), the practical
sensitivity limit is 1 ng/g dry sediment or 1 part per billion (1 ppb). For
10 g of wet mussel tissue it is about 20 ppb (dry wt). Anticipated improve-
ments in the GC sample introduction apparatus should improve these sensitivi-
ties from 5- to 100-fold.
The objective of these analyses was to determine the abundance and variation
of residual hydrocarbons in marine intertidal sediments and biota. Compara-
tive differences between these hydrocarbon levels from a relatively unpolluted
and a polluted site were investigated. Table 1 shows the comparative residual
levels of the hydrocarbon classes found in sediments at Port Angeles harbor
and Dungeness Bay. The supposed greater exposure to petroleum related hydro-
carbon pollution at Port Angeles is indicated in the higher levels of satu-
rated hydrocarbon (chromatography fraction 1) found in Port Angeles sediments
(samples P-l to P-5) compared to those found in Dungeness (D-l to D-5).
These differences are also reflected in the comparative levels of the
selected individual saturated hydrocarbons (alkanes) shown in Table 2. The
total urtsaturated hydrocarbons (chromatography fraction 2) at Port Angeles
and Dungeness (Table 1) mirror this situation, although not to the same degree
of difference. This lesser difference is also borne out by the comparative
levels of selected predominant arenes (aromatic hydrocarbons) in the unsatu-
rated fraction (see Table 3).
The evidence of the Port Angeles microgravimetric data (Table 1) strongly
suggests that these hydrocarbon class levels are related to known hydrocarbon
contamination. Specifically, Peabody Creek, which empties into Port Angeles
13
-------
Table 1. Microgravimetric analysis of hydrocarbons extracted from intertidal
sediment, Port Angeles harbor (P) and Dungeness Bay (D).
Sample
P-l a
b
P-2 a
b
P-3 a
b
P-4 a
b
P-5 a
b
c
d
e
D-l a
b
D-2 a
b
D-3 a
b
c
d
D-4 a
b
D-5 a
b
Silica Gel Chromatography Fraction
1. (saturated)
950
770
76
78
1000
1100
170
130
110
100
100
110
73
2.5
2.7
5.5
3.7
2.8
1.5
2.9
1.8
2.7
3.2
3.9
4.2
(yg/g dry sediment)
2. (unsaturated)
360
350
95
94
430
400
160
190
360
84
180
180
140
34
37
29
27
23
24
33
29
29
17
36
33
14
-------
Table 2. Alkanes extracted from intertidal sediment, Port Angeles harbor (P) and Dungeness Bay (B).
Alkane* concentration, ng/g of dry sediment
Sample
P-l a
b
P-2 a
b
P-3 a
t
P-4 a
b
P-5 a
t
D-l a
b
D-2 a
b
D-3 a
t
D-4 a
I
D-5 a
b
C14
290
210
<2
8.6
170
160
30
25
22
12
9.1
9.0
12
10
4.8
5.7
12
11
4.3
4.1
C15
370
380
19
20
200
190
43
35
35
17
42
45
87
73
31
31
71
72
20
23
C16*
300
300
15
15
180
170
38
32
35
27
11
11
18
15
8.5
7.7
16
17
6.5
6.6
C17
420
410
25
24
240
240
64
50
50
42
26
30
31
29
13
11
32
34
15
13
C18
290
300
17
16
130
120
33
31
35
25
11
12
16
14
11
6.5
15
16
6.5
7.0
C19
370
370
13
10
34
32
<2
<2
30
55
19
20
27
26
17
13
31
36
11
10
C20
230
250
18
18
110
110
28
26
28
26
11
10
15
12
30C
5.7
16
19
7.5
8.2
C21
220
250
28
30
32
32
22
2
33
22
15
16
21
21
10
11
26
33
12
13
C22
180
210
22
23
74
60
16
2
28
28
8.0
9.8
12
13
10
6.5
18
32
9.2
9.8
C23
450
480
250
150
170
170
14
21
47
76
33
34
42
43
42
30
76
100
36
35
C24 C25
270 1700
290 1400
9.9 1200
9.5 1000
120 630
92 570
3.4 140
6.6 130
35 130
43 140
13 130
14 120
15 110
17 140
23 130
16 110
25 220
50 300
14 110
16 120
C26
630
670
120
230
210
170
78
69
50
46
13
13
13
26
25
18
8.9
21
8.4
17
C27
1700
1800
2300
1400
930
870
250
240
210
190
160
160
190
220
230
160
350
450
210
220
C28
320
350
270
150
290
240
160
180
40
20
56
19
48
23
100
25
21
40
10
20
c29
870
750
1400
1400
470
450
150
170
120
66
110
100
150
70
c 180
120
68
71
66
49
C30
94
240
150
150
88
88
<2
<2
<2
<2
4.8
2.5
13
14
12
8.8
1.9
2.6
8.3
3.9
C31
1100
1600
870
820
430
360
160
180
110
130
44
41
62
88
110
66
82
57
55
77
Pristane
400
370
47
47
240
230
77
74
74
58
33
36
45
41
16
16
59
62
23
22
Phytane
290
280
24
25
94
99
43
42
42
37
12
15
11
10
5.2
4.5
16
16
5.1
5.5
Eodd-C
7200
7140
6205
5554
3136
2914
843
826
765
738
580
570
720
710
763
552
960
1150
540
560
£even-C
7200
2820
602
520
1372
1210
386
370
277
230
140
100
160
140
234
94
130
210
80
90
* normal alkane denoted where chain length given as CL,, C^., etc.
c denotes contaminated peak
-------
Table 3. Selected aromatic hydrocarbons extracted from intertidal sediment,
Port Angeles harbor (P) and Dungeness Bay (D).
Sample
P-l a
b
P-2 a
b
P-3 a
b
P-4 a
b
P-5 a
b
c
d
e
D-l a
b
D-2 a
b
D-3 a
b
c
d
D-4 a
b
D-5 a
b
Concentration,
Phenanthrene
320
240
43
50
190
230
320
110
37
18
30
36
87
11
6.7
11
5.4
1.2
3.8
3.3
3.1
11
—
-
3.9
ng/g dry sediment
Fluoranthene
730
1000
120
110
660
1100
600
370
180
190
180
150
230
17
18
24
13
10
25
7.5
8.0
22
-
6.4
7.0
Pyrene
450
480
110
100
340
510
390
300
110
190
200
150
260
13
17
24
13
13
13
13
15
11
-
6.4
7.0
16
-------
harbor at the sampling site (Fig. 1), was reported to have been previously
contaminated by a long-standing fuel tank leak. Table 1 shows that sediment
samples taken adjacent to the Peabody Creek stream bed (samples P-l and P-3)
contained up to 10-fold higher levels of saturated hydrocarbon residues
(chromatography fraction 1) than samples on the same beach farther from the
stream bed (samples P-2, P-4, and P-5). The unsaturated hydrocarbon fraction
(chromatography fraction 2), containing the arenes also reflected this trend,
though to a lesser degree. The data in Table 3 suggest that the three
selected arenes (phenanthrene, fluoranthene, and pyrene) may be possible
petrogenic indicators. Again, the highest values were adjacent to the
Peabody Creek, Port Angeles, and the lowest values were at Dungeness. In
addition to data on levels of hydrogen classes and specific compounds (Tables
1-9), the actual gas chromatograms (GC charts) can aid in diagnosing a
pristine vs. contaminated situation. For example, the n-alkanes from sediment
samples show an odd-carbon predominance over even-carbon in the range of
ii-C26 to C3i for both Port Angeles (Fig. 4a) and Dungeness (Fig. 4b). This
odd-carbon predominance is believed to be due to terrestrial biogenic input
in both cases; however, in the region around ii-Ci5 the comparison changes.
