Reconnaissance Field Study For Verification Of Equilibrium Partitioning: Nonpolar Hydrophobic Organic Chemicals


Untied State*	Office of Water	aan

Environmental Protection	Regulation} and Standard*	JNOVentDet" 1987

Afiency	Criteria and Standard* Diviiior.	SCD# 15

Washington DC 20460			

Water

w CDA'

\7

RECONNAISSANCE FIELD STUDY FOR VERIFICATION
OF EQUILIBRIUM PARTITIONING: NONPOLAR
HYDROPHOBIC ORGANIC CHEMICALS

-------
gppi os"/7

RECONNAISSANCE FIELD STUDY FOR VERIFICATION
OF EQUILIBRIUM PARTITIONING: NONPOLAR
HYDROPHOBIC ORGANIC CHEMICALS

Work Assignment 77, Task 13

November 1987

Prepared by:

Roger D. Kadeg and Spyros P. Pavlou
Envirosphere Co., Inc.-
Bellevue, Washington

For:

U.S. Environmental Protection Agency
Criteria and Standards Division
Washington, D.C.

Submitted by:

BATTELLE

Washington Environmental Program Office
Washington, D-.C.

-------
ABSTRACT

A reconnaissancs field study was performed to test methodology for
interstitial water sampling and analytical testing, and to provide
preliminary data on field-determined sediment/water partitioning
coefficients for representative nonpolar hydrophobic organic
contaminants. The objectives were to determine appropriate sampling
and analysis procedures and to ascertain the validity of the
equilibrium partitioning approach towards sediment quality criteria
development as applied to field sediments and associated interstitial
water. Elliott 8ay in Puget Sound, adjacent to Seattle, Washington,
was selected as the study area.

A box corer and custom pressurized interstitial water extractor were
tested and proved satisfactory, although the extractor required
modifications to reduce plugging. Approximately 400-800 ml of
interstitial water were extracted from 0.3 to 0.5 ft' of sediment.
Standard analytical solvent extraction and gel permeation
chromatography c-leanup techniques were used, followed by standard
Environmental Protection Agency gas chromatography/mass spectrometry
(1625) and gas chromatography/electron capture detector (608) analysis
with additional extract preconcentration. Detection limits of
approximately 0.05-0.4 ppb in interstitial water and 4-17.3 ppb in
sediment for the selected hydrophobic contaminants were achieved.

Samples were collected at 12 stations and analyses performed for
6 stations representing four general areas. Results on supporting
parameters of pH, salinity/conductivity, and temperature showed little
variability and were representative of oreviously reported values for
Elliott Bay. Distribution of hydrophobic contaminants in the sediment
among locations agreed with previously reported spatial patterns,
although levels of contamination were generally lower. Mo relationship
between sediment total organic carbon and interstitial water dissolved
organic carbon could be demonstrated.

7120a

-------
Field organic carbon-normalized Dartitioning coefficients (K ) were

oc

determined for naphthalene, phenanthrene, pyrene, anthracene, and
fluoranthene. The log Kqc values are, respectively, 4.8, 5.2, 5.7,
4.8, and 5.3. These values are in agreement, within uncertainty, of
previously predicted values as well as other field study results
obtained for Puget Sound marine waters. This preliminary investigation

provides support for the application of the equilibrium partitioning
approach in developing sediment quality criteria.

