United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/600/R-93/113a
July 1993
&EPA Chemical Shoreline
Cleaning
Agents
Evaluation of Two
Laboratory Procedures for
Estimating Performance
-------
-------
EPA/600/R-93/113a
July 1993
CHEMICAL SHORELINE CLEANING AGENTS
Evaluation of Two Laboratory Pr xedures for Estimating Performance
John Clayton, Siu-Fai Tsang, Victoria Frank.
Science Applications
by
Paul Marsden, Nellie Chau, and John Harrington
International Corporation
San Diego, California 92121
EPA Contract No. 68-C8-0062
Work Assignment No. 3-48
SAIC Project No. 01-0895-03-1000
Proje :t Officer
Choud iry Sarwar
Risk Reduction Engineering Laboratory
Releases Control Branch
Edison, New Jersey 08837
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAJL PROTECTION AGENCY
CINCINNATI, OHIO 45268
Printed on Recycled Paper
-------
Disclaimer Notice
The information in this document has been funded wholly or in pan by the United States Environmental
Protection Agency under EPA Contract No. 68-C8-0062 to Science Applications International Corporation. It
has been subjected to the Agency's peer and administrative review, and it has been approved for publication as an
EPA document. Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
-------
Foreword
Today's rapidly developing and changing technologies and industrial products and practices frequently
cany with them the increased generation of materials that, if improperly dealt with, can threaten both public
health and the environment The U.S. Environmental Protection Agency is charged by Congress with protecting
the Nation's land, air, and water resources. Under a mandate of national environmental laws, the Agency strives
to formulate and implement actions leading to a compatible balance between human activities and the ability of
natural systems to support and nurture life. These laws direct the EPA to perform research to define our
environmental problems, measure the impacts, and search for solutions.
research, development, and demonstration programs
support of the policies, programs, and regulations of
arV i
The Risk Reduction Engineering Laboratory is responsible for planning, implementing, and managing
:o provide an authoritative, defensible engineering basis in
he EPA with respect to drinking water, wastewater.
pesticides, toxic substances, solid and hazardous was :es, and Superfund-related activities. This publication is one
of the products of that research and provides a vital c ammunication link between the researcher and the user
community.
This report presents data from studies desigred to evaluate characteristics of selected bench-scale test
methods for estimating cleaning performance of chemical agents for removal of oil from substrate surfaces. Such
agents have the potential to be used to remove oil that might strand on shorelines and cause adverse effects to
impacted ecosystems. In order to mitigate the effect of stranded oil with chemical cleaning agents, however, an
on-scene coordinator must have information and an understanding of performance characteristics for available
cleaning agents. Performance of candidate cleaning agents can be estimated on the basis of laboratory testing
procedures that are designed to evaluate performance
Data presented in this report are intended to
of different agents.
assist the U.S. EPA in evaluation of candidate test methods
for estimating performance of cleaning agents. Two i est methods were selected for evaluating performance:
Environment Canada's Inclined Trough test and a Swirling Coupon test developed in this program. Tests with
each method were performed with two substrates (sta nless steel and porcelain tile), two oil types (Prudhoe Bay
crude and Bunker Q, and three commercial cleaning agents that have potential for use on oiled shorelines
(Corexit 9580, Citrikleen XPC, and Corexit 7664). The testing procedures are compared on the basis of (1)
determination of the precision of experimental result! obtained for multiple test runs with each procedure, (2)
costs associated with both acquisition of necessary equipment and the actual conduct of tests, and (3) a non-
quantitative evaluation of the ease of conducting a given test (i.e., how many individual test runs can be
performed in a given period of time, the complexity of performing a given testing procedure, the necessary skill
level required of an operator, and the cost of equipment for a particular testing procedure).
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
111
-------
Abstract
Environmental effects of .crude and refined oil products that could strand on shorelines might be
mitigated by the use of chemical cleaning agents under appropriate conditions. However, short-term, reliable
laboratory tests are required to establish the efficacy or performance of various cleaning agents in order to
evaluate their usefulness as elements for oil spill response strategies. Science Applications International
Corporation (S AIC) was tasked by the Releases Control Branch of the U.S. EPA Risk Reduction Engineering
Laboratory (RCB/RREL) to evaluate laboratory efficacy tests for cleaning performance. Tests selected for
evaluation included Environment Canada's Inclined Trough test and the Swirling Coupon test that was developed
in this program. Test oils used in the study included Prudhoe Bay crude and Bunker C. Chemical agents for the
study included Corexit 9580. Citrikleen XPC. and Corexit 7664. Primary objectives in the study included
evaluation of the precision and repeatability of test results, comparison of performance values obtained with each
test, and estimation of the complexity and costs associated with the conduct of each testing procedure.
Considerations of precision of the test results and costs associated with acquisition of necessary equipment favor
selection of the Inclined Trough test, although related values for the Swirling Coupon test are not substantially
worse. Relative trends in cleaning performance among the different chemical agents and oils used in this study
show considerable similarities among the testing procedures and substrates evaluated (i.e.. stainless steel and
porcelain tile). However, absolute values for cleaning performance do differ. Absolute values for cleaning
performance are consistently higher with Corexit 9580 and Citrikleen XPC as opposed to Corexit 7664.
This report was submitted in partial fulfillment of EPA Contract No. 68-C8-0062 by Science
Applications International Corporation under the sponsorship of the U.S. Environmental Protection Agency.
This report covers a period from September 1991 to July 1992, and work was completed as of 12 August 1992.
IV
-------
TABLE
3F CONTENTS
Foreword.
Abstract...
Figures
Tables
Page
.111
. iv
SECTION 1: INTRODUCTION
SECTION 2: CONCLUSIONS '.".",
SECTION 3: RECOMMENDATIONS '.
SECTION 4: EXPERIMENTAL APPROACH
DESCRIPTION OF METHODS
Inclined Trough Test.
Swirling Coupon Test
EQUIPMENT AND MATERIALS
EXPERIMENTAL DESIGN
MEASUREMENT OF RELEASED OIL..
SECTION 5: RESULTS AND DISCUSSION ^
EVALUATION OF SPECTROPHOTOMETMC WAVELENGTHS ->2
CLEANING PERFORMANCE FOR TWO OILS IN ALL TESTING PROCEDURES 28
Cleaning Performance Among Testing Procedures 28
Cleaning Performance for Different Cleaning Agents 38
Rankings for Cleaning Performance! for Chemical Agents Among Testing Procedures 38
STATISTICAL ESTIMATION OF PRECISION ASSOCIATED WITH INDIVIDUAL
TESTING PROCEDURES | 33
INCLINED TROUGH TEST: EFFECT OF WASH-WATER FLOW ON PH^ORMANCE
. VH
viii
... 1
...4
.10
.11
.11
.11
.11
.13
.15
:i6
22
RESULTS.
.46
fc. QUALITY CONTROL CHECKS ......J 48
SPECTROPHOTOMETRIC STANDARDS J. .. 48
REPLICATE SPECTROPHOTOMETRIC MEASUREMENTS' OF'SAMPLE EXTRACTS !!!!!!!!!! 48
OIL STANDARDS 1 5 j
CLEANING PERFORMANCE VALUES " ^ANs"FROM TflS ANALYTrcAL
WAVELENGTHS VERSUS INDIVIDUAL WAVELENGTHS 51
METHOD BLANKS IN TESTING PROCEDURES •• • • , ^
DUPLICATE TEST SET MEASUREMENTS !!!!!!!!!!!!! 53
NON-CRITICAL EXPERIMENTAL VARIABLES 53
COMPLETENESS gJD
REPRESENTATIVENESS " 60
COMPARABILITY I ^$0
SECTION ?: REFERENCES I!!!!!!!!!!!!!!!!!!!!!!!!!!!! " 6i
APPENDIX A: STANDARD OPERATING PROCEDURESi (SOPS) 62
APPENDK B; MSDS/INFORMATION - OILS, CHEMICAL CLEANING
SEAWATER
APPENDIX C: OIL STANDARD CURVES
87
___ 104
APPENDIX D: SUMMARY - INCLINED TROUGHJTEST•RESULTS!!!!!!!!!! !!!!!!!!!!!!!!.!!! 113
APPENDIX E: SUMMARY - SWIRLING COUPON TEST RESULTS. 118
APPENDIX F: SUMMARY - METHOD BLANKS..J.... . 124
APPENDIX G: SUMMARY - DUPLICATE TEST SET RESULTS 126
-------
TABLE OF CONTENTS (continued)
APPENDIX H: SUMMARY - TRIPLICATE SPECTROPHOTOMETRIC MEASUREMENTS ON
SAMPLE EXTRACTS •
Easa
,133
VI
-------
Figure
1
T
3
4
5
6
7
8
9
10
11
12
13
.14
15
16
Description
Figures
S3SS.
Cleaning performance for four testing >rotocols with two oils and three cleaning agents .......... 5
Schematic representation of the Inclined Trough test apparatus ............................. .7. .............. 12
Swirling Coupon test apparatus ........... I [[[ 14
Summary of experimental-test design.! [[[ 17
Example of spectrophotometric data for Prudhoe Bay crude oil ............................................. 19
Wavelength scans of two test oils and tihree chemical cleaning agents .................. . ................. 23
Effect of cleaning agents and water extraction on absorbance of oil standards ................... 24-27
Gravimetric oil weights and spectrophotometric absorbances at analytical wavelengths
as functions of weathering time for (a) Bunker C oil and (b) Prudhoe Bay crude oil .......... 29
Oil distributions in extracts of wash water and substrate. Inclined Trough/stainless
steel .......................................... ... I ..................................... . ........ , ................................... 32
Oil distributions in extracts of wash water and substrate. Inclined Trough/porcelain
33
Oil distributions in extracts of wash water and substrate. Swirling Coupon/stainless
steel I 34
Oil distributions in extracts of wash water and substrate. Swirling Coupon/porcelain
tile [ 35
Summary of cleaning performance for chemical agents relative to control samples 37
Cleaning performance for separate testing protocols with two oils and three cleaning
agents { 39
-------
Tables
Table
4
5
6
10
11
12
13
14
15
Description
Page
Summary of Features of Laboratory Methods for Testing Performance of 2
Oil-Cleaning Agents
Information Summary for Shoreline-Cleaning-Agent Performance '-— 6
Testing Procedures
Shoreiine-CIeanmg-Agent Testing Apparatus: Equipment Costs - Single Testing 7
Apparatus
Shoreline-Cleaning-Agent Testing Apparatus: Costs - Single Test Run 8
Oil Concentrations for Oil Calibration Standards ^
Summary: Oil-Distribution Values for Cleaning Performance 31
Summary: Cleaning Performance for Chemical Agents Relative to Control Samples 36
Component of Variance Analysis: Inclined Trough Test with Stainless Steel 42
Component of Variance Analysis: Inclined Trough Test with Porcelain Tile 43
Component of Variance Analysis: Swirling Coupon Test with Stainless Steel 44
Component of Variance Analysis: Swirling Coupon Test with Porcelain Tile 45
Spectrophotomerric Absorbances for Calibration Check Solutions 49
Summary: Triplicate Spectrophotomerric Measurements on Sample Extracts 50
Summary: Method Blanks for Shoreline-Cleaning-Agent Performance Tests 52
Summary: Duplicate Test Set Measurements 54-59
vm
-------
SEi
:TION
INTRODUCTION
Releases of oil into coastal waters may result in the stranding of oil along shorelines. One option for
mitigating effects of this stranded oil is to apply chemical cleaning agents that are designed to facilitate release of
oil from substrate surfaces with washing with water. Chemical agents for cleaning oiled shorelines qan be
included in three categories: (1) non-surfactant-based solvents, (2) chemical dispersants, and (3) surfactant
formulations especially designed to release stranded oil from shoreline substrates (i.e., shoreline-cleaning-
agents). The intended purpose in applications of agents in all three groups is to facilitate mobilization or release
of stranded oil from shoreline surfaces. Depending on die specific circumstances, it is generally desirable that
chemical agents used for shoreline cleaning release oil from shoreline substrate(s) to (offshore) surface waters
where recovery of the oil is accomplished by mechanical procedures such as booming and skimming operations.
In biologically sensitive environments, chemical cleaning agents should neither facilitate dispersion of the treated
oil into the offshore water column nor enhance penetration of the oil further into permeable shoreline substrates.
Cleaning-solvents and shoreline-cleaning-agents (i.e.j groups 1 and 3, respectively) are designed to minimize
dispersion of treated oil as small droplets into associated water columns. In contrast, chemical dispersants not
only will promote dispersion of oil into water (i.e., their intended purpose) but also can produce elevated
concentrations of oil in permeable sediment substrates under appropriate conditions. Hence, use of chemical
dispersants for purposes of cleaning must be done selectively depending on the particular circumstances inherent
to an oiled shoreline. Their use may be appropriate on beaches with low permeability and offshore waters in
which the dispersed oil can be rapidly diluted to non-problematic concentrations. Regardless of the desired fate,
however, the fundamental purpose of a chemical cleaning agent is to promote release of oil from a substrate
surface.
Laboratory tests to evaluate the ability of che tiical agents to promote release of oil from substrate
surfaces need to be developed as well as evaluated. Ifj use of a chemical cleaning agent on an oiled shoreline is
determined to be appropriate at a spill site, a Federal On-Scene Coordinator, (FOSC) can order the application of
an effective cleaning agent with consideration of the specific conditions present at the site. For such an action,
the FOSC must have information regarding the relath e performance capabilities of available chemical agents, for
removing oil from surfaces. Information on the perfo inance capabilities of candidate cleaning agents can be
obtained from standardized laboratory testing procedures intended to evaluate performance of candidate agents.
The Oil Spills Research Program was initiated to satisfy the Oil Pollution Act (OPA) of 1990, which
included monies dedicated to oil spill research. The cjurrent Work Assignment is an element of the research
program that supports the EPA work group concerned with sub-part J (dispersant effectiveness and toxicity) of
the National Contingency Plan (NCP). As a portion of this Work Assignment, SAIC was directed to evaluate!
selected candidate protocols for laboratory testing of performance of chemical agents for removing oil from
surfaces. The work was performed for the Releases Control Branch, Risk Reduction Engineering Laboratory
(RCB/RREL) of EPA. Laboratory tests to evaluate performance of chemical cleaning agents are reviewed by
Clayton (1992). Table 1 is derived from the latter reference and summarizes features for a number of candidate
testing procedures. For the studies described in this report, two of the testing procedures in the table were
selected for evaluation and comparison of test results:
Swirling Coupon test
Environment Canada's Inclined Trough test and the
Primary objectives in the evaluations of the procedures are to determine the precision for test results
related to cleaning performance with each testing procedure, evaluate comparability of results obtained with the
different procedures for selected cleaning agents and oils, and summarize the qualitative ease of conducting each
testing procedure (i.e., how many individual test runs ban be performed in a given period of time, the complexity
of the testing procedure in relation to the required training time and skill level of an operator, and associated
-------
T"
0
a
"~
V)
c
o
<
O)
c
c
CO
0
o
O
"o
0)
o
c
CO
E
0 ^i.
r >
5
w
DO
C
1
o
ir>
'I.
.g;
o
*•*
1
"c
'3
CO
-1
u
l>
"S
==
.c
s
^
§
OO
.c
•5
'1
2
"I
o
u
.22
|
2
o
OO
stainless
°° <
25 w
.^
Q
C=
1
1
5
CO
f
g
2
•g
^^
g
—
s
>S
C3
^
C
'a
o
CN
1
C
s
glass/quartz
__J
'oi
S
U
00 e
0 1
— en
e
c
' i
"ob
o
—
g
o
G
>v
n
So
o
•" ^
^^
in
(N
^^
2
s-
^
ll
^ -o
ll
I 2
§• eo
CS
§«"
j= 2
*•• 4>
S o
=• ^ 5\
-5E-25
00
J"i
>
m
OO
o
'o
'3
<0
C9
>
_?»,
0
agent-to-oil rati
00
s
1
• O
i
"5
1
u
.06
%
I
^0
0\
C5
>-,
i
(v:v; assume oil
_o
.1
1
3=
2
'o
M
O
a £>
_
-------
costs for both necessary equipment and the conduct of individual
the suitability of a testing procedure for use as a routine
tests). All of these objectives have relevance to
laboratory testing method.
-------
SECTION 2
CONCLUSIONS
Tests were performed with two laboratory procedures for evaluating cleaning performance of candidate
agents for removing oil from substrate surfaces: Environment Canada's Inclined Trough test and a Swirling
Coupon test developed in this program. Common factors through all tests that were evaluated for their effects on
cleaning performance included the following:
o test type (Inclined Trough and Swirling Coupon tests),
o substrate type (stainless steel and porcelain tile),
o oil type (Prudhoe Bay crude and Bunker C oils),
o cleaning agent type (Corexit 9580. Citrikleen XPC. Corexit 7664, and "no agent" controls),
o analytical wavelength (340. 370. and 400 nanometer or nmeter absorbance).
o duplicate measurements for particular groups, and
0 water temperature (not a specified variable of interest for these studies, but one that did exhibit only slight
variations).
Statistical analyses of results show that the greatest effects on cleaning performance are introduced by differences
in the types of oil or cleaning agents used in a particular testing procedure. Smaller amounts of the overall
variance for results are attributable to the procedure itself. The latter variabilities, which are indicative of the
precision inherent to a given procedure, are approximately 4-7% and 9-12% for the Inclined Trough and
Swirling Coupon tests, respectively. Differences in analytical wavelengths contributed essentially 0% to the
overall variance in experimental results.
General trends in cleaning performance due to chemical agents for different testing procedures and
substrates are illustrated in Figure 1 for the two test oils (Prudhoe Bay crude and Bunker C) with four testing
protocols (Inclined Trough-stainless steel. Inclined Trough-porcelain tile, Swirling Coupon-stainless steel, and
Swirling Coupon-porcelain tile) and three cleaning agents (Corexit 9580, Citrikleen XPC, and Corexit 7664).
Data for the figure are overall means for particular test-substrate-oil-cleaning agent combinations. As illustrated,
cleaning performance is consistently higher with Corexit 9580 and Citrikleen XPC for both test oils and all
testing procedures. Relative rankings of cleaning performance for the three chemical agents are generally similar
among the testing procedures for the two test oils. For example, general trends in performance values are
Corexit 9580 - Citrikleen XPC > Corexit 7664 for both Prudhoe Bay crude and Bunker C. In contrast,
differences occur in absolute values of cleaning performance among the procedures.
The primary objective of the study is to evaluate the efficacy of the laboratory testing methods for
determining cleaning-performance. Major criteria for assessing the procedures include (1) determination of the
precision of experimental results obtained for multiple test runs with each procedure, (2) comparison of
performance values obtained with each test, and (3) a non-quantitative evaluation of the ease of conducting a
given procedure. Data for these criteria are summarized in Table 2. Detailed breakdowns of estimated costs for
equipment acquisition and conduct of individual tests are presented in Tables 3 and 4, respectively. As indicated
in Table 2, lower (Le., good) values for precision of results are indicated for the Inclined Trough test
Furthermore, costs for acquiring equipment necessary to perform a given test are lower for me Inclined Trough
procedure. Based on results for precision of measurements and costs associated with acquiring necessary
equipment, the Inclined Trough appears to be the more attractive of the two procedures. The requirement for an
orbital shaker table is the primary item responsible for cost differences between the Inclined Trough and Swirling
Coupon procedures. Both procedures show similar values for costs to run an individual test, the number of tests
that can be performed in an 8-hour period, and non-quantitative criteria. The non-quantitative criteria include
the number of tests that can be performed in 8 hours; costs associated with equipment acquisition, conduct of
tests, and waste disposal; and qualitative items such as necessary skill level of an operator and overall complexity
-------
Prudhoe Bay crude
cleaning performance clue to agent (°/
-------
Table 2
Information Summary for Shoreline-Cleaning-Agent Performance Testing Procedures
standard deviation for
oil recovery in fraction complexity of operator
test protocol
inclined trough-stainless
inclined trough-tile
swirling coupon-stainless
swirling coupon-tile
water
4.4%
7.2%
12.0%
10.3%
substrate
3.8%
4.8%
10.4%
8.9%
no. tests/8 hr
24
24
24
24
equip. cost
$305
$305
$1,570
$1,570
cost/run
$32
$32
$32
$32
procedure
low
low
low
low
skill lavai
low
low
low
low
NOTE: standard deviations for water and substrate fractions are from Tables 8,9,10, and 11
NOTE: italicized values for standard deviations are estimates because variances among groups are
heterogeneous by Bartietfs test (see Section 5)
-------
Tables
Shoreline-Cleaning-Agent Testing Apparatus
Equipment Costs - Single Testing Apparatus
General laboratory items:
UV-visible spectrophotome er, each
spectrophotometer cells, 1 :m path, spectrocil, pair
spectrophotometer chart paper, $63/pack, 2 packs
spectrophotometer calibration solutions, each
Calibration weights, Class S, set
stopwatch, each
magnetic stirrer, each
positive displacement pipet
miscellaneous glassware
miscellaneous supplies
Inclined Trough apparatus:
trough, steel/tile, 12 inches
10-mL syringe
support stand
angle indicator for support stand
miscellaneous supplies
tes + supplies, 4 total
total->
long, purchase + machining
total->
Swirling Coupon apparatus:
coupons, steel/tile, 1 x1 inc i, purchase + machining
support rack and clamps foi
shaker table + supplies
miscellaneous supplies
• coupons
total->
$10,000
$300
$125
$100
$550
$115
$170
$1,300
$1,800
$110
$14,570
$40
$5
$150
$10
$100
$305
$20
$250
$1,200
$100
$1,570
-------
Table 4
Shoreline-Cleaning-Agent Testing Apparatus
Costs - Single Test Run
Inclined Trough apparatus:
labor: oil standard curve, 2 analysts, $45/hr, (24 runs per day/2) $30.00
seawater (salt + distilled water), $0.40/L, 20 ml/run $0.01
extraction solvent, 120 mL methylene chloride/run $1 -00
waste water disposal, $1.20/L, i 0 mL $0.02
waste solvent disposal, $6/L, 120 mL $0-75
total-> $32
Swirling Coupon apparatus:
labor: oil standard curve, 2 analysts, $45/hr, (24 runs per day/2) $30.00
seawater (salt + distilled water), ;$0.40/L, 300 mL/run $0-12
extraction solvent, 120 mL methylene chtoride/run $1 -00
waste water disposal, $1.20/L, 0.3 L $0.3.6
waste solvent disposal, $6/L, 120 mL $0-75
total-> $32
-------
of a testing procedure. For the final two non-quanti ative criteria (i.e., necessary skill level of an operator and
overall complexity of a testing procedure), all tests f|3r all procedures were performed by the same qualified
analysts. None of the analysts had performed any of the procedures prior to this program. The analysts had to be
experienced bench-scale chemists with the necessary, experience in conduct of experiments, preparation of
chemistry samples and standards, complete conduct bf residue analyses in sample extracts by UV-visible
absorption spectrophotometry, and performance of all required data reduction and validation. Estimates for the
level of complexity associated with a given procedure are relative and were provided by the analysts who
performed all of the procedures. Factors in the estimates for complexity include the effort for the initial setup of
the complete testing apparatus, the attention to details necessary during the actual conduct of the procedure, the
degree of time and effort required for clean-up of the testing apparatus at the conclusion of a test, and the degree
of effort required to document and package wastes for proper disposal following tests.
On issues relating to analytical measuremei
evaluate appropriate UV-visible wavelengths for the
ts of oil in sample extracts, studies were conducted to
spectrophotometric detection and quantitation of test oils.
Studies were performed with the two test oils: Prudhoe Bay crude and Bunker C. Appropriate analytical
wavelengths were determined to be 340,370, and 400 nanometers (nm). Analytical measurements at
wavelengths less than 340 nm and greater than 400 nm suffered from limitations including lack of linear
response of the spectrophotometer with varying oil concentration, poor reproducibility of absorption values, and
poor measurement sensitivity. Statistical analyses performed with results from the study indicate that no
significant difference exists between measurements of performance results at the three wavelengths. Therefor
average performance values from the three wavelengths are used to evaluate cleaning performance with other
experimental variables.
-------
SECTION?
RECOMMENDATIONS
The following are recommendations based on results and experience gained during the studies summarized in
this report.
o The Inclined Trough test appears to be well suitedfor evaluating cleaning performance of candidate cleaning
agents for removing oil from substrate surfaces. The test is relatively simple and straightforward, requires a
minimum of equipment, generates minimal waste requiring disposal at the conclusion of tests, and is
characterized by relatively good precision for test results. It should be noted, however, that the same
characteristics for the Swirling Coupon test were comparable or only slightly less desirable than those for the
Inclined Trough test.
o Based on observed variabilities among testing procedures and substrates, consideration might be given to
establishing a specified substrate-reference oil-cleaning agent combination that would be included in any
routine laboratory tests for evaluating new cleaning agents. Inclusion of results from a standard substrate-oil-
cleaning agent combination with those for candidate cleaning agents would allow for a degree of calibration
or intercomparability between test results for different agents.
o It is premature to provide a pass/fail criterion for acceptability of cleaning performance in a given testing
procedure. In order to provide such a criterion, additional laboratory studies with the selected procedure
would be required. Such studies should be conducted to establish Cleaning-Performance-Indices for a large
set of chemicals (e.g.. 30-40 in number; including chemicals that are and are not specifically identified as
cleaning agents) using EPA standard reference oils. The distribution of Performance-Indices for a large set of
chemicals would provide a better rationale for establishing pass/fail criteria in a given testing procedure.
o The EPA should provide a source of standard cleaning agents, perhaps through the Environmental
Monitoring Systems Laboratory-Cincinnati (EMSL-CI) Industrial Chemicals Repository. Laboratories could
provide performance data for standard oils and standard cleaning agents to facilitate intercomparison of test
results.
o The effect of non-critical variables (e.g., water and air temperature, weathering state of oils, substrate types,
cleaning agent-to-oi! ratios and "soak" times, and wash-water flow rates and volumes) on Cleaning-
Performance-Indices should be evaluated in greater detail. The studies presented in this report are not
intended to evaluate the broad ranges of possibilities inherent to these variables. However, such information
would be valuable in determing appropriate conditions under which cleaning agents might be used or not in
the field.
-------
SECTION 4
EXPERIMENTAL APPROACH
DESCRIPTION OF METHODS
Two laboratory testing procedures were evaluated for estimating the performance of chemical cleaning
agents to remove oil from substrate surfaces: (1) Environment Canada's Inclined Trough test and (2) a Swirling
Coupon test that was developed in this program. Two substrates (stainless steel and porcelain tile) were
evaluated in each procedure. The testing procedures were selected because of their relative, simplicity and
adaptability for routine testing purposes. Additional attractive features include the well defined nature of
components used for the tests and their ready availability from commercial sources. Standard Operating
O«*s-ut As4i***Afi /O/"^T5«\ f«_ A~._L. l * . » _. r O
in Appendix A.