Although the n-alkanes from Dungeness sediments display a marked odd-carbon
predominance over even, this situation is not reflected in the n-alkanes
of Port Angeles sediment. Since it is known that diesel oil has contaminated
Peabody Creek for a long time at Port Angeles, it is not surprising that the
Ci4~C20 n-alkanes show less alternation. The size of the extensive hump of
unresolved compounds above the usual GC baseline rise in Figure 3a also
suggests petrogenic contamination. Finally, Fig. 4b is much less complex
than Fig. 4a. Thus, significant qualitative information can be gained by
visual inspection of the chromatograms.
Chromatograms of the arenes from sediment (Fig. 5) are also significantly
simpler from Dungeness (Fig. 5b) than from Peabody Creek, Port Angeles
,(Fig. 5a). Unfortunately, visual differences are not as clear with Mytilus
or Thais arene gas chromatograms; biogenic hydrocarbons complicate the GC
picture such that it may be preferable to rely on tabular abundancy data of
selected compounds (Tables 5,6,8, and 9).
GC/MS was used to identify or verify the identity of all hydrocarbon compounds
listed in the tables. Table 11 lists the unsaturated compounds identified
from sediments sampled adjacent to Peabody Creek, Port Angeles. Most are
arenes commonly occurring in petroleum and its products. Pinene, a natural
alkene, and dichorobenzene, a petrochemical, were also identified.
Composite intertidal samples of mussels (Mytilus edulis) could not be
obtained*immediately adjacent to the Peabody Creek bed, therefore, where they
were found could be presumed to reflect more the general harbor pollution at
Port Angeles than the specific fuel tank contamination. Saturated hydro-
carbon levels (chromatography fraction 1) at Port Angeles (Table 4) predomi-
nate over those at Dungeness in all but one instance (Sample D-2). On the
other hand, microgravimetric data on the unsaturated hydrocarbons (chroma-
tography fraction 2) from mussels seem to bear little relationship to hydro-
carbon contamination due to the predominance of biogenic olefins (10) in this
fraction.
17
-------
SATURATED HYDROCARBONS
C27 ^8 C29
I \ I
PORT ANGLES SEDIMENTS
DUNGENESS SEDIMENTS
27 c,
%5
&
U
1. STD.
r
Pristane Phytane
<*3
C26
'29
C,n 31
"32
Figure 4. Gas chromatograms of saturated hydrocarbons extracted from (a) Port Angeles
sediments and (b) Dungeness sediments.
-------
UNSATURATED HYDROCARBONS
vo
PORT ANGELES SEDIMENTS
a = Naphthalene
b = 2-Methylnaphthalene
c = 1-Methyl naphthalene
d = Biphenyl
e = C2-Naphthalenes
f = Cj-Naphthalenes
g = Fluorene
h = Phenanthrene
i = Methyl phenanthrenes/anthracenes
j - Pyrene
k - Chrysene - , STO
u
DUNGENESS SEDIMENTS
I. STD.
Figure 5. Gas chromatograms of unsaturated hydrocarbons from (a) Port Angeles sediments
and (b) Dungeness sediments.
-------
Table 4. Microgravimetric analysis of hydrocarbons extracted from Mytilus
Sample
P-l a
b
P-2 a
b
c
d
e
P-3 a
b
P-4 a
b
P-5 a
b
D-l a
D-2 a
b
D-3 a
b
c
d
e
edulis tissue, Port Angeles harbor
Silica Gel Chromatography Fraction
1. (saturated)
350
310
350
220
460
300
350
450
380
260
250
340
370
31
123
340
21
51
58
35
63
(P) and Dungeness Bay (D) .
(yg/g dry tissue)
2 . (unsaturated)
1200
710
810
420
550
780
880
410
490
360
700
540
450
720
340
630
1000
560
690
670
620
20
-------
Many of the individual alkanes (viz., n-Ci4, B.-Ci8> B^ig* £"C20» nrc22>
n.~c24> n.~c26j pristane and phytane) extracted from mussels are much more
abundant in the Port Angeles samples than in the Dungeness samples (Table 5).
This may reflect petrogenic hydrocarbon contamination. In contrast, most
levels of the prominent, odd-numbered n-alkanes do not differ sufficiently
between Port Angeles and Dungeness mussels to indicate petroleum contamina-
tion. Under the GC analysis conditions, the n-C28 alkane in mussels co-
chromatographs with a major biogenic (terpanoid) hydrocarbon and, therefore,
cannot be used as a petrogenic indicator. The data also show n-C3Q to be of
little aid. The comparative levels of the selected arenes in the mussels
from the two areas (Table 6) indicate that these compounds could be signifi-
cant in determining petrogenic contamination.
Among snails studied from both areas, the microgravimetric data on the hydro-
carbon classes (Table 7) show no significant differences. Except for pristane
and phytane in Table 8, the same is true for the alkanes. As with mussels,
the abundance of selected arenes in snails (Table 9) reflect the assumed
contamination at Port Angeles when compared to Dungeness. However, Thais
lamellosa is not promising for baseline studies because its abundance at a
given place and time is unpredictable.
In this particular study, it was convenient to use three predominant arenes
in the unsaturated fraction (phenanthrene, fluoranthene, and pyrene) to indi-
cate the degree of possible petrogenic contamination. For the follow-on
baseline study, GC conditions should be optimized to cover the expanded list
of arenes in Table 12. These compounds were selected because (a) they are
found in crude and refined petroleum, (b) they are recovered efficiently by
our procedures, and (c) they cover a wide range of arenes from 1-5 rings, yet
can be determined in a single gas chromatogram. The n- and ^-propylbenzenes
are about 60% recoverable, naphthalene about 70%. Because of their lower
volatilities, the rest of the arenes in Table 12 are over 80% recoverable in
sample workup.