7120a

11 i

-------
CONTENTS

Page

ABSTRACT	i i

1.0 INTRODUCTION		1.1

1.1	BACKGROUND AND RATIONALE		1.1

1.2	OBJECTIVES AND APPROACH		1.3

2.0 FIELD METHODS . . . .		2.1

2.1	.SITE SELECTION		2.1

2.2	VESSEL SELECTION 		2.8

2.3	SEDIMENT SAMPLING AND SAMPLE PROCESSING 		2.9

2.3.1	Box Corer			2.9

2.3.2	Station Positioning 		2.12

2.3.3	Sample Processing 			2.12

2.3.4	Pressure Extraction Equipment . •		2.14

2.3.5	Safety Plan		2.16

3.0 ANALYTICAL METHODS 		3.1

3.1	SAMPLE STORAGE AND EQUIPMENT PREPARATION 		3.1

3.2	DEWATERED SEDIMENT EXTRACTION		 . . .	3.1

3.3	INTERSTITIAL WATER EXTRACTION 		3.4

3.4	GEL PERMEATION CHROMATOGRAPHY OF SEDIMENT EXTRACTS . .	3.4

3.5	GAS CHROMATOGRAPHY/MASS SPECTROMETRY ANALYSIS ....	3.5

3.6	GAS CHROMATOGRAPHY/ELECTRON CAPTURE DETECTOR ANALYSIS	3.6

4.0 DISCUSSION AND RESULTS			4.1

4.1	EFFECTIVENESS OF SAMPLING PLAN		4.1

4.2	PERFORMANCE OF EQUIPMENT AND ONBOARD INSTRUMENTATION .	4.1

4.3	SAMPLE ARCHIVING 		4.3

4.4	MEASUREMENT OF ANCILLARY PARAMETERS . 			4.4

4.5	CHEMICAL ANALYSIS			4.4

4.6	FIELD-DETERMINED PARTITION COEFFICIENTS 		4.11

5.0 CONCLUSIONS AMD RECOMMENDATIONS 		5.1

5.1	SUMMARY AND CONCLUSIONS		5.1

5.2	RECOMMENDATIONS		5.3

5.2 1 Sampling Equipment		5.3

5.2.2 Sample Processing 		5.4

5.2.J Sample Analysis/Analytical Methods		5.5

6.0 REFERENCES		6.1

APPENDIX A - ORIGINAL DATA			A.l

APPENDIX B - HEALTH AND SAFETY PLAN		B.l

7120a

iv

-------
FIGURES

Page

2.1a Sediment Sampling Stations Locations 	 2.5

2.1b Sediment Sampling Stations Locations 	 2.6

2.2	Sediment Box Core Sampler	 2.10

2.3	Pressurized Interstitial Water Extractor	 2.15

4.1 Comparison of Field Koc and Predicted Koc Ranges . . . 4.21

7120a

v

-------
TABLES

Page

2.1	Selected Contaminants for Analysis 		2.2

2.2	Sediment Sampling Stations Locations 		2.7

3.1 Sample Surrogates 		3.2

4.1	Summary of Sampling Activity 		4.2

4.2	SupDortinq Parameter Data From Overlying Bottom Water .	4.5

4.3	Summary of Chemical Analyses 		4.7

4.4	Concentrations of Detected Contaminants in Extracted
Sediment		 4.8

4.5	Concentrations of Detected Contaminants in Extracted
Interstitial Water 	 4.9

4.6	Summary of Field-Determined Sediment/Water

Partition Coefficients for Naphthalene 	 4.12

4.7	Summary of Field-Determined Sediment/Water

Partition Coefficients for Phenanthrene 	 4.13

4.8	Summary of Field-Determined Sediment/Water

Partition Coefficients For Pyrene 	 	 4.14

4.9	Summary of Field-Determined Sediment/Water

Partition Coefficients for Anthracene 	 4.15

4.10	Summary of Field-Determined Sediment/Water

Partition Coefficients for Fluoranthene 	 4.16

4.11	Comparison of Field-Determined and Predicted
Partition Coefficients for Selected Nonpolar

Hydrophobic Chemicals 	 4.18

4.12	Comparison of Field-Measured Koc and

Predicted Koc Distributions .... 	 4.20

4.13	Comparison of Field-Measured PAH Koc Values 	 4.22

7120a

vi

-------
1.0 INTRODUCTION

1.1 BACKGROUND AND RATIONALE

This report is the third in a series of documents prepared by
Envirosphere Company for the Environmental Protection Agency (EPA)
Criteria and Standards Division in support of the agency s program to
develop sediment quality criteria. The report presents the results of
a reconnaissance field study to test methodology for interstitial water
sampling, to perform analytical testing, and to obtain preliminary data
on field-determined partitioning coefficients for selected nonpolar
hydrophobic organic chemicals. Field partitioning data will be used to
verify the theoretical basis for the equilibrium partitioning (EP)
approach towards criteria development.

Because the field verification is intended to validate the theoretical
basis of the EP approach, a summary of the basic elements of the EP
approach is provided below.

• The concentration of contaminant at the water/sediment interface
is assumed to be at equilibrium, and therefore can express
partitioning by the equation

Cs

K =	(1)

T~
w

concentration of the contaminant in the sediment
concentration of the contaminant in the interstitial water
the equilibrium distribution constant or sediment/water
partition coefficient.

where C$ =

Cw =
K =

7120a

1.1

-------
•	Bioavailable form C is the "free" concentration of a

w

contaminant in the aqueous pnase, i.e., the concentration of the
contaminant in the interstitial water that is not bound to liqands
in the water such as dissolved organic carbons.

•	Sorption is controlled by the physical chemical properties of the
sediment and chemical composition of the aqueous phase.

•	Proper normalization of contaminant concentration is required to
reflect the specific sorptive characteristics of the sediment.

The EP approach is used to develop sediment criteria for nonpolar
organic compounds by modifying Equation (1) as follows:

By setting Cw = CWgr in Equation (1), the criterion can be
expressed as

CSQC = KCWQC

(2)

where K
C

WQC
'SQC

= Kqc for nonpolar organic contaminants
= contaminant chronic water quality criteria concentration
= sediment quality criteria concentration (permissible
contaminant concentration value)

In the previous reports (Kadeg, Pavlou, and Duxbury 1986, Pavlou et al.
1987), organic carbon-normalized sediment/water partition coefficients

(Koc) were used with existing chemical water quality criteria to
determine permissible sediment contaminant concentrations (PCCs) for
different nonpolar contaminants. The KQC values were oredicted from
empirically-determined regression equations utilizing octanol-water

partition coefficients (K ) or were 1aboratory-measured K values

ow	oc

reported in the literature. These KQC values must be verified bv
both laboratory and field studies to confirm the validity of the EP
approach in the sediment criteria development process. This
verification can be achieved by computing K from-direct measurements of

7120a

1.2

-------
the concentration of the contaminant in the sediment and its
corresponding concentration in the interstitial water. The K is
subsequently normalized to the measured sediment organic carbon content
to compute	The field-determined Kqc values may then be

compared with the previously predicted or laboratory-determined KQC
values.

1.2 OBJECTIVES AND APPROACH

The objectives of this study are to determine the validity of the
equilibrium partitioning approach as applied to field sediments and
interstitial water samples and, because this study is at the
reconnaissance level, to determine the appropriate sampling and
analysis methods. The following steps outline the approach taken to
meet these objectives:

•	Determine the concentration of nonpolar hydrophobic contaminants
in sediment and associated interstitial water samples, and
determine the total -organic carbon (TOC) of sediment. Using these
data, calculate the field Kqc values for each contaminant.

•	Apply existing and/or develop new sampling and analytical methods
for measuring these contaminants in sediments and interstitial
water.

•	Compare preliminary field Kqc values for each contaminant with
the previously reported values.

•	Obtain information on the suitability of the selected sampling and
analytical methods, and how sampling and analysis can be improved.

•	Based on the above results, provide recommendations for an
intensive field verificatior, program.

7120a

1.3

-------
2.0 FIELD METHODS

This chapter describes the rationale for the selection of the sites,
vessel, and sampling equipment, and specific methods for sample
extraction and processing.