Procedures (SOPs) for each procedure are presentee
Inclined Trough Test
The general procedure for the Inclined Tro igh test has been developed by Fingas, et al. (1989) at
Environment Canada. The basic apparatus for the test consists of (1) stainless-steel or porcelain-tile troughs
measuring approximately 25-30 cm in length and 2J3 cm wide and having a shallow V-shaped crossrsection, (2)
an adjustable frame capable of holding multiple troughs at angles of 0° and 90° from the horizontal, (3) 5-1.0 cc
syringes (plungers removed) for delivery of wash-water, (4) vials or beakers for collection of water washes, and
(5) a programmable timer. A schematic illustration
apparatus is shown in Figure 2.
of troughs in the frame holder for the Inclined Trough test
For the test, a clean, dry trough is placed in a horizontal position. A 0.150-mL volume of oil is
deposited as a "slick" (ca. 5 cm long) in the bottom Central portion of the trough using a positive-displacement
pipette. The oiled trough is then allowed to stand at kmbient temperature for approximately 18 hours (i.e.,
overnight). The 18-hour oil-to-substrate contact tide (OSCT) is intended to (1) allow for spreading of the oil on
the substrate surface, (2) allow for the formation of an adhesive bond between the oil and substrate, and (3)
introduce a component of evaporative weathering to the oil. Such weathering is a phenomenon that occurs when
oil strands on a shoreline in the real world. After thfe 18-hour contact period, 0.050 mL of a chemical cleaning
agent is applied uniformly to the surface of the oil in the trough. A 10-minute contact, or "soak," time is
observed to allow the agent to penetrate and mix into the oil while the trough is maintained in the horizontal
position. At the end of this cleaning agent-to-oil contact time (AOCT), the trough is transferred to the test frame
shown in Figure 2 and the inclination angle is adjusted to 45°. The wash receiver-vessel is set in place at the
lower end of the trough. A wash-syringe containing 5.0 mL of synthetic seawater is positioned at the upper end
of the trough and the seawater is allowed to How acr yss the treated oil surface at a rate controlled by the exit
aperture of the syringe. For the present studies, this was determined to be 1 mL/second (63 mL/minute),
although limited tests also were performed with a floW rate of 7.5 mL/minute to evaluate the effect of flow rate on
cleaning performance (see Section 5). Following a 10-minute interval after the first seawater wash, the oil on the
trough is washed with a second 5-mL volume of seawater from the syringe. After a final 10-minute interval, the
syringe and receiver-vessel are removed from the apparatus frame. Both the trough and the wash water are
separately extracted with methylene chloride (dichlotomethane or DCM) to recover the oil that remains on the
trough and is released to the wash water, respectively. The DCM extracts for the trough and wash-water
fractions are analyzed spectrophotometrically for oil ^content
Swirling Coupon Test
The basic apparatus for the Swirling Coupon test consists of (1) a variable speed orbital shaker table
with retaining brackets to hold test beakers, (2) an apparatus frame with coupon mounting arms capable of being
raised and lowered for attachn -nt and immersion of coupons during the test, (3) stainless-steel or porcelain-tile
11
-------
Figure 2. Schematic representation of the Inclined Trough test
apparatus.
12
-------
coupons measuring 2.5 cm x 2.5 cm, (4) 400- or 600
the test coupons are immersed, and (5) a programmable
test apparatus is shown in Figure 3.
mL beakers that serve as wash-water containers in which
timer. A schematic illustration of the Swirling Coupon
For the test, a 0.048-mL volume of oil is ev( inly deposited as a "slick" on the surface of a clean, dry
coupon using a positive-displacement pipette. The oiled coupon is maintained in a horizontal position for
approximately 18 hours (i.e., overnight). As with th^ Inclined Trough test, the 18-hour oil-to-substrate contact
time (OSCT) is intended to (1) allow for spreading ohhe oil on the substrate surface, (2) allow for the formation
of an adhesive bond between the oil and substrate, and (3) introduce a component of evaporative weathering to
the oil. After 18 hours, a 0.016-mL volume" of a chemical cleaning agent is applied uniformly to the surface of
the oil on the coupon. The coupon is maintained in i horizontal position for an additional 10-minute contact or
"soak" time to allow for penetration and mixing of the agent into the oil. The coupon is then attached to one of
the mounting arms of the support frame shown in Figure 3. A beaker containing 250 mL of synthetic seawater is
inserted into the retaining bracket on the shaker tablS beneath the mounted coupon. The shaker table is then
turned on at a rate of 150 RPM and an orbital diameter of 2.0 cm. The mounted coupon is lowered into the
seawater in the beaker for two minutes. During this time, the coupon is swirled in the seawater due to the motion
of the shaker table. At the end of the 2-minute swirling period, the shaker table is turned off and the coupon is
withdrawn and allowed to drain over the test beaker.
Both the coupon and the wash water are separately
extracted with methylene chloride (DCM) to recover the oil that remains on the coupon and is released to the
wash water, respectively. The DCM extracts for the coupon and wash-water fractions are analyzed
spectrophotometrically for oil content
EQUIPMENT AND MATERIALS
The following equipment and materials were used in the conduct of all tests for this study.
UV-visible Spectrophotometer. A Hitachi Model U-2000 UV-visible spectrophotometer was used for all
analytical measurements of oil extracts. The instrument is capable of measuring absorbances from 200 to 900
nmeters.
Spectrophotometer Check Standards. Certified spect ophotometric reference standards were analyzed on a daily
basis to monitor relative absorbance accuracy and long-term stability of the instrument The standards consist of
Oxford Spectro-Chek Solutions I, II, III, and IV (Oxford Labware, St Louis, MO). Solutions I and H contain
cobalt ammonium sulfate at concentrations of 32.7 arid 16.3 g/L, respectively. Solutions in and IV contain
potassium chromate at concentrations of 0.04 and 0.02 g/L, respectively. Maximum absorbances for the cobalt
ammonium sulfate (Solutions I and II) and potassium chromate (Solutions HI and IV) are 510 and 370 nm,
respectively.
Synthetic Seawater. A commercial salt recipe for syn Jietic seawater (product name: Instant Ocean;
manufacturer Aquarium Systems, 8141 Tyler Blvd., Mentor, OH) was used (or all tests. The synthetic seawater
solution was prepared by dissolving 68 g of the salt mixture in 2.00 L of distilled water (i.e., for a salinity of 34
parts per thousand or ppt). A table of the ionic composition of the salt mixture in the seawater is presentediih
Appendix B.
Test Oils. Two EPA standard reference oils were used in the studies: (1) Prudhoe Bay crude and (2) Bunker C
Residual (No. 6) fuel oil. These oils were obtained from the Industrial Chemicals Repository, EPA
Environmental Monitoring Systems Laboratory, Cincinnati, OH (James Longbortom, custodian). The oils have
been thoroughly homogenized as well as characterizec physically and chemically for previous EPA studies.
Information sheets for each oil are presented in Appet dix B. For the current studies, the density of each oil was
determined by gravimetric measurements of a known • rolume of the oil.
13
-------
orbital motion for beakers
on shaker table
stationary
support rack
for coupons
Figure 3. Swirling Coupon test apparatus.
14
-------
Chemical-Cleaning-Agent Formulations. Three commercial cleaning agents were used in the studies: (1)
Corexit 9580; (2) Corexit 7664; and (3) Citrikleen XjPC. These chemical agents are obtained from the
Emergencies Science Division, Environment Canada, Ottawa, Ontario, Canada (courtesy of Mervin F. Fingas)
and from Exxon Chemical Company, Houston, Texas (courtesy of Kenneth W. Becker). Two of the agents
(Corexit 9580 and Corexit 7664) are currently listed on the U.S. National Contingency Plan (NCP) Product
Schedule. Product Information sheets for all of the £ gents are presented in Appendix B. The density of each
cleaning agent was determined by gravimetric measirements of a known volume of agent
Extraction Solvent. Pesticide-analysis-quality methjjlene chloride (DCM) was used for extraction of all samples
(i.e., residual substrate and wash-water) as well as all oil-standard-water samples.
EXPERIMENTAL DESIGN
The primary objective of this study was to e/aluate the efficacy of the testing procedures (the Inclined
Trough and Swirling Coupon protocols with stainless steel and porcelain tile substrates) for estimating
performance of chemical cleaning agents for removing oil from substrate surfaces. Efficacy was assessed by
conducting parallel experiments for cleaning performance with the different testing procedures and substrates.
Major criteria for the assessment included (1) determination of the precision of experimental results obtained for
multiple test runs with each test procedure, (2) comparison of performance values obtained with each test, and (3)
a non-quantitative evaluation of the ease of conducting a given test (i.e., how many individual test runs can be
performed in a given period of time, the complexity of performing a given testing procedure, the necessary skill
level required of an operator, and the costs of equipment and individual test runs for a particular testing
procedure). For the purpose of the current report, precision (or repeatability) is defined as the standard deviation
for repeated measurements of cleaning performance with the same testing apparatus and procedure under
operating conditions that were maintained as constart as possible. AU results were obtained in SAIC's testing
laboratory only.
Precision of results with a given testing pnx edure is important. A procedure that is characterized by
better precision in its results (i.e., lower values for standard deviations of replicate measurements) will be better
able to differentiate between cleaning-performance vklues for different candidate chemical agents. Furthermore,
fewer test runs may be required to confirm differentiation between cleaning-performance values for candidate
agents if the testing procedure is characterized by better precision in its test results.
The importance of the non-quantitative criteria identified above for a given test derives in part from the
fact that the conduct of laboratory tests to evaluate performance of candidate cleaning agents can become tinie-
critical in the aftermath of spill events and stranding bf oil on shorelines. As such, the ability to rapidly generate
results from a laboratory testing procedure may become important for mitigation response to a spill. Factors that
can influence the time-critical availability of test results include (1) the number of tests that can be performed in
a specified period of time (e.g., 8 hours), (2) the overall cost (and, hence, availability) of the necessary testing
apparatus, (3) the complexity of the testing procedure, and (4) the skill level required of an operator for the
procedure. The latter two items can be important for both the time required to sufficiently train an operator for a
given procedure as well as the depth of background understanding and knowledge required of that operator to
learn and perform the procedure in an acceptable and reproducible manner.
Relative rankings of cleaning performance a mong testing procedures for a given test oil and different
cleaning agents also is of interest. Variability among
specific differences such as (1) the chemical composi
results with different procedures can be attributable to test-
ion and weathering-state of the test oil, (2) characteristics
of the substrate on which oil is stranded, (3) the method of application and soak-time for the cleaning agent to
the oil, (4) the ratio of cleaning agent to oil, and (5) the degree of flushing or turbulence that characterizes the
wash water in the test procedure. In summary, absolute values for cleaning performance with a given oil and
15
-------
cleaning agent will likely differ among testing procedures due to characteristics inherent to specific testing
procedures.
The sequence of tests conducted in the current study is illustrated in Figure 4. The majority of tests were
to be performed in Test Series 1, which yields a "2 X 2 X 4" matrix (i.e., two oils by two substrates by three
cleaning agents and a "no agent" control) for both the Inclined Trough and Swirling Coupon procedures.
Furthermore, measurements in each of the matrix cells were to be performed in at least triplicate. This yields a
total of 48 experiments for a specified procedure (2 oils X 2 substrates X 4 test agents X 3 replicates per matrix
cell). For the two testing procedures (i.e.. Inclined Trough and Swirling Coupon), a total of 96 separate
experiments are indicated. For Quality Control (QC) purposes, duplicate test sets of data (3 replicates per test
set) were to be collected for two matrix cells for both the Inclined Trough and Swirling Coupon procedures
during Test Series 1. This yields an additional 12 experiments (i.e.. 4 matrix cells X 3 replicates per cell). The
sum of all of these anticipated experiments for Test Series 1 yields 108 separate tests. In addition to the
experiments for Test Series 1. an additional set of experiments were to be performed for Test Series 2 in Figure 4.
In the latter, a single oil/cleaning agent/substrate combination with a cleaning performance greater than 20% in
each testing procedure was to be selected. Prudhoe Bay crude, Corexit 9580. and stainless steel were chosen as
the oil. cleaning agent, and substrate, respectively. For the oil, consideration also was given to the fact that
Prudhoe Bay crude can be released into coastal waters of the U.S. (e.g., the EXXON VALDEZ spill). For
experiments in Test Series 2, seven replicate measurements of cleaning performance were to be performed for
both the Inclined Trough and Swirling Coupon procedures. In summary, the combined measurements from Test
Series 1 and 2 yields a total of 122 separate tests for the complete effort of the laboratoiy program.
MEASUREMENT OF RELEASED OIL
Released oil in experimental samples is measured following extraction of either the wash-water sample
or the trough/coupon with DCM. Quantities of oil in extracts are determined with UV-visible spectrophotometry
by comparing absorbance measurements of experimental-sample with those of external-oil-standard extracts.
The volumes of DCM, the number of extractions, and final extract volumes for samples are presented in the
detailed Standard Operating Procedures (SOPs) for the methods in Appendix A as well as the brief descriptions
of the testing procedures at the beginning of this section. Following determination of appropriate wavelengths
for analytical measurements (see Section 5), spectrophotometric measurements on DCM extracts for all samples
and standards were performed at 340,370, and 400 nm. Initial recommendation for measurements at these
wavelengths was based on the work of Fingas et al. (1987). The latter investigators determined that
spectrophotometric measurements at these wavelengths provided the best quantitative results for oil
concentrations in terms of sensitivity, consistency, and repeatability. Analytical measurements at wavelengths
less than 340 nm and greater than 470 nm suffered from limitations including lack of linear response in the
spectrophotometer with varying oil concentration, poor reproducibility of absorption values, and/or poor
sensitivity of the spectrophotometer. As presented in Section 5, similar conclusions are arrived at in this study.
Calibration of the UV-visible spectrophotometer was accomplished using external standards at the six
concentrations for each test oil that is shown in Table 5:
16
-------
Inclined Trough Test
Swirling Coupon Test
Procurement and Fabrication of Test Equipment
Test Series 1: Performance Tests for All
Oil/Agent/Substrate Combinations
2 oils/2 substrates/3 cleaning
2 sample extracts per test (1
at least 3 replicate tests for e
agents + 1 "No Agent" control;
substrate +1 water);
ach combination
Selection Criteria:
a single oil/substrate/cleaning agent with a
cleaning performance >20% for both procedures
Test Series 2: Performance Tests with Additional
Emphasis Toward Precision
1 oil/1 substrate/1 cleaning agent;
2 sample extracts per test (1 kubstrate + 1 water);
7 replicate tests for the combi
nation
Figure 4. Summary of experimental-test design.
17
-------
Table5
Oil Concentrations for Oil Calibration Standards3
Bunker
C Standard
6
5
4
3
2
1
Prudhoe
Bav Standard
6
5
4
3
2
1
Final Oil
Concentration
1.00 mg/mL
0.50 mg/mL
0.20 mg/mL
0.10 mg/mL
0.05 mg/mL
0.02 mg/mL
0.01 mg/mL
'assuming a nominal oil density of 0.9 g/mL and an extraction efficiency of 100% for the oil from the oil-
water standard
Slightly lower concentrations were used for Bunker C because it is darker and spectrophotometrically more
sensitive than Prudhoe Bay crude. The presence of water and some dispersants in DCM can affect absorbance
readings in a spectrophotometer (Fingas et at, 1990). Therefore, oil-standard calibration curves for quanutauon
of oil in samples were developed in a manner similar to that of experimental water samples. For example, eacn
oil standard was added to 30 mL of seawater. If a cleaning agent was used in an experiment, premixed cleaning
asent-to-oil ratios for the "standard oil" solutions were prepared at ratios equivalent to those in the actual testing
procedure (Le., 1 part cleaning agent to 3 parts oil). The oil-(cleaning agent)-water solutions were extracted with
DCM. (3 x 5-mL volumes) and diluted to a specified volume (20.0 mL) with DCM.
Calibration standards for the UV-visible spectrophotometer for a particular test-set of data were prepared
for each oil and associated cleaning agent (if present) being tested. Response of the spectrophotometer was
determined to be linear by calculating response factors for the standards. Response factors (RFs> for oil-cleaning
agent standards were determined at each of the three analytical wavelengths (i.e., 340,370, and 400 nm). KTS
are calculated as:
RFX = (oil concentration)/Ax ^ '
where RFX = response factor at wavelength x (x=340,370, or 400 nm),
oil concentration = mg of oil/mL of DCM in standard solution, and
Ax = spectrophotometric absorbance at wavelength x (x=340,370, or 400 nm).
RFs had to be less than 20% different ftom the mean RF for those standard concentrations that were below
absorbance saturation and above the detection level of the spectrophotometer (approximately 3.5 and O.OSU
absorbance units, respectively, in this study). Examples of plots of RF values and absorbances ^ the three
analytical wavelengths (340,370, and 400 nmeters) versus oil concentration are shown in Figure 5. As long as
absorbance values were less than 3.5 absorbance units, points for standard curves were always in the linear range
for plots of absorbance versus oil standard concentration and RF values were always within ±20% ot the mean
RF. Examples of data' reduction sheets for oil standards generated in this testing program (including RF values
for individual standards relative to the mean RF) are presented in Appendix C.
Concentrations of oil in final DCM-extracts of wash-water and substrate samples were calculated from
the UV-visible absorbance readings
-------
1.0
"c
I0-8
0
CD
P .-
RF at 400 nm (mg oil/abs unit) RF at 340 nm (r
0 p P P P - -* oooc
o to » 0> o» b lo b KJ *. b
RF linearity: 340 nm absorbance |
•
(3 n
D
™
individual RF* mean_RF memn HFW-20%
0.05 0.10 020 0.50 1.00 2.00
oil standard concentration (mg/mL)
RF linearity: 400 nm absorbance |
n a a
a
a
-
indMdwl RF* moan RF moan RFW-20%
RF at 370 nm (mg oil/abs unit)
Io o o o o -• -»
b (o *. -»l\]MUU.fe
a w o a\ o en b bi b
absorbance saturation
: / //
: //
- I tt s
-/>•""
j4f»' 340jm 37Q.nm 400jm
".I.I.I. L . 1 ^_
0.05 0.10 020 0.50 1.00 2.00
oil standard concentration (mg/mL)
0.0 0.5 1.0 1.5 2.0 2.5 3.0
oil standard concentration (mg/mL)
Figure 5. Example of spectrophotometric data for Prudhoe
Bay crude oil.
19
-------
amount of oil in a test fraction (i.e.. wash-water or substrate extract) was calculated from absorbance
measurements of experimental extracts at analytical wavelengths "x" (Cx):
Cx = (Ax) x (RFX) x (VDCM) (2)
where Cx = total mass of released oil in test vessel at wavelength x (x=340. 370, or 400 nm) ,
Ax = spectrophotometric absorbance in experimental water-sample extract,
RFX = mean response factor for oil at wavelength x (equation 1), and
VDCM = final volume °f DCM-extract of experimental water sample.
Three concentration values for oil in each experimental sample were obtained: €349, C37Q and €400-
As discussed in Section 5. concentration values at these three wavelengths were determined to be not
significantly different from each other. Consequently, mean values for oil from the three wavelengths were used
for most calculations of cleaning performance:
cmean = = 100 X [1 - (CmellrM.OOIIDl.n4litntfe X
20
-------
*nere Pcontroi = fraction of oil (as %) in the wash-water and substrate fractions for the control test Values of
Pcontroi measure the effects of water washing separatb from oil removal due to a specific chemical cleaning agent
In the third step, the cleaning performance due to a cleaning agent alone is calculated as the difference
in the oil values in the wash-water samples with and without a cleaning agent:
3 - P
agent-water control-water
where
= net cleaning performance due to tht chemical cleaning agent.
(5)
21
-------
SECTIONS
RESULTS AND DISCUSSION
EVALUATION OF SPECTROPHOTOMETRIC WAVELENGTHS
Preliminary studies were conducted to not only identify appropriate spectrophotometric wavelengths for
analytical measurements but also determine the effect of initial evaporative weathering of test oils on absorbance
measurements. For analytical wavelengths, studies performed at Environment Canada for evaluating
performance of chemical dispersant agents with oils have identified wavelengths of 340,370, and 400 nm as
appropriate for measurements of oils (Fingas et al., 1987). Oil concentrations are then reported as means of
measurements for the three wavelengths. Other studies report spectrophotometric measurements for oil in
sample extracts at a variety of wavelengths (e.g., 580 nm for the Exxon Beach Washing test, Fiocco, 1991; 620
nm for the Revised Standard EPA test for dispersant performance, 40 CFR Part 300, EPA, 1984). To evaluate
spectrophotometric absorbances at relevant wavelengths for test oils and cleaning agents in this program,
absorbance-scans from 200-700 nm were performed for DCM solutions of the two test oils (Prudhoe Bay crude
and Bunker C) and the three cleaning agents (Corexit 9580, Citrikleen XPC, and Corexit 7664). The results are
presented in Figure 6. Absorbance values on the y-axes are presented in logarithmic format to aid in
distinguishing the traces for the various oils (Fig. 6a) and cleaning agents (Fig. 6b). Note that absorbance values
for all oils and cleaning agents are higher at wavelengths of 340, 370, and 400 nm than at 580 or 620 nm.
Therefore, sensitivity for detection and quantitation of oil in sample extracts will be higher at the three lower
wavelengths (i.e., 340,370, and 400 nm versus 580 or 620). Furthermore, reproducibility of absorbance values
should be better at the lower wavelengths (i.e., 340,370, and 400 nm) for the concentrations of oil used in this
study because measurements at 580 or 620 nm would be much closer to the detection limit (and associated
background noise) of the spectrophotometer. Hence, the conduct of absorbance measurements at 340,370, and
400 nm were determined to be suitable for the measurement-sensitivity objectives of this project
Note that the absorption maximum for the spectrophotometer is approximately 3.6 absorbance units, as
indicated by the "plateau" appearance of measurements at this value in Figures 6a and b. Therefore, all
measurements with absorbances greater than 3.5 were treated appropriately (i.e., oil-standard calibration
absorbances greater than 3.5 were excluded from determinations of spectrophotometric response to an oil, and
extracts of experimental samples with absorbances approaching this value were diluted accordingly). Note that
absorbance (A) is related to transmittance (T) by Beer's Law (-log T = A). T has a maximum value of 1.0 when
A = 0. When A = 3.5, then T = 0.000316 or 0.0316%.
Further tests were conducted to evaluate the effects of cleaning agents and water extraction on
absorbance values of oil standards. Results are shown in Figure 7. The figure contains standard curves of
spectrophotometric absorbances for Prudhoe Bay crude with and without the cleaning agents Corexit 9580
(Figures 7a through f), Citrikleen XPC (Figures 7g through i), and Corexit 7664 (Figures 7j through 1). Figures
7a through c are standard curves at 340,370, and 400 nmeters, respectively, for standards prepared with water
extractions (i.e., as occurs for all wash-water samples). Figures 7d through f are standard curves at 340,370, and
400 nmeters, respectively, for standards prepared without water extractions (i.e., as occur for all substrate-extract
samples). Figures 7g through i and Figures 7j through 1 are standard curves at 340,370, and 400 nmeters for
standards prepared with water extractions for Citrikleen XPC and Corexit 7664, respectively. The tests used to
generate the results in the figure were prepared using the SOPs presented in Appendix A. The plots in Figure 7
indicate that the cleaning agents used for the tests in this report are essentially transparent to the method of
standard preparation (i.e.. absorbances with and without the cleaning agents are essentially the same). Therefore,
variation in the concentrations of the cleaning agent (i.e., from no agent to the 3:1 oil-to-agent ratio used in tests
with an agent) should not significantly affect measurements for oil at the analytical wavelengths used in this
project
22
-------
(a) two test oils in methylene chloride (DCM)
E E E
c c e
T — ' - 1 - ' - 1
350 400 450 500 550
wavelength (nmeters)
T
600
650
700
Tvcavyic7i ivju i ^i ii i itsiei oy
(b) three shoreline cleaning agents in DCM
0)
u
e 0.5
OJ
Cor exit 9580
Corexit_7664
CitriWeenXPC
DCM
I ' 1 « 1
500 550 600
0.05 -
0.02
200
250
700
Figure 6.
wavelength (nmeters)
Wavelength scans of two test oils and three
chemical cleaning
agents relative to air.
23
-------
(a) 340 nmeters
2.0
1.6
1.2
0.8
0.4
°'°
oil only
oil+Conydt 9580
(b) 370 n meters
2.0
0.0 0.2 0.4 0.6 0.8 1.0
oil standard concentration (mg/mL)
(c) 400 nmeters
2.0
1.6
rt 0.8
0.4
0.0
oil only
[ oil-t-Core^it 9580 i
-«_ _____ — — — —'
0.0 02 0.4 0.6 0.8 1.0
oil standard concentration (mg/mL)
o.O 0.2 0.4 0.6 0.8 1.0
oil standard concentration (mg/mL)
standards prepared
with water extraction
oil type: Prudhoe Bay crude
cleaning agent: Corexit 9580
Figure 7. Effect of cleaning agents and water extraction on
absorbance of oil standards (oil to cleaning agent ratio
of 3:1).
24
-------
(d) 340 nmeters
2.0
1.6
g 1.2
I
o
u>
«0.8
0.4
0.0
0.0 0.2 0.4 0.6 0.8 llo
oil standard concentration (mg/mL)
(f) 400 nmeters
2.0
1.6
8 1.2
co 0.8
0.4
0.0
oil only ,
Oil+Cor«y
-------
(g) 340 nmeters
3.5
(h) 370 nmeters
2.0
1.6
S1.2
I
o
<9
«0.8
"0.0 0.2 0.4 0.6 0.8 1.0
oil standard concentration (mg/mL)
(i) 400 nmeters
2.0
1.6
51-2
•50.8
0.4
0.0
j oil+Citrik£en XPC \
0.0 0.2 0.4 0.6 0.8 1.0
oil standard concentration (mo/mL)
0.4
O.Q
i oil^jply ,
\ oil+Citrikj^en XPC J
"0.0 0.2 0.4 0.6 0.8 1.0
oil standard concentration (mg/mL)
standards prepared
with water extraction
oil type: Prudhoe Bay crude
cleaning agent: Citrikleen XPC
Figure 7. (continued)
26
-------
G) 340 nmeters
3.5
3.0
2.5
I1-5
ro '
1.0
0.5
0.0
0.0 0.2 0.4 0.6 0.8
oil standard concentration (mg/mL)
(1) 400 nmeters
2.0
1.6
s1-2
0.8
0.4
0.0
i oiljjnly |
J oil+Corgjdt 7664 j
0.0 0.2 0.4 0.6 0.8 1.0
oil standard concentration (mg/mL)
Figure 7. (continued)
(k) 370 nmeters
2.0
1.6
g1.2
0.4
0.0
oil+Core^cit 7664 J
0.0 0.2 0.4 0.6 0.8 1.0
oil standard concentration (mg/mL)
standards prepared
with water extraction
oil type: Prudhoe Bay crude
cleaning agent: Corexit 7664
27
-------
In summary, spectrophotometric measurements of DCM-extracts of oil-standards and experimental
samples were made at wavelengths of 340, 370, and 400 nm for this study. Absorbance responses of the
spectrophotometer at these wavelengths were sufficiently sensitive and stable over anticipated oil concentration
ranges to satisfy the necessary measurement objectives for the study.