The benzenes and naphthalenes are the most abundant arenes in crude oil, as
well as the most water-soluble, volatile, and acutely toxic. Hence, the
benzenes and naphthalenes deserve a prominent place in a baseline survey.
Unfortunately, beyond the C2 substituted members, the GC pattern becomes
difficult to manage with all the possible chemical isomers, thus the list is
limited to members which are prominent in pollution and well separated by GC.
The polycyclic arenes, though less abundant and less water-soluble, are
important for their relationship to possible chronic biological effects.
More information is needed on their accumulation in the environment. The
polycycllcs include: fluorene, phenanthrene and anthracene, methyl phenan-
threnes and anthracenes (3-ring arenes); fluoranthene, pyrene, chrysene,
benzanthracene (4-ring arenes); and the benzpyrenes, and perylene (5-ring
arenes).
21
-------
Table 5. Alkanes extracted from Mvtilis edulis,
to
Port Angeles harbor (P) and Dungeness Bay (D).
Alkane* concentration, ng/g dry tissue
Sample
P-l
P-2
P-3
P-4
P-5
D-l
D-3
a
b
a
b
c
d
e
a
b
a
b
a
b
a
a
b
a
b
c
d
e
C14
360
150
190
170
340
190
450
140
150
190
290
150
220
—
_
-
33
35
33
32
C15
150
180
370
330
330
430
220
140
130
180
130
150
170
110
/•£
DO
100
200
210
210
210
210
C16
170
140
92
59
92
540
59
68
65
220
-
200
130
—
_
-
-
-
33
32
C17
33
300
120
59
150
340
400
170
130
330
210
230
290
76
i f\r\
J.UU
120
100
180
70
120
120
C18
100
33
61
30
61
280
30
27
65
-
-
-
-
-
-
-
-
-
C19 C20
- 170
160
61 180
30 170
210 340
170 280
- 420
-
-
100
-
- 100
-
-
-
-
-
-
C21
-
-
310
270
700
440
440
-
-
-
-
-
-
100
620
70
320
32
C22 C23
100
65
120 220
120 130
180 190
240 280
59 180
34 68
48 65
66
-
66
- <20
33
-
-
33
32
C24
<20
30
120
130
150
220
89
-
33
33
-
-
-
-
-
-
-
-
C25
60
98
340
120
280
400
410
190
100
200
130
100
160
-
65
100
130
110
C26
<20
65
250
120
280
280
240
68
65
130
100
-
-
-
-
-
<20
32
C27
130
130
310
59
120
220
270
-
33
130
100
130
160
170
100
240
290
220
C28
370
390
64
30
61
93
510
-
65
260
510
700
250
630
590
920
1600
610
C29
230
200
310
-
-
130
180
-
-
230
300
-
190
100
100
100
220
150
C30 C31
70 -
65 30
-
-
-
-
-
-
33 33
33 -
-
-
-
~ ~
-
100
65
35
33 -
33
Pristane
1000
1000
250
810
640
1000
1200
1700
1700
830
130
660
870
390
-
-
-
-
100
110
Phytane
350
280
150
280
430
420
450
510
590
260
170
230
260
~
-
-
-
-
-
-
Eodd-C
703
1003
2041
998
1980
2410
2100
568
491
1136
870
676
970
186
i (\f\
.LOO
703
1275
790
1323
874
£even-C
1240
1033
1077
829
1504
2123
1857
337
491
966
900
1150
600
0
0
730
688
990
1699
739
* normal alkane denoted where chain length given as C ,, C _, etc.
-------
Table 6. Selected aromatic hydrocarbons extracted from Mytilus edulis tissue,
Sample
P-l a
b
P-2 a
b
c
d
e
P-3 a
b
P-4 a
b
P-5 a
b
D-l a
D-2 a
b
D-3 a
b
c
d
Port Angeles harbor (P) and
Concentration ,
Phenanthrene
130
170
86
100
64
71
120
110
180
190
57
150
170
-
-
-
-
-
__
Dungeness Bay (D) .
ng/g dry tissue
Fluoranthene
1000
1400
1000
740
950
740
540
370
290
560
640
1100
1100
—
-
-
-
-
-
Pyrene
400
590
460
360
400
340
230
130
160
170
140
140
170
-
-
-
-
-
-
23
-------
Table 7. Microgravimetric analysis of hydrocarbons extracted from Thais
lamellosa tissue, Port Angeles harbor (P) and Dungeness Bay (D),
Silica Gel Chromatography Fraction (]ag/g dry tissue)
Sample 1. (saturated) 2. (unsaturated)
P-l a 480 210
b 22 180
P-2 a 85 270
D-l a 18 230
b 12 300
c 20 250
d 46 470
D-2 a 1.5 690
24
-------
Table 8. Alkanes extracted from Thais lamellosa, Port Angeles harbor (P) and Dungeness Bay (D).
Alkane* concentration, ng/g dry tissue
Sample
C14 C15
P-l a 28 50
b 14 51
P-2 a - 31
NJ
Cn
D-l a 42
b - 34
c - 34
d - 50
D-2 a 16
C16 C17 C18 C19
28 150 66 82
28 130 33 -
30 - 34 -
17 96 -
76 - -
11 140 14
- 97 -
- 24 -
C20 C21 C22 C23 C24 C25 C26 C27 C28
26 - 26 35 25 58 47 110 520
23 20 - 26 100 240
62 - - - - - 60 130
________ 390
________ 280
_______ 45 820
100 - 31 42 65 53 160 310
- 26 ------ 180
C29 C30
60
50 33
65
65
43 -
100
140
64 -
C. Pristine Phytane
25 1900 620
1300 360
500 130
210
250
330
74 350
150
Eodd-C
570
354
156
203
153
319
717
130
leven-C
766
394
256
407 ,
280
845
405
180
* normal alkane denoted where chain length given as C..., C.,. etc.
-------
Table 9. Selected aromatic hydrocarbons extracted from Thais lamellosa
tissue, Port Angeles harbor (P) and Dungeness Bay (D).
Concentration, ng/g dry tissue
Sample
Phenanthrene Fluoranthene Pyrene
P-l a 160 210 200
b 100 160 160
P-2 a 280 200 130
D-l a
b - -
c - - -
d - - -
D-2 a - - -
26
-------
Table 10. Reproducibility of replicate GC sample injections (N = 5)
Alkane
C14
C15
C16
C17
C18
C19
C20
C21
C22
C23
C24
C25
C26
C27
C28
C29
C30
C31
Pristane
Phytane
Rel. Std. Dev.
2.3
4.4
1.9
2.5
2.8
3.4
6.6
6.0
5.9
8.3
10.0
12.4
12.0
16.9
12.3
14.1
51
38
2.9
3.6
27
-------
Table 11. Unsaturated compounds identified in Peabody Creek sediment.