2.1 SITE SELECTION

The study area and sampling sites within the study area were selected
based on the following criteria. For the study area the criteria wore:

•	typical coastal and/or estuarine environment receiving chronic
contaminant input

•	known sediment contamination with nonpolar hydrophobic compounds

t extensive historical data

•	proximity to the survey team

•	availability of selected sampling equipment and sampling vessel.
For specific sampling sites within the area the criteria were:

t proximity to contaminant sources

t historical data for the site, particularly for the trace organic
constituents listed in Table 2.1

t a gradient or variety of sediment textures and availability of
ancillary sediment chemistry data, especially total organic carbon
as well as pH and salinity

7121a

2.1

-------
TABLE 2.1. Selected Contaminants for Analysis

CAS No.	Compound

78-59-1

Isophorone

91-20-3

Naphthalene

91-58-7

2-Chloronaphthalene

208-96-8

Acenaphthylene

83-32-9

Acenaphthene

86-73-7

Fluorene

85-01-8

Phenanthrene

120-12-7

Anthracene

206-44-0

Fluoranthene

129-00-0

"Pyrene

56-55-3

Benzo(a)anthracene

213-01-9

Chrysene

205-99-2

Benzo(b)fluoranthene

207-08-9

Benzo(k)fluoranthene

50-32-8

Benzo(a)pyrene

53-70-3

Dibenzo(a,h)anthracene

193-39-5

Indeno(1,2,3-cd)pyrene

53-70-3

8enzo(ghi)perylene

91-57-6

2-Methylnaphthal ene

132-64-9

Dibenzofuran

319-84-6

A1pha-BHC

319-85-7

Beta-BHC

319-86-8

Delta-BHC

58-89-9

Gamma-BHC (Lindane)

76-44-8

Heptachlor

309-00-2

Aldrin

1024-57-3

Heptachlor Epoxide

959-98-8

Endosulfan I

60-57-1

Dieldrin

72-55-9

4,41-ODE

72-20-8

Endrin

33212-65-9

Endosulfan II

72-54-8

4,4'-DDD

1031-07-8

Endosulfan Sulfate

50-29-3

4,4'-DDT

72-43-5

Methoxychlor

53494-70-5

Endrin Ketone

57-74-9

Chlordane

8001-35-2

Toxaphene

12674-11-2

Aroclor-1016

11104-28-2

Aroclor-1221

11141-16-5

Aroclor-1232

53469-21-9

Aroclor-1242

12672-29-6

Aroclor-1248

11097-69-1

Aroclor-1254

11096-82-5

Aroclor-1260

7121a

2.2

-------
• level of contamination in the sediments is high $nouqh to ensure
that the corresponding concentrations in the interstitial water
could be detected with state-of-the-art analytical methods.

Puget Sound, Washington, was selected as the best area for this
reconnaissance study based on the above selection criteria. Puqet
Sound contains s.everal major estuary embayment systems which have
received chronic contaminant inputs in areas adjacent to metropolitan

sites such as Everett, Seattle, and Tacoma, Washington. There is
demonstrated sediment contamination with nonpolar hydrophobic
compounds. Through a review of the literature and ongoing work in the
area, and through discussions, with agencies and organizations currently
conducting studies in Puget Sound, several key investigations
containing relevant data were identified. These data sources
included: Pavlou and Dexter (1977), Dexter (1976), Pavlou et al.
(1973), Krogslund (1976), Pavlou et al. (1977),'Mai.ins et al. (1980),
Dexter et al. (1981), Konasewich et al. (1982), Romberg et al. (1984),
Stober and Chew (1984), and Tetra Tech, Inc. (1985). All members of
the survey team, comprised of Envirosphere personnel, live in the
greater Everett/Seattle/Tacoma area adjacent to Puget Sound. Finally;
all sampling equipment and the vessel were readily available through
use of local vessel contractors, the University of Washington
Department of Oceanography, and Envirosphere's Bellevue Regional Office.

From the historical information it became immediately apparent that two
embayments had the most extensive! database and were best suited for
specific sampling sites. These embayments were Commencement Bay,
adjacent to Tacoma, Washington, and Elliott Bay, adjacent to Seattle,
Washington. The most extensive data for the nonpolar hydrophobic
compounds in Table 2.1 were presented in the Tetra Tech (1985) and
Metro investigations (Romberg et al. 1984). These data were used to
select the station locations for this study.

7121a

2.3

-------
Of over 90 stations evaluated from the Tetra Tech and Metro databases,
approximately 55 percent were in Commencement Bay and 45 percent were
in Elliott Bay and adjacent areas. However, 15 of the 24 stations
indicating the highest potential for detecting a broad range of
compounds were located in Elliott Bay. These stations covered a wide
spatial distribution. The areas of highest contamination in
Commencement Bay appeared to occur for the most part in the rather
constricted waterways (e.g., City Waterway and Hybelos Waterway) which
are not representative of a typical marine embayment. On this basis,
Elliott Bay was selected as the sampling area. Figures 2.1a and 2.1b
illustrate the locations of the stations that were sampled during the
field survey and Table 2.2 provides details on these locations and
cross references to station identification numbers from previous
investigations.

To determine which sites should have sufficiently elevated interstitial
water concentrations to allow detection of the target compounds in the
interstitial water, the following procedure was used. Using the
equilibrium partitioning relationship shown previously (Equation [1]),
the interstitial water concentration, Cw, was set at the EPA method
detection limit (MDL) for the compound of interest. The corresponding
Kqc value for the compound was assigned using the data in Kadeg,

Pavlou, and Duxbury (1986). Equation (1) was solved for C , the
minimum sediment concentration required for detection of the compound
in interstitial water. Because the TOC of the sediments was not known,
to express this value in units of gram of contaminant/gram dry
sediment, the KQC values were converted to (dry weight
normalized partition coefficients) using a series of TOC values (0.5,
1.0, 2.0, 3.0, and 5.0 percent). These projected minimum
concentrations were then compared with the measured values in the two
embayments to determine which contaminants would be above detection
limits.

7121a

2.4

-------
"« M m

I m

*
¦»

T.r-"o_

*4 '

«•

C"!

" &

> ^
-:¦* -

#?\.
'«£•

ii- /
*. *

- ~?V:

xr /

*¦ !