As for effects of evaporative weathering on absorbance values, test oils were always.allowed to stand at
ambient temperature for approximately 18 hours (i.e., overnight) following their application to substrates in both
the Inclined Trough and Swirling Coupon tests. This initial oil-to-substrate contact time was intended to (1)
allow for spreading of oils on substrate surfaces. (2) allow for, formation of adhesive bonds between the oil and
substrate, and (3) introduce.a component of evaporative weathering to the oil. Such weathering is a phenomenon
that occurs when oil strands on a shoreline in the real world. However, selective losses of certain compounds
from the oil during the weathering might affect spectrophotometric absorbances of the oils. To evaluate the effect
of evaporative weathering on test oils, standard amounts of Bunker C and Prudhoe Bay crude oils (150 uL) were
placed in a series of tared aluminum weighing pans. Initial weights of all pans with oil were determined to
obtain starting oil weights. Triplicates of the pans were then weighed at selected time intervals ( hours) followed
by quantitative collections of the oil with methylene chloride (DCM) washes. Final volumes of the washes were
adjusted with DCM to that of a typical sample and the extract analyzed by UV-visible spectrophotometry at the
three analytical wavelengths (340,370, and 400 nm). The results were used to determine the effect of
evaporative weathering on gravemetric weights and spectrophotometric absorbances of the test oils. Results are
illustrated in Figure 8. Significant gravimetric losses in oil weights over time were observed for both Bunker C
and especially Prudhoe Bay crude. The greater weight losses from Prudhoe Bay crude reflect the greater
abundance of low molecular weight, volatile components in this oil. Of particular importance, however, is the
lack of change in spectrophotometric absorbances over time for the three analytical wavelengths, which indicate
that the chromophores responsible for absorption at the three wavelengths were conserved during the weathering
process. Therefore, preparation of oil standards for estimating cleaning performance in sample extracts did not
have to include a component for evaporative weathering when spectrophotometric measurements were used to
quantify oil.
CLEANING PERFORMANCE FOR TWO OILS IN ALL TESTING PROCEDURES
All measurements for cleaning performance are summarized in Appendices D and E for the Inclined
Trough and Swirling Coupon tests, respectively. The appendices include information for the type and substrate
used for a particular test (Tro-S = Inclined Trough-stainless steel; Tro-T = Inclined Trough-porcelain tile; Cou-S
s Swirling Coupon-stainless steel; Cou-T = Swirling Coupon-porcelain tile), the cleaning agent (C9580 =
Corexit 9580; XPC = Citrikleen XPC; C7664 = Cbrexit 7664), and the type of oil (PBay = Prudhoe Bay crude;
BunC = Bunker Q. These designations for testing procedures, cleaning agents, and oils will be intermittently
used in discussions, tables, and figures below. For all tests, the following are major variables that were
incorporated into the overall testing design:
o two substrate types: stainless steel and porcelain tile,
o two oils: Prudhoe Bay crude and Bunker C, and
o three chemical cleaning agents (Corexit 9580, Citrikleen XPC, and Corexit 7664) plus a "no agent"
control)
Discussions of results are presented below for purposes of satisfying specific objectives of the testing program.
Cleaning Performance Among Testing Procedures
Results for all testing procedures with the two testing procedures (Inclined Trough and Swirling
Coupon), two substrates (stainless steel and porcelain tile), two oils (Prudhoe Bay crude and Bunker Q, and the
Cleaning agents (Corexit 9580, Citrikleen XPC, and Corexit7664, as well as a "no agent" control) are
28
-------
(a) Bunker C oil
110
1.6
1.4 I!31**
jj J ab.§£7°
1.2 I !ao-S
«
,o
1.0 %
I
- 0.8
o.
p
___ m- 0.6 ^
Q.
CO
10
20
30
46
time (hours)
50
60
70
0.4
(b) Prudhpe Bay crude oil
100
.S> 90
° 85
•^
"c
-------
summarized in Table 6. The table includes information for mean values of oil recovered in the wash-water and
substrate samples, as well as the sum of the oil (i.e., mass balance) for the two fractions. The value reported for a
given test/substrate/oil/cleaning agent combination is obtained from all replicates for that combination and the
means of the measurements for the three analytical wavelengths (340,370, and 400 nm) (see Appendices D and
E). Table 6 also presents descriptive statistics for each experimental combination—i.e., the standard deviation
(sd) and number (n) of replicate measurements as well as the relative standard deviation (RSD; sd/mean).
To assist in visualization of information from Table 6, the distributions of oil in the (a) wash-water, (b)
substrate, and (c) wash-water plus substrate fractions for .the different oil/cleaning agent .combinations in
particular testing procedures are illustrated in Figures 9,10,11, and 12 for the Inclined Trough/stainless steel,
Inclined Trough/porcelain tile. Swirling Coupon/stainless steel, and Swirling Coupon/porcelain tile protocols,
respectively. The following items can be noted from the figures.
o Mass balances for oil (i.e., the wash-water plus substrate extracts) approach 100% in all testing protocols.
o Only minor amounts (i.e., less than 10%) of Prudhoe Bay crude in the Swirling Coupon test and Bunker
C in all tests are removed from stainless steel or porcelain tile in the absence of chemical cleaning agents.
In contrast, approximately 40% of Prudhoe Bay crude is washed off stainless steel and porcelain tile in the
absence of cleaning agents in the Inclined Trough test.
o Treatment with either Corexit 9580 or Citrikleen XPC generally produces greater release of oil from
substrate surfaces (i.e., higher water values) than is observed for Corexit 7664.
As defined in equation (5) in Section 4, the cleaning performance due to a chemical cleaning agent
alone (i.e., PCLEAN) 's calculated as the difference in oil contents of wash-water samples from tests with and
without a cleaning agent. The information presented in Table 6 is used to calculate values for PCLEAN ror
particular test/substrate/oil/cleaning agent combinations. Results are presented in Table 7. Values of means and
standard deviations (sd) for"+ agent" and "control (no agent)" samples are obtained from Table 6. Values for
PCLEAN ("due to agent alone") are calculated by difference. Standard deviations for values of PCLEAN are
estimated by equation (6):
sd
p-CLEAN =
(6)
Trends in values for PCLEAN from Table 7 sre illustrated in Figure 13 by test oil and protocol for the 3.cleaning
agents. Effects of different cleaning agents alone for removing test oils from substrate surfaces in particular
testing procedures is more apparent in Figure 7. For example, the following items can be noted from the figure
among the testing protocols.
o Mean values for cleaning performance (PCLEAN) are consistently higher for both test procedures (trough
and coupon) and substrates (stainless steel and porcelain tile) with Corexit 9580 and Citrikleen XPC as
opposed to Corexit 7664.
o Mean values for PCLEAN arc generally higher for porcelain tile as opposed to stainless steel, although the
trend is more pronounced and consistent for Prudhoe Bay crude than Bunker C.
The higher values for PCLEAN witn Corexit 9580 and Citrikleen XPC are likely the result of better mixing of
these cleaning agents with the test oils than is the situation for Corexit 7664. As indicated from the Product
Information Sheets for the three cleaning agents in Appendix B, Corexit 9580 and Citrikleen XPC are
formulated with hydrocarbon-based solvents, whereas Corexit 7664 is a water-based formulation. Hence, C9580
and XPC are more inclined to mix into slicks of the test oils during the cleaning agent-to-oil contact or "soak"
times before the washing step. The result is that C9580 and XPC are more effective than C7664 as cleaning
agents for removing oil from the test substrates. The reason for the general trend of higher values for PCLEAN
with porcelain tile as opposed to stainless steel is not clear. However, the trend emphasizes the importance of the
30
-------
© ]§
i I
tffi o
o «>
a.
CO
O 2
S 3
i
I
>
i
9
o
CO
1
i
Tw^1. -
I & 2 35 8 8 S 8 I S; 8 % 8 3
3?
oovo
cici
no «I2
•i
1'
00 OJ C^ ^? S v 8
«O °? 10 CO
1
f^ — O>CVIO>^.T-OO.
§ X I
x
O
1
o
I
1
9 £
n ^ o Q_
X g g X
s £- <» s
X O £000
o
i
cc
I
o
i
10
i
31
-------
C9S80
XPC
07864
C9680
XPC
C7684
cleaning agent
IPrudhoe Bay OBunker C
cleaning agent
IPaidhoe Bay EBunker C
test procedure: Inclined Trough,
stainless steel
test oils: Prudhoe Bay crude
Bunker C
test agents: Corexit 9580
Citrikleen XPC
Corexit 7664
C8680 XPC C7B84
cleaning agent
BPrudhoo Bay HBunker C
Figure 9. Oil distributions in extracts of wash water and substrate.
Inclined Trough/stainless steel. Vertical line indicates
+/-1 standard deviation around mean, (a) wash-water,
(b) substrate, (c) total oil (wash water + substrate)
32
-------
C9580
XPC C7664
cleaning agent
•Prudhoe Bay HBunker C
C9580
XPC
C78A4
cleaning agent
IPrudhoe BayHBunkerC
test procedure: Inclined Trough,
porcelain tile
test oils: Prudhoe Bay crude
Bunker C
test agents: Corexit 9580
Citrikleen XPC
Corexit 7664
CKM XPC C7M4
cleaning agent
HPrudho* Bay HBunker C
Figure 10. Oil distributions in extracts of wash water and substrate.
Inclined Trough/porcelain tile. Vertical line indicates
+/-1 standard deviation around mean, (a) wash-water,
(b) substrate, (c) total oil (wash water + substrate)
33
-------
(c)
120
C9S80
XPC
C7M4
C9580
cleaning agent
IPrudhoe Bay dBunker C
XPC C7M4
cleaning agent
•Paidhoe Bay EJBunker C
test procedure: Swirling Coupon,
stainless steel
- s%_*
test oils: Prudhoe Bay crude
Bunker C
test agents: Corexlt 9580
Citrikleen XPC
Corexit 7664
caew
XPC C7684
cleaning agent
•Prudhoe Bay ElBunker C
Figure 11. Oil distributions in extracts of wash water and substrate.
Swirling Coupon/stainless steel. Vertical line indicates
+/-1 standard deviation around mean, (a) wash-water,
(b) substrate, (c) total oil (wash water + substrate)
34
-------
C9680
XPC
07664
cleaning agent
IPrudhoe Bay EUBunker C
C968O
XPC C7884
cleaning agent .
Prudhoe BayEDBunkerC
test procedure: Swirling Coupon,
porcelain tile
test oils: Prudhoe Bay crude
Bunker C
test agents: Corexit 9580
Citrikleen XPC
Corexit 7664
XPC C7M4
cleaning agent
•Prudhoe Bay E3Bunker C
Figure 12. Oil distributions in e:(tracts of wash water and substrate.
Swirling Coupon/porcelain tile. Vertical line indicates
+/-1 standard deviation around mean, (a) wash-water,
(b) substrate, (c) to
al oil (wash water + substrate)
35
-------
Table?
Summary: Cleaning Performance for Chemical Agents Relative to Control Samples
cleaning performance (% of initial oil on substrate)
test substrate oil agent
trough steel Prudhoe Bay Corexit 9580
Citrikleen XPC
Corexit 7664
Bunker C C9580
XPC
C7664
tile Prudhoe Bay Corexit 9580
Citrikleen XPC
Corexit 7664
Bunker C C9580
XPC
C7664
coupon steel Prudhoe Bay Corexit 9580
Citrikleen XPC
Corexit 7664
Bunker C C9580
XPC
C7664
tile Prudhoe Bay Corexit 9580
Citrikleen XPC
Corexit 7664
Bunker C C9580
XPC
C7664
+ agent
mean sd
76.3
77.6
62.1
35.2
48.0
12.6
88.5
88.7
74.5
49.4
53.0
6.9
57.6
47.5
7.7
56.2
47.2
0.7
65.9
58.2
15.9
46.1
50.8
7.6
4.9
0.6
2.1
0.5
5.5
4.4
9.0
4.9
11.0
7.8
4.9
2.2
11.2
12.9
2.9
14.9
7.5
0.3
11.9
12.3
11.5
2.1
3.3
7.0
control (no agent)
mean sd
44.1
44.1
44.1
3.6
3.6
3.6
37.9
37.9
37.9
6.8
6.8
6.8
2.2
2.2
2.2
0.4
0.4
0.4
3.4
3.4
3.4
0.2
0.2
0.2
8.8
8.8
8.8
1.1
1.1
1.1
8.1
8.1
8.1
3.1
3.1
3.1
0.6
0.6
0.6
0.4
0.4
0.4
2.1
2.1
2.1
0.1
0.1
0.1
due to agent alone
mean sd
32.2
33.5
18.0
31.6
44.4
9.0
50.6
50.8
36.6
42.6
46.2
0.1
55.4
45.3
5.5
55.8
46.8
0.3
62.5
54.8
12.5
45.9
50.6
7.4
10.1
8.8
9.0
1.2
5.6
4.5
12.1
9.5
13.7
8.4
5.8
3.8
11.2
12.9
3.0
14.9
7.5
0.5
12.1
12.5
11.7
2.1
3.3
7.0
NOTE: sd for"+ agent (A) - control sample (B)" = (sdA*2 + sdBA2)A0.5
36
-------
Prudhoe Bay crude:
Trough
Coupon
C9580
XPC
cleaning agent
stainless steel E3porcelain tile
C7664
Bunker C:
Trough
C9680 XPC C7864
cleaning agent
•stainless steel IfcorceJain tile
C9580
XPC
cleaning agent
stainless steel Qporcelain tile
Coupon
C76C4
C9680
XPC
C7864
cleaning agent
•stainless steel ^porcelain tile
Figure 13. Summary of cleanir g performance for chemical agents
relative to control samples (i.e., performance with
agent minus performance without agent). Vertical
line indicates +/-1 standard deviation around mean.
37
-------
substrate to estimates for cleaning performance. Further studies of the effect of substrate types on cleaning
performance are warranted.
As for absolute values of cleaning performance, means for PCLEAN m the Inclined Trough protocol with
the stainless steel substrate and Bunker C oil are 32%, 44%, and 9% for C9580, XPC, and C7664, respectively.
For comparison, equivalent results reported in Fingas et al. (1989) for the Inclined Trough test with a stainless
steel substrate. Bunker C oil, and seawater washes are 42%, 36%, and 27% for C9580, XPC, and C7664,
respectively. Reasons for differences in results for. the current tests and those of Fingas et.al. are not clear.
Cleaning Performance for Different Cleaning Agents
Data from Table 7 also is used to prepare Figure 14, which illustrates cleaning performance for the
separate testing procedures as functions of the three cleaning agents for the two test oils (Prudhoe Bay crude and
Bunker C). As shown, relative trends (but not absolute values) in cleaning performance show many similarities
for the three cleaning agents for the two testing procedures, two substrates, and two oils. For example, the
general trend in mean values for cleaning performance follow similar trends of C9580 ~ XPC > C7664. These
trends appear more pronounced in the Swirling Coupon test with both stainless steel and porcelain tile substrates.
Rankings for Cleaning Performance for Chemical Agents Among Testing Procedures
Using data from Table 7, Figure 15 shows cleaning performance for the four test protocols (Inclined
Trough and Swirling Coupon with stainless steel and porcelain tile substrates). Very similar trends among the
cleaning agents (i.e., C9580 ~ XPC > C7664) are again in evidence for both test oils. The information in the
figure indicates that relative rankings for cleaning performance show similarity among testing procedures and
substrates, although absolute values differ.
STATISTICAL ESTIMATION OF PRECISION ASSOCIATED WITH INDIVIDUAL TESTING
PROCEDURES
Best estimates for precision for a given testing protocol were originally intended to be obtained by
performing seven replicate measurements for a given procedure with a selected substrate, oil, and cleaning agent
(e.g., see Section 4 and Figure 4 in particular). As indicated by the data in Table 6, however, estimates of
precision for measurements (i.e., standard deviations about means) show substantial differences among test
protocols, sample fractions (i.e., water and substrate samples), substrates, oils, and cleaning agents.
Furthermore, increasing the number of replicate measurements for a given combination does not necessarily yield
smaller values for standard deviations about means. The observations suggest that precision associated with
measurements for cleaning performance will likely have components that are due to a number of experimental
variables including:
o the test protocol and substrate (Inclined Trough/stainless steel, Inclined Trough/porcelain tile, Swirling
Coupon/stainless steel, and Swirling Coupon/porcelain tile),
o the test oil (Prudhoe Bay crude and Bunker C),
o the cleaning agent (Corexit 9580, Citrikleen XPC, and Corexit 7664), and
o the analytical wavelength for measurements (340,370, and 400 nm).
As such, selection of a single substrate, oil, and cleaning agent may not be the most appropriate means to obtain
estimates of precision for measurements with a particular testing protocol.
An alternative approach that is adopted for estimating precision of a method with the present data set
utilizes analysis of variance (ANOVA) with determination of components of variance for defined experimental
variables. Specifically, ANOVA with component-of-variance analysis is performed for each of four separate
38
-------
Inclined Trough:
Stainless Steel
Piudho.Bay Bunk«rC
-oil type
•C9580 HXPC DC7664
Swirling Coupon:
Stainless Steel
Piu*o«B«y
oil type
•C9580 HXPC QC7664
Bunk«rC
Porcelain Tile
Prudho* Bay BuntorC
oil type
•C9580IIXPC DC7664
Porcelain Tile
Piudho«Bay
Bunker C
oil type
•C9580 IEXPC DC7664
Figure 14. Cleaning performance for separate testing protocols
with two oils and three cleaning agents. Vertical
line indicates +/-1 standard deviation around mean.
39
-------
(a) Prudhoe Bay crude
trough-stainless
(b) Bunker C
trough-tile coupon-stainless
test protocol
C9580
DC7664
coupon-tile
trough-stainless trough-tile coupon-stainless coupon-tile
test protocol
• C9580 E3 XPC D C7664
Figure 15. Cleaning performance for four testing protocols
and three cleaning agents. Vertical line
indicates +/-1 standard deviation around mean.
40
-------
protocol-substrate combinations: Inclined Trough/ste inless steel, Inclined Trough/porcelain tile, Swirling
Coupon/stainless steel, and Swirling Coupon/porcelain tile. Data for the four protocol-substrate combinations are
treated separately because each combination can be defined as a unique testing protocol whose evaluation is the
major focus of this laboratory effort. Separate ANOVA/component-of-variance analyses are performed for results
of oil measurements in wash-water and substrate sarr pies. Within each protocol-substrate and sample-fraction
combination, determinations are made of the fraction of the overall variance due to the following factors: total
variance (which will always be 100%), oil type, cleaning agent type, analytical wavelength, and residual (or
error). The variance of the residual or error (or its associated standard deviation) is indicative of the precisian of
measurements in a particular method (i.e., protocol-sjibstrate combination) after the removal of variance
components due to differences in the oils, the cleaning agents, and the analytical wavelengths. All statistical
measurements for this effort are made with procedures from either of two computer-based statistical programs
(SAS and SYSTAT) or computational procedures obtained in Sokal and Rohlf (1981) and Rohlf and Sokal
(1981).
Results of the component of variance analyses for the four combinations of protocol and substrate (i.e.,
Inclined Trough/stainless steel, Inclined Trough/porc'plain tile. Swirling Coupon/stainless steel, and Swirling
Coupon/porcelain tile) are presented in Tables 8,9, ip, and 11, respectively. Separate summaries exist in each of
these tables for the substrate and wash-water sample fractions. Data used for each analysis is grouped into 18
subpopulations for the different combinations of oils (Bunker C and Prudhoe Bay), cleaning agents (C7664,
C9580, and XPC), and wavelengths (340,370, and 400). An assumption for the conduct of ANOVA is that
standard deviations (or variances) of groups be homogeneous (i.e., not significantly different from each other).
Bartlett's test for homogeneity of group variances (performed with the computer-based program SYSTAT and
computational procedures in Sokal and Rohlf, 1981) s used to evaluate this assumption. The results for this test
are presented in Tables 8 through 11 for each test-substrate and sample-fraction combination. Variances among
groups are determined to be homogeneous (at the alpha=0.05 or 95% confidence level) for the following five test-
substrate/sample-fraction combinations in Tables 8 through 11: (1) Inclined Trough-stainless steel-substrate
fraction, (2) Inclined Trough-porcelain tile-substrate jraction, (3) Inclined Trough-porcelain tile-water fraction,
(4) Swirling Coupon-porcelain tile-substrate fraction, and (5) Swirling Coupon-porcelain tile-water fraction.
Heterogeneity of group variances is indicated for three test-substrate/sample-fraction combinations: (1) Inclined
Trough-stainless steel-water fraction, (2) Swirling Coupon-stainless steel-substrate fraction, and (3) Swirling
Coupon-stainless steel-water fraction. Mathematical transformation of data for the latter combinations to natural
logarithmic and square-root formats did not improve results of the homogeneity tests. Consequently, results of
component of variance analyses and values for precision (i.e., standard deviations or variances) for the latter test-
substrate combinations must be viewed as estimates only.
r
Values of precision for particular testing protocols are contained in the sections for "Component of
Variance Analysis" in Tables 8 through 11. Specifically, variance components and the percent of the total
variance are presented for the following: (1) different oils, (2) different cleaning agents, (3) different analytical
wavelengths, and (4) the residual or error. The latter is the variance associated with the particular testing
procedure itself, separate from contributions of the oil type, the cleaning agent type, and the analytical
wavelength. This residual or error term for variance ik equivalent to the square of the standard deviation, which
is indicated by the "sd" at the bottom of the Component of Variance Analysis section. The following summarizes
results from Tables 8 through 11 for the precision values for the four testing protocols:
test protocol
Inclined Trough-stainless steel
Inclined Trough-porcelain tile
Swirling Coupon-stainless steel
Swirling Coupon-porcelain tile
sa-water
4.4%*
7.2%
12.0%*
10.3%
sd-substrate
3.8%
4.8%
10.4%*
8.9%
41
-------
V
a>
55
en
CO
^>
_c
(0
CO
CO
£
J=
O)
o
CO O
p £
2 "o
co c
f™*
• »
CO
"35
_>.
CO
<
0)
o
(0
UJ
a.
co
a:
UJ
I
OOOOO
p
OO
p
OOOO
•if ^ ^ O O O "& *F ^t O O O
(0c0c0cdcocoOOO CD cr>
OOOOOO
_
s<* f^ ^^ f^ CD CD CD
OOOOOO
ooooooooo c2 3
2 |
So
15 -c
S E
co
^r CD o g
: ra«
! >
CM O CO CM o>
I
^ a
55 3
IS
.£
"o
"o
CO
o
§
a.
o
O
(Ocqr.pcDcqoqcDp^cnencMcocMinioi^
CO
UJ
co
§
OO
Sh-.
co
OOOOO
OOOOOOOO
r^ovh^o^r^o
co^coco ^r co co ^
*«D
t^. Is* i*^
ooo
O O O ^ ^J" ^* O O O
u>Q_Q_cL
CD CD 01 ^* ^^ N^ r^* f1^ ^^ o CD os ^x ^x ^>
ooooooooo
ooooooooocSaicSiSiS &&££
333333333"D "O "O "O T3 -O "O TJ ^
mmmoammmmmsSSSSSESS
0.0.0.0.0.0.0.0.0.
CO !•••
SCM CO
o 55
3
co
o § g
C io CO
I
I
to to Is: ^ 2
CM CM in • ™
CO O CO *T ^
»- «- CM *-
O ^ C •£
12 ff'
I
C TJ >
g "O
II
u
42
-------
a
g
"S
0
s«
cc
LU
O>
f^ ^» o) O) ^ft
O 0 00 0
O ' *^ ^f ^^ f*i
a>OOOu}u>'. >. >. >. >-
_33333333"DT3T3'a'O
ffimcacDommcam 2 2 E S S
Q. Q. CL Q. CL
COCOCOWOOJCOCOTOCOOOCOCOCO
a
§
co
ai
i
.«
I I
S
ooooooo
'--
o
OOOOOO
,.
000000000|g(SlS(S(3
Ifgllllllllll
cQcDmcticnmfflcacQ22225
a. a. a. a. a.
co o>
a> ut -f -v
CM ^ oq cq
co co oo co
§000
•*r r^ S
T eo c> ir
§ O O O
in Q. a. Q.
g X X X
>,>.>.>.
W W W IV
Illl
ZZZZ
y- <3> CO
»- »- O CM
S
E
CD
•*. °
8§
J§]
m E
«> CM
.2
co
5 S S
°|i
c S CO
8.
O
CM CO
T- to
3
cS 5
ID O> fc
^i^
s
CJ <£) •<.
&&&&
11
-a TJ
5 5
a. a.
h- "?
»- o
^3
O
»r o
o -c
£
1 a
i I
vp •
a E
>3
CO
CO
8 S
"i co
|
I
g -
Is!f
"5
43
-------
CO
en
IE
co
55
in
£
c
o
Q.
3
O
O
U» tai
« ,1
<
CO
cc
Ul
•3
ooooo
--
S^*<.>»>%>»>.>.>.>.:
mcomcommcamffiSSSESSSSS
0.0.0.0.0.0.0.0.0.
oo
s
in
m
I
(9
5O
«
§•§>
2 g
0> S
if
Is
to
£
S
T3
£
CO
si
£
«
1.
CO
5 S
'o .co
c S
<*> cj •
C\JCO'"O>.
CO
c
0}
§
J5
1
g
o
Q.
O
O
s
^•io oo <°. oo
S S0
2
CO
CO
CD
o
to
to
1
?
~— g
O
E X
a «
e
_«
CO
0 C
"S -
S S
i: _a>
P
o
it
ca
a
I I
| 1
I cS
-^ «
•O „ a
^ J S
.a "3 g
44
-------
CO
H
C
Us
«
o
o
,2
c
o
a.
o
o
O)
I
0)
13
<
OJ
a>
i
45
-------
The "*" in this summary indicates values for a test protocol combination in which variances among the
contributing groups are heterogeneous by the Bartlett's test for homogeneity. As noted above, such values must
be vlswed as best estimates only for precision. However, the preceding results do indicate that the precision for
measurements of cleaning performance is better with the Inclined Trough test.