Compound Number of isomers found
toluene
xylene 3
pinene*
C~ - benzene 6
C, - benzene 10
GS - benzene 4+
napbthalene
C, - benzene 1+
dichlorobenzene* 1
methylnaphthalene 2
C_ - naphthalene 7
C_ - benzene
C_ - naphthalene 14
fluorene
C, - naphthalene 18
phenathrene
anthracene
methyl fluorene 2
methyl phenanthrene and/or 5
methyl anthracene
C_ - (phenanthrene and/or 10
anthracene)
fluoranthene
pyrene
benzanthracene
chrysene
benzofluoranthene 1
benzpyrene 2
perylene
* non-petrogenic
28
-------
Table 12. Selected aromatics suggested to be reported in the baseline study
1. n_-propylbenzene
2. ij-propylbenzene
3. naphthalene
4. 1-methylnaphthalene
5. 2-methylnaphthalene
6. biphenyl
7. dibenzothiophene
8. phenanthrene
9. anthracene
10. methylphenanthrene
11. fluoranthene
12. pyrene
13. chrysene
14. benz(a)anthracene
15. benzo(e)pyrene
16. benzo(a)pyrene
17. perylene
29
-------
RECOMMENDATIONS
General
This pilot study has demonstrated that methodology exists to detect and meas-
ure a number of hydrocarbons in sediments, mussels (Mytilus edulis and
Mytilus califomianus) , and a snail (Thais lamellosa) . The use of this meth-
odology in an area relatively polluted with oil (Port Angeles) and in a
relatively unpolluted area (Dungeness Bay) has revealed substantial quantita-
tive differences in these compounds. During the first year, major emphasis
should be given to seasonal variations and broad geographical coverage with
minor effort devoted to widening the list of compounds and trophic levels
under study. Work should continue to emphasize analysis of sediments since
this is where indications of the accumulation of petroleum contamination can
be expected. Past and current problems with water column analyses preclude
main reliance on water as a sample matrix but further study of it is war-
ranted.
A number of parameters important in baseline studies are undefined. For ex-
ample, the optimum interval for sampling for baseline studies has not been
established; seasonal differences are unknown. Areas having the highest and
lowest probable petrogenic contamination thus should be sampled more fre-
quently (e.g., twice quarterly). If hydrocarbon levels in these areas fluc-
tuate significantly, the program should be flexible enough to allow even more
frequent sampling. If possible, compounds such as the cycloalkanes (naph-
thenes) and a larger number of aromatic compounds including heterocyclics,
should be surveyed. Also, compounds which survive weathering and biodegrada-
tion or result from these processes should be included in the analyses. Al-
though trophic levels were treated minimally in the pilot study (snails which
feed on the mussels), this area of study could be expanded. Attention could
be given to studying special situations such as times of physiological stress
in biota (e.g., spawning cycles) and areas where microbiological sampling is
planned.
First-Year Recommendations
As a result of the pilot study and in consultation with the workshop panels
of experts, the following recommendations are made for the first year of a
PETROLEUM HYDROCARBON BASELINE INVESTIGATION FOR NORTHERN PUGET SOUND AND
STRAIT OF JUAN DE FUCA.
Sample environment. Sampling of the intertidal zone should be emphasized.
The intertidal zone is a major point of contact between surface-borne oil pol-
lution and marine biota and sediment through the action of wind, surf, and
tides. The intertidal zone normally serves as a vital shelter for a multitude
of juvenile and mature biota, some of which are particularly vulnerable to
parts-per-billion (ppb) levels of certain petroleum hydrocarbons. Through
various processes petroleum hydrocarbons are sorbed by intertidal sediments
30
-------
and by biota exposed to them. These contaminants eventually can be purged
from such intertidal substrates by complex, incompletely-understood processes.
However, many hydrocarbons are retained in the environment weeks or months
after introduction. A logistical advantage of the intertidal zone is its ac-
cessibility for reproducible, periodic sampling by land.
Target samples. Sampling of sediment and Mytilus sp. should be empha-
sized. Sedimentary beaches exist throughout the study region and Mytilus sp.
is ubiquitous in the intertidal zone. Contamination of the intertidal zone
by petroleum hydrocarbons will be reflected in these substrates. Both sub-
strates are amenable to reproducible and periodic sampling and should be sam-
pled simultaneously at the same site to allow correlation of results.
Site selection. Sampling areas should be evenly distributed throughout
the region. Maps of suitable areas are given in Appendix B. Sampling sites
should include representatives from zones presumed to be relatively uncon-
taminated. Such intertidal sites provide the lowest hydrocarbon baseline data
and, therefore, allow the clearest early indication of a change due to incipi-
ent petroleum pollution. Appendix B covers a number of such prospective sites
(a-h) throughout the region, which have been examined by road maps, marine
charts, and by seaplane to establish their general suitability (e.g., access,
beach type and extent). These sites comprise a geographical grid covering
the region. Sampling areas also should include those believed to be contam-
inated (i-k, Appendix B).
Sampling areas should include a variety of sedimentary beach types. Beach
types vary according to their slope, exposure to wave action, sediment grain
size, and density and diversity of biota. All of these variables have been
shown to affect the disposition of hydrocarbon contaminants in other areas.
Sampling areas should coincide, where possible, with those of other related
studies in the region. Each sampling area should contain both sample types
(sediment and mussel) in close proximity to permit correlation of results.
Sampling. Samples should be collected quarterly at constant tidal eleva-
tions. Sediment grain size generally increases with increasing tidal eleva-
tion while biotic density declines generally below the zero-foot tide level.
Because small sediment grains tend to retain petroleum contaminants and since
knowledge of the interrelation of petroleum contamination in biota is desired,
the range of 0 to +3 feet tidal elevation should be sampled.
Sediment sampling should be carried out according to a "systematic-stratified"
scheme. Core samples (20-100) should be collected to a constant depth (e.g.,
3 cm) at regular intervals along at least two different tidal elevations over
a distance of 50 meters or greater and combined to make a single composite
sample. A second composite sample should be collected according to the same
procedure. This scheme should provide adequate statistical representation
of the sample site within normal analytical variability.
Specimens of Mytilus sp. should not be collected from pilings. This study
showed that these organisms absorb aromatic hydrocarbons from creosote treated
pilings. Specimens of Mytilus sp. should be within a given size range (e.g.,
2-4 cm length).
31
-------
Field measurements and laboratory analyses. Field conditions at each
sampling time should be described, including wave exposure, weather conditions,
and air, water, and sediment temperature. Sediment samples should be de-
scribed according to physical characteristics (color, layering, etc.)- The
location and substrate in which the mussels are found should be described.