" * " S -i§—~

"-.'.•V. ST/A-sr ,: _•¦]•,
;V* I • '„¦

^:F

*« l l ATTntstifH-'itfr®

-;. • •.-~~f==; !t "«- itTip;.i^^p«t;rrn^fesa

ft ^Z*'i sfl^/aa-rtCrT'-rrrr- M «?¦ ¦'i- ¦* • ¦i

:ifS£

i **- *

* P*lCAUT
-------
2.6
-------
01

TABLE 2.2. Sediment Sampling Stations Locations

Date

Col lected

Depth
(feet)

loran Coordinates

Metro/NOAA

Station

Equivalent

Description

3/2S/87

28004.6

42299.7

401603

So. Denny Wy. CSO

3/25/87

115

28003.9

42299.7

401810

So. Denny Wy. CSO

3/25/37

28004.3

42299.8

near 401603

So. Denny Wy. CSO

3/25/87

27991.8

42294.4

S0034

West Duwamish Head

3/26/87

28045. S

42293.5

SO 104

Off West Point

3/26/87

28004.7

42Z99.7

401603

So. Denny Wy. CSO

3/26/87

27993.0

42301.6

S0090

near Piers 54, 56

3/26/87

27977.6

42296.8

near 0149

So. Harbor Is. Duwamish Y

3/27/87

28039.5

42297.7

near S0028

Salmon Bay entrance

3/27/87

28001.3
•

42300.1

A061

off Pier '69

3/27/87

28005.8

42299.3

(Tear S0031

No. Denny Wy. CSO

3/27/87

28004.7

42299.7'

401603

So. Denny Wy. CSO

2.7
-------
2.2 VESSEL SELECTION

Vessels available in the Puget Sound area were screened for this study
based on the following criteria:

• large A-frame boom and winch with sufficient capacity for the
large box corer required for adequate sample volumes

• sufficient deck space and clearance to operate the box corer

• sufficient work space for pressure extraction equipment (including
high pressure gas cylinders) (see Section 2.3.4) and sediment
processing

• storage space for gear, equipment, and collected samples

• experienced skipper, preferably with knowledge of sediment
sampling equipment, and general background in marine sciences

• midsized vessel having sufficient maneuverability in nearshore
areas and yet capable of navigating in adverse weather conditions
often encountered in Puget Sound

• full electronics including radio, ship-to-shore communication
system, Loran, and radar

• running seawater for deck and equipment washdown

• reasonable rental rates

• scheduling availability.

Based on these criteria, the Research Vessel (R/V) Kittewake, which
satisfied all of the above requirements, was selected. The vessel has
a 57-ft beam and was custom built as a research vessel for Puget Sound

7121a

2.8
-------
applications by its skipper/owner Charles Eaton. Mr. Eaton holds a
degree in oceanography and has extensive experience with sediment
sampling equipment.

2.3 SEDIMENT SAMPLING AND SAMPLE PROCESSING

This section describes procedures
processing sediment samples.

2.3.1 Box Corer

and equipment used in collecting and

Based on the review of the sediment sampling methods used in previous
sediment studies in Puget Sound and the large volume of material
required for extracting sufficient quantities of interstitial water,
the best suited sampling device was a large box-type corer.
Specifically, a custom designed corer, maintained by the University of
Washington Department of Oceanography, was selected due to its
versatility, proven reliability, ease of operation, and availability.
Figure 2.2 illustrates the basic components of the corer.

During the survey, the corer was placed on the stern of the vessel and
attached to the A-frame boom and winch. The sampling blade and lever
arm assembly was manually raised and locked in the up or "armed"
position, permitting access to the box mount as shown in Figure 2.2.
The removable stainless steel sampling box, approximately
9 x 11 x 22 in., was slid into place using the box guides and then
attached to the sampling assembly by four pressure bolts, two on each
wide side of the box. The water flaps, which permit free passage of
water during descent, were tied in the up position with a piece of
string. The armed box corer was then raised with the winch and
carefully guided over the stern of the boat and suspended in a fixed
position with the bottom of the frame approximate.y 1 ft above the
water surface. The trigger pins, which hold the movable weighted main
shaft, were removed. Removing the pins allows the movable shaft to
drive the box into the sediment when reaching bottom.

7121a

2.9
-------
FIGURE 2.2. Sediment Box Core Sampler

2.10
-------
When the "clear" signal was given, the winch was released, and the box
corer was released to free-fall to the bottom; driving the stainless
steel sample box into the sediment. The winch was engaged, slowly
raising the corer, which pulled the lever arm, causing the sampling
blade to cut through the sediment and seal an intact sediment core in
the sample box. The movement of the lever arm also broke the string,
causing the water flaps to drop and seal the overlying water in the
sampler. The box corer was then winched back to the surface where the
frame bottom was again positioned approximately 1 ft above the surface,
and the trigger pins reinserted. The excess bottom sediment (mud) was
quickly hosed off the frame and other portions of the corer using
running seawater, taking care not to contaminate the collected core.
The boom was then adjusted and the corer lowered onto the stern of the
boat.

The contents of the corer were visually inspected for signs of an
improper or incomplete sample, e.g., entrained material in the sample
olade (sticks, rocks, plastic, trash, etc.). The sample was discarded
if such materials were found during this inspection. Next, the water
flap was lifted and, using a hose made of inert material, the overlying
water was siphoned off the top of the sediment core. A portion of this
overlying water was collected for analysis of ancillary parameters
and/or archiving.

Next, the sampling blade and lever arm were raised, and the stainless
steel sampling box loosened to place the collected core directly on the
fiberglass working deck for immediate subsampling. Alternatively, a
stainless steel plate was driven between the sampling blade and latched
to the sample box, and the lever arm raised. This stainless steel
sampling box could then be removed, containing an intact sample for
temporary storage prior to subsampling. The sample was visually
inspected again for obvious signs of integrity violation, (e.g.,
washout, trash, bottles, cans) and was discarded if it failed
inspection. Typically, a suitable core length sample was from H to
22 in., the maximum depth of the stainless steel box.

7121a

2.11
-------
2.3.2 Station Positioning

During samoling,. the vessel was anchored over the designated site using
Loran and radar positioning equipment. The Loran coordinates of the
site were recorded in the field notebook. These coordinate's were
periodically monitored, and the vessel repositioned as necessary to
adjus.t for any drift or anchor drag. Surface water at the area of
sample entry was visually inspected for any floating debris or oil
sheen. Samples were not taken if debris or sheen were seen. The corer
was lowered and triggered as described previously. Following sample
retrieval, the corer was raised and the sample box visually inspected.
If there was any indication of integrity failure, the sample was
rejected. A repeated failure to pass either debris, sheen, or sample
integrity inspections at a given anchor point resulted in a
repositioning of the vessel to new coordinates.