In addition to the preceding information for precision associated with a given test procedure, the results
of the Component of Variance Analyses in Tables 8 through 11 also provide insight into the effects of different
oils, cleaning agents, and analytical wavelengths on measurements for cleaning performance. The following
summarizes information from the tables.
source of variance
oil
cleaning agent
wavelength
residual (error)
> of overall variance-Trough
66-83%
15-32%
0%
1-4%
% of overf1^ variance-Coupon
0-34%
61-82%
0%
6-25%
This indicates that the percent of the overall variance due to differences in the analytical wavelengths (i.e., 340,
370. or 400 nm) is always 0% for the four testing procedures. This conclusion provides justification for using
averages of results from the three wavelengths for determinations of cleaning performance. For the Inclined
Trough test, differences in test oils contributes the most to the overall variance (i.e., 66-83%). In contrast,
differences in cleaning agents provide the largest contribution to the overall variance in the Swirling Coupon test
(61-82%). The reason for the latter contrast among the two procedures is not known. For all tests, however,
differences associated with two factors (i.e., test oils and cleaning agents) contribute to at least 75% of the overall
variance in test results (i.e., the sum of variance components due to the residual or error and the analytical
wavelengths is never greater than 25%).
INCLINED TROUGH TEST: EFFECT OF WASH-WATER FLOW ON PERFORMANCE RESULTS
The Inclined Trough test uses a flow of water down the,surface of the trough to promote release of oil
from the substrate surface. The flow rate for most tests in this program was established as 1.05 mL/sec (63
mL/min). However, tests also were performed to evaluate the effect of flow rate on cleaning performance.
Experiments were performed at two flow rates (63 mL/min and 7.5 mL/min) for two separate substrate-oil-
cleaning agent combinations: stainless steel-Prudhoe Bay crude-Corexit 9580 and porcelain tile-Bunker C-
Corexit 9580. Tests were performed in triplicate for each flow rate at each substrate-oil-cleaning agent
combination. The slower flow rate (7.5 mL/min) was obtained by attaching an 18-gauge needle to the syringe
barrel from which wash-water flowed down a trough. The normal flow rate for tests (63 mL/min) was obtained
with no restricting needle attached to the syringe barrel.
Results of the effect of the different flow rates for wash-water on cleaning performance are illustrated in
Figure 16. Data are presented for means and one standard deviation unit around the means for triplicate
measurements. The figure contains information for the fractions of oil recovered in the wash-water and substrate
fractions as well as the sum of the two fractions (i.e., mass balance). As indicated, cleaning performance
increases in both tests at the lower flow rates, although the effect is more pronounced in the tests with porcelain
tile and Bunker C. The lower flow rate presumably allows for more efficient washing of treated oil on the
substrate surfaces. While results shown in Figure 16 must be viewed as preliminary (i.e., they represent only
triplicate measurements in each instance), the importance of evaluating the effect of flow rate on cleaning
performance is apparent.
46
-------
(a) stainless steel-Prudhoe Bay-Corexit 9580
120 i
water fraction
substrate traction
test fraction
• flow-63 mL/mip U flow-7.5 mUmin
(b) porcelain tile-Bunker (b-Corexit 9580
•<«v\ _____ , I _._
water+substrate
120
c 100
(8
80
8
rt 60
E
o
1 «>
O.
O)
-I 20
water fraction
• fraction
test fraction
I flow-63 mLVmin 13 ftow-7.5 mL/min
water-f substrate
Figure 16.
Cleaning performance in Inclined Trough protocol
with different wash-water flow rates. Vertical line
indicates +/-1 standard deviation around mean.
47
-------
SECTION 6
QUALITY CONTROL CHECKS
As can be determined from Appendices D and E. this report summarizes data for a total of 182
experiments for cleaning performance in the different testing systems: 82 tests for the Inclined Trough (stainless
steel and porcelain tile substrates) and 100 tests for the Swirling Coupon (stainless steel and porcelain tile
substrates). Each test yields two sample fractions (i.e.. separate wash-water and substrate-extract fractions).
Consequently, the total number of sample extracts analyzed for oil content in the program is 327. The number of
tests does not include method blanks, which were run for each procedure.
In all testing procedures, absorbance measurements made with the UV-visible spectrophotometer are
critical for achieving project objectives involving determination of oil quantities in experimental samples. The
spectrophotometric measurements are used to quantify the amount of oil released into the wash-water fractions of
test samples. Accuracy of absorbance readings for oil standards and the precision of oil measurements in wash-
water and substrate-extract samples are obtained from extracts of oil-water standards and experimental test
samples, respectively, that have been subjected to routine analytical procedures to which experimental samples
are submitted. Accordingly, final spectrophotometric data for samples include variability associated with not
only the analytical instrumentation but also the preparation of standards and samples (i.e., standards and wash-
water samples are extracted from a seawater matrix with comparable procedures; see SOPs in Appendix A). As a
consequence of the critical nature of the spectrophotometric readings, quality control (QC) checks that were
incorporated into the experimental testing program included routine analysis of spectrophotometric standards,
replicate measurements of sample extracts, linearity checks for oil standards in the spectrophotometer, and
evaluation of values associated with cleaning performance for the individual analytical wavelengths (340, 370,
and 400 nm) relative to the mean value for all wavelengths. Data for these QC checks are discussed in the
following sections.
In addition to QC checks related to spectrophotometric measurements, additional QC efforts included
routine analysis of procedural method blanks and collection of duplicate test sets of data (at least triplicate
experimental runs per test set). These data also are discussed below.
SPECTROPHOTOMETRIC STANDARDS
Prior to and during analysis of standards and samples on each day of testing, one or more
independently-prepared spectrophotometric reference standards (Oxford Spectro-Chek Solutions I, n. III, and IV)
were analyzed to monitor relative absorbance accuracy and long-term stability of the spectrophotometer.
Maximum absorbances occur at 510 nm for Solutions I and II (cobalt ammonium sulfate) and 370 nm for
Solutions in and IV (potassium chromate). Examples of the reproducibility of absorbance readings for these
check standards are shown in Table 12. In the table, values obtained during the conduct of measurements for the
present study (i.e., 9 March 1992 and later) are indicated in bold print To show the stability of the
spectrophotometer over extended periods of time, results from measurements in other studies (i.e., from 3
October 1991 and later) are included. As indicated, variations in absorbance readings for a given check solution
varied by no more than 2% from overall mean values for the six-month period (October 1991 through March
1992).
REPLICATE SPECTROPHOTOMETRIC MEASUREMENTS OF SAMPLE EXTRACTS
In addition to the spectrophotometric reference standards, individual sample extracts were periodically
analyzed in triplicate in the spectrophotometer. The purpose was to evaluate the reproducibility of
spectrophotometric readings as defined by relative standard deviations (RSDs) for values about the means of the
triplicate measurements. Results at the three analytical wavelengths are summarized in Table 13, while complete
48
-------
s
2 co
— «
^
>E
c
CO
o
J3
O
en
.*
o
Q)
O s-E
c
o
^^
2
.a
15
O
8
o
0
o
Q.
O
«^
O
Q)
a.
co
= 8 P
jll
en £2 w
_ o
= o E
S *"
1
rom
dif
cS.i
O 0
= p o
a s =•
o «
CO O
(9
•a
? 9 9" 9 o d 9 o' d d • ? 9 9 i-: Y d o' i-: d
oooddddddddddddddd d
cD
1
- 2 ° o o o o o o o f 4- <B^ - - -• t* J - - ^ff — -
090999909,130000090 d 9d9
sisissiisisssssssllsiiss
"*•: op •» p p p p « «
9 d d 9 d o' d d d •r-
o o W ^ o
dddo'do'ddddddddo'dd
C? ^ •$ 03 0> 0» O)
, •--:»ju7pi_c: co
i! I
CO Q
IE i
I- •=
i S i
™B §
•s **• 2
ii i
IT |.
o> »
•o "5
EC *
w J8 -g
ii I
— — »
"o "o "<&
COCO O
49
-------
Table 13
Summary: Triplicate Spectre-photometric Measurements on Sample Extracts
anal
data
3-26
3-26
3-26
3-16
3-16
3-16
3-17
3-17
3-17
3-16
3-16
3-16
3-23
3-23
3-23
3-27
3-27
3-27
3-23
3-23
3-23
3-23
3-23
3-23
3-27
3-27
3-27
3-23
3-23
3-23
anal expt
time type
1524 Cou-S
1527 Cou-S
1532 Cou-S
15:45 Cou-S
15:53 Cou-S
15:57 Cou-S
1624 Cou-T
1628 Cou-T
1636 Cou-T
1323 Cou-T
1327 Cou-T
1332 Cou-T
13:17 Tro-S
1321 Tro-S
1329 Tro-S
1024 Tro-S
1028 Tro-S
1032 Tro-S
1639 Tro-S
16:42 Tro-S
16:45 Tro-S
1438 Tro-T
14:41 Tro-T
14:46 Tro-T
12:04 Tro-T
12:08 Tro-T
12:12 Tro-T
10:15 Tro-T
10:19 Tro-T
1024 Tro-T
lab
qua!
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
MB
no
S1
S1
S1
S4
S4
34
T3
T3
T3
T4
T4
T4
TS1
TS1
TS1
TS5
TS5
TS5
TS3
TS3
TS3
TT2
TT2
TT2
TT5
TT5
TT5
TT1
TT1
TT1
dean
agent
XPC
XPC
XPC
C7664
C7664
C7664
XPC
XPC
XPC
C7664
C7664
C7664
C9580
C9580
C9580
C7664
C7664
C7664
C9580
C9580
C9580
C9580
C9580
C9580
C7664
C7664
C7664
C9580
C9580
C9580
oil
type
BunC
BunC
BunC
PBay
PBay .
PBay
BunC
BunC
BunC
PBay
PBay
PBay
BunC
BunC
BunC
PBay
PBay
PBay
PBay
PBay
PBay
BunC
BunC
BunC
PBay
PBay
PBay
PBay
PBay
PBay
340 nm absorbance
inH2O
2.1%
0.7%
0.9%
0.8%
1.9%
0.4%
22%
1.6%
1.6%
02%
0.4%
0.4%
0.4%
0.5%
0.6%
0.6%
0.5%
0.9%
13%
1.4%
0.9%
0.9%
0.6%
1.0%
1.1%
2.1%
12%
03%
1.9%
2.7%
onSubstr
12%
0.3%
2.0%
2.5%
1.3%
1.8%
2.5%
1.8%
23%
02%
0.6%
0.1%
13%
0.5%
03%
03%
0.6%
0.5%
23%
0.9%
05%
0.9%
03%
0.6%
0.5%
03%
03%
23%
25%
4.0%
370 nm absorbanca
inH2O
1.5%
0.7%
0.7%
0.9%
2.0%
0.5%
2.1%
1.5%
12%
03%
02%
03%
03%
0.7%
0.7%
0.5%
0.6%
0.9%
12%
1.6%
03%
03%
0.6%
03%
13%
2.5%
13%
0.7%
1.6%
£5%
onSubstr
1.1%
0.4%
22%
2.5%
1.4%
1.9%
2.5%
1.5%
2.4%
02%
0.7%
02%
0.9%
0.7%
0.1%
12%
03%
0.6%
23%
0.9%
03%
0.9%
0.8%
0.6%
0.4%
03%
0.6%
3.0%
3.5%
4.9%
400 nm absorbanca
inH2O
23%
0.4%
0.7%
1.0%
2.0%
0.6%
2.1%
1.5%
1.1%
0.4%
02%
0.4%
0.5%
03%
0.7%
0.7%
0.6%
12%
1.2%
1.6%
03%
0.8%
0.7%
03%
1.4%
32%
1.7%
0.7%
1.7%
2.4%
onSubsr
03%
03%
23%
2J&Y*
1.4%
15%
2.4%
1.5%
2S%
03%
1.7%
0.2%
1.0%
05%
0.1%
1.6%
1.0%
05%
23%
03%
0.6%
05%
03%
0.6%
0.5%
0.5%
05%
33%
4.0%
5.4%
NOTE: % values are relative standard deviations (RSDs) about the means for the triplicate measurements.
50
-------
results are presented in Appendix H. The table com ains measurements for 30 extracts for a frequency of 16.5%
(30 out of 182 total sample extracts). As indicated n the table, only 1 of 180 values has an RSD greater than
5%. The absorbance readings for the single sample with the one RSD greater than 5% are determined to be
acceptable because the RSDs at the two remaining wavelengths are less than 5%.
OIL STANDARDS
The UV-visible spectrophotometer was calibrated with external standard solutions of the specific oil and
cleaning agent (if applicable) that was used in an experiment prior to measurements for sample extracts. As
defined in Section 4, linearity of instrument response over concentration ranges of oil standards was evaluated
with response factors (RFs). RFs for standards had to be less than 20% different from the mean RF for those
standards that yielded absorbances below instrument saturation and above the detection level of the
spectrophotometer (approximately 3.5 and 0.080 absorbance units, respectively). An example of plots of RF
values and absorbances at the three analytical wave engths (340,370, and 400 nm) versus oil concentration has
been presented earlier in Figure 5. Examples of linearity data for standard curves are presented in Appendix C.
Data in the appendix indicate that instrument linearity was acceptable for the defined project objectives (i.e..,
individual RFs less than 20% different from mean RFs) on almost all occasions (i.e., 39 of 43 oil standard curves,
and only single solutions of six standards falling outside of limits in the four aberrant curves).
CLEANING PERFORMANCE VALUES - i
VERSUS INDIVIDUAL WAVELENGTHS
MEANS FROM THREE ANALYTICAL WAVELENGTHS
As described in Section 4, concentration values for oil in sample extracts are calculated as the mean of
values (Cmean) for the three analytical wavelength^ (€340, C370, and C40o). If one of the values for €340,
C370»or C400 was more than 30% different from Cmean, the value was to be flagged. Complete data for all
samples for the percent of total oil in wash-water and substrate-extract fractions as well as differences of these
values from the mean at individual analytical wavelengths (% diff mean) are presented for each testing procedure
in Appendices D (Inclined Trough) and E (Swirling Icoupon). As indicated in these appendices, quantities of oil
in the wash-water and substrate-extract fractions determined at the individual wavelengths are always less than
4% different from the corresponding mean value with the exception of a single extract.
Statistical evaluation of the effect
performance also has been performed. As indicated
experimental results that could be attributed to differences
average values for the three wavelengths are used
variables (e.g., cleaning agent types, oil types, and
of indivk ual analytical wavelengths on results of cleaning
in Tables 8 through 11, the fraction of the total variance in
in analytical wavelengths was always 0%. Hence,
to evaluate cleaning performance with other experimental
testing procedures).
METHOD BLANKS IN TESTING PROCEDURES
Method blanks were analyzed in conjunctio i with experimental samples for each of the testing
procedures. A method blank involves analysis of a blank (i.e., wash water and substrate extracts) through am
entire testing and analytical procedure. Spectrpphotometric measurements of the final extracts are used to
estimate the equivalent amount of oil that would correspond to the resulting absorbance values for a specified oil.
Detailed summaries of method blanks for all testing procedures are presented in Appendix F. Table 14 presents a
brief summary of these results for the mean values frbm the three analytical wavelengths. Values for oil in the
method blanks are always less than 0.5% of that for associated oil-experiment samples. A total of 24 method
blanks (i.e., 6/test-substrate combination) were analyzed for the four testing procedures: Inclined Trough-
stainless steel. Inclined Trough-porcelain tile, Swirli ig Coupon-stainless steel, and Swirling Coupon-porcelain
tile. This yields a frequency of 13% (i.e., 24 method blanks/182 runs for the various test-substrate combinations).
51
-------
Table 14
Summary: Method Blanks for Shoreline-Cleaning-Agent Performance Tests
mean % oil
anal
date
anal
time
expt
type
lab
qual
MB
no
clean
agent
oil
type
OilMass
Add(mg)
Tile Coupon:
3-10
3-11
3-11
3-13
3-13
3-13
17:08
11:31
11:46
09:06
0957
09:46
Cou-T
Cou-T
Cou-T
Cou-T
Cou-T
Cou-T
T1
T2
T3
T4
T5
T6
T1
T2
T3
T4
T5
T6
none
(C9580)
(C9580)
(C7664)
(C7664)
(C7664)
BunC
BunC
BunC
PBay
PBay
PBay
46.7
46.7
46.7
42.9
42.9
42.9
for 3
-inH20
0.0
0.1
0.2
0.2
0.1
0.2
Steel Coupon:
3-10
3-11
3-11
3-13
3-13
3-13
17:15
11:39
12:06
09:18
0931
09:56
Cou-S
Cou-S
Cou-S
Cou-S
Cou-S
Cou-S
S1
S2
S3
S4
S5
S6
S1
S2
S3
S4
S5
S6
none
(C9580)
(C9580)
(G7664)
(C7664)
(C7664)
BunC
BunC
BunC
PBay
PBay
PBay
46.7
46.7
46.7
42.9
42.9
42.9
0.0
0.1
0.1
0.1
0.1
0.3
wavelengt
onCoup
0.0
0.1
0.0
0.1
0.1
0.1
0.0
0.1
0.0
0.1
0.1
0.1
hs
Total
0.1
0.2
0.2
0.2
0.1
0.2
0.1
0.1
0.1
0.2
0.2
0.4
mean % oil
anal
date
anal
time
expt
type
lab
qual
MB
no
clean
agent
oil
type
OilMass
Add(mg)
for 3 wavelengths
inH20
Tile Trough:
3-20
3-25
3-25
3-26
3-26
3-27
16:42
10:59
11:14
09:05
09:08
12:53
Tro-T
Tro-T
Tro-T
Tro-T
Tro-T
Tro-T
TT1
TT2
TT3
TT4
TT5
TT6
TT1
TT2
TT3
TT4
TT5
TT6
(XPC)
none
none
(C9580)
(C9580)
(C7664)
PBay
BunC
BunC
PBay
PBay
PBay
134.1
145.8
145.8
134.1
134.1
134.1
0.0
0.2
0.1
0.1
0.1
0.1
Steel Trough:
3-20
3-25
3-25
3-26
3-26
3-27
16:49
10:51
11:07
09:17
09:37
13:00
Tro-S
Tro-S
Tro-S
Tro-S
Tro-S
Tro-S
TS1
TS2
TS3
TS4
TS5
TS6
TS1
TS2
TS3
TS4
TS5
TS6
(XPC)
none
none
(C9580)
(C9580)
(C7664)
PBay
BunC
BunC
PBay
PBay
PBay
134.1
145.8
145.8
134.1
134.1
134.1
0.0
0.2
0.1
0.1
0.2
0.1
onTrou
0.0
0.2
0.1
0.1
0.1
0.1
0.0
0.2
0.1
0.1
0.1
0.1
Total
0.0
0.3
0.2
0.2
0.3
0.2
0.0
0.3
0.2
0.2
0.3
0.2
52
-------
DUPLICATE TEST SET MEASUREMENTS
As described in Section 4, at least two dup icate measurements for cleaning performance were required
during conduct of all tests with not only the Incline^ Trough but also the Swirling Coupon procedures. Duplicate
measurements involve collection of two separate date sets with at least triplicate test runs per data set (i.e.,
resulting in two sets of triplicate or greater values). In fact, duplicate measurements were performed at a greater
frequency than defined above in an effort to evaluatkresults from duplicate data sets. Detailed summaries of the
duplicate measurements are presented in Appendix G.
)f replicate groups is used to evaluate differences between
Analysis of variance (ANOVA) for means
duplicate test set measurements. All statistical evaluations were performed with the a computer-based statistical
program (S YSTAT) and computational procedures jabtained in Sokal and Rohlf (1981) and Rohlf and Sokal
(1981). Results of the analyses are summarized in Table 15 for various test-substrate-oil-cleaning agent
combinations. Separate analyses are performed for Ml measurements from substrate-extract and wash-water
samples. In the table, data are presented for descriptive statistics (i.e., means, standard deviations, and number
of samples) for each group contributing to an ANOVA. An assumption for ANOVA is that standard deviations
(or variances) of groups should be homogeneous (i.e., not significantly different from each other). Bartlett's test
for homogeneity of group variances is used to evalukte this assumption. The results for the Bartlett's tests are
included in the table for each combination of test-slibstrate-oil-cleaning agent and sample-fraction. Variances
among groups for the given duplicate data sets are determined to be homogeneous (at the alpha=0.05 or 95%
confidence level) for 19 of the 22 groups of replicate measurements. Therefore, use of ANOVA to evaluate
differences between duplicate test set measurements is determined to be acceptable. Results of these ANOVAs'
indicate that means for duplicate test set measurements are not significantly different from each other
(alpha=0.05 or 95% confidence level) for 13 of the 22 groups of replicate measurements.
NON-CRITICAL EXPERIMENTAL VARIABLES
Critical measurements for satisfying projec
UV-visible spectrophotometer. However, additional non-critical
in the specific testing procedures. These non-critica
o Inclined Trough test
o S wirling Coupon test •
objectives are the absorbance measurements made with the
variables also are important for measurements
variables include the following.
—flow rate of wash water.
angle of the trough for washing
temperature of the wash water
ambie U temperature of the testing laboratory
shaking on the shaker table
length of shaking on the shaker table
— rate of
stroke
temperature of the wash water
ambient temperature of the testing laboratory
Information for each of these variables in the respective testing procedures is included with each test result in
Appendices D (Inclined Trough test) and E (Swirling Coupon test). In general, acceptance ranges for each of the
variables as defined in the Quality Assurance Project Plan (QAPjP) for the laboratory study were satisfied. Water
temperatures were between 19 and 22°C during all testing measurements. Air temperatures were between 18 and
24°C during all testing measurements. The rate of shaking and the stroke length of the shaker table for Swirling
Coupon tests were measured as 150 RPM and 2.0 cml respectively. How rates for all general test measurements
(except when variation was intended) and trough angles for all measurements in the Inclined Trough were 63
mL/min and 45°, respectively.
53
-------
03
c
o
E
3
(0
in
ID
0
CO CO
o co
•of o
O O
o *— o co
8 =5 ££
j 1
8-
f <5
Q. •
§t
a> 5
'"S
^•§
fe*
°
1^
•9 in
S o
Q- if
Q.S
.
S« 2
a, -5 o3
< CO
a.
•3
O
«
2-a
CD O
JU
^
Q
• «
&•
3
CO
1
55
I
55
54
-------
o
m
55
-------
o
o co
Q E
co a.
oc
oo
QC
HI
•32
i-3
eo co co
cq o« ". a>
co •* ° O>"
•» CD •* CM
S 5 S S
Q O O
o o o o
(TJ fB CO
CO CO CO CQ
~ Q. a. a.
S8S-
4 to ^
co r^ co
^ co o r^
;= II
•m 'B
73
| 2-
i\
II
•5-=
11 ^
9« m
! I- -i
; j? 1
is i
CO
o-S
StQ
TJ
CO 00 CO CO -CO
co r~ co
c\i ^ (o ^ od
o uj T- o oo
co »- co in to
co co uS u> tA
O O O O O
0.0.0.0.0.
X X X X X
OS A (Q (Q ^0
m gi. m CQ
Q- Q- Q- O- Q-
CO CO
O O>
O
C
m
1
<0 S? .£ -S
•B Q-£0
i 8 I 5
< CO
in
i
o
I
CO
t CO CO CO
r- CB . o
co ™ ui
o to q
0.0.
>
m
0.
i-~ CO C3 |C
S^I-ST
I 1*
I Is
± Q- »
o as
i"=B « 5!^
fls
M V
S--5
si
^2 c Ji
CD jj (9 CB CB
^ "5 S S o
ll|f
o 3
O E
a
_c
V
o
#
^
=5
CO O CO CO CO
CO CM T °?
co oj 2f £J
!»; CO O CO CO
S S 8 S S
o o o o o
o. o. a. o. a.
CO CO CO •
o> ^
• o en
§
U.
56
-------
c
— at
o a>
Q E
en a
^If
LLJ =
cc
Ui
il a
CO CO CO CO
°°. a> *H uj
.
l>^ <» i^ -^
(O
8888
CO
UJ
o co
co uj
^~ »~
O) CO
o K
CO
o!
Q E
CO CX
c o
; o
III!1
- 'jg » »J
zS
< CO
57
-------
o
a
co
DC
fS
a|
§i "5
1*
DC
HI
ii
S 11
!O CO CO CO
•* *•; «
i-g
CXI C\l
p
U)
es
_£
X X
!?S
m m
a. a.
V?
CO •*
N. r-- •»
p p cq
CO i- P CO
«=5-g'?
I 14
si
i°
-f
Q. g.
ll
o> g
P
!l
^
o.5
i
28
Q- tf
I
S
"S
CM
u-55
to
o •*
I p p
I
••* t
o
in
«!
a e
CO
DC
^
3
a.
1
in
i
s
CJ CO CO
CO IO CO
CO 10 ^
CD to CD
0.0.0.0.
to o r-
i
CO CO
8
I
I
2
F
o
If
Ol
"8
^s
li
3
_•
P
CO CO
Is
r- p
3Si
ft
^8
co4
•s
u
CO
MS
^«. _
S p;
•* "^
« p
) ^ w (d
•* co x:
a.
sv
S
|c
Hi
S S
i!
2 2-i
O> O) I
58
-------
O
U)
9
s
o S
Q E
w a
DC 2
"S
*
-
5 §
58
p 01
00
00
oo
O O
n co
Sin •r-
»- ^
oj to oq
c> ^ o cj
3
o
s
d
to *-
§
•5
8
S
a.
•3
ta .
0.3
o
o
to
cr
as
»5
m*
i
CO CO
00
o o
CO CO
> O CO
-333?
« * 1^ »~ H
§ =5
^-s
J«SjJ
a.
3
=5-*?
a.
•3
»
«
5
2 r- v. vj -^
II 111
'S CD OJ O
> g = l
•35*0
11
in
a.
59
-------
COMPLETENESS
Completeness entails an assessment of the total amount of valid data obtained for a program relative to
the amount of data that was planned to achieve project objectives. As summarized in Section 4, a total of 122 test
results for the various testing procedures was anticipated for the program. In fact, the total number of test results
collected in the program is 182 (see discussion at the. beginning of Section 6). Consequently, the completeness
objective for the project as defined by the number of experiments anticipated for completion has been satisfied.
The additional samples that were generated during the conduct of the laboratory study were directed primarily at
QC issues (e.g., additional duplicate measurements) and the effect of flow rates for wash water in the Inclined
Trough test.
REPRESENTATIVENESS
Representativeness of results is a non-quantifiable concept However, the intent is that all critical
measurements should be made so that results are representative for all samples and system conditions being
measured in a particular testing procedure. For this purpose, each sample in a specific testing procedure (e.g.,
the Inclined Trough and Swirling Coupon tests) was collected and processed in exactly the same way and manner
every time. The exact testing protocols are described in detail in the SOPs in Appendix A.
COMPARABILITY
Comparability also is a non-quantifiable concept that is intended to describe the capacity for test results
to be compared with each other for purposes of satisfying project objectives. Toward this end, samples were
always collected and processed in exactly the same way and manner every time for a given testing procedure to
yield data for the procedure that was comparable to another testing procedure with similar reporting units.
Furthermore, all experimental test results for a particular oil-cleaning agent-testing procedure combination were
performed at least in triplicate.