Each sampling location should be documented each sampling time by photography.
Sediment composites should be characterized according to grain size. This is
necessary for comparing results between samples since the retention of hydro-
carbons is related to grain size. Total organic carbon should be determined
in sediment composites and total lipids in mussel composites. Analytical re-
sults should be reported in terms of sample dry and wet weight.
Samples should be extracted and analyzed for residual hydrocarbons according
to the procedures developed in the pilot study (Appendix A). Analytical re-
sults should include: (a) microgravimetric determination of total extract-
ables, (b) total saturated and unsaturated hydrocarbons from adsorption
chromatography, and (c) gas chromatographic determination of n-alkanes from
0^4 to €3^, pristane, phytane, and specific aromatic compounds listed in
Table 12. Compound identities should be verified by mass spectrometry.
Special projects. Additional studies to be considered include:
1. Continued sampling at the Port Angeles/Peabody Creek site to
provide information on changes in hydrocarbon concentrations
with time at a site where chronic input is believed to have
been stopped.
2. Evaluation of water column sampling and analysis techniques.
3. Sampling of sediment and Mytilus at one Pacific Ocean intertidal
site (south of Cape Flattery) as a comparison.
32
-------
ACKNOWLEDGEMENTS
¥e wish to acknowledge the generous assistance of:
D.C. Malins, N. Karrick, R.C. Clark, Jr., and J.S. Finley of the
Environmental Conservation Division, Northwest and Alaska
Fisheries Center, Seattle, Washington, and
H. Harris and E. Long of the Marine Ecosystems Analysis Program,
Pacific Marine Environmental Laboratory, Seattle, Washington
We are also grateful to the contributions of the workshop participants
(in addition to the above):
J. Anderson, Battelle Northwest Laboratories (Sequim)
R. Carpenter and C. Nyblade, University of Washington
D. Jamieson, H. Tracy, and E. DeNike, Washington State Department
of Ecology
J. Cummins, U.S. Environmental Protection Agency
C.S. Wong and W. Cretney, Canadian Department of the Environment
D. Wolfe, E. Myers, M. Stansby, D. Worlund, R. Kappenmann,
R. Kopensky, J. Cline, R. Feeley, J. Larrance, and J. Mattson
from the National Oceanic and Atmospheric Administration
University of Washington students, R. L. Dills and S. M. Price, provided
valuable laboratory assistance.
33
-------
REFERENCES
1. Oceanographic Commission of Washington (1972). Risk analysis of the oil
transportation system and (1974) Offshore petroleum transfer systems for
Washington State. Oceanographic Institute of Washington, Seattle, WA;
and Pac. Northwest Sea 7(3,4):3-23 (1974).
2. Clark, R.C., Jr. and J.S. Finley (1976). Unpublished data, Northwest
and Alaska Fisheries Center, NMFS, NOAA, Seattle, WA.
3. Bradley, M.P.T. (1975). Hydrocarbons. In: Petroleum. (J.M. Fraser,
ed.), Anal. Chem. 47(5):169R; see also previous odd-year reviews listed
therein (ref. 1A-11A).
4. Farrington, J., C.S. Giam, G.R. Harvey, P. Parker, and J. Teal (1972).
Analytical techniques for selected organic compounds: petroleum* In:
Marine Pollution Monitoring: Strategies for a National Program.
(E.D. Goldberg, Ed.), Allan Hancock Found., Santa Catalina Mar. Biol.
Lab., Univ. S. Calif.
5. Workshop (May 21-25, 1973) on Inputs, Fates and Effects of Petroleum in
the Marine Environment (1975). Petroleum in the Marine Environment.
Natl. Acad. Sci., Washington, DC.
6. Chesler, S.N., B.H. Gump, H.S. Hertz, W.E. May, S.M. Dyszel, and
D.P. Enagonio (1976). Trace hydrocarbon analysis: The Natl. Bur.
Stand. Prince William Sound/Northeastern Gulf of Alaska Baseline Study.
Tech. Note 889, Washington, DC.
7. Harper, D.E., Jr., R.J. Scrudato, and C.S. Giam (1976). Pilot study of
the Buccaneer oil field (benthos and sediments) - A preliminary environ-
mental assessment of the Buccaneer oil/gas field. Gulf Fish. Center,
Galveston, TX.
8. Rohrback, B.C. and W.E. Reed (1975). Evaluation of extraction
techniques for hydrocarbons in marine sediments. Pub. No. 1537, Inst.
of Geophysics and Planetary Physics, U of Calif., Los Angeles, CA.
9. So. Atlantic Benchmark Study (1976). Bur. of Land Management,
Washington, DC.
10. Warner, J.S. (1976). Determination of aliphatic and aromatic hydrocar-
bons in marine organisms. Anal. Chem. 48:578; also private commmunica-
tion.
11. Gordon, D.C., P.O. Keizer, and J. Dale (1974). Estimates using fluores-
cense spectroscopy of the present state of petroleum hydrocarbon con-
tamination in the water column of the Northwest Atlantic Ocean. Mar.
Chem. 2:251.
34
-------
APPENDIX A
ANALYTICAL PROCEDURES
*
Materials
Materials contacting the sample were confined to glass, Teflon, metal or
residue-free solvents and reagents. This includes the liners of caps and
lids. All glassware was washed in hot laboratory detergent, dried, and
rinsed in sequence with reagent grade acetone and methylene chloride sol-
vents dispensed from previously cleaned Teflon wash bottles. Teflon and
metal foil sheeting and metal implements were also rinsed sequentially with
acetone and methylene chloride before use. Highest purity reagents such as
hydrochloric acid, anhydrous sodium sulfate, coarse sand, sodium hydroxide,
silica gel, and glass wool were extracted with methylene chloride before use.
Solvents employed in this study were the highest purity obtainable from
Burdick and Jackson Laboratories, Inc., or Mallinckrodt Chemical Works.
They were employed without further purification because they gave no meas-
urable residues in procedural blank analyses. Other items are listed as
follows:
Teflon wash bottles, 500 ml
Laboratory scalpels
Homogenizer - Tekmar Tissumizer No. SDT-182EN or Virtis Model 23
Test tube racks - A. H. Thomas Co., Cat. No. 9266-N32
Centrifuge tubes, 40 ml, with screw caps - Corning Glass Works,
Cat. No. 8122
Teflon cap liners - A. H. Thomas Co., Cat. No. 2390H
Centrifuge - International Equipment Company, Model C5
Glass bottles, 1 oz. with screw caps and Teflon liners
Concentrator tubes, 25 ml - Kontes Glass Co., No. K570050,
size 2525
Reflux columns - Kontes Glass Co., Cat. No. K569351, or VWR 1 mm
Ebullators (boiling tubes) - Kontes Glass Co., Cat. No. K569351,
or VWR 1 mm glass tubes, VWR Cat. No. 32829-020 (cut to ca.