2.3.3 Sample Processing

Immediately upon collection, the sediment sample was processed in one
of several ways. If the sample was to be archived, it was placed in
either precleaned amber glass jars or precleaned, 4-in. diameter, 2-ft
length stove pipe tubes sealed with aluminum foil. The sample was
transferred directly from the core to these containers using stainless
steel trowels. The sediment core sample was manually segregated over
depth, placing approximately the upper third layer in one set of
containers, the middle third in another set, and the bottom third in a
third set of containers. Material that had come in contact with the
vessel deck was not used. Under no circumstances was the sediment
stirred, mixed, or otherwise homogenized in a manner which might
disrupt equilibrium conditions or unduly elevate particulates in the
interstitial water. These containers were then placed in storage
coolers on ice until archived. Samples for archiving were typically
placed in tiie freezer within 12 h of collection and frozen to
approximately -lo'C. Appropriate data sheets and chain-of-custody
forms were completed for each sample.

7121a

2.12
-------
If the sediment core sample was to be used for extraction, it was
transferred in a similar manner (using stainless steel trowels) to a
temporary, precleaned covered storage bucket, or placed directly into
one of three pressure extraction devices, which are described in
Section 2.3.4. Extracted interstitial water was collected in
precleaned 2 1 amber glass jars. In addition, 50-75 ml of methylene
chloride (ACS spectroscopic grade or better) was volumetrically added
to the extracted interstitial water as a preservative and to minimize
sample loss to glass walls. The extracted sediment was removed from
the teflon pressure extractors using stainless steel trowels and placed
in 2 1 amber glass jars. Both the extracted water and sediments were
placed -in storage coolers on ice for transfer to the laboratory for
analysis. Appropriate data sheets and chain-of-custody forms were
completed for each sample.

As noted in Section 2.3.1, overlying water samples were also
collected. Immediately upon collection, these samples were analyzed
for temperature, salinity, conductivity, and pH using a Yellow Springs
Instrument Company (YSI) Model 33 SCT meter and an Orion Model 231 pH
meter equipped with a Fischer standard gel-filled combination pH
electrode. These measurements were not made on the extracted
interstitial water due to the small volumes of interstitial water
collected. Subsampling of interstitial water for ancillary
measurements was avoided to provide the largest possible volumes for
total sample analysis for organic contaminants. The data for the
overlying water, however, did provide an indication of the interstitial
water conditions because the overlying water was collected from less
than 5 cm above the sediment water interface. All overlying water
samples were archived by placing in a freezer within 12 h of collection
and freezing to approximately -10*C. Appropriate data sheets and
chain-of-custody forms were completed for these samples.

7121a

2.13
-------
2.3.4 Pressure Extraction Equipment

Interstitial water was extracted from the sediment using customized
teflon pressure extractor equipment designed by Dr. Eric Crecelius of
Battelle Marine Research Laboratory at Sequim, Washington.^

Figure 2.3 is a schematic diagram of the equipment. The device was
operated as described below.

A precleaned, 4-in. diameter 0.45 u glass fiber filter was placed in
the bottom filter holder. The tube, or sidewall, was then positioned
over the filter and seated with a teflon Q-ring. The sediment to be
extracted was added until the tube was from one-fourth to three-fourths
full (approximately 0.5-2 1 of sediment). The tube lid, with a
matching 0-ring, was then placed over the four exterior ready-bolts,
sealing the sediment in the tube. The lid was tightened down using
four wing nuts on the ready-bolts: The lid contains the input line
from a medium pressure inert (nitrogen) gas source, and a safety
pressure release pop-up valve.

To operate the extractor, high purity nitrogen gas was fed from a
two-stage regulator to a bank of control valves, one valve for each
extractor tube. The pressure to this bank could be varied from 0 to
the maximum operating pressure of approximately 80 psiq. After setting
the operating pressure, the control valve was ODened, allowing the gas
to pass through a final line cleanup filter and into the extractor.
The pop-up valve was left open for a few moments to purge the system of
atmospheric gas. The pop-up valve was then closed to pressurize the
system. The high pressure gas was forced through the sediment, causing
the interstitial water to pass through the filter and subsequent teflon
water line in the bottom of the extractor to the sample collection
.jar. The first few milliliters of interstitial water sample were, not
collected, but were used to purge the interstitial water line.

Pressure was adjusted at the two-stage regulate, during the extraction

(a) Telephone conversations with Dr. Eric Crecelius, Battelle Marine
Research Laboratory, Sequim, Washington, September 1986.

7121a

2.14
-------
Teflon tube
body

Inert pressurized
gas

Extracted interstitial water

FIGURE 2.3. Pressurized Interstitial Water Extractor

2.15
-------
process to maximize interstitial water -flow. Maximization of flow was
usually accomplished by starting at a relatively low pressure, e.g.,
20 psig, and subsequently increasing the pressure to the maximum
operating pressure. After several hours, when the interstitial water
ceased to flow, or flowed at a minimal rate, the pressure to the system
was shut off, the pop-up valve was opened to release internal pressure,
and the lid was removed. As stated previously, the extracted sediment
was then carefully removed using a stainless steel trowel and placed in
2 1 or 1 gal sample jars. The equipment was then ready for cleaning
and reuse. A new filter was required for each extraction sequence.

2.3.5 Safety Plan

To satisfy corporate requirements and to ensure good, safe field
practices, a detailed safety plan was written prior to the cruise.

This plan addressed potential chemical exposure and risk as well as
standard safe-boating practices. The plan was reviewed and signed by
all field personnel. The plan also covered potential diving
activities, an option that was not exercised due to the s'atisfactory
operation of the sediment sampling equipment. A copy of the safety
plan is presented in Appendix B. All procedures identified in this
plan were followed during the sampling cruise.

7121a

2.16
-------
3.0 ANALYTICAL METHODS

This chapter describes the analytical methods that were used to
determine the contaminant concentrations in the sediment samples and
associated interstitial water samples. The analysis for trace organic
contaminants was performed by Analytical Resources Inc. (ARI), Seattle,
Washington, and the TOC and DOC measurements were made by AT-Amtest„
Inc., Redmond, Washington. These laboratories were selected based on
their extensive experience in Puget Sound sediment analysis, their
proximity to the -study area, and cost effectiveness.