60
-------
SECTION 7
REFERENCES
Clayton, J.R., Jr. 1992. Chemical Shoreline Clean
Mechanisms of Action and Factors Influencing
Protection Agency, Risk Reduction Engineering Laboratory
and RREL, Office of Research and Development,
International Corporation, San Diego, CA 9212
ng Agents for Oil Spills: Update State-of-the-Art on
Performance. Final Report to The U.S. Environmental
(RREL), Releases Control Branch, Edison, NJ
Cincinnati, OH 45268 from Science Applications
48 p.
Fingas, M.F., K.A. Hughes, and M.A. Schweitzer.
Technology Division. Proc. Tenth Arctic Marir
Edmonton, Alberta, Canada. Conservation and
Fingas, M.F., G. Stoodley, G. Harris, and A. Hsia.
studies. Proc. Thirteenth Arctic Marine Oil Spi
1987a. Dispersant testing at the Environmental Emergencies
e Oilspill Program Technical Seminar. 9-11 June 1987,
Protection, Environment Canada, pp. 343-356.
1989. Evaluation of Chemical Beach Cleaners. Paper
presented at Cleanup Technology Workshop, Anchorage, AK, 28-30 November 1989.
Fingas. M.F., B. Kolokowski, and EJ. Tennyson. 1990. Study of oil spill dispersahts effectiveness and physical
1 Program Technical Seminar. 6-8 June 1990, Edmonton,
Alberta. Environment Canada, Ottawa, Ontario, pp. 265-287.
Fiocco, RJ. 1991. A new chemical beach cleaner for oiled shorelines - Corexit 9580. Personal correspondence
from R.J. Fiocco (Environmental, Marine & Safety Division, Exxon Research and Engineering Company,
Florham Park, NJ) to J. Clayton (Science Applications International Corporation, San Diego, CA). 35 p.
Rohlf, FJ. and R.R. Sokal. 1981. Statistical Tables. W.H. Freeman and Company, San Francisco. 219 p.
Sokal, R.R. and FJ. Rohlf. 1981. Biometry. W.H.
Freeman and Company, San Francisco. 859 p.
61
-------
APPENDIX A
STANDARD OPERATING PROCEDURES (SOPS)
62
-------
1. INCLINED TROUGH TEST FOR SHORELINE CLEANING AGENT PERFORMANCE
1.1 Summary of Method
The Inclined Trough Test is a modification of a procedure developed in Environment Canada's laboratory by
Fingas, et al. (Reference 1). In the procedure, a measured quantity of a reference test oil is applied to the inner
surface of a stainless steel or ceramic trough approximately 25-30 cm in length. The oiled trough is placed in a
horizontal position in a fume hood for 18 hours (ije., "overnight") to allow for (1) oil spreading on the substrate
surface, (2) evaporative loss of the volatile fractions in the oil, and (3) formation of an adhesive bond between oil
and substrate. After the 18-hour oil-to-substrate contact time, a measured quantity of a chemical shoreline
cleaning agent is applied to the oil in the trough a| a cleaning agent-to-oil ratio of 1:3 (v/v). Following a 10-
minute contact (or "soak") time (in a horizontal position) to allow for the cleaning agent to penetrate and mix
with the oil, the trough is inclined at a 45° angle. The surface of the treated oil is rinsed with two successive 5-
mL volumes of seawater, and the seawater runoff is collected in a small vial or beaker. The trough and the
seawater washings are separately extracted with methylene chloride (a.k.a. dichloromethane, or DCM), and the
extracts are analyzed by UV-visible absorption spectrophotometry at 340,370, and 400 nm analytical
wavelengths (as per the recommendations of Fingas, et al., Reference 2). Two separate estimates of shoreline
cleaning agent performance (estimated from (1) oil remaining on the trough and (2) oil released into the seawater
washes) are obtained from the trough and seawater extracts, respectively.
1.2 Experimental Apparatus and Chemicals
Inclined Trough Test Apparatus. The basic appar itus for the Inclined Trough Test consists of (1) sets of small
stainless steel or ceramic troughs measuring approximately 25-30 cm in length and 2-3 cm wide and having a
shallow V-shaped cross-section, (2) an adjustable prame capable of holding multiple troughs at angles from the
horizontal between 0° and 90°, (3) a set of positive-displacement pipettes for delivering precise, repeatabfe
volumes of the reference oils and shoreline cleaning agents, (4) a set of 5-10 cc syringes serve as the wash water
delivery systems, (5) a set of 250-mL beakers to act as receiving vessels for water washes, and (6) a
programmable timer. A proposed setup for the Inclined Trough Test apparatus is shown in Figure 1.
Test Monitoring Equipment. Mercury/glass thermometers graduated to the nearest 0.5 °C or better are used to
measure the ambient temperature of (1) the air over the troughs and (2) the synthetic seawater solution used in
the washing process.
Extraction Equipment VOA vials (40 mL capacity) are used to collect the DCM extracts from the troughs. Two
successive 10.0-mL volumes of DCM from a 10.0 mL gas-tight syringe are used to extract all remaining oil from
a trough into the VOA vial. Seawater extractions
•equire 250-mL separatory runnels with Teflon stopcocks.
UV-Visible Spectrophotometer. A UV-visible abs xption spectrophotometer capable of measuring absorbances at
340,370, and 400 nm is required for analytical measurements. An Hitachi Model U-2000 (or equivalent) is
acceptable.
Reference Test Oils. Two EPA standard reference
oils are used: (1) Bunker C; and (2) Prudhoe Bay crude.
These oils are obtained from me Industrial Chemicals Repository, EPA Environmental Monitoring Systems
Laboratory, Cincinnati, OH (James Longbottom, custodian). These oils have been thoroughly homogenized and
physically and chemically characterized for previous EPA studies. The density of each oil has been determined
by gravimetric measurements of a known volume of the oil at a measured temperature.
Shoreline Cleaning Agent Formulations. Three commercial shoreline cleaning agents are tested: (l)Corexit
9580; (2) Corexit 7664; and (3) Citrikleen XPC. These chemical agents are obtained from the Emergencies
Science Division, Environment Canada, Ottawa, Ontario, Canada (courtesy of Mervin F. Fingas) and from
63
-------
Figure 1. Schematic representation of the Inclined Trough test
apparatus.
64
-------
Exxon Chemical Company, Houston, Texas (courtesy of Kenneth
cleaning agent is determined by gravimetric measurements of a known
measured temperature.
Synthetic Seawater. A commercial salt recipe for j
Aquarium Systems, 8141 Tyler Blvd., Mentor, OH
aquarium store. The synthetic seawater is prepare
water to give a salt-water solution having a salinit)
in the sea salt mixture is included on the following page
Extraction Solvent. Pesticide-analysis quality methylene chloride
methanol are used for all oil extractions and standard
W. Becker). The density of each shoreline
volume of the cleaning agent at a
ynthetic seawater (Product: Instant Ocean, manufactured by
44060, Tel. (216) 255-1997) is obtained from a salt-water
by dissolving 68 g of the salt mixture in 2.00 L of distilled
of 34 parts per thousand (ppt). The list of major ingredients
(aJc.a. dichloromethane. or DCM) and
preparations for trough and seawater-wash samples.
1.3 Experimental Setup and Conduct
1.3.1 Pre-Test Preparations
The synthetic seawater used in the test procedure is prepared according to manufacturer's specifications with
distilled water and the synthetic sea salt mixture. Following its preparation, the seawater solution is allowed to
equilibrate to the ambient temperature in the laboratory. The temperature of the seawater is measured (to (the
nearest °C) and recorded immediately prior to use
in order to be used.
The reference oils and shoreline cleaning agents to
temperature in their respective, sealed containers.
1.3.2 Inclined Trough Test
A clean, dry trough is placed at an inclination angl
in the testing procedure and must be in the range of 20 ± 3 °C
be used in the test are also allowed to equilibrate to ambient
e of 0° to the horizontal. A 0.150-mL standard oil "slick" (ca.
5 cm long) is deposited in the bottom central portion of the trough using a positive-displacement pipette. The
trough is placed in a fume hood and allowed to stand at ambient temperature for 18 hours (i.e., "overnight"). The
temperature of the air over the trough is measured md recorded and must be in the range of 20 ± 3°C in oirder to
proceed with the test (The 18-hour oil-to-substrats contact time (OSCT) under ambient-temperature, flowing-air
conditions is intended to simulate "real-world" conditions and to allow for (1) oil spreading on the substrate
surface, (2) evaporative loss of the volatile fractions from the oil, and (3) formation of an adhesive bond between
oil and substrate.) At the end of this oil-substrate contact period, the temperature over the trough is again
measured and recorded. A 0.050-mL volume of a cleaning agent is applied uniformly to the surface of the oil in
the trough and a 10-minute contact (or "soak") time is observed to allow for the cleaning agent to penetrate and
mix with the oil. At the end of this cleaning agentjoil contact time (AOCT), the trough is attached to a test-
apparatus frame (e.g., see Figure 1) on which the inclination angle for the trough is set to 45°. The wash syringe
and wash receiver-vessel are set in place at the upper and lower ends of the trough, respectively. A 5-rnL volume
of synthetic seawater is added to the open barrel of the wash syringe and allowed to flow across the oil surface in
the trough at a rate determined by the aperture of the syringe exit port (ca. 1 mL/sec). After a 10-minute interval,
the water washing procedure is repeated with a second 5-mL volume of seawater. After a final 10-minute
interval, the syringe and receiver are removed fron
the apparatus frame.
The trough and the wash solution are now ready to be separately extracted with methylene chloride (DCM) in
preparation for UV-visible spectrophotometric analysis.
65
-------
MAJOR ION
HMOR ION COMPOSITION
OF
INSTANT OCEAN* SYNTHETIC SEA SALT
XTOTAL WEIGHT
IONIC CONCENTRATION
AT 34 °/oo SALINITY
Chloride
Sodium
Sulfate
Magnesium
Calcium
Potassium
Bicarbonate
Boron
Strontium
SOLIDS
rfatar
(CT)
(Na+)
(S04-)
(Hg^)
(Ca++)
(K+)
(HCQ3-)
(8)
(Sr**)
TOTAL
(H20)
47.470
26.280
6.S02
3.230
1.013
i.ois
0.491
0.015
.001
86.11*
13. aa
18,740
10,454
2,631
1,256.
400
401
194
6.0
7.5
34,089.SO
TOTAL
99.99X
66
-------
1.3.3 Summary of Test Operating Parameters
The operating parameters which must be monitor sd daily for the Inclined Trough Test are (1) me ambient air
temperature in the fume hood, and (2) the temperature of the synthetic seawater used in the washing process.
Acceptance ranges for these parameters are as follows:
Operating Parameter
Fume Hood Air Temperature:
Seawater Wash Temperature:
Acceptance Range
17 - 23f
17 - 23V
Air and seawater temperatures are measured at least once each day during the testing period with mercury/glass
thermometers graduated in 0.5°C increments or better. Actual performance tests on shoreline cleaning agents
should not be started if measured values for the preceding variables are not within the indicated acceptance
ranges.
1.4 Sample Collection and Sample Extraction
All trough and seawater wash samples from the Ir clined Trough Test are extracted with methylene chloride
(DCM) for analysis by UV-visible absorption spectrophotometry. The extraction procedures are described briefly
below.
1.4.1 Trough Extractions
After completion of the seawater washing procedi re, a 10-mL gas-tight, calibrated syringe is used to rinse all
remaining oil from the trough into a 40-mL VGA vial with two successive 10.0-mL volumes of DCM (i.e., total
DCM volume = 20.0 mL). The vial is fitted with a septum-seal screw cap to await analysis by UV-visible
spectrophotometry. As needed, dilution of the final extract with DCM is performed to assure that the absorbance
of the extract does not exceed that of the highest-cjoncentration oil standard.
1.4.2 Seawater Wash Extractions
Wash-water samples are extracted by a separatory funnel method that involves initial transfer of the water to a
250-mL glass separatory funnel (fitted with a Teflon stopcock). The inner surfaces of the wash-water collection
beaker is carefully rinsed with three small volumes of DCM (-2 mLs each), which also are added to the
separatory funnel. The funnel is vigorously shaken for 15 seconds and then allowed to stand in a stationary
position (stopcock end down) for 2 minutes to allow for phase-separation of the water and DCM. The DCM layer
is then drained from the separatory funnel into a 40-mL VGA vial (calibrated for 20 mLs). The DCM-exllraction
process of the water sample is repeated with two additional 5-mL volumes of DCM. All DCM volumes are
combined in the VGA vial, and the final volume adjusted to 20 mLs with additional DCM. The VGA vial is
fitted with a septum-seal screw cap to await analysis by UV-visible spectrophotometry. As needed, dilution of the
final extract with DCM is performed to assure that the absorbance of the extract does not exceed that of the
highest-concentration oil standard.
1.4.3 Sample Extraction Summary
Table 1 summarizes the sample extract volumes and their expected range of oil concentrations for a
corresponding range of shoreline cleaning agent performance values.
67
-------
Table 1. Expected Oil Concentration Ranges for Test Sample Extracts31
Sample
Type
Trough
Seawater
Total
Water
Volume
10 mL
Total
Oil
Volume
0.150 mL
0.150mL
Final DCM
Extract
Volume
50mLd
50mLd
Oil Concentration @
Pn. RAN =100%°
470 nm were
found to suffer from such limitations as lack of linear response on the spectrophotometer with varying oil
concentration, poor reproducibility of absorbance values, and/or poor sensitivity on the speatropnotorneter.
1.5.1 Preparation of Oil Standards
Calibration of the UV-visible spectrophotometer (Hitachi Model U-2000, or equivalent) is accomplished using six
concentrations of a particular reference oil. Since the presence of water and certain chemical agents in DCM can
affect absorbance readings, the oil/DCM calibration standards used for quantitation of oil in sample extracts are
prepared in a manner similar to that of the corresponding experimental sample.
In order to minimize the volumes of DCM required to prepare the calibration standards from reliably measured
amounts of reference oils, stock solutions of each reference oil and oil/cleaning agent combination are prepared
by mixing 51 uL of oil, 17 uL of cleaning agent, and 442 uL of DCM (i.e., a 1:10 v/v dilution of true oil and a
1:30 dilution of cleaning agent) for stock solutions used to prepare standards containing the cleaning agents
Corexit 9580 and Citrikleen XPC (hydrocarbon-based agents). Stock-standard solutions containing Corexit 7664
(a water-based agent) are prepared by mixing 51 uL of oil, 17 uL of cleaning agent, 100 uL of methanol, and 342
uL of DCM because of the water-based nature of this cleaning agent The shoreline-cleaning-agent-to-oil ratio
(SOR) in all standard solutions is 1:3 v/v, which is the same as that used for tests. A stock solution for oil
without a cleaning agent is prepared by mixing 51 uL of oil and 459 uL of DCM (i.e., a 1:10 v/v dilution of the
oil).
Calibration standards for the seawater wash samples and coupon extracts are prepared at the required
concentrations by (1) adding an appropriate quantity of a reference oil stock solution to a 30-mL volume of
seawater in a 250-mL separatory funnel, extracting the seawater with 3 sequential 5-mL volumes of DCM, and
68
-------
adjusting the combined volume for all DCM extracts to 20 mLs with additional DCM. Oil standards are made
for six concentrations in final extracts. Information on the calibration standard concentrations and volumes, and
on the quantities of reference oil and oil stock solutions involved in their preparation, is summarized in Table 2.
Table 2. Nominal Oil Concentrations for Crude Oil Calibration Standards3
Bunker C
Standard
6
5
4
3
2
1
PrudhoeBay
Standard
6
5
4
3
2
1
Final Oil
Concentra-tion
1.00 mg/mL
0.50 mg/mL
0.20 mg/mL
0.10 mg/mL
0.05 mg/mL
0.02 mg/mL
0.01 mg/mL
Final
Standard
Volume
20 mL
20 mL
20 mL
20 mL
20 mL
20 mL
20 mL
Total Oil
Mass in
Standard
19.8 mg
9.9 mg
4.0 mg
2.0 mg
1.0 mg
0.4 mg
0.2 mg
Oil-DCM Stock
Solution Volume
^ 220 uL
L_ 110 uL
44 uL
22 uL
lluL
4.4 uL
2.2 uL
a Assuming a nominal oil density of 0.9 g/mL and an extraction efficiency of 100% for oil from the seawater
matrix. Actual concentrations will differ slightly according to the measured densities of specific test oils (i.e.,
Bunker C and Prudhoe Bay crude).
The oil concentrations for the standards have been chosen for the following reasons: (1) they span the
measurable linear ranges for all of the analytical wavelengths; (2) they bracket oil concentrations that can
reasonably be expected in experiments for cleaning! agent performance values (e.g., approximately 5% to 100%;
see Table 2.1); and (3) the necessary volumes of a parent oil-DCM standard solution can be prepared quickly and
accurately using low-volume positive-displacement pipettes or gas-tight syringes. Furthermore, a common set of
standard solutions are used for both coupon and wash-water samples because preliminary studies have shown that
comparable absorbances are found for standards prepared with and without water extraction for the oils and
cleaning agents used for this study.
1.5.2 UV-Visible Spectrophotometric Measurer icnts
Absorbance measurements for all reference oil calibration standards and test sample extracts are made at tfoe
three analytical wavelengths - 340 nm, 370 nm, anjd 400 nm - using a double-beam scanning UV-visible
spectrophotometer Hitachi U-2000 or equivalent). The spectrophotometer is operated in double-beam mode.
Measurements are made using matched cuvettes, with extracts in the sample-beam cuvette and a reagent-DCM
blank in the reference-beam cuvette to correct for absorption by reagent-DCM. For all absorbance measurements,
the standards, sample extracts, and DCM blanks are introduced into the specttophotonieter in a standard 1.0-cm
cuvette.
j ' _ .... _
Prior to analysis of standards or samples, a scan of I he wavelength region that includes the selected analytical
wavelengths is conducted on a certified spectrophot ametric reference standard (Oxford Spectro-Chek Solutions I,
II, ffl, or IV; or equivalent) to monitor relative absorbance accuracy and long-term instrument stability. The
peak absorbance for the standard scanned must be within ±5% of the initial peak absorbance recorded for tiiiat
standard in order to proceed with external oil standard calibration and sample analysis, specifications for trie
Oxford Spectro-Chek solutions are included below.
69
-------
1.5.3 Calibration Linearity and Stability
The UV-visible absorption spectrophotometer must pass a calibration linearity criterion before analyses of sample
extracts can begin. Instrument response is demonstrated to be linear by calculating response factors for the
standard concentrations specified in Table 2 for each oil and oil/cleaning agent combination. Response factors
(RFs) for these oil calibration standards are determined at each of the three analytical wavelengths (i.e., 340.370,
and 400 nm). RFs are calculated as in equation (1):
RFX = oil concentration/A,
(1)
where RFX = response factor at wavelength x (x=340,370, or 400 nm),
oil concentration = mg of oil/mL of DCM in standard solution, and
Ax = spectrophotometric absorbance at wavelength x (x=340,370, or 400 nm), corrected for DCM
background.
The calibration linearity is acceptable when individual RFs are less than 20% different from the mean RF for
absorbance values that are greater than the background absorbance for a DCM blank and less than the
absorbance saturation of the spectrophotometer (i.e., an absorbance of 3.5) at each of the three analytical
wavelengths. The calibration linearity criterion is defined by the following inequality:
Initial Calibration: -0.20 < (RFj - RFm)/RFm < 0.20
where RFj = individual UV-visible response factor measured at a specific wavelength, and
Fm = mean of the UV-visible response factors plotted on the calibration curve.
If the linearity criterion defined above is satisfied, analyses of sample extracts may begin. Figure 2 illustrates
examples of plots of oil calibration standard RF values and absorbances versus oil concentration, for the three
analytical wavelengths. ,
If the linearity criterion is not satisfied by one or more of the relevant calibration standards, each standard in
question can be prepared a second time and the new standard tested. If the new standard passes the linearity test
when used to replace the failed standard (i.e., if its RF values differ from the corresponding mean RF values by
less than 20%), then the linearity criterion is satisfied and sample analysis can begin. If the replacement
standard also results in a failure of the linearity test, sample analyses cannot begin until the source of the problem
(e.g., the preparation protocol for the oil standards, instrument instability, etc.) has been determined and
corrected.
The initial six-point calibration of the UV-visible spectrophotometer is required at least once per day.
The stability of the spectrophotometer calibration is determined by reanalyzing a midrange calibration standard
after every 20 or fewer sample analyses. Calibration stability is acceptable if the individual RF for the midrange
standard is less than 20% different from the mean RF for absorbance values that are greater than tlte background
absorbance for a DCM blank and less than the absorbance saturation of the spectrophotometer (i.e., an
absorbance of 3.5) at each of the three analytical wavelengths.
If the stability criterion defined above is satisfied, analyses of sample extracts may continue. If the RF for the
midrange standard is > 20% different from the mean RF, the analysis of the midrange standard may be repeated.
If the stability criterion is still not satisfied, a new six-point calibration should be performed, and flic samples
analyzed after the last satisfactory calibration stability check must be reanalyzed.
70
-------
RF at 400 nm(mg oil/abs unit) RF at 340 nm (mg oil/abs unit)
° . S 2 8- S . S C. § S 2 S SS C
RF linearity: 340 nm absorbance |
-
a n
a
• !
1 indHm^al RF* me«n_RF mean RFW-20%
i . i
0.05 0.10 050 0.50 1.00 2.00
oil standard concentration (mg/mL)
RF linearity: 400 nm absorbance |
•
n a a
a
a
-
1 indvfcfcaf RF* m*jn_RF mm RFy-20%
at 370 nm (mg oil/abs unit)
O 0 O -• —
*• b> in o (o
u_
cr
05
0.0
RF linearity: 370 nm absorbance |
-
n 0
a
individyJ RF* mean RF mean RF+/-20%
0.05 0.10 050 0.50 1.00 2.00
oil standard concentration (mg/mL)
standard absorbances at 3 wavelengths |
JJ3.0
«
absorbance (blank-corn
3 P -* -. j\> (v,
3 (n o cn o cn
absorbance saturation
• //V"°
6f»*' 34Qnm 370jyn 40
-------
1.5.4 Oil Quantitation in Test Samples
Concentrations of oil in the final DCM extracts for both trough and seawater samples are calculated from the
UV-visible absorbance readings (Ax) at the three analytical wavelengths
(x=340.370. or 400 nm) using equation (2):
(2)
where Cx = concentration of oil in sample extract at wavelength x (x = 340.370. or 400 nm),
Ax = absorbance of sample extract measured at wavelength x. corrected for DCM background.
RFX = mean response factor for oil extract standards at wavelength x (equation 1),
Vf = final dilution volume for sample extract requiring dilution, and
Vj = initial aliquot volume before dilution.
The oil concentrations at each of the three analytical wavelengths, C34Q, C37Q, and CQQQ, are then averaged
using equation (3):
-mean
(3)
The wavelength-averaged mean concentration values. Cmean, obtained with equation (3) are used for all
subsequent cleaning performance calculations. Sample results for which one of the values for C34Q, C37Q. and
C4oo is more than 20% different from Cmean should be flagged. For all sample results so flagged, the raw data
should be reviewed and evaluated to establish that the measurements are valid. In addition, attempts will be
made to correlate the difference to the oil type, substrate type, cleaning agent, or testing procedure used. If no
errors or correlations are apparent and <10% of all oil measurements are flagged, the mean concentration data
should be used in the cleaning performance calculations, and the subject data should be flagged.
1.5,5 Cleaning Agent Performance Calculations
Determinations of shoreline cleaning agent performance are made by calculating the percentage of the total mass
of oil on a substrate surface that is removed from that surface as a result of the application and action of a
chemical cleaning agent Calculations for cleaning performance determinations for each oil/substrate/cleaning
agent combination and each testing procedure are made in three steps. First, performance values are determined
from the results for each of the sample extracts (i.e., for substrate and water samples) from a given cleaning agent
test. Cleaning agent performance values are calculated using equation (4a) for water sample extracts and
equation (4b) for substrate extracts:
<4a>
(4b)
pagent (%> = 10° x (Cmean x vDCM)/mTOT
Pagent <%> = 100 x [1 -
where Pagent = cleaning performance value for the cleaning agent test,
cmean = wavelength-averaged concentration of oil in the sample extract (from equation (3)),
VDCM = fi"31 voton* of DCM extract of the sample (i.e., 50 mL, as specified in Section 1.4) , and
m JOT = total mass °f °H on t"6 substrate prior to cleaning agent application.
In the second step, performance values are calculated from the results for each of the sample extracts (i.e., for
substrate and water samples) from the corresponding "no cleaning agent" control test (i.e., for the same
oil/substrate combination and testing procedure) using equation (4c) for the water extract and equation (4d) for
the substrate extract:
72
-------
Pcontrol (%> = 10<> x (Cmean * vDCM)/mTOT (4c)
Pcontrol (%> = "JO* [1 - 10% of all mean oil concentrations (Cmean) based on the absorbances at 340,370, and 400 nm (Cj^, C37o.
and C4oo) are flagged (i.e., Cx more than 30% different from Cmean), data from each wavelength should be
treated as a separate data set in calculating cleaning performance. Separate treatment of each wavelength will
require that cleaning performance data are generated separately for absorbance measurements at 340,370, and
400 nm. Cleaning performance data from individual wavelengths should then be compared to see if there is any
bias associated with an individual wavelength. Cleaning performance values (Px) determined from the
concentrations measured at each individual wavelength (x = 340,370, and 400 nm) should be compared on the
basis of their relative percent difference (RPD) values with the cleaning performance determined using the mean
concentration (Pmean), as given in equation (6):
RPD (%) = 100 x [(Px - Pmean)/Pmean] (6)
L6 Quality Control Procedures
In addition to the calibration linearity and stability checks described in Section 1.5, quality control (QQ for the
Inclined Trough Test for shoreline cleaning agent
it pert
formance is established by the following two procedures.
1.6.1 Spectrophotometric Measurement Precision
At least 5% of all UV-visible Spectrophotometric n teasurements are performed in triplicate as a QC check on the
precision of the analytical measurement method. Triplicate absorbance measurements of a randomly selected
*_...« . • . .__ ..
sample are to be non-consecutive and are required
1.6.2 Method Blank Analysis
to have a relative standard deviation (RSD) of <5%.
Experimental method blanks are performed at a frequency of at least one blank for every eight oil/chemical agent
tests performed (i.e., a frequency of ca. 12.5%). An experimental method blank involves the complete conduct of
the test protocol (i.e., the full execution of a test with seawater, trough, etc.), except that no oil or shoreline
cleaning agent are used for the test The oil concentration measured in an experimental method blank must be
<5% of that occurring with a cleaning performance of 100%. Method blanks with oil concentrations meeting the
acceptance criterion (i.e., <5%) are used to indicate that the testing system is sufficiently clean of contamination
between experimental runs.