2.5 cm length and flame sealed at one end in laboratory)
Tube heater, 6-tube - Kontes Glass Co., Cat. No. K720003
Tube heater control unit - Kontes Glass Co., Cat. No. 720001
Adsorption chromatography columns - Kontes Glass Co., Cat.
No. 42028
Glass (Pyrex) wool - Corning Glass Works, No. 3950
Silica gel, 100-200 mesh - MCB Cat. No. SX0144-06
Copper, fine granular - Mallinckrodt, Cat. No. 4649
Sand, coarse, reagent grade
GC sample vials - Hewlett-Packard, Cat. No. 5080-8712
Reference to a company or product does not imply endorsement by the U. S.
Department of Commerce to the exclusion of others that may be suitable.
35
-------
GC Teflon lined vial caps - Hewlett-Packard Cat. No. 5080-8703
Vial capper - Hewlett-Packard Cat. No. 871-0979
Dish, aluminum, utility, 57 mm diameter
Ether peroxide test paper - EM Laboratories, Inc., Cat. No. 10061-9G
Sediment extraction glass bottle, 1 liter - Scientific Products
Cat. No. B 7573-IL
Ball mill tumbler - Model 8-RA, Scott-Murray, 8511 Roosevelt Way NE,
Seattle WA 98115
Automatic gas chromatograph - Hewlett-Packard Model 5840, dual FID
Automatic GC sampler - Hewlett-Packard Model 7671A
GC columns, 30 m L x 0.25 mm ID, wall coated, glass
capillary (SE-30) - J & W Scientific, P. 0. Box 216,
Orangevale CA 95662
Gas Chromatograph/Mass Spectrometer and Data System, Dual EI/CI -
Finnegan, Model 3200
36
-------
Dry Weight Determination
Sediment. Thaw sediment and remove pebbles by spatula or sieve. Thor-
oughly mix by spatula. Add 10-20 g of the sediment to a tared aluminum dish.
Weigh and record the weight of dish and sample. Cover the dish and sample
loosely with aluminum foil. Dry the sample in an oven at 120°C for 24 hr,
then remove and cool for 30 min in a dessicator. Reweigh and record dried
weight. Calculate percent dry weight as:
weight (final) - weight (tare)
weight (initial) - weight (tare) x
Tissue. Place ca. 3 g clean coarse sand and a glass spatula in an alu-
minum dish and dry overnight in a 120°C oven. Cool the dish in a dessicator
for 30 min. Weigh and record as tare weight.
Weigh into the dish, to the nearest mg, 0.5 g of sample. Using the spatula,
mix the sample thoroughly with the sand, taking care to avoid loss of sand
granules. Dry the sample in a 120°C oven for 24 hr, then remove and cool in
a dessicator for 30 min. Reweigh and record the dried weight. Calculate
percent dry weight as:
weight (final) - weight (tare)
weight (initial) - weight (tare)
Tissue Extraction
Mussels. Pry open the shells with a clean spatula and separate the two
halves by severing the adductor muscle. Scrape the tissue from the shell
into a tared 100 ml beaker for compositing with other individuals.
Snails. Place specimens between several sheets of clean foil and crack
the shells by striking them firmly with a hammer. Remove the shell frag-
ments (with clean forceps or spatulas), peel off the foot and deposit the tis-
sue in a tared 100 ml beaker. The remainder of the procedure is identical
for both molluscs.
Transfer the tissue sample to a homogenizer tube and blend with the homoge-
nizer at medium speed for at least 30 seconds. Return the tissue to the
original pre-tared beaker and weigh to assure that the sample amount is suf-
ficient for the procedure. Weigh 10 g (to nearest 0.1 g) of sample into a
tailed 40 ml screw-capped centrifuge tube. Add 6 ml of 4N sodium hydroxide to
each sample and to one empty tube for a reagent blank. Cap each tube tightly
with a Teflon-lined screw cap, shake vigorously for 1 min, and place each
sample tube in an oven at 30°C for 18 hr (overnight). Cool the samples to
room temperature and shake to check completeness of digestion. If well di-
gested, add 15 ml of peroxide-free diethyl ether, recap tubes tightly, and
shake vigorously for 1 min. Check the caps for tightness, then centrifuge
the tubes at 3000 RPM for 10 min. If the upper ether phase is clear, trans-
fer it with a Pasteur pipet to a 1 oz sample bottle equipped with a Teflon-
lined screw cap. Avoid any carryover of the lower aqueous plase. If the
37
-------
supernatant ether phase is not transparent, see note 1 below before proceed-
ing. Add approximately 0.5 g of anhydrous sodium sulfate to each bottle
without agitation or swirling. Repeat the extraction with 10 ml of ether
and combine the extracts. Cap the bottles tightly, swirl briefly and allow
to settle for 10-15 min. A persistent turbidity indicates the presence of
residual water which must be removed by additional anhydrous sodium sulfate
before proceeding (see note 2 below).
Transfer the dried ether extracts to a 25 ml concentrator tube, attach the
reflux column, and add a micro-ebullator (boiling tube). Place the appara-
tus in the tube heater at 80°-85°C (see note 2). Shroud the apparatus with
aluminum foil to enhance distillation. Concentrate the solution to 2 ml
and remove concentrator tubes from the heater. Add 2 ml of hexane and a
second micro-ebullator, and concentrate to 1.8 ml to completely remove the
ether. If the extract is turbid or viscous, column flow will be restricted.
'Such a sample should be dissolved in methylene chloride and filtered through
a short (1-2 cm) silica gel column with methylene chloride. A bed of silica
gel on a small fritted-glass (coarse) Buchner funnel is suitable. The
methylene chloride solvent in the eluate should then be concentrated and
displaced by hexane. The sample is now ready for microgravimetry and silica
gel chromatography.
Notes:
1. If the emulsion layer is small, remove clear ether layer and pro-
ceed to the second extraction. If the emulsion is extensive, add about 1 g
anhydrous sodium sulfate to the mixture and shake and centrifuge as before.
Transfer the clear supernatant ether phase to the 1 oz bottle and proceed
with the second extraction.
2. Care must be taken to avoid bumping during evaporation to avoid
loss of the sample. Incomplete removal of water is the principal cause of
this bumping, as indicated by the turbidity noted earlier. During evapora-
tion, residual water comes out of solution as a separate phase at the bottom
of the concentrator tube. This phase plugs the ebullator and halts boiling,
leading to overheating and bumping. The remedy is to dry over more anhydrous
sodium sulfate for a longer time and to carefully transfer the ether phase,
avoiding any aqueous phase. In extreme cases, a double-ended ebullator may
be used.