3.1 SAMPLE STORAGE AND EQUIPMENT PREPARATION

The dewatered sediment samples were delivered to the analytical
laboratory in large mouth 2 1 or 1 gal glass jars. Extracted
interstitial water samples were delivered to the laboratory in 2 1
amber glass jars. Sediment and water samples were stored in the dark
at 4"C from time of receipt at the laboratory until they were
solvent-extracted for analysis. Samples were typically extracted
within 48 h. All samples were extracted within 15 d of collection.

All glassware, sodium sulfate (NaS04), and 1aboratory utensils used
during the extraction were heated to 450"C for 6 h and/or rinsed with
clean methylene chloride (C^C^) prior to use.

3.2 OEWATERED SEDIMENT EXTRACTION

Each dewatered sediment sample was thoroughly stirred in the jar prior
to removing material for analysis. After homogenization, 30-50 g (wet
weight) of sediment was placed in a tared 300 ml beaker with
approximately 50 g of Na2S04. The sample was then covered with
100 ml of solvent (CH2Cl2/acetone [1:1], and spiked with the
base/neutral and Desticide/PC8 surrogates listed in Table 3.1.

7122a

3.1
-------
2

TABLE 3.1. Sample Surrogates

Compound

Amount Added

Sediment

Intersti tial
Water

D5-Phenol (Q.M. = 71)

D4-2-Chloropheno1 (Q.M. = 132)
D4-2-Nitrophenol (Q.M. = 143)

D3-2,4-Dimethyl phenol (Q.M. = 125)
2,3,5,5-D4-pCresol
D3-2,4-Dichlorphenol (Q.M. = 167)
D2-4-Chloro-3-Methylphenol (Q.M. = 109)
D2-2,4,6-Trichlorophenol (Q.M. = 200)
D2-2,4,5-Trichlorophenol (Q.M. = 200)
D3-2,4-Dinitrophenol (Q.M. = 187)
D4-4-flitrophenol (Q.M. = 143)
D2-2-Methyl-4,6-Dinitrophenol (Q.M. = 200)
*C6-Pentachlorophenol (Q.M. = 256)
D8-Bis(2-Chloroethyl)Ether (Q.M. = 101)
1,3-Dichlorobenzene (Q.M. = 152)
D4-1.4 Dichlorobenzene (Q.M. = 152)
D4-l,2-Dichlorobenzene (Q.M. V 152)
D12-B1s(2-Chloroisopropyl)Ether (Q.M. = 131)
*C13-Hexachloroethane (Q.M. = 204)
D5-Nitrobenzene (Q.M. = 128)

06-Isophorone (Q.M. = 88)
D3-l,2,4-Trichlorobenzene (Q.M. = 183)
08-Maphthalene (Q.M. = 136)
0a-l,2,3-Trichlorobenzene (Q.M. = 183)
C4-Hexachlorobutadiene (Q.M. = 231)
D7-2-Chloronaphthalene (Q.M. = 169)
Q8-Acenaphthylene (Q.M. = 160)
C4-Hexach1orocycl opentadiene (Q.M. = 241)
D4-0imethylphthalate (Q.M. = 167)
03-2,6-Dinitrotoluene (Q.M. = 167)
QlO-Acenaphthene (Q.M. = 164)
03-2,4-0initrotoluene (Q.M. = 168)
010-Fluorene (Q.M. = 176)

05-4-Chlorophenylphenyl ether (Q.M. = 209)
Q4-0iethyl phthalate (Q.M. = 153)
D8-8is(2-Chloroethyoxy)Methane (Q.M. = 97)
DIO-Ofphenyl amine (Q.M. = 179)

06-Nitrosodiphenyl amine (Q.M. = 175)
010-1,2-0iphenyl hydrazine (Q.M. = 32)
C6-Hexachlorobenzene (Q.M. 292)
05-4-Bromophenyl Phenylether (Q.M. = 253)
010-Phenanthrene (Q.M. = 188)

010-Anthracene (Q.M. = 188)

80 ug/sample
80 ug/sample
80 ug/sample
80 ug/sample
80 ug/sample
80 ug/sample
80 ug/sample
80 ug/sample
80 ug/sample
80 ug/sample
80 ug/sample
80 ug/sample
80 ug/sample
40 ug/sample
40 ug/sample
40 ug/sample
40 ug/samp1-
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug/samp
40 ug./ samp
40 ug/samp

8
8

ug/samp
ug/samp
8 ug/samp
8 ug/samp
8 ug/samp
8 ug/samp
8 ug/samp
8 ug/samp
8 ug/samp
8 ug/samp
8 ug/samp
8 ug/samp
8 ug/samp
4 Ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp
4 ug/samp

3.2
-------
38

100

102

104

106

108

110

112

114

116

118

TABLE 3.1. Sample Surrogates (Continued)

Compound

Amount Added

Sediment

Intersti ti al
Water

D8-Carbazole (Q.M. = 175)

D4-Di-N-Butyl Phthalate (Q.M. = 153)
DlO-Fluoranthene (Q.M. = 212)

DIO-Pyrene (Q.M. = 212)

D4-Benzyl 3utyl Phthalate (Q.M. = 153)
D8-8enzidine (Q.M. = 192)
012-Benzo(a)Anthracene (Q.M. =240)
D12-Chrysene (Q.M. = 240)
D8-3,3'-DichlorObenzi'Sine (Q.M. = 258)
D4-Bis(2-£thylhexyl) Phthalate (Q.M. = 153)
D4-Di-N-0ctyl Phthalate (Q.M. = 153)
D12-Benzo(b)Fluoranthene (Q.M. = 264)
D12-Benzo(k)Fluoranthene (Q.M.-= 264)
D12-8enzo(A)Pyrene (Q.M. = 264)
D14-Dibenzo(a,h)Anthracene (Q.M. = 292)
D14-Dibenzo(a,h)Anthracene (Q.M. = 292)