73
-------
1.7 References
1. Fingas, M.F., G. Stoodley, G. Harris, and A. Hsia. 1989. Evaluation of Chemical Beach Cleaners. Paper
presented at Cleanup Technology Workshop, Anchorage, AK, 28-30 November 1989.
2. Fingas, M.F.. K. A. Hughes, and M.A. Schweitzer. 1987. Dispersant Testing at the Environmental
Emergencies Technology Division. Proc. Tenth. Arctic Marine Oilspill Program Technical Seminar. 9-11 June
1987, Edmonton, Alberta, Canada. Conservation and Protection, Environment Canada, pp. 343-356.
3. Clayton, J.R., Jr. 1992. Chemical Shoreline Cleaning Agents for Oil Spills: Update State-of-the-Art on
Mechanisms of Action and Factors Influencing Performance. Final Report to The U.S. Environmental Protection
Agency, Risk Reduction Engineering Laboratory (RREL), Releases Control Branch, Edison, NJ and RREL,
Office of Research and Development, Cincinnati, OH 45268 from Science Applications International
Corporation, San Diego, CA 92121. 48 p.
74
-------
2. SWIRLING COUPON TEST FOR SHORELINE CLEANING AGENT PERFORMANCE
2.1 Summary of Method
The Swirling Coupon Test for shoreline cleaning agent performance has its conceptual origin in the Swirling
Flask Test for oil dispersant performance developed in Environment Canada's laboratory (Reference 1), and in
the recent oil dispersant test evaluations conducted by SAIC for the EPA and summarized by Clayton, et aJL
(Reference 2). In this procedure, a measured quantity of a reference test oil is applied to the surface of a small
stainless steel or ceramic "coupon" approximately 2 J cm x 2.5 cm. The oiled coupon is placed in a fume hood
for 18 hours (i.e., "overnight") to allow for (1) oil spreading on the substrate surface, (2) the evaporative loss of
the volatile fractions in the oil, and (3) the formation of an adhesive bond between oil and substrate. After this
18-hour oil-to-substrate contact time, the coupon isj removed from the hood and attached to one of the mounting
arms of the test apparatus. A measured quantity of a chemical shoreline cleaning agent is applied to the oil at a
cleaning agent-to-oil ratio of 1:3 (v/v). Following a 10-minute contact (or "soak") time to allow for the cleaning
agent to penetrate and mix with the oil, the coupon is mounted on a retaining bracket and lowered into a beaker
containing 250 mL of seawater. The coupon is me ;hanically swirled for 2 minutes in the seawater, using an
orbital shaker table to provide the mechanical swirl ing motion. Following the swirling period, the coupon is
withdrawn and allowed to drain into the seawater beaker. The coupon and the seawater are separately extracted
with methylene chloride (aJc.a. dichloromethane, o
: DCM), and the extracts are analyzed by UV-visible
absorption spectrophotometry at 340,370, and 400 nm analytical wavelengths (as per the recommendations of
Fingas, et al.. Reference 1). Two separate estimate 5 of performance of the shoreline cleaning agent (estimated
from (1) oil remaining on the coupon and (2) oil re eased into the seawater matrix) are obtained from the coupon
and seawater extracts, respectively.
2.2 Experimental Apparatus and Chemicals
Swirling Coupon Test Apparatus. The basic appan tus for the Swirling Coupon Test consists of (1) a variable-
speed orbital shaker table, (2) an apparatus frame with coupon mounting arms capable of being raised and
lowered (for attachment and immersion of substrate coupons), (3) sets of stainless steel or ceramic coupons;
measuring 2.5 cm x 2.5 cm, (4) a set of positive-displacement pipettes for delivering precise, repeatable volumes
of the reference oils and shoreline cleaning agents, (5) a set of 400- or 600-mL beakers to serve as wash-water
containers for the swirling process, (6) a set of retaining brackets to hold the beakers in place during swirling,
and (7) a programmable timer. A schematic setup for the Swirling Coupon Test apparatus is shown in Figure 1.
Monitoring Equipment. Mercury/glass thermometers graduated to the nearest 0.5 °C or better are used to
measure the ambient temperature of (1) the air over the test coupons and (2) the synthetic seawater solution used
in the washing process.
Extraction Equipment VOA vials (40 mL capacity) are used to collect the DCM extracts from the coupons.
Two successive 10.0-mL volumes of DCM from a 10.0-mL gas-tight syringe are used to extract all remaining oil
from a coupon into the VOA vial. Seawater extractions require 250-mL separatory funnels with Teflon
stopcocks.
UV- Visible Spectrophotometer. A UV-visible abso ption spectrophotometer capable of measuring absorbainces at
340,370, and 400 nm is required for analytical measurements. An Hitachi Model U-2000 (or equivalent) is
acceptable.
Reference Test Oils. Two EPA standard reference oils are used: (1) Bunker C; and (2) Prudhoe Bay crude.
These oils are obtained from the Industrial Chemicalls Repository, EPA Environmental Monitoring Systems
Laboratory, Cincinnati, OH (James Longbottom, custodian). These oils have been thoroughly homogenized and
physically and chemically characterized for previou; EPA studies. The density of each oil has been determined
75
-------
orbital motion for beakers
on shaker table
stationary
support rack
for coupons
Figure 1. Swirling coupon test apparatus.
76
-------
by gravimetric measurements of a known volume of the oil at a measured temperature.
Shoreline Cleaning Agent Formulations. Three commercial shoreline cleaning agents are tested: (l)Corexit
9580; (2) Corexit 7664; and (3) Citrikleen XPC. These chemical agents are obtained from the Emergencies
Science Division, Environment Canada, Ottawa, Ontario, Canada (courtesy of Mervin F. Fingas) and from
Exxon Chemical Company, Houston, Texas (courtesy of Kenneth W. Becker). The density of each shoreline
cleaning agent is determined by gravimetric measurements of a known volume of the cleaning agent at a
measured temperature.
Synthetic Seawater. A commercial salt recipe for synthetic seawater (Product: Instant Ocean, manufactured by
Aquarium Systems, 8141 Tyler Blvd., Mentor, OH
aquarium store. The synthetic seawater is prepared.
water to give a salt-water solution having a salinity of 34 parts per thousand (ppt). The list of major ingredients
contained in the sea salt mixture is included on the
following page.
Extraction Solvent. Pesticide-analysis quality metHylene chloride
methanoi are used for all oil extractions and standa
44060, Tel. (216) 255-1997) is obtained from a salt-water
by dissolving 68 g of the salt mixture in 2.00 L of distilled
(aJc.a. dichloromethane, or DCM) and
d preparations for coupon and seawater-wash samples.
2.3 Experimental Setup and Conduct
2.3.1 Pre-Test Preparations
The synthetic seawater used in the test procedure is prepared according to manufacturer's specifications with
distilled water and the synthetic sea salt mixture. Following its preparation, the seawater solution is allowed to
equilibrate to the ambient temperature in the laboratory. The temperature of the seawater is measured (to the
nearest °Q and recorded immediately prior to use in the testing procedure and must be in the range of 20 ± 3°C
in order to be used.
The reference oils and shoreline cleaning agents to
temperature in their respective, sealed containers.
2.3.2 Swirling Coupon Test
A 0.048-mL standard oil "slick" is deposited evenlj
be used in the test are also allowed to equilibrate, to ambient'
for 18 hours (i.e., "overnight"). The temperature 01
under ambient-temperature, flowing-air conditions
on the surface of a clean, dry coupon using a positive-
displacement pipette. The oiled coupon is placed in a fume hood and is allowed to stand at ambient temperature
the air over the coupons is measured and recorded, ami must
be in the range of 20 ± 3°C in order to proceed with the test (The 18-hour oil-to-substrate contact time (OSCT)
is intended to simulate "real-world" conditions and to allow
for (1) oil spreading on the substrate surface, (2) th; evaporative loss of the volatile fractions in the oil, and (3)
the formation of an adhesive bond between oil and jsubstrate.) At the end of this oil-substrate contact period, the
temperature over the coupon is again measured and recorded. The coupon is then removed from the hood and
0.016 mL of the cleaning agent being tested is applied uniformly to the surface of the oil on the coupon^ A 10-
minute contact (or "soak") time is observed to allow for penetration and mixing of the cleaning agent into oil.
The coupon is then attached to a mounting arms over the shaker table. A beaker containing 250 mL of synthetic
seawater is inserted into the retaining bracket on the shaker table beneath the mounted coupon. After the coupon
is lowered into the beaker, the shaker table is immediately turned on and die coupon is swirled in the seawater for
2 minutes. At the end of this swirling period, the shaker table is turned off and the coupon is withdrawn and
allowed to drain over the beaker. After draining, the coupon is detached from its mounting arm and the beaker is
removed from its retaining bracket
The coupon and the wash solution are now ready to be separately extracted with methytene chloride (DCM) in
77
-------
MAJOR ION COMPOSITION
OF
INSTANT OCEAN* SYNTHETIC SEA SALT
IONIC CONCENTRATION
AT 34 °/oo SALINITY
(mo/IV
Chloride
Sodium
Sulfate
Magnesium
Calciua
Potassiua
Blcartonatt
Boron
Strontium
SOLIDS
tattr
(cr)
(Ma*)
(S04-)
(Mg*+)
(Ca~)
(K+)
(HC03-)
(8)
(Sr^)
TOTAL
(H20)
47.470
26,280
6,602
3.230
1,013
1.015
0.491
0,015
rooi
85.11%
is. aa
18.740
10,454
2.631
1,256
400
401
194
6.0
7.5
34,089,50
TOTAL
99.9951
78
-------
preparation for UV-visible spectrophotometric an; Jysis.
2.3.3 Summary of Test Operating Parameters
The operating parameters that are monitored dailj for the Swirling Coupon Test are (1) the ambient air
temperature in the fume hood, (2) the temperature of the synthetic seawater used in the washing process, and (3)
the shaker table rate. Acceptance ranges for these
Operating Parameter
Fume Hood Air Temperature:
Seawater Wash Temperature:
Shaker Table Rate:
three parameters are as follows:
Acceptance Range
17-23°C
17 - 23
-------
2.43 Sample Extraction Summary
Table 1 summarizes the sample extract volumes and their expected range of oil concentrations for a
corresponding range of shoreline cleaning agent performance values.
Table 1. Expected Oil Concentration Ranges for Test Sample Extracts3
Sample
Type
Coupon
Seawater
Total
Water
Volume
—
250 mL
Total .
Oil
Volume
0.050 mL
0.050 mL
Final DCM
Extract
Volume
20 mL
20 mL
Oil Concentration @
Pn.F.AN=100%b
470 nm were
found to suffer from such limitations as lack of linear response on the spectrophotometer with varying oil
concentration, poor reproducibility of absorbance values, and/or poor sensitivity on the spectrophotometer.
2.5.1 Preparation of Oil Standards
Calibration of the UV-visible spectrophotometer (Hitachi Model U-2000, or equivalent) is accomplished using six
concentrations of a particular reference oiL Since the presence of water and certain chemical agents in DCM can
affect absorbance readings, the oil/DCM calibration standards used for quantitation of oil in sample extracts are
prepared in a manner similar to that of the corresponding experimental sample.
In order to minimize the volumes of DCM required to prepare the calibration standards from reliably measured
amounts of reference oils, stock solutions of each reference oil and oil/cleaning agent combination are prepared
by mixing 51 uL of oil, 17 uL of cleaning agent, and 442 aL of DCM (i.e., a 1:10 v/v dilution of the oil and a
1:30 dilution of cleaning agent) for stock solutions used to prepare standards containing the cleaning agents
Corexit 9580 and Citriklecn XPC (hydrocarbon-based agents). Stock-standard solutions containing Corexit 7664
(a water-based agent) are prepared by mixing 51 uL of oil, 17 uL of cleaning agent, 100 uL of methane!, and 342
uL of DCM because of the water-based nature of this cleaning agent The shoreline
-------
Calibration standards for the seawater wash sample s and coupon extracts are prepared at the required
concentrations by (1) adding an appropriate quantity of a reference oil stock solution to a 30-mL volume of
seawater in a 230-mL separatory funnel, extractinglthe seawater with 3 sequential 5-mL volumes of DCM, and
adjusting the combined volume for all DCM extracts to 20 mLs with additional DCM. Oil standards are made
for six concentrations in final extracts. Information on the calibration standard concentrations and volumes, and
on the quantities of reference oil and oil stock solutions involved in their preparation, is summarized in Table 2.
Table 2. Nominal Oil Concentrations for Crude Oil Calibration Standards3
Oil Standard
6
5
4
3
2
1
Final Oil
Concentration
l.OOmg/mL
0.50 mg/mL
0.20 mg/rnL
0.10m«/mL
0.05 msJmL
0.02 mz/mL
J Final
Standard
Volume
]20mL
!20mL
|20mL
|20mL
|20mL
|20mL
Total Oil
Mass in
Standard
19.8 mg
9.9 mg
4.0 mg
2.0 mg
1.0 mg
0.4 mg
Oil-DCM Stock
Solution Volume
220 uL
110 uL
44 uL
22 uL
lluL
44uL
a Assuming a nominal oil density of 0.9 g/mL and ah extraction efficiency of 100% for oil from the seawatcr
matrix. Actual concentrations will differ slightly according to the measured densities of specific test oils (Le.,
Bunker C and Prudhoe Bay crude).
The oil concentrations for the standards have been c hosen for the following reasons: (1) they span the
measurable linear ranges for all of the analytical wavelengths; (2) they bracket oil concentrations that can
reasonably be expected in experiments for cleaning kgent performance values (e.g., approximately 5% to 100%;'
see Table 1); and (3) the necessary volumes of a parent oil-DCM standard solution can be prepared quickly and
accurately using low-volume positive-displacement pipettes or gas-tight syringes. Furthermore, a common set of
standard solutions are used for both coupon and wash-water samples because preliminary studies have shown that
comparable absorbances are found for standards pre pared with and without water extraction for the oils and
cleaning agents used for this study.
2.5.2 UV-Visible Spectrophotometric Measuren ents
Absorbance measurements for all reference oil calib ration standards and test sample extracts are made at the
three analytical wavelengths - 340 nm, 370 nm, and 400 nm - using a double-beam scanning UV-visible
spectrophotometer Hitachi U-2000 or equivalent). The spectrophotometer is operated in double-beam mod.
Measurements are made using matched cuvettes, widi«xtracts in the sample-beam cuvette and a reagent-DCM
blank in the reference-bean cuvette to correct for absorption by reagent-DCM. For all absorbance measurements,
the standards, sample extracts, and DCM blanks are introduced into the spectrophotometer in a standard 1.0-cm
cuvette.
Prior to analysis of standards or samples, a scan of the wavelength region that includes the selected analytical
wavelengths is conducted on a certified Spectrophotometric reference standard (Oxford Spectro-Chek Solutions I.
II, III, or IV; or equivalent) to monitor relative absorbance accuracy and long-term instrument stability. The
peak absorbance for the standard scanned must be within ±5% of the initial peak absorbance recorded for that
standard in order to proceed with external oil standard calibration and sample analysis. Specifications for the
Oxford Spectro-Chek solutions are included below.
81
-------
2.5.3 Calibration Linearity and Stability
The UV-visible absorption spectrophotometer must pass a calibration linearity criterion before analyses of sample
extracts can begin. Instrument response is demonstrated to be linear by calculating response factors for the
standard concentrations specified in Table 2 for each oil and oil/cleaning agent combination. Response factors
(RFs) for these oil calibration standards are determined at each of the three analytical wavelengths (Le., 340,370,
and 400 nm). RFs are calculated as in equation (1):
RFX = oil concentration/Ax (1)
where RFX = response factor at wavelength x (x=340,370, or 400 nm),
oil concentration = mg of oil/mL of DCM in standard solution, and
Ax = spectrophotometric absorbance at wavelength x (x=340,370, or 400 nm), corrected for DCM
background.
The calibration linearity is acceptable when individual RFs are tess man 20% different from the mean RF for
absorbance values that are greater than the background absorbance for a DCM blank and less than the
absorbance saturation of the spectrophotometer (i.e., an absorbance of 3.5) at each of the three analytical
wavelengths. The calibration linearity criterion is defined by the following inequality:
Initial Calibration: -0.20 < (RFj - RFm)/RFm < 0.20
where RFj = individual UV-visible response factor measured at a specific wavelength, and
RFm = mean of the UV-visible response factors plotted on the calibration curve.
If the linearity criterion defined above is satisfied, analyses of sample extracts may begin. Figure 2 illustrates
examples of plots of oil calibration standard RF values and absorbances versus oil concentration, for the three
analytical wavelengths.
If the linearity criterion is not satisfied by one or more of the relevant calibration standards, each standard in
question can be prepared a second time and the hew standard tested. If the new standard passes the linearity test
when used to replace the failed standard (i.e.. if its RF values differ from the corresponding mean RF values by
less than 20%). then the linearity criterion is satisfied and sample analysis can begin. If the replacement
standard also results in a failure of the linearity test, sample analyses cannot begin until the source of the problem
(e.g.. the preparation protocol for the oil standards, instrument instability, etc.) has been determined and
corrected.
The initial six-point calibration of the UV-visible spectrophotometer is required at least once per day.
The stability of the spectrophotometer calibration is determined by reanalyzing a midrangc calibration standard
after every 20 or fewer sample analyses. Calibration stability is acceptable if the individual RF for the midrange
standard is less than 20% different from the mean RF for absorbance values that are greater than the background
absorbance for a DCM blank and less than the absorbance saturation of the spectrophotometer (i.e., and
absorbance of 3.5) at each of the three analytical wavelengths.
If the stability criterion defined above is satisfied, analyses of sample extracts may continue. If the RF for the
midrange standard is > 20% different from the mean RF. the analysis of the midrange standard may be repeated.
If the stability criterion is still not satisfied, a new six-point calibration must be performed, and tfie samples
analyzed after the last satisfactory calibration stability check must be reanalyzed.
82
-------
RF linearity: 340 nm absorbance j
15
1.0
1
o
0.8
.0.6
O
C? 0.4
00
Li-
ne
02
0.0
individjj^ HF« ™«n_HF maw RFW-20%
0.05 0.10 050 0.50 1.00 2.00
oil standard concentration (mg/mL)
RF linearity: 400 nm absorbance I
1.0
•5
o>
£0.6
0.4
a
u.
ac
05
0.0
RF* mMn RF rraan RFW-20%
'
0.05 0.10 050 0^0 150 2.00
oil standard concentration (mg/mL)
Figure 2. Example of spectrophotometric data for
RF linearity: 370 nm absorbance j
!.£
1.0
C
I0-8
'o
O)
,§.0.6
E
c
CO 0.4
a»
u.
QC
nn
-
-
• =-'-_
a
a "
"
-
individual RF* nuanHF me*n RFV-20% I
' • • • -i 1 • — «—
0.05 0.10 050 0.50 1.00 2.00
oil standard concentration (mg/mL)
standard absorbances at 3 wavelengths ||
4.0
3.5
|,.
i
I"
CO *5 ft
1.5
1.0
0:5
absorbance saturation
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
oil standard concentration (mg/mL)
Prudhoe Bay crude
oil.
83
-------
2.5.4 Oil Quantitation in Test Samples
Concentrations of oil in the final DCM extracts for both coupon and seawater samples are calculated from the
UV-visible absorbance readings (Ax) at the three analytical wavelengths(x=340, 370, or 400 nm) using equation
(2):
Cx = (Ax)x(RFx)xCvYVi) (2) .
where Cx = concentration of oil in sample extract at wavelength x (x = 340, 370. or 400 nm),
Ax = absorbance of sample extract measured at wavelength x. corrected for DCM background,
RFX = mean response factor for oil extract standards at wavelength x (equation 1),
Vf = final dilution volume for sample extract requiring dilution, and
Vj = initial aliquot volume before dilution.
The oil concentrations at each of the three analytical wavelengths, C^Q, C37o, and C400, are then averaged
using equation (3):
Cmean- = 10° x
= 100 x [1 - (Cmean x V1Xatf)/mTOT] <4b)
agen
where PaKent = cleaning performance value for the cleaning agent test,
Cmean = wavelength-averaged concentration of oil in the sample extract (from equation (3)),
VDCM = final volurae of DCM extract of the sample (i.e., 20 mL, as specified in Section 2.4) , and
mTOT = total mass of oil on "* substrate P™* to cleaning agent application.
In the second step, performance values are calculated from the results for each of the sample extracts (i A., (at
substrate and water samples) from the corresponding "no cleaning agent" control test (i.e., for the same
oil/substrate combination and testing procedure) using equation (4c) for the water extract and equation (4d) for
the substrate extract:
84
-------
Pcontrol (*> = "JO x (Cmean x
(4c)
Pcontrol (*) = 100 x [1 - (Cmean x VDGM)/mTOr](4d)
where Pcontrol= cleaning performance value for the control test.
Values of Pcontrol measure the effects of water washing action independent of effects due to a chemical agent
In the third and final step, the cleaning performance due only to the application and action of the cleaning agent
is determined as the difference of the performance values from the corresponding cleaning agent and control
is determined as the difference of the performance
tests, using equation (5):
= pagent ' pcontrol <5>
where PCLEAN = net cleaning performance valu
s for the shoreline cleaning agent.
If > 10% of all mean oil concentrations (Cmean) has ed on the absorbances at 340, 370, and 400 nm ^349, C-yjQ,
and C4Q0) are flagged (i.e., Cx more than 30% different from Cmean), data from each wavelength should be
treated as a separate data set in calculating cleaning performance. Separate treatment of each wavelength will
require that cleaning performance data are generatep separately for absorbance measurements at 340, 370, and
400 nm. Cleaning performance data from individual wavelengths should then be compared to see if there is any
bias associated with an individual wavelength. Cleaning performance values (Px) determined from the
concentrations measured at each individual wavelength (x = 340, 370, and 400 nm) should be compared ori the
basis of their relative percent difference (RPD) values with the cleaning performance determined using the mean
concentration (Pmean), as given in equation (6):
RPD (%) = 100 x [
-------
2.7 References
1. Fingas, M.F., K.A. Hughes, and M.A. Schweitzer. 1987. Dispersant Testing at the Environmental
Emergencies Technology Division. Proc. Tenth Arctic Marine Oilspill Program Technical Seminar. 9-11 June
1987, Edmonton, Alberta, Canada. Conservation and Protection, Environment Canada, pp. 343-356.
2. Clayton, J.R., Jr. and J.R. Payne. 1992. Chemical Oil Spill Dispersants: Update State-of-the-Art on
Mechanisms of Action and Factors Influencing Performance with Emphasis on Laboratory Studies. Final Report
to The U.S. Environmental Protection Agency, Risk Reduction Engineering Laboratory (RREL), Releases
Control Branch, Edison, NJ and RREL, Office of Research and Development, Cincinnati, OH 45268 from
Science Applications International Corporation, San Diego, CA 92121. 83 p.
3. Clayton, J.R., Jr. 1992. Chemical Shoreline Cleaning Agents for Oil Spills: Update State-of-the-Art on
Mechanisms of Action and Factors Influencing Performance. Final Report to The U.S. Environmental Protection
Agency, Risk Reduction Engineering Laboratory (RREL), Releases Control Branch, Edison, NJ and RREL,
Office of Research and Development, Cincinnati, OH 45268 from Science Applications International
Corporation, San Diego, CA 92121. 48 p.
86
-------
APPENDIX B
MSDS/INFORMATION - OILS, CHEMICAL CLEANING AGENTS, AND SEA WATER
87
-------
REFERENCE VALUES
Prudhoe 3ay Crude Oil
This oil Has been analyzed by skilled oil testing and research laboratories to
characterize it and to ensure that.substantial comoositional changes have not
occurred during storage and sample preparation. Results for various selected
parameters were as follows:
Analvte
Result
Specific gravity*
API gravity*
Sulfur
Sulfur compounds, profile.
Nitrogen
Vanadium
Nickel
Simulated distillation profile
Infrared spectrum
UV fluorescence spectrum
Pour point
Viscosity,
at 40°C
at 100°C
Index
0.894 <
-------
Fuel Oil No. 6, 3unker
This oil has been analyzed by skilled.
REFERENCE VALUES
C Residual fhigh viscosity}
oil testing and research laboratories to
inii oil piQ3 'jecii aiiaijricu
-------
Corexit 9580
Shoreline Cleaner
General Information
Corexit 9580 is a balanced formulation of spe-
cially selected biodegradable surfactants in a
low toxicity. de-aromatized hydrocarbon solvent
system. Corexit 9580 has been specifically
designed to remove a wide range of crude oil
and petroleum product residues from contami-
nated shoreline surfaces. The uniqus formula-
tion characteristics of Corexit 9580 promote the
re-mobilization ana displacement of thick, vis-
cous hydrocarbon residues from coated sur-
faces, especially those involving asphaltic and
paraffinic components.
Fundulu* hmrecinus
ArtaxTua Mlina
Corexit 9580 has been designed to have a low
degreeof toxicity to marine and shoreline organ-
isms. Additionally, in response to environmental
concerns. Corexit 9580 has been formulated to
leave the treated hydrocarbons in a fonn which
can readily be recovered by typical skimming TOXIClty
methods. Application of .Corexit 9580 will not
disperse the oil residue into the inter-tidal water
column. . Species
Application Techniques
Corexit 9580 can be applied by the use of hand-
held sprayers or small hoses directly to the oiled
shoreline surface. The recommended dosage
is approximately one gallon per 100 square feet
but ihis can vary depending on the amount of
oiling. TheproductsnouUbeapptiedfullstrength
as supplied. Since the product is hydrocarbon
base, it should not be diluted with water during
application as this will greatly reduce effective-
ness.
After a soak time of approximately 30 minutes.
the cleaner and the oil released from the shore-
line surface is washed by hose into the water
where it can be readily recovered by conven-
tional means such as skimmers -or absorbents.
The soaK time may vary with ambient tempera-
ture, oil density and degree of weathering.
Corexit 9580 is useful on shorelines in fresh or
salt water. It is effective on all types of oil
including heavily weathered and emulsified oil
("chocolate mousse") containing up to 50 per-
cent water.
Typical Physical Properties
Specific Gravity at 60°F/15.6°C 0.820
Density, ib/gal at 60°F/15.6°C .....S-8
Rash Point °F/°C (SETA CC) 195/90.5
Pour Point. °F/"C - -42M1
Viscosity:
GST at 1508F/65.6°C 8.4
GST at 100°F/37.8°C 4.8
Solubility:
Hydrocarbons «:"SoluSe
Fresh Water Slightly Oispersible
Sea Water Slightly Dispersitole
•LC.
>87.600 com » 96 hours
2.400 cpm at 48 hours
Packaging and Availability
Corexit 9580 is packaged in unlined 5S-galion
non-returnable drums. This product is available
from Exxon Chemical Company manufactunng
and distribution points throughout the worra.
Storage
The shelf Bfe of unopened drumsofCorexitSSBO
is unlimited. No unusual storage precautions
are necessary. The containers should always
be capped when not in use to prevent con-
tamination and evaporation. Unless evapora-
tion is allowed to occur. Corexit 9580is not
adversely affected by changes in storage tem-
perature.