Sediment Extraction
Accurately weigh 100 g of pebble-free sediment into a 1 liter bottle fitted
with a Teflon-lined screw cap. Add 50 ml of 0.1 N hydrochloric acid and
100 ml of ethanol-free, peroxide-free diethyl ether to the sample in the
1 liter bottle. Roll the sample on the ball-mill tumbler for 18 hr (over-
night) . Decant the supernatant ether phase through a glass-wool plug in a
powder funnel into a 500 ml erlenmeyer flask. Add another 100 ml of ether
to the slurry and roll again for 1 hour. Decant the ether extract into the
same flask. Concentrate the extracts to ca. 15 ml by swirling the 500 ml
erlenmeyer in a pan of warm (tap) water in a well-ventilated hood. Transfer
38
-------
the ether extract with washing to a 25 ml concentrator tube, add an ebullator
and a reflux column. Concentrate to 2 ml, then add 2 ml of hexane and a
second ebullator and concentrate to 1.8 ml to remove ether. The extract is
now ready for microgravimetry and silica gel chromatography.
Silica Gel Chromatography
Column Preparation. Prepare columns immediately prior to use. Fill a
column to the flare in the reservoir with methylene chloride. Push a 0.5 cm
glass-wool plug to the bottom of the column with a glass rod. Measure 15 ml
(7 g) of 100-200 mesh silica gel (activated at 150°C for 24 hr, then cooled
in a dessicator) into a 25 ml graduated cylinder and transfer to a 250 ml
erlenmeyer flask. Add 25 ml of methylene chloride and swirl vigorously to
make a slurry. Place a long-stem funnel into the column such that the tip
rests off-center on the bottom of the reservoir just below the surface of
the methylene chloride.
Quickly pour the slurry into the funnel and wash the residual slurry into
the funnel with methylene chloride from a Teflon wash bottle. The adsorb-
ent particles should quickly settle to the bottom of the column with little
turbulence at the settling front. When the settling front extends upward
about 1 cm from the glass-wool plug, slowly open the stopcock to a flow of
1-2 drops per second. Collect the eluate in an erlenmeyer flask to minimize
solvent vapor escape. Swirl the column reservoir gently to wash the parti-
cles into the column. When the settling front reaches the top of the sus-
pended particles, open the stopcock all the way to complete the settling.
Add about a 1 cm layer of clean sand through a funnel to the top of the gel,
followed by an equal amount of anhydrous sodium sulfate.
When the methlene chloride surface is just above the top of the column, add
a ml of petroleum ether with a Pasteur pipet and allow to drain. When the
liquid level again almost reaches the column top, add 40 ml of petroleum
ether and continue to elute. Close the stopcock when the solvent meniscus
almost reaches the top of the column. Discard the rinse elutes. Cover the
column with aluminum foil until use.
Sample Chromatography. The sample extract should be in 1-2 ml of hexane
in the concentrator tube. Crush the ebullator with a glass rod and rinse
the rod with a small amount of petroleum ether. Carefully transfer the
extract solution with a Pasteur pipet to the top of the column and elute.
Never allow the liquid meniscus to go below the upper surface since air will
be entrapped, which will disrupt the column. Rinse the concentrator tube
with 0.5 ml of petroleum ether and add to the column. Open the stopcock and
collect the *eluate in a clean 25 ml concentrator tube. When the meniscus
just reaches the column top, carefully add 15 ml of petroleum ether. Care
must be exercised not to disturb the upper surface of the column during each
addition. When the meniscus again just reaches the sand, add 3 ml of 20%
(V/V) methylene chloride in petroleum ether. Elute solvent at 2-4 ml/min to
separate the saturated from the unsaturated hydrocarbons. When 18 ml has
eluted into the concentrator tube receiver, replace it with a second tube.
This 18 ml eluate, referred to as fraction 1, contains the saturated
39
-------
hydrocarbons. As the meniscus again just reaches the top, add 25 ml of 40%
(V/V) methylene chloride in petroleum ether. This eluate, fraction 2, will
contain the unsaturated and aromatic hydrocarbons. A transparent extract,
when applied to the column, will elute in less than 30 minutes.
Sediment Desulfurization. Silica gel fractions of sediment extract are
treated with activated, fine granular copper to remove elemental sulfur.
Prior to use, activate the copper with concentrated hydrochloric acid (HC1).
Rinse the activated copper five times with acetone to remove the HC1 and
then five times with petroleum ether to remove the acetone. Activated cop-
per should be prepared fresh daily and stored under petroleum ether until
used. Activated copper should not be washed with water or heated. To remove
elemental sulfur from the sample, place the eluate (not more than 1 ml in
volume) in a 40 ml conical centrifuge tube and add about 0.5 ml of activated
copper. Stir for 2 minutes on a vortex mixer. Centrifuge to settle any
sulfide particles in the mixture. Transfer the sample with a Pasteur pipet
to a clean concentrator tube. Rinse the copper once with 1 ml of petroleum
ether and combine the rinse with the eluate sample. Reconcentrate the sample
to a 0.5 ml and continue to microgravimetry and GC analysis.
Microgravimetric Peterminations
The first and second silica gel fractions are weighed on a Cahn microbalance.
In an efficient hood, transfer 25 yl from a known volume of eluate (or ex-
tract) onto the balance pan and allow the solvent to evaporate. Record the
weight and normalize the value to yg/g dry weight of sample.
Gas Chromatography (GC)
GC Sample Preparation. Attach the reflux column to the concentrator
tube containing the eluate from silica gel chromatography. Evaporate the
solvent in the heater block as previously described. After concentrating to
0.5 ml, remove from heat. Add 1.0 ml of internal standard solution (4 ng/yl
hexamethylbenzene in carbon disulfide) and concentrate to 0.5 ml. If neces-
sary, adjust final volume to 0.5 ml with carbon disulfide. Transfer the sam-
ples to the GC vials and crimp on the Teflon-lined septum caps. Replace the
cap each time it is pierced by a syringe to avoid evaporative losses.
GC Apparatus and Modifications. GC analysis is performed on a micropro-
cessor-controlled gas chromatograph (Hewlett-Packard model 5840A) equipped
with: an automatic sample injector (model 7671A); a wall-coated, open tubular
(WCOT) glass capillary column (20-30 m length, 0.25 mm inside diameter); and
a hydrogen flame-ionization detector (FID).
The GC sample injection port is modified to split the carrier gas as shown in
Figure 1. Inlet carrier gas (helium) pressure is adjusted to provide 2 ml/
min flow through the column at 60°C, as determined on a bubble flow-meter.