Resp. for Indeno(l,2,3-CD)Pyrene
D12-Benzo(qhi)Perylene (Q.M. = 288)

40 ug/sample 4

40 ug/sannle 4

40 ug/sample 4

ug/sample

ug/sample.

ug/sample

ug/samole

ug/sample

ug/samole

ug/sample

40 ug/samole 4 ug/sample

3.3
-------
The sample, solvent, and surrogate mix was then stirred using a 3/4 in
ultrasonic probe (Branson Sonifer, 350 W) at full power using a
50 percent pulse for 3 min. The solvent was decanted through a glass
wool plug into a 300 ml Erlenmeyer flask. The stirring was repeated
twice with fresh solvent mixture. The comDosited extract was then
dried by passing it through a chromatography column containing
approximately 35 g of MaSO^. The dried extract was collected in an
assembled Kuderna-Danish (K-D) concentrator flask attached to a three
ball Snyder column and the solvent volume reduced to approximately 8 ml
on an 80"C water bath. The concentrator tube was removed from the K-D
flask and sample extract placed under a gentle stream of clean, drv
nitrogen and reduced to 2 ml. The extract was then cleaned uo using
gel permeation chromatography (GPC) as described in Section 3.4.

3.3 INTERSTITIAL WATER EXTRACTION

The entire aqueous interstitial water sample (300-500 ml) was placed in
a 2 1 separatory funnel and 60 ml of CH2C12 added. The sample was
spiked with base/neutral and pesticide surrogates (Table 3.1). The
sample was shaken vigorously for about 2 min, the water and CH2C1?
allowed to separate, and the solvent drawn off into a 300 ml Erlenmever
flask containing approximately 20 g of NaS04. The extraction was
repeated twice and the composited extract dried and concentrated to
1 ml using the same method as described in Section 3.2 for sediment
extracts. The sample extracts were stored in the dark at 4°C pending
gas chromatography/mass spectrometry (GC/MS) analysis. Analyses were
typically performed within 3 d of extraction, and in no cases later
than 5 d after extraction.

3.4 GEL PERMEATION CHROMATOGRAPHY OF SEDIMENT EXTRACTS

The sediment extract cleanup was performed to reduce the quantitv of
extraneous contaminants, including sulfur, in the extract. The method
used a chromatography system consisting of a Waters pump, a Reodyne
sample injector valve, a Waters refractive index detector connected to

7122a

3.4
-------
a strip chart recorder, and a Spectrum chromatography column. The

2.5 cm ID column was packed with 40 cm (bed height) BioBeads SX-3
(BioRad Inc.).

The column was eluted with at 5 ml per min and calibrated

with a mixture of corn oil, bis-ethylhexylphthalate, and
pentachlorophenol. The entire sample (approximately 2 ml) was injected
and effluent was collected from 30 to 60 min after injection. The
collected sample from the GPC system (approximately 150 ml) was then
concentrated using a K-D concentrator flask on an 80°C water bath to
approximately 5 ml, followed by sample reduction under a dry nitrogen
stream to 250 ul for GC/MS analysis.

3.5 GAS CHROMATQGRAPHY/MASS SPECTROMETRY ANALYSIS

The sample extracts were analyzed for the PAH subset of contaminants

listed in Table 2.1 using the GC/MS isotopic dilution method after EPA

method 1625 (40 CFR 136 Appendix A). Specific instrument conditions
were as follows:

Gas Chromatograph - Hewlett-Packard 5790 Injector Temp. 200°C

Column - 0.32 cm ID by 30 m DB-5 0.25 u phase thickness (J

and W) column connected directly to MS source
through transfer oven at 260°C

Temperature Program - 40°C (4 min) then 8°C/min ramp to 310*C for 9 min

Mass Spectrometer - Finnigan 4000 with Incos Data System

Mass Range - 35 to 500 AMU
Scan Time - 1 sec

Sample extracts were concentrated to 250 ul by sap"vie reduction under a
dry nitrogen stream, and 1 ul of the reduced extract was injected into
the GC/MS. This sample volume resulted in detection limits in the

7122a

3.5
-------
range of 0.05-0.4 ppb for interstitial water and 4-123 ppb for sediment
depending on sample size and GC/MS resDonse. The interstitial water
detection limits compare favorably to the average detection 1imit of
0.1 ppb reported by Socha and Carpenter (1987).

3.6 GAS CHROMATOGRAPHY/ELECTRON CAPTURE DETECTOR ANALYSIS

A one-tenth volume fraction of all the sediment sample extracts and

selected interstitial water extracts were analyzed for the pesticide
and PCB subset of contaminants on Table 2.1. Analysis was by gas

chromatograDhy/electron capture detector (GC/ECD) analysis after EPA
method 608 (40 CFR 136 Appendix A), following GPC cleanup and before
final concentration. The aliquot was diluted 1:2 with acetone and
cleaned up by passing it through an alumina (Activity III) column
(0.8 x 5 cm). The column was eluted with 10 ml of n-hexane:methylene
chloride (5:1). The 10 ml pesticide/PCB fraction was concentrated to
1 ml using nitrogen blowdown and analyzed on a Hewlett Packard 5890 gas
chromatograph using DB-5 megabore (0.32 mm ID x 15 cm) column with a
phase thickness of 1 urn. The temperature was programmed'from 140°C
(1 min hold) to 250°C at 3.5uC/min with a post temDerature of 270°C for
10 min. The pesticide/PCBs were quantified using a single point
calibration with all standards run immediately before the analysis
runs. For a few samples, pesticide/PCB analyses were performed on a
separately extracted aliquot of interstitial water to which the
base/acid-neutral (BAN) surrogates required for GC/MS PAH analysis were
not incorporated. This procedure eliminated interference of the BAM
PAH surrogates in the GC/ECD analysis. In these cases approximate!-/
10 percent of the total interstitial water was separately extracted',

concentrated, cleaned up with alumina, and analyzed by GC/ECD using'the
methods described above.