90
-------
Storage temoerature guidelines are as follows:
Ootimum 40 to 100°F/4to38?C
Maximum 170°F/76.7°C
Minimum -30°F/-34.4°C
Storage should comply with prevailing local or
international guidelines. Corexit 9580 contains
no cnemicals known to be harmful to marine
eauipment.
Handling
Avoid contact with eyes, skin and clothing. Wash
thorouanly after handling. If contact with the
eyes should occur, immediately flush them with
plenty of water for at least 15 minutes. Call a
pnysiaan. Skin should be flushed with water.
Wash clothing before reuse. Refer to the mate-
rial safety data sheet for additional information.
Regulatory Submission
Corexit 9580 is on the U.S. Environmental Pro-
tection Agency's NCP Product Schedule. This
listing does NOT mean that EPA approves.
recommends, licenses, certifies, or authorizes
the use of Corexit 9580 on an oil discharge. This
listing means only that data have been submit-
ted to .EPA as required by Subpart H of the
National Contingency Plan 300.85.
91
-------
MATERIAL SAFETY DATA SHEET
CXXON CMSMICAL AMiHICAS. • O. »OX J37J. HOUSTON. TIXA» 770»l
A O>»oie« cl CXXON CHI MICAL COMPANY. A Oi»iti«I It (XXON COHPOHATIOM
COREXIT 9580
PAGE: 1
DATE PREPARED: MOV 2. 1S9O
MSOS NO. : 79380000
SECTION 1 PRODUCT IDENTIFICATION & EMERGENCY INFORMATION
PRODUCT NAME: COREXIT 958O
FORMERLY: COREXIT 958O
CHEMICAL NAME:
Not applicable: Bland
CHEMICAL FAMILY:
Shoreline Cleaner
PRODUCT DESCRIPTION:
Clear Straw Colored Liquid
Hydrocaroon Odor
EMERGENCY TELEPHONE NUMBERS: EXXON CHEMICAL AMERICAS
CHEMTREC
T13-STO-€OOO
•OO-4Z4-93OO
SECTION 2 HAZARDOUS INGREDIENT INFORMATION
The composition of this mixture may be proprietary information. In tne «vont of a
n>*d(ca1 am«rg«ncy. con«os1t1ona1 information will b« provided to a pnyaiclan or nur*«.
This product 1* hazardous as defined in 29 CFR191O.12OO. Based on tn« followtng
compositional infornation: „.,.„,
COMPONENT QSHA HAZARD
Paraffime Solvent CoaDu»TiB«« UlquUS
P»raff 1nic Solvent. Organic Esters Ey« «ne» Skin Irritant
P.r,fflnic Solvent Vapor. Irritant te, Eyea
and H«apir«tery Tract
For additional information see Section 3.
SECTION 3 HEALTH INFORMATION & PROTECTION
NATURE OF HAZARD
EYE CONTACT:
Irritating, but does not injure eye tissue.
SKIN CONTACT:
Low order of toxldty.
Frequent or prolonged contact may Irritate and cause dermatitis.
INHALATION:
Hlgn vapor concentrations are irritating *« *"• eyes end trie resp^atory
tract, may cause headacnes and dizziness, are anesthetic and nay nave
other central nervous systeti effects.
INQESTION:
Small aawunts of the liquid aspirated into the respiratory system during
ingastlon. or from vomiting, may cause broncniopneunonia or pulmonary
edena.
92
-------
MATERIAL SAFETY DATA SHEET
COREXIT 9580
PAGE: 2
DATE PREPARED: NOV 2. '39O
MSDS NO.: 7958OOOO
EYE CONTACT:
FIRST AID
.1
lusn eyes with :arge amounts of witar until -rr,ration subsides
r«f-»*»»»^M^ **AM«I**B __.*. _.. I , —•—«•«• W4 .
= lusn with 'arge amounts of *ater; ua«
Remove grossly contaminated clothing.
-euse.
-* •--'ration persists
INHALATION:
soap • * ava i l ab 1 e.
•Deluding snoes. and launder before
attention.
'Jsing proper -esoiratory protection. ;iim«diatal y -emov. th« af ec-«a
?*C™ r e^BOSure- 4««"n,,t.r frtif.cial r.SD,rat1On if br.atning
INGESTION* ° at rMt- CaH f?r Pro"P^ m«d1Cal attention.
attention"*0' °° N°T ' naUC" vomit1"a- K~P a^ -••«. S«t promot
WORKPLACE EXPOSURE LIMITS
EXXON RECOMMENDS THE FOLLOWINS OCCUPATIONAL EXPOSURE LIMITS:
3OO ppm total nydrocaroon oas«o on composition.
PRECAUTIONS
ik«ly. wear safety glasses witn side
resistant gloves.
y glasses wi tn side smelds.
PERSONAL PROTECTION:
?or open systems where contact is
shields, iong sleeves, and cnemica
Where contact may occur, wear safe v glasses with side smelds
Where concentrations ,n air may exceed the limits given -in this
Section and engineering, work practice or other means of exposure
-eduction are not adequate. NIOSH/l5sHA approved respirators may
be necessary to prevent overexposure by inhalation.
VENTILATION:
sf mechanical dilution ventilation is recommended wnenever this
or 's'agitatad '" a C0nf1n*a sP»ce. is heated above ambient temperatur«
CHRONIC EFFECTS:
studies have snown that prolonged and repeated inhalation
hydrecareon vapors in tha same napntha boiling rang* as
produce adverse kidney effects in mala rats. However, these
observed 1n similar studies with female rats and mala'and
and in limited studies]with other animal species. Additionally.
- of human studies, there was no clinical evidence of sucn effects
occupational levels. Itlia therefore mgnly unlikely that tn«.
_„__ -^ II * observed in mala ra^« nave significant implications for human*
exposed at or below recommended vapor limits in the workplace.
CHRONIC TOXICITY DATA IS AVAILABLE UPON REQUEST
SECTION 4 FIRE & EXPLOSION HAZARD
FLASHPOINT: -74 0-g p METHOD- Seta CC
FLAMMABLE LIMITS: LSL- 0 S U6L- 7*0
AUTOISMITIOM TEMPERATURE: U
av,,,.bl.
93
-------
MATERIAL SAFETY DATA SHEET
=XXCN CHEMICAL AMERICAS. ' O. SOX 3272. "OUSTON. TEXAS 77001
A Civilian of EXXON CHEMICAL COMPANY. .A Division at EXXON CORPORATION
COREXIT 9580
PAGE: 3
DATE PREPARED: NOV 2. "99O
MSDS NO. : 7958OOOO
GENERAL HAZARD:
CGmousfBle _:quia. can 'arm comDust:ble mixtures at temperatures at
or above tre flasnpomt.
Toxic gases will *orm upon coniBust :on.
•=
-------
MATERIAL SAFETY DATA SH^ET
COREXIT 9580
PAGE: 4
DATE PREPARED: NOV 2. 199O
MSOS NO.: 73SaOOGO
This product -nay contain tr-ace amounts of etny:«ne oxide
(-AS No. 75-21-8). a conoit:on wnicn creates tre ootent-al 'or
accijnulation of etnylene oxide -h tne reaa space of snipping
ana storage containers ana in enclosed areas wnere tne proauct
-To?*'"9 na"al€<3 or ^*«a- HthylLn* oxide is considered By OSHA.
-ARC. and NTP as a potential carcinogen 'or *uraans. -thylene oxide
Tiay also present -eproauct:ve. -nLtagemc, genotoxic. -eurologic
ana sensitization "azaras :n numkns. If tnis proauct -s nanalna
itn adequate vent-'at ion. tne
•s not expected to result :n any
HAZARD RATING SYSTEMS:
This -nformation is for oeople t
National Paint & Coatings Associ
•esence of tnese trace amounts
snort or long term nazaras.
•amed in:
ition-s (NPCA)
Hazardous Mater'als I dent ifitat ion System (HMIS)
National F•re Protection Association (NFPA 7Q4)
Identification of tne F* re Hazards of Materials
HEALTH
"LAMMABILITY
REACTIVITY
NPCA -HMIS NFPA 704
2 • E
o
SECTION 7
REGULATORY
n 313 Reportaol« Ingredients.
SECTION 8 TYPICAL H.HYSICAL £ CHEMICAL PROPERTIES
SPECIFIC GRAVITY:
O.81 at SO
Density: 6.8 ibs/gal at 6O
SOLUBILITY IN WATER. WT. X AT *F:
Insoluble
VAPOR PRESSURE. mmttQ at
2 at 100 Calculated
'f:
VISCOSITY OF LIQUID. CST AT
3 at 100 Cannon-Fenake
2 at -SO Cannon-Fenake
95
-------
MATERIAL SAFETY DATA SHEET
rXXON CHEMICAL AMIA1CAS. " C. SOX 32T2. ^CUSTOM. TEXAS 7-001
A Oivi«.on o> EXXON CHCMICAL COMPANY. A O.V...OT at EXXON COKPOHATION
COREXIT 9580
PAGE: S
DATE PREPARED: NOV 2. '99O
MSOS NO . ; 7938OOOO
SP. GRAV. OF VAPOR, at 1 ata (A1r=1): FREEZING/MELTING POINT, 'F:
= CC -35 Pour
EVAPORATION RATE. n-Bu Acetate'1:
2,0 Calculated
BOILING POINT.
38O to 478
"F:
SECTION 9 REACTIVITY DATA
STABILITY:
Staola
CONDITIONS TO AVOID INSTABILITY
None
HAZARDOUS POLYMERIZATION:
W111 ^ot occur
MATERIALS AND CONDITIONS TO AVOID INCOMPATIBILITY:
Strong Oxidizing Agents
HAZARDOUS DECOMPOSITION PRODUCTS:
None
SECTION 10 STORAGE AND HANDLING
ELECTROSTATIC ACCUMULATION HAZARD:
Unknown, -3e prooer grounding procedure
STORAGE TEMPERATURE, 'F:
Ambient
STORAGE/TRANSPORT PRESSURE. MHg:
Atmospner-c
LOADSNG/UNLOAOTNG TEMPERATURE. F:
Anoient
VXSC. AT LOADING/UNLOADING TEMP..
Not available
cST:
REVISION SUMMARY:
Stnce OCTOBER 23.199Q tn1» MSDS has been ravlaod in Section(s):
€
REFERENCE NUMBER:
HOHA-A-12OO3
DATE PREPARED:
Nevenber 2.199O
SUPERSEDES ISSUE DATE:
October 23.198O
FOR ADDITIONAL PRODUCT INFORMATION. CONTACT YOUR TECHNICAL SALES REPRESENTATIVE
FOR ADDITIONAL HEALTH/SAFETY INFORMATION. CALL 713-870-6885"
THIB IMPORBATlal ««UAT»« fO TTm 3*KCIrIC •BTBHIAb UMlBWllmB Am BAT WOT Vft VAklB FOH 9UQI WTVJIIAb U««B I" *.*<•»•»••• «MM
ami! utixiALS p*J*J*lnr **ocsu. >UCH in>rc«nimoii is JO^TMB «t«T OP oun IMOWUOCI AW IILIW. ACCUUTI ""j-Si/JliS.
TMI UIU'S ««j»Otm»ILITT TO SATIirV HIIKMLP A* TO TM SUITAIIL1TV AMD CCMVLITtMSa OP WCM IwoMUTlOH PCM «'»_2'P» IJJ
UK. rt 00 HOT ACaVTLIAIILITT PON AMT LOO OH OAWGI 1MAT IUV OCOM PBOH TMI UM OP THIS IHPOMWTION NO* 00 « OPPM
-------
CMOttCAI. PRODUCTS Of CAMAOA LTD.
••oowrr* CXIMIOUCS wesr ou CANADA
•MMMCAMT. m.l.f B'UUOU. P.O. NU1U
MATERIAL SAFETY
DATA SHEET
SECTION 1. PRODUCT IDENTIFICATION
•RACE NAME
CITRIXLEEH XPC
CHEMICAL NAME
CHEMICAL INGREDIENTS
! Terpene hydrocarbon, allcyl aryl polvach
; TELEPHONE MOI( 5 14) 355-4660
IN eMePGENCYf 6131906-6666 C..
•r,
PHQOUCT use
Liquid 'dctcrcenc
glycol athar. and imina.
SECTION II. HAZARDOUS INGREDIENTS
HAZARDOUS INGREDIENT
3 Jpropylene glycoi c-icr.yl cchcr
J-Limonene
i
7-
70-
1
SECTION
EVAPORATION RATE (ETHER - 1|
SOLUBILITY IN H,CH» 3Y WT <9 ZOPC
Forms i cable te
SPECIFIC cnAviTv JH,O « n 9 a«e
161
% CAS NO. ' EXPOSURE ] LQ» LC«
Or UN NO ' uMirs -SHtats AKO BOUTIE
13
90
34590-94-3 K 0. T KNOWN
5989-27-5 , NONE Rats'/ Oral
1
III. PHYSICAL DATA
-------
Pigs 2 •! a
SECTION VI. TQXICOtOGICAL PROPERTIES
THRESHOLD LIMIT VALUE
' Hoc applicable
C OflONOGtMtC C BEPTOOUCTIVE 70XK3TV
j 3 UUTAGEN»C O TEHATOGENIC ..nj.f_fffrtt-
j =r=ECTS OF OVEPEXPOSUP.E: Dizziness ar loss on equilibrium from overexposure in enclosed
I HALATION jrnns '-ncli L.iujrooc-.r veuiilation. ;
I SKIN
! Dryness or slight irricac±on on contact only. —
I EVES"
I Irrieanc co eves upon contact only. ;
j CHHONIC OVEBEXPOSUne EFFECTS
• Soc dec era ued. .^
SECTION VII. REACTIVITY DATA
i '
CONDITIONS CONTRIBUTING TO INS7AGILIYY
Produce la -cable. •—
INCOMPATIBILITY
Strong oxidizing ap.encs (e.g. 1 iqmd chlorlua, concentrated oxygen).
HAZARDOUS OECOMWJSHION HHOUUCTS
Mon« knotm
CONDITIONS CONTBIBUnNG TO POLYMERISATION
noc occur
SECTION VIII. SPILL OR LEAK PBOCEDUHES
' STEPS TO BETAKEN IP MATERIAL !S RELEASED O« SPILLED
Provide adequate vcncxlacion. -Ceeo aw«y £roa he«c sources. MOD uo or ab«orfr, rln*a
Hot
NEUTRALISING CHEMICALS
L:l»
WASTE DISPOSAL MCTMOO
Dispose in accordance uich lor.il ,in erf
/Erase rr*i«ai«b«n
QATg;
8Ch. 19M
Th« «tarm»i(«« «c»«iw«
w«f mnn i»o w
hn Man
ir. •lexis or «"•«•«. rcg«i*n« •>• (cmney •«»••*«<.
98
-------
COREXIT 76640
Concur "664 Oil Oispersant is
a water-eased system. CCflExir
and shoreline vegetation, it is a
L DISPERSANT
a carefully baiancao bieno of non-ionic surfactants :n
7664 Has very low toxicity to fisn. manne animals.
ffective on a wide range of oatroieum products.
Property applied. COROCIT 7664J renders an oil slick disoersiole Oy wave action or
mechanical agitation. Decomposition at the resulting oil drooiets Dy natural 3'oiogi-
=ai and physical processes s greatly accelerated. The amount of Coneor 7664
ieedeo ceoends on tne vncatity of me soil) (the lighter the oil. tne lets COREXIT
"664 reouireol. Experience riai snown mat one oart of COHBOT 7664 wul usually ;reat
• 0 to 20 parts o* oil.
The aoility of COREMT 7664 to Water-wet mineral surfaces also maXes it vaiuaoie as a
snoreiine orateetor or cleaner.
Prior regulatory aporovai may cje reauired lor any aooiication of this oroduct
COPEXIT 7664 is recommended for use as an oil spill control agent in three different
aopiications. These are:
• AS a OlSPEBSANT of lignt to medium oil or petroleum products on either fresn or
sail water.
• AS a SHORELINE PROTECfOR to orcvem oil from adhering 10 beaches, rocks.
marsnes or snorelme structures.
• As a POST-SPILL CLEANEF
aauiomant
RecomrrMnaad procedures for each cf these are given beiow.
DISPEHSANT
Setter results are always octai
after oil has came asnore or contaminated
,ned
with disoersants if treatment is begun early.
COHBOT 7664 Oil Otspersar* should be applied to me Hotting oil. not to tht wafer
arouna it.
Depending on t*« equipment a\|ailaoie. COPOir 7664 may oe used full-strength as
supplied or in dilution with water, usual dosage is 10 to 30 U.S. gallons (USG) per
acre. Regardless of me method [employed, the cnemical snould oe appii«d in tne
form of small droplets — n«ver as a tog or mist. This is necassa/y to aid in rapid
penetration of me oil by the active surfactants, which are effective only wnen oriented
at the Oii/wa»r interface.
Hand-H«M Equipment
. Small spills can often De_narx3lid by spraying CORCXIT 7664 from handheld
insecticide-woe pressure can*. Ifrom oackpack sprayers, or from noses attached to
undiluted w'uh nny commercial spraver with a
cane.
pumps. Use Ccncxir 7664
99
-------
Tests nave snown mat an altitude ot 30 to 50 '«« is oot.mum 'or acm nahcocters and
h«ed-*.nq aircrat. S=e«O ~,,l va/v «.tn d.rcraft tvpt*. sorav ccom -.OUT. d.Jd cu."O
:aDaorv Aithougn an o.l SI.CK treated «.m CSWJCIT .'684 .s ,-rrore eas,,y oroxen .n,=
ircoiets ov waves and me normal motion of tne sea. en caimer waters aquation ov
• coats fay co necessary alter aanai treatment.
SHORELINE PROTECTION
- ^RE»T 7664 Oil Oisoersant used 'n very low concentration ( l to 3% .n water). w.u
orotect beaches. marsnes. rocks, and snoralina structuras from o.t contamination if
aooiiod t»fo« tne o.l com«s asnom. Th« ailotea solution may oe sorayea on tn«
snoralin* or tn« fuii-strengtn cnamicai may t» sprmyed •mo me surt wn.cn wasnes
me snor.iin.. Dosage .s not cnt.cal but soray.no, 1 gallon oi 1-3% aoict.on oer 10-1 5
sauar* t«at is r»asonaoie. A« a result o« m* water-wening action ot CCBCXII 756^ c,i
csming asnor* w.il not stick to tnose cam ot tnesnorei.ne wnicn nave oeen ireatM.
ana m«cnan.cai recovery w.ii Oe easier.
POST-SPILL CLEANUP
COReor 766* u excaiiem for wasn.ng o.l from oeacnes. seawalls, rocxs. dccxs and
-MM For mi« aewucatiom. usa a 1 to 3% soiut.on of COREXIT 786* .n water it me on
dnoait .s oiraeuuny neavy or weaiwrea. a prs-scax w.tn a 5^«"t;°"»a7^e^'
• sucn 13 OwexiT 8667} is recommended, tonowea oy ma 1 to 3% C~HE»T 7564 i ne
COTOOT r«8* solution can be aooiied from nand-neid sorayeis or oy high-pressure
water eduction systems as me 100 raouires.
TYPICAL PHYSICAL PROPERTIES
So«cific Gravty.
Run Point SETA CC. "F/-C
Poor Point
viaeoaitim
7/"14
3*
25
21
118
Seiubtiity:
SoluDia in tresn watar and saawatar.
Dr«O4nuW« m nydrocartxma.
AVAILABILITY AND STORAGIE
COMDor 7694 )s av«il«bi* >n 55-aalkxi. lined, nen-vtumao;* drums.
Th« tMlf lite of uoooeoed drums of COfleaT 7664 is unlimited. No unusual storage
precautions ara nccMaaiy.
Th* eoniairwn snould alwaya O» capped wnan net m us* to prevent contamination
«nd •vaporatian. Units* evaporation i« ailowwl to occur. COHGOT 7684 is not ad-
v«fMty attecicd by cnanges in storage temperaiura. Storage temperature guidelines
arr
Optimum storage tempecatunw: or to 10CTR4' to 38"C)
Maximum storage temperature: about 130*P(5**Q
Minimum storage temperafta: about KrR-i2*C)
Storage oetow ceek on vessels snould be m accord wnn prevailing local or interna-
tional guidelines. Cowxrr 7684 contains no cnetmcais known to be harmful 10
manne equipment
100
-------
caoacitv ot 3 to 5 gallons. Aooiy as a fine soray over tne surface of tra soul ana then
..— ...... , <..- ... -. a 30,^5 crooenSr to aia in disoer*ma me on
!c 50% in water if use ot nana sprayers is ;o oe
Booma
•••jsn ««m a lire *>cse or agitate witr
-CfiexiT 7664 may o« aiiutea to 25
extensive.
Workfaoat Fitted with Spray
"'satment of large areas may oe ae comgusned (ram me water s surface oy using
•vorfcooais equtooea with scray oooms mountea aneaa of me oow wave or as far
'crwara as oossibie. Soray eauiomj«nt on worxboats must be designed to orovide a
•ailutec disoersant solution to tne soray Dooms. This is best accomplished by use of
an eauctor with a metenng adjustment to teed cnemicai into me seawater stream
oroaucaa oy a oumo ooerating at aoout 80 to ?00 csi witn a total caoacicy of aoout
'00 to :50gpm.
=ductors can aiso b« used wtm lire hose systems out tne total watef flow snould be
'eaucso as aoorooriate. With low-woiume. low-oressure oumos (with wnic.i eduction
is not oossibie). me chemical can oe fed to tne water stream with a sman metering
3ump. '
"he concentration of chemical reaujirsa must be calculated 'rom: tne total oumo
-acacity: me swath wioth covered oy tne booms emoioyed: me speed ot me ooai:
and (possibly) me tmcxness of tne si ex or tne amount ot oil to M (reittd over a
area.
•n any case, some agitation or mixing energy mu« oe aopiied to tn« treated oil.
usually tne normal motion of tne sea will surtice. as will tna waves generated by the
wake of tne worfc&oat. in calmer conditions, power coats can provide mixing energy
as necessary, or fire hoaea can be Used to aid in dispersing small spills.
ana areas a/e tnrettenea. neitfier tgiation not cramicti concenrrar/on
should oe //wreased simply te cause mom rap/d ef«app«a/ance of f/r» oil.
les 'or spray booms should be carefully selected to give a uniform spray. The
soray pattern snouid be flat, sinking tne watttr in a line perpendicular to me direction
of tne boats travel. The nozzle's spray angle should be sucn mat me fan-shaped
sprays from acjacent nouies overlap just aoove tne water.
Helicopter/ Airplane Spraying
=or aenai spraying. CORSMT 7664 is used unailutea. Depending on me type of on
oetng treated and tne spray equipment used, the dose required will usually be aoouJ
1 5 to 30 USG per acre, out excellent results nave been obtained witn dosage as low
as 7 to 10 USG per acre. A variety of fixed-wing aircraft can be used tor spraying
over a large area. These range from vety small to very large aircraft equipped for
carrying 100 to over 3.000 U.S. gallons of chemical.
The soray nozzles used are most or tical, since droplet size must be controlled.
Many nozzles used for agncuitural spraying are ol low capacity and produce too fine
a spray (actually a mist or fog) whie Hs not cesiraWe for dispersant spraying. Care-
ful selection ot nozzle capacity to achieve desired dose levels cannot oa overem-
onasized. (See cor Oil Spill Chemicals Application Guide tor additional information).
Calculations similar to those mentio iea orevnousiy for worttooat spraying should also
be made for aerial application. Consideration must oe given to: me speed and al-
titude ot me plane: the capacity of tne chemical pumps: me pressure at which m*y
operate: the effective swath width obtained: and any windage losses. Bights should
be made directly into me wind for best results.
-•eiicooters may also be employed in aenai spraying, either with fixed spray booms
y oy attachment of a siung-oucket system eauipped wtin spray ocoms. if a bucket is
jsed. it anouid be stabilized against rotating and swaying.
101
-------
COREX1T 7684 OIL DISPERSANT
WARNING!
CAUSES SKIN IRRITATION
COMBUSTIBLE
MAY CAUSE EYE IRRITATION
VAPORS IRRITANT
PRECAUTIONS — Ke«o away from neat. sparks ana open flames. Avoid
eye contact. Avoid contact with skin and clewing. Avoid breaming mists
or vapors. Use with ventilation equal to unobstructed outdoon m moder-
ate breeze. Keep container closed. Wa«h thoroughly after handling.
FIRST AID — Wasn skin with soap and water. Hush eye* witn plenty of
water until irritation suostde*. Remove to fr«sn air. If not breaming, apply
artificial respiration and CALL A PHYSICIAN.
P1RE RQHTINO — Use waterspray to cool flra-exoosed surfaces and era-
tecs personnel. Extinguish preferentially wiui waterspray. waterlog, dry
cnemical or alconoi-iype foam.
CHEMICAL SPILL CONTROL—Keep public away. Siminate sources of
ignition. Shut off source, if possible to do safely. Prevent liquid from entering
sewers, watercourses or low areas. Advise autnonties if material has entered
a sewer or watercourse ornas contaminated ioil or vegetation. Contain spilled
liquid with sand or eartn and dilute with water. Recover cy pumping or with
suitable absorbent Consult an expert on disposal of recovered material and
ensure conformity with local disposal regulations.
US! INSTRUCTIONS — Use of tins material, in any specific application.
may create hazards or nsfcs to human health. Before commencing use.
consult our company's Technical Sales Representative for appropriate
recflnwnendatfdria'and" precautionary instructions tor any intended appli-
cation.
102
-------
MAJOR ION
MAJOR ION COMPOSITION
OF
INSTANT OCEAN* SYNTHETIC SEA SALT
".TOTAL WEIGHT
IONIC CONCENTRATION
AT 34 °/oo SALINITY
Chloride
Sodium
Sulfate
Magnesium
Calcium
Potassium
Bicarbonate
Boron
Strontium
SOLIDS
later
(CT) 41
.470
(Na+) 26.280
(S04-) 6.602
(Hg**J :
(Ca~)
(K*)
(HC03-)
(B)
(Sr+*J
TOTAL 8
1.230
..013
1.015
>.491
J.015
.001
5.11X
(H?0) 13.88
18.740
10,454
2,631
1,256
400
401
194
6
7
34,089.
.0
.5
50
TOTAL
99.99X
103
-------
APPENDIX C
OIL STANDARD CURVES
NOTE: This appendix contains examples of the standard curves for two oils (Prudhoe Bay crude and Bunker
O. three cleaning agents (Corexit 9580. Citrikleen XPC, and Corexit 7664),iand "oil plus no
cleaning agent" controls.