By adjusting the needle valve to allow 20 ml/min bypass flow, a split ratio
of 10:1 is obtained. Although 90% of the injected sample is sacrificed, the
inlet system is rapidly purged of injected solvent and sample. This
40
-------
maintains sharp solvent and sample peaks. This inlet system (Fig. 1) fea-
tures low dead volume and a glass inlet liner that is readily removable for
cleaning. The inlet end of the glass capillary column must be positioned
inside the glass liner near the location of the inserted sampling needle tip
to gain best sample transfer to the column with the least GC peak broaden-
ing. A charcoal trap absorbs compounds from the vented split stream which
avoids contaminating the needle valve.
Because of the low, carrier gas flow through capillary GC columns, it is
necessary to add make-up gas at the FID (Fig. 1). The flame jet has been
flared to allow the GC column outlet end to be inserted about 2 cm into the
jet. This effectively eliminates any potential dead volume effects with the
make-up gas @iO ml/min) plus hydrogen (24 ml/min.) rapidly sweeping eluted
compounds directly into the flame.
GC Sample Analysis. Analysis is carried out according to conditions
listed in Table 1. GC samples in crimp-sealed, septum-capped vials are
loaded into the automatic sampler. Then the desired operating conditions
(Table 1) are programmed into the microprocessor memory. A sample volume of
2 ul are injected per analysis with the column temperature held at 60°C.
After 10 min, the column temperature is programmed at 2° or 4°C/min to 250°C
and held for 30 minutes. Depending on the program rate, the compounds of
interest are eluted in 1% to 2% hours. Separated compounds are detected by
the FID as they emerge from the GC column. The gas chromatogram is con-
structed by the microprocessor, which prints compound retention times along-
side each peak.
Peak areas are automatically computed using "valley to valley1' mode base-
line correction. Areas are printed in tabular form at the end of the GC run
according to retention times. The quantities of compounds represented by
the peak areas are also computed automatically by ratio of the individual
peak areas to the area of the known amount of internal standard peak. If
reference samples are available for compounds of interest, relative response
factors for these compounds with respect to the internal standard should be
determined experimentally under identical conditions.
Gas Chromatography/Mass Spectrometry (GC/MS). The identity and relative
abundance of compounds detected and measured by GC are periodically confirmed
by GC/MS analysis. A capillary column similar to that used in GC analysis is
employed in conjunction with a Grob sample inlet system. Effluent from the
GC column is fed directly into ion source. Table 2 lists analysis conditions.
A sample of 1-2 yl is injected into the GC/MS while the ion source filament
and electron multiplier voltage are turned off. Passage of the solvent peak
from GC to MS is noted on the instrument high vacuum gage as a transient rise
and fall in pressure. After this, the source filament and multiplier voltage
are restored to normal settings and data acquisition by the computers is ini-
tiated for mass scans every 2 sec. The GC column is subjected to virtually
the same analytical parameters for the GC/MS confirmation run as in the GC
detection and measurement run. At the end of the run, the chromatogram is
reconstructed (RGC) from the total ion current of each individual scan.
Specific ion chromatograms featuring ion abundancies of ions characteristic of
a particular molecular configuration may also be produced. Primarily,
41
-------
Table A-l
Gas Chromatography Conditions
Column
type
Column:
Liquid phase:
Film thickness:
30 m x 0.25 mm ID wall-
coated glass capillary
SE-30 GC (dimethyl siloxane
polymer)
-4
4-5x10 mm
Gases
Inlet
Detector
Carrier gas:
Split ratio:
Column flow:
Bypass flow:
Makeup (
Air
Hydrogen
He
10:1
2 ml/min
20 ml/min
(bypass:column)
30 ml/min
240 ml/min
24 ml/min
Temperatures
Initial Temp:
Program delay:
Program rate:
Final temp:
Injector:
Detector:
60°C
10 min
2 or 4°C/min
250°C
250°C
300°C
42
-------
compounds shown to be present in the GC/MS chromatogram are identified by com-
paring their mass spectrum (background subtracted) with standard reference
tables of mass spectra or laboratory spectra of reference compounds.
Table A-2
GC/MS Analysis Parameters
GC: Same as Table A-l, except no make-up gas
GC/MS interface temp.: 250°
MS:
Filament emission: 500 uA
Electron multiplier voltage: 1600 V
Electron energy: 70 eV
Data acquisition:
Mass range: 80-280 (aromatic samples)
50-300 (alkane samples)
Integration time: 6 msec/scan
Scan time: 2 sec
43
-------
Needle
valve
Charcoal
trap
l/ie'stainless
steel ~
Silver Solder
Septum
Washer
Notch
Injector block
He (22ml/min)
Silanized glass wool
Glass liner
Viton 0-ring
ii H
1/4-1/16 Reducer
I mm id. Graphite ferrule
Collector
air (240ml/min)
Flame jet
N2 makeup
30ml/min)
Figure A-l. Schematic details of the GC sample train:injector, column and
detector.
44
-------
APPENDIX B
DETAILED MAPS OF RECOMMENDED SAMPLING AREAS
(Intertidal sampling areas found suitable from aerial survey, August 9, 1976)
Relatively unpolluted
Code Area
a. Western Strait
b. Central Strait
c. Eastern Strait
d. Whidbey Island
e. Fidalgo Island
f. West Lummi Bay
g. Birch Bay
h. San Juan Island
Site Page
Chito Beach or Kydaka Point B-3
Pillar Point, Agate Bay or B-3,4
Freshwater Bay
Dungeness Bay or Jamestown B-4
One of several beaches from B-5
Partridge Point north to the
Naval Air Station
Telegraph Bight or Langley Bay B-5
Sandy Point B-6
Birch Point B-6
Cattle Point or False Bay B-7
Relatively polluted
i. Port Angeles
j . Fidalgo Island
k. Central Strait
Peabody Creek or Morse Creek
Shannon Point or Green Point
Crescent Bay
B-8
B-8
B-4
45
-------
X-sr;: / ••.:'"""
Northern Puget Sound and Strait of Juan de Fuca
-------
a. Western Strait: Chito Beach or Kydaka Point
b. Central Strait: Pillar Point
47
-------
.Maslj
.|-«ed
GO
b. Central Strait: Agate Bay or Freshwater Bay
k. Central Strait: Crescent Bay
-------
DUNGENES3 L. H.
r*
c. Eastern Strait: Dungeness Bay or Jamestown
-------
•L
d. Whidbey Island
e. Fidalgo Island: Telegraph Bight or Langley Bay
50
-------
f. West Lummi Bay: Sandy Point
g. Birch Bay:* Birch Point
31
51
-------
h. San Juan Island: Cattle Point or False Bay
29
20
j. Fidalgo Island: Shannon Point or Green Point
52
-------
Ul
ANGELES HARBOR
i. Port Angeles: Peabody Creek or Morse Creek
-------
# 799-033
------- |