7122a

3.6
-------
4.0 DISCUSSION AND RESULTS

This chapter presents the field survey results, a discussion of the
effectiveness of the sampling plan, the performance of the equipment
and onboard instrumentation, and the computational methods used to
determine the field partition coefficients for the nonpolar hydrophobic
chemicals measured in the sediments and interstitial water.

4.1 EFFECTIVENESS OF SAMPLING PLAN

The chosen station locations were accessible and easily located using
the electronic locating equipment. The sediment textures of samples
taken from these sites were within the expected range, i.e., primarily
consolidated silt and clay material characteristic of Elliott Bay. The
sediments collected off the Denny Way combined sewer overflow (CSO)
station contained extraneous material such as glass bottles, PVC pipe,
cans, and plastic bags. The presence of this material necessitated
resampling Stations 03, OS, and 12 to obtain satisfactory sediment
grabs. Resampling was.also required at Station 07 due to'the presence
of an oily sheen at the bottom of the'sediment core indicating an
extremely contaminated sample. The odor of the sample and the presence
nearby of ferry dock pilings suggested that the material was highly
contaminated with creosote residues. The number of stations occupied
and the type and number of samples collected are summarized in Table
4.1.

4.2 PERFORMANCE OF EQUIPMENT AND ONBOARD INSTRUMENTATION

The box corer functioned as anticipated; core depths were approximately
14 to 18 in. The bottom seal plate on the removable stainless steel
box, however, was difficult to hammer into place. Instead of removing
the sample box intact and disassembling it, or subsampling with hand
corers, investigators placed the core sample on a clean portion of the
fiberglass deck and subsampled with stainless steel shovels.

7142a

4.1
-------
TABLE 4.1. Summary of Sampling Activity

Number

Day 1 Day 2 Day 3 Total

Date (3/25/87) (3/26/87) (3/27/87)

Locations
occupi ed

Stations occupied

Sediment cores
collected

Overlying water
samples

Sediment samples
archived

Overlying water
archived lb)

Samples extracted/
shipped for
analysis

Cores shipped
off-site l^

(a) For Day 1 stations: 1, 2, 3, 4
For Day 2 stations: 5, 6, 7, 8
For Day 3 stations: 9, 10, 11, 12

(b) Located at Envirosphere's Offices, Bellevue, WA, includes
composites and individual horizons.

(d) To Dr. I.H. Suffet, Drexel University, Philadelphia, PA.

4 4 4 12

(see footnote a)

6 9 7 22

6 9 4 19

15 15 io. 40

6 9 4 jg

0 3 2

7142a

4.2
-------
During this subsampling, care was taken to minimize disruption of the
sediment and core horizons. The cores taken could readily be divided
into upper third, middle third, and bottom third using this procedure.
A few cores (two or three) were archived by these horizons; the balance
were archived as composites.

The sediment extraction equipment did not work as well as expected.
The equipment (three tubes) was originally reported to be able to
extract approximately 0.5-2 1 per hour of interstitial water, based on
previous operating experience for similar sedimentary material. The
actual rates were significantly lower, typically around 0.1-0.15 1 per
hour, with a maximum rate of about 0.3 1 per hour in the more sandy,
coarse sediment encountered at Station 05. As a consequence, the
collection of interstitial water required excessive time .(greater than
3 h/sample). The sample volumes were only about one-third of the
originally anticipated quantities (approximately 0.4-0.8 1 instead of
the estimated 1.5-2.0 1). The lower extraction rates were attributed
to the packing of the fine mud material on the 0.45u filter, forming a
semi-impermeable plug. The low extraction rates were the'limiting
factor in sample processing and constrained the number of samples
processed for analysis. Reducing the sediment quantity loaded in the
extraction tubes and adjusting the overpressure (initially at 20 psig
and increasing incrementally to a final value of approximately 80 psig)
slightly improved extraction rates. These alterations in procedure
increased the extraction equipment maintenance and required continual
monitoring and changing/reloading of filters. While the basic
extraction procedure is sound and workable, some operational and design
modification, as suggested in Chapter 5, could improve overall
performance and thereby increase the volume of interstitial water
extracted within a reasonable period of time.

4.3 SAMPLE ARCHIVING

Raw sediment samples were archived in precleaned glass jars with teflon
sealed lids. Sediment was transferred directly from the core into the
jars, using stainless steel trowels. An alternate procedure involved

7142a

4.3
-------
filling 2-ft lengths of 3- or 4-in. diameter metal stove pipes with the
sediments. The tubes were precleaned with spectroscopic grade
methylene chloride prior to use. Heavy gauge aluminum foil and duct
tape was used to seal the bottom of the tube, and the tube was placed
upright on the deck and then directly filled with sediment using the
trowels. The top of the tube was also sealed with heavy gauge aluminum
foil and duct tape. The tubes fit horizontally in a standard size ice
chest for storage until freezing. Each tube contained approximately
3-5 1 of sediment. The use of these tubes eliminated concern over
breakage during freezing.

4.4 MEASUREMENT OF ANCILLARY PARAMETERS

Due to the difficulties with the extraction equipment, there was
insufficient interstitial water to measure in situ ancillary
parameters. Therefore, these quantities were measured in a subsample
of the collected overlying water, which should exhibit similar
characteristics for the given parameters because the sample can be
considered as representative of the chemistry of the sediment/water
interface. Table 4.2 presents the results of these measurements.

These values are consistent with those reported from previous studies
(e.g., Romberg et al. 1984) and reflect the characteristic
well-buffered, cold, moderately saline marine water of a typical
industrialized embayment in Paget Sound. The lowest conductivity and
salinity values were measured at Station 09 in the Salmon Bay Canal
area, reflecting the conditions resulting from the freshwater input
from the Lake Washington/Lake Union Ship Canal lock system As the
data illustrate, there is little variation in pH, conductivity/
salinity, and temperature values among the sampling stations.

4.5 CHEMICAL ANALYSIS

Due to the small volumes of interstitial water, only selected sample
were submitted for analysis. The stations and cores analyzed and the

7142a

4.4
-------
TABLE 4.2. Supporting Parameter Data from Overlying Bottom Water

Oepth Conductivity Salinity Temperature^ ^

Station Core Date Tine (m) (units) Umhos) (pot) (*C)

3/25