104
-------
SPECTROPHOTOMETRIC ABSORBANCES— INIT
date: 3-12-92
oil: Prudhoe Bay
0.894 <-«oil density (g/mL)
89.4 <— parent oil/DCM cone (mg/mL)
340 nmeter measurements:
vol parent
std oil/DCM in
no. std (uL)
1
2 4.4
3 11
4 22
5 44
6 110
7 220
370 nmeter measurements:
vol parent
aid oil/DCM in
no. std (uL)
2 4.4
3 11
4 22
5 44
6 110
7 220
400 nmeter measurements:
vol parent
std oil/DCM in
no. std (uL)
1
2 4.4
3 11
4 22
5 44
6 110
7 220
oil std
cone absorber*
(mg/mL) ini OCM
0.00 0.000
0.02 0.000
0.05 0.000
0.10 0.000
0.20 0.000
0.49 0.000
0.98 0.000
oil std
cone absorbanc
(mg/mL) ini OCM
0.00 0.000
0.02 0.000
0.05 0.000
0.10 0.000
0.20 0.000
0.49 0,000
0.98 0.000
oil std
cone absorbanc
(mg/mL) ini DCM
0.00 0.000
0.02 0.000
0.05 0.000
0.10 0.000
0.20 0.000
0.49 0.000
0.98 0.000
AL OIL STANDARDS
»* at 340 nm:
fin DCM std
0000
0.000 0.062
0.000 0.153
0.000 0.288
0.000 0.550
0.000 1 .407
0.000 3.419
es at 370 nm:
fin DCM std
0.000
0.000 0.032
0.000 0.082
0.000 0.155
0.000 0.298
0.000 0.766
0.000 1 .556
»s at 400 nm:
fin DCM std
0.000
0.000 0.021
0.000 0.053
0.000 0.100
0.000 0.194
0.000 0.500
0.000 1.012
105
adjstd
0.062
0.153
0.288
0.550
1.407
3.419
adjstd
0.032
0.082
0.155
0.298
0.766
1.556
adjstd
0.021
0.053
0.100
0.194
0.500
1.012
RF RF
(mgoil/ %dffi
abe) mean
0.317 -3.6%
0.321 -2.4%
0.341 3.7%
0.358 8.7%
0.349 6.2%
0.288 -12.6%
0.329 <-meon
0.026 <-std
7.8% <%R80
RF RF
(mgoil/ %dHf
abe) mean
0.615 -2.5%
0.600 -4.9%
0.634 0.6%
0.660 4.7%
0.642 1.8%
0.632 0.2%
0.630 <-me«in
0.021 <-std
33% <%RSD
**•** «» ^ ^vnwv
RF RF
(mgoil/ %dffi
abe) mean
0.937 -3.4%
0.928 -4.3%
0.983 1.4%
1.014 4.6%
0.983 1.4%
0.972 0.2%
0.969 <-mean
0.032 <-«td
3.3%<%RSO
-------
SPECTROPHOTOMETRIC ABSORBANCES-INITIAL OIL STANDARDS
date: 1-23-92
oil: Prudho* Bay + C9580
0.894 <—oil density (g/mL)
89.4 <—parent oil/DCM cone (mg/mL)
340 nmetar measurements:
std
no.
370
std
no.
400
std
no.
1
2
3
4
5
g
7
vol parent
oil/DCMin
std (uL)
4.4
11
22
44
110
220
nmeter measurements:
1
2
3
4
5
6
7
vol parent
oil/DCMin
std (uL)
4.4
11
22
44
110
220
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.10
0.20
0.49
0.98
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.10
0.20
0.49
0.98
absorbances at 340 nm:
ini DCM fin DCM std
0.043
0.043
0.043
0.043
0.043
0.043
0.043
0.039
0.039
0.039
0.039
0.039
0.039
0.039
absorbances at 370 nm:
ini DCM fin DCM
0.039
0.039
0.039
0.039
0.039
0.039
0.039
0.036
0.036
0.036
0.036
0.036
0.036
0.036
0.109
0,206
0.397
0.716
1.747
3.747
std
0.073
0.127
0.231
0.406
0.959
1.807
adjstd
0.068
0.165
0.356
0.675
1.706
adjstd
0.036
0.090
0.194
0.369
0.922
1.770
nmeter measurements:
1
2
3
4
5
6
7
vol parent
oil/DCM in
std(uL)
4.4
11
22
44
110
220
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.10
0.20
0.49
0.98
absorbances at 400 nm:
ini DCM fin DCM
0.035
0.035
0.035
0.035
0.035
0.035
0.031
0.033
0.033
0.033
0.033
0.033
0.033
0.029
std
0.057
0.093
0.161
0.275
0.635
1.177
adjstd
0.023
0.059
0.127
0541
0.601
1.147
RF RF
n* "*
(mg oil/ % dHf
abs) mean
0.289 0.2%
0.298 3.3%
0.276 -4.3%
0.291 1.0%
0.288 -0.1%
0.289 <-moan
0.008 <-std
2.7%<%RSD
RF RF
(mg oil/ % dHf
abs) mean
0.554 2.8%
0.549 1.9%
0.508 -5.7%
0.534 -1.0%
0.534 -1-0%
0.556 3.1%
0.539 <-rrw*n
0.018<-std
3.3% <%RSD
RF RF
nr^ • "•
(mgoil/ %dHf
abs) mean
0855 3.6%
0.833 0.9%
0.774 -65%
0.816 -15%
0.818 -0.9%
0.857 3.8%
0.826 <-mMn
0.031 <-ttd
3.7% «%RSO
106
-------
SPECTROPHOTOMETRIC ABSORBANCES— INITI/
data: 3-20-92
oil: Prudhoe Bay + XPC
0.894 <~o« density (g/mL)
89.4 <~parent oii/OCM cone (mg/mL)
340 nmeter measurements:
vol parent
std oil/DCM in
no. std (uL)
1
2 4.4
3 ,-11
4 22
5 44
6 110
7 220
370 nmeter measurements:
vol parent
std oii/OCM in
no. std (uL)
1
2 4.4
3 11
4 22
5 44
6 110
7 220
400 nmeter measurements:
vol parent
std oil/DCM in
no. std (uL)
1
2 4.4
3 11
4 22
5 44
6 110
7 220
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.10
0.20
0.49
0.98
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.10
0.20
0.49
0.98
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.10
0.20
0.49
0.98
absorbanct
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
absorbanc*
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
absorbance
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
L OIL STANDARDS
s at 340 nm:
finDCM std
0.000
0.000 0.060
0.000 0.154
0.000 0.315
OiOOO 0.601
0.000 1 .554
0.000 2.998
s at 370 nm:
finDCM std
0.000
0.000 0.031
0.000 0.082
0.000 0.170
0.000 0.328
0.000 0.854
0.000 1 .669
a at 400 nm:
fin DCM std
0.000
0.000 0.020
0.000 0.052
0.000 0.111
0.000 0.214
0.000 0.559
0.000 1 .092
adjstd
0.060
0.154
0.315
0.601
1.554
2.998
adjstd
0.031
0.082
0.170
0.328
0.854
1.669
adjstd
0.020
0.052
0.111
0.214
0.559
1.092
RF
(mgoil/
abe)
0.328
0.319
0.312
0.327
0.316
0.328
0.322
0.007
2.1%
RF
(mgoil/
aba)
0.634
0.600
0.578
0.600
0.576
0.589
0.596
0.021
3.6%
RF
(mgoil/
abs)
0.983
0.946
0.886
0.919
0.880
0.901
RF
%dtff
mean
1.9%
-0.8%
-3.0%
1.7%
-1.7%
1.9%
<-mean
<-std
<%RSD
RF
%dr»
mean
6.4%
0.6%
-3.0%
0.6%
-3.4%
-1.2%
<-mean
<-std
<%RSD
RF
%cfiff
mean
7.0%
2.9%
-3.6%
0.0%
-4.3%
-2.0%
0.919 <-mean
0.040 <-s*d
4.3% <%RSD
107
•^^I^IM
-------
SPECTROPHOTOMETRIC ABSORBANCES-INITIAL OIL STANDARDS
data: 3-13-92
CM): Prudbo* Bay -i- C7664
0.894 <-~otl d«n«ity (g/mL)
89.4 <---parent oil/DCM cone (mg/mL)
340
std
no.
370
std
no.
nmeter measurements:
•j
2
3
4
5
g
7
voJ parent
oil/DCM in
std(uL)
4.4
11
22
44
110
220
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.10
0.20
0.49
0.98
absorbances at 340 nm:
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
finDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
std
0.065
0.156
0.301
0.589
1.484
3.397
adjstd
0.065
0.156
0.301
0.589
1.484
3.397
nmeter measurements:
•j
2
3
4
5
g
7
400 nmeter
std
no.
4
2
3
4
5
6
7
vd parent
oil/DCM in
std (uL)
4.4
11
22
44
110
220
measurements:
vol parent
oil/DCM in
Sld(uL)
4.4
11
22
44
110
220
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.10
0.20
0.49
0.98
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.10
0.20
0.49
0.98
absorbances at 370 nm:
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
finDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
std
0.036
0.087
0.165
0.323
0.814
1.627
adjstd
0.036
0.087
0.165
0.323
0.814
1.627
absorbances at 400 nm:
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
finDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
std
0.024
0.057
0.108
0.212
0.533
1.061
adjstd
0.024
0.057
0.108
0.212
0.533
1.061
r^r- nE
RF RF
(mgoil/ %diff
abs) mean
0.303 -4.4%
0.315 -0.4%
0.327 3.2%
0.334 5.5%
0.331 4.7%
0.289 -8.5%
0.317 <-mean
0.018 <-std
56% <%RSD
nc DC
RF Hr
(rng oil/ % dHf
abs) mean
0.546 -7.0%
0.565 -3.8%
0.596 1.4%
0.609 3.6%
0.604 2.8%
0.604 2.9%
0.587 <-mean
0.026 <-std
4.4% <%RSD
DC DC
Hr Mr
(mgoil/ %dffi
abs) mean
0.820 -8.4%
0.863 -3.6%
0.911 1.7%
0.928 3.7%
0.923 3.1%
0.927 3.6%
0.895 <-nwan
0.044 <-std
5.0%<%RSO
108
-------
SPECTROPHOTOMETRIC
date:
oil:
0.972
97.2
3-9-92
Bunker C
ABSORBANCES-- INITl
< — oil density (g/mL)
<— parent oii/DCM cone (mg/mL)
340 nmeter measurements:
vol parent
std oii/DCM in
no. std (uL)
1
2
3
4
5
6
7
370 nmeter
std
no.
1
2
3
4
5
6
7
400 nmeter
std
no.
1
2
3
4
5
6
7
4.4
11
22
44
110
220
measurements:
vol parent
oil/DCMin
std(uL)
4.4
11
22
44
110
220
measurements:
vol parent
oil/DCMin
std (uL)
4.4
11
22
44
110
220
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.11
0.21
0.53
1.07
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.11
0.21
0.53
1.07
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.11
0.21
0.53
1.07
absorbanc
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
absorbanc*
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
absorbance
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
U_ OIL STANDARDS
is at 340 nm:
finDCM std
0.000
0.000 0.191
0.000 0.444
0.000 0.868
0.000 1 .699
0.000 3.742
0.000 3.742
s at 370 nm:
finDCM std
0.000
0.000 0.109
0.000 0.257
0.000 0.507
0.000 0.995
0.000 2.606
0.000 3.734
adjstd
0.191
0.444
0.868
1.699
adjstd
0.109
0.257
0.507
0.995
2.606
RF
(mgoii/
aba)
0.112
0.120
0.123
0.126
0.120
0.006
5.0%
RF
(mgoii/
«*»)
0.196
0.208
0.211
0.215
0.205
0.207
RF
mean
-7.0%
0.0%
2.4%
4.6%
<-meaii
<-std
<%RSD
^ /wFlWl,^
RF
mean
•5.2%
0.5%
1.9%
3.8%
•0.9%
<-meari
0.007 <-std
a AV. *v*asn
i at 400 nm:
finDCM std
0.000
0.000 0.070
0.000 0.168
0.000 0.333
0.000 0.656
0.000 1.670
0.000 3.357
adjstd
0.070
0.168
0.333
0.656
1.670.
3.357
RF
(mgoii/
abe)
0.305
0.318
0.321
0.326
0.320
0.318
RF
%dM
mean
•4.0%
•0.0%
0.9%
2.4%
0.6%
0.1%
0.318 <-me«n
0.007 <-std
109
-------
SPECTROPHOTOMETRIC ABSORBANCES--INIT1AL OIL STANDARDS
date: 4-6-92
oil: BunkerC + C9580
0.972 <—oB denerty (g/mL)
97.2 <—parent oil/DCM cone (mg/mL)
340
std
no.
370
std
no.
400
std
no.
nmeter measurements:
vol parent
1
2
3
4
5
6
7
oil/DCM in
lid (uL)
2.2
4.4
11
22
44
110
220
oil std
cone
(mg/mL)
0.01
0.02
0.05
0.11
0.21
0.53
1.07
absorbances at 340 nm:
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
finDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
std
0.090
0.177
0.429
0.877
1.793
3.732
adjstd
0.090
0.177
0.429
0.877
1.793
3.732
nmeter measurements:
1
2
3
4
5
6
7
vol parent
oil/DCM in
std(uL)
2.2
4.4
11
22
44
110
220
oil std
cone
(mg/mL)
0.01
0.02
0.05
0.11
0.21
0.53
1.07
absorbances at 370 nm:
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
finDCM
0.000
0.000
0.000
0.000
0:000
0.000
0.000
std
0.052
0.103
0.253
0.518
1.072
2.609
adjstd
0.052
0.103
0.253
0.518
1.072
2.609
nmeter measurements:
1
2
3
4
5
6
7
vol parent
oil/DCM in
std(uL)
2.2
4.4
11
22
44
110
220
oil std
cone
(mg/mL)
0.01
0.02
0.05
0.11
0.21
. 0.53
1.07
absorbances at 400 nm:
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
finDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
std
0.034
0.068
0.167
0.344
0.716
1.717
adjstd
0.034
0.068
0.167
0.344
0.716
1.717
RF HF
(mgotl/ %dffi
abe) mean
0.119 -1.9%
0.121 -0.2%
0.125 2.9%
0.122 0.7%
0119 -1.5%
0.121 <-moan
0.002 <-«tid
1.9%«%RSD
nr~ DC
HF HP
(rngoil/ %dHf
abs) mean
0.206 -0.1%
0.208 0.8%
0.211 2.6%
0.206 0.3%
0.199 -3.1%
0.205 -0.5%
0.206 <-msan
0.004 <-atd
1.9%«%HSD
nr~ OC
RF HP
(mg oil/ % dHf
abe) mean
0.314 0.9%
0.314 0.9%
0.320 2.7%
0.311 -0.3%
0599 -«^%
0.311 -0.1%
0.312 <-mean
0.007 <-etd
2.3%<%HSO
110
-------
SPECTROPHOTOMETRICABSORBANCES-INITIAI
date: 3-10-92
oil: Bunkar C + CHriWeen XPC
0.972 <-~oil eternity (g/mL)
97.2 <~-parant oil/OCM cone (mg/mL)
340 nmeter measurements:
voJ parent
std oiVDCM in
no. atd (uL)
1
2 4.4
3 11
4 22
5 44
6 110
7 220
370 nmeter measurements:
vd parent
std oii/DCM in
no. std(uL)
1
2 4.4
3 11
4 22
5 44
6 110
7 220
400 nmeter measurements:
vci parent
s(d oil/DCM in
no. std (uL)
1
2 4.4
3 11
4 22
5 44
6 110
7 220
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.11
0.21
0.53
1.07
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.11
0.21
0.53
1.07
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.11
0.21
0.53
1.07
absorbance
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
absorbance
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
absorbance!
iniDCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
*
.OIL STANDARDS
( at 340 nm:
fin DCM std
0.000
0.000 0.158
0.000 0.415
0.000 0.846
0.000 1 .495
0.000 3.736
0.000 3.736
» at 370 nm: '
fin DCM std
0.000
0.000 0.094
0.000 0.246
0.000 0.503
0.000 0.891
0.000 2.439
0.000 3.730
t at 400 nm:
fin DCM std
0.000
0.000 0.062
0.000 0.165
0.000 0.335
0.000 0.595
0.000 1.608
0.000 3.144
111
adjstd
0.158
0.415
0.846
1.495
adjstd
0.094
0.246
0.503
0.891
2.439
adjstd
0.062
0.165
0.335
0.595
1.608
3.144
RF
(mgoil/
abs)
0.135
0.129
0.126
0.143
0.133
0.007
5.6%
RF
(mgoil/
abs)
0.227
0.217
0.213
0.240
0.219
0.223
0.011
4.8%
RF
(mgoil/
abs)
0.345
0.324
0.319
0.359
0.332
0.340
0.337
0.015
4.4%
RF
%drtf
mean
1.5%
-3.4%
-5.3%
7.2%
<-mean
<-std
<%RSD
RF
%diff
mean
1.9%
-2.7%
-4.8%
7.5%
-1.8%
<-mean
<-std
<%RSD
RF
%diff
mean
2.4%
-3.8%
-5.2%
6.8%
-1.2%
1.0%
<-mean
<-«td
<%RSO
-------
SPECTROPHOTOMETRIC ABSORBANCES—INITIAL OIL STANDARDS
data: 3-10-92
oil: Bunker C + C7664
0.972 <~oil density (gfmL)
97,2 <—parent oil/DCM cone (mg/mL)
340
sid
no.
nmeter measurements:
vot parent
1
2
3
4
5
6
7
oil/DCM in
std(uL)
4.4
11
22
44
110
220
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.11
0.21
0.53
1.07
absorbancas at 340 nm:
ini DCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
fin DCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
std
0.156
0.411
0.836
1.602
3.740
3.740
adjstd
0.156
0.411
0.836
1.602
RF
(mgoil/
abs)
0.137
0.130
0.128
0.133
0.132
0.004
RF
%dtff
mean
3.7%
-1.6%
-3.2%
1.0%
<-rhean
<-std
3.0% <%RSD
370
std
no.
nmatar measurements:
1
2
3
4
5
6
7
vol parent
oil/DCM in
std (uL)
4.4
11
22
44
110
220
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.11
0.21
0.53
1.07
absorbancas at 370 nm:
ini DCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
fin DCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
std
0.089
0.245
0.500
0.956
2.440
3.733
adjstd
0.089
0.245
0.500
0.956
2.440
RF
(mgoil/
abs)
0.240
0.218
0.214
0.224
0.219
0.223
0.010
RF
%cHf
mean
7.7%
-2.2%
-4.1%
0.314
-1.8%
<-mean
<-std
4.6% <%RSD
400
std
no.
nmeter measurements:
1
2
3
4
5
6
7
voi parent
oil/DCM in
std(uL)
4.4
11
22
44
110
220
oil std
cone
(mg/mL)
0.00
0.02
0.05
0.11
0.21
0.53
1.07
absorbancas at 400 nm:
ini DCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
fin DCM
0.000
0.000
0.000
0.000
0.000
0.000
0.000
std
0.057
0.163
0.334
0.638
1.607
3.501
adjstd
0.057
0.163
0.334
0.638
1.607
3.501
RF
(mgoil/
abe)
0.375
0.328
0.320
0.335
0.333
0.305
0.333
0.023
7.0%
RF
%dc(f
mean
12.7%
-1.4%
-3.8%
0.7%
-0.0%
-8.2%
<-mean
<-«td
<%RSD
112
-------
APPENDIX D
SUMMARY - INCLINED TROUGH TEST RESULTS
NOTE: Samples with "lab qual" followed by a "-pi"
flow-rate tests; samples with "lab qual" followed
7.5 mL/min for flow-rate tests.
(e.g., A-pl) indicate a syringe flow rate of 63 mL/min for
' by a "-18" (e.g., A-18) indicate a syringe flow rate of
113
-------
5555SS&
53SS
3333333
55
SS
si si
33
a a
55
33
(38
333
aaa
233
S385
iisss!
SSI
as
99
I slsfeteistels -
*
SS^i
90
28
S
3S
000
N. O
»sl
SRRR
§2*
99
28
:1
58
88
V1 V « PS,
aSgg
888^5
SaS
855
igs*
SS
Is
370nn
*crt K<«*
nTnu
9999999
R5888SS-
99
88
588
8
,
8*8
9999999
99
fi!S5S
S3
aa
33$
gsg
S5|5^£
88
gsS
§1
II--
888882
99
p ,
88
« «! 0<
999
5S?a
999
08
RR
RRwS
*< . . a K
zsz2~
99|9
a
?5
*^ *: *f
8S3
dB=!
"*. "' W ~ W
I§iii
p PI^ i
°°°-~
* ™ ™ a C
358°*>
888
SB
888
8S = ,
58S
RRRv
ZZ8
29$
S
8'&22
a*
3885
zss
sss
fill
II
II
444
»«
388
2383
38
Hi
II
ggs
PM •
JJ
£2
£ S S
88
823
ma
S5
n
HI
II
aas
«S
»•>
332
igg
«».(
114
-------
-------
*=
if
i
33
38
38
Ss^i
38
-*
-s
333
85
Isife:
382!:
2*8
33
SS
teasS*
$«2
I "8
8S
SS^*
S~*
88382;
S-*
o
999
P»l*25*
« *> » r n c
8S28»«
55
-'8
»•-*
. j
a*
S|i|"!z
n o
1717
33
ii:
<7<77
33SS-
99
99
§2$
.-*
•*-«
3^
J'-
iSaaS
n^
• - ™ • AJ r
388X22
5-*
880
23
sstss
sa
!li
i»
li
§2
1
I
1
III
id
III
ill
i!
833
33
li
II
11?
11
fit
B2
9*
88
14^
SSS
116
-------
I
!'
gfl
Ii!
lf
* 1
%ori
onTnu
S*6
fi
II
II
II
sift
38
«R
88
88
$8
•? V
99
!!
n-lf
S-
^
IF ••
ssa
sas
8838
S3S!2
•-•-£.
117
-------
APPENDIX E
SUMMARY - SWIRLING COUPON TEST RESULTS
118
-------
-------
SB
1
8S-:
si a
-*
as
3 as
s
*- M jf
ri2 •*
«„*
«
a a
999
3
*
88-j
.,,2
N « £
88-S
88g8S5
88
99
999
za^:
n
882;
3
*
38
3B^
as
15
3§SS5
^
s?
:
3S8
iil
«*•
HO
1
•
I
jitlt
£
EE
SII
ti
II
II
sa
33
sss
233
44
9.99
Ann
«?
|*
(I Si
888
120
-------
ana
aaa
SSS
si a a
555
is!
8™ «
** n
§:•;
^-
28
MM
HdM
IS
%
52
u
*>£
828
33i
*-55
!2
$8
9
«
SfSlS
8'
g «• «
53^
* * CB
!«!
I
***
iuy
S^S
ill
i
NS?.
11
88
i
»«
M N
33
121
-------
|f!
!H
88
SB
Sti
88
5S
88
55
88
88
P
5|5
I
IS
1a'l
S§
88
I'S
338' =
ii
2&
35
55
aS
888
1=5
99
<*•» .
8ZZ
B3
K si
S
38
31*3*
i!
SB
§8<
3S
is!
88
«Si:<
I'i
iaS
il
i_ t
e3^
111
99
as
99
§S5
p:=
99
ss
98
a?:
55!
xx
His*
ss
88
33fe
SS^S
. P"!
383
MS
as
8!
MO
%OH
99
$18
ii
99
99
59
SliiP'
M
as
999
o np-i
iiis!
K5*
si
*
^ •
38
<•»£
^ *
MI
i^l
S8§
R8
98
^•z
iaS
,
s-'i
:*•*
3X
'•s 2
si
£sss=:
s~
s*
ii
5**
ii
s§
I
ii
8S
Sl5
III
5S5
H
II
II
m
!??
A A A
S«?
122
-------
I
" I *
£ 1
«< «
i-
?!l
HI
if 5
IfS
II
jsl
58
*i
if
pi
Ml
** l
5$
II
III
!5i
38
i1
t n
18
||
*J
11
is
sl
II
11
11
8
8
P
5
§
|
•
§
s
i
3
8
0
0
n
0
|
i
0
b
n
b
i
p
s
S
s.
i
I
«
«
S
5
3
S
8
P
S
p
n
b
•>
S
§
b
n
?
8
O
b
b
A
i
N
|
O
b
b
i
§
$
1
I
M
D
E
F»
0
S
8
p
S
p
a
|
5
§
§
5
S
n
S
§
b
o
•>
K
O
b
b
S
s
b
K
I
I
N
O
i
i
0
MS
M!
2s|
!s»S
|aS
35|
MS
MS
25*
MS
p * *
Mj
123
-------
APPENDIX F
SUMMARY - METHOD BLANKS
124
-------
1!
WP* ID*
oo»>»
iMM
*
-------
APPENDIX G
SUMMARY - DUPLICATE TEST SET RESULTS
126
-------
»•
F
k
a^
s
« '
83
2*^
ag
s!
M
39*
5-
I
« i
«i
•8
/
S
/
3
3--
8 ••
sat
•»,,*
,
-J3
^
"
i
1*SECSS2S
•44444444
II
II
127
-------
i
il
1
Xdrt
cc
O f
s
2*
**
II
Ii
33
88
38
3
°S
!§'-
it
as
99
88-5
38<
38
-*
" « •• «*
5S35
33S
|S2S«
S83S--
is*
3883"
Al ^r ~ "^
asss
3S
?*
II!
3 j
S
3-1
*«-»£
»
! •. »
. • ^ *
S -
ii
ii
!fs
is*
M
128
-------
IT
Q
fr
-------
IP
ill
Ifs
S8S
o-i*
?*«
s i
51
Sta$35*«
i
s
5 1
11
38
a.
P
I**
HI
IP
II
51
S3
K^SXS
»st
S3
82
823
S3
az
SB
3«S
*l
aaa
§5
« £
•»«-*
€)_ /
88$
as
•-rS
;>«»
es
•P$
3!
*•
*?i
p
130
-------
-------
1
i
sas
ass
si si
S88
si si a
88
siaa
s §
388
aaa
ss°::
288
51 a
as
83
z~2
«_*
?3S
»)
*l £
9
40
3 It
3S
SX3-S
f> •* F^
-I
sa
sat
3S
3S3S
pis
999
*«
99
S
^P,/
as'o
• 9
385-1
999
99
E«»2
999
SS3
^ £
SSS
9|9
n
2
8« 2
99
CO
S
999
^ * «*
i^SSKJK
p
as
»
§8*
SSJJo
83
138
999
a
ss
999
999
*S?3~
ii
999
99
3S
J539JS8
n — £
999
xs
M
S8S
ȣ
?i
285
388
s»a
33
«;**
s
S8
!!
ii
33*
1
!
*»3
«S*
888
Hm
U
IS
?!I
11
8
U
5515
533
SS?
1>I«
III
a EI
iii
823
/ ?
9^9
999.
m
333
132
-------
-------
rn
i
o
S
to
SU
m
f a-
ffg
-1 at
fi>
CD (Q
CO CD
CD !3
o
03
i
•o
,mr-
03
1
O
------- |