innHHhmm
¦¦¦¦¦('¦¦¦¦¦¦¦¦¦¦
¦¦¦¦¦¦¦¦¦¦¦run
¦¦¦¦¦¦¦¦¦¦¦hann
¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦¦
Dispersant Effectiveness, In-Situ Droplet Size
Distribution and Numerical Modeling to Assess
Subsurface Dispersant Injection as a Deepwater Blowout
Oil Spill Response Option
and
Evaluation of Oil Fluorescence Characteristics to
Improve Forensic Response Tools
£EPA
United States
Environmental Protection
Agency
EPA/600/R-16/152 | September 2016 | www.epa.gov/research
Office of Research and Development
National Risk Management Research Laboratory
Land Remediation and Pollution Control Division
-------
EPA/600/R-16/152
September 2016
IAA No. E12PG00037
Final Report
Dispersant Effectiveness, In-Situ Droplet Size Distribution and Numerical Modeling to Assess
Subsurface Dispersant Injection as a Deepwater Blowout Oil Spill Response Option
and
Evaluation of Oil Fluorescence Characteristics to Improve Forensic Response Tools
Submitted to:
Bureau of Safety and Environmental Enforcement
Submitted by:
Robyn N. Conmy, Ph.D.
U.S. Environmental Protection Agency, Office of Research and Development
26 W. Martin Luther King Drive, Cincinnati, OH 45268
Tel: 513-569-7090, Fax: 513-569-7620, Email: conmy.robyn@epa.gov
Thomas King. M.Sc.; Brian Robinson. M.Sc; Scott Ryan, M.Sc; Youyu Lu. Ph.D.;
Department of Fisheries and Oceans Canada, Bedford Institute of Oceanography
1 Challenger Drive, Dartmouth, Nova Scotia, Canada B2Y 4A2
Tel: 902-426-4172, Tom.King@mar.dfo-mpo.gc.ca
Mary Abercrombie, M.Sc.
University of South Florida, College of Marine Science
140 7th Ave South, St. Petersburg, FL 33701
*Tel: 727-553-1140, Email: mabercrombie@marine.usf.edu
Michel Boufadel, Ph.D.
New Jersey Institute of Technology, Civil and Environmental Engineering
University Fleights, Newark, NJ 07102
Tel: 973-596-5657, Email: michel.boufadel@njit.edu
Flaibo Niu. Ph.D.
Dalhousie University, Department of Engineering
PO Box 550, Truro, N.S. Canada
Tel: 902-893-6714, Email: haibo.niu@dal.ca
September 30, 2016
IA-E12PG00037 Final Report
Page ii
-------
EPA/600/R-16/152
September 2016
Notice/Disclaimer
The U.S. Environmental Protection Agency (EPA), through its Office of Research and
Development, along with the Department of Fisheries and Oceans Canada (DFO Canada)
conducted the research described herein. This report contains scientific observations from a
series of subsurface oil injection experiments and high resolution fluorescence analyses
which were funded from the Bureau of Safety and Environmental Enforcement (BSEE). It has
been subjected to peer and administrative review through the EPA, DFO Canada and BSEE,
and has been approved for publication as an EPA document, thus the information provided
here should not be parsed. Approval does not signify that the contents reflect the views of
the U.S. EPA, DFO Canada, or BSEE, nor does the mention of trade names or commercial
products constitute endorsement or recommendation for use.
IA-E12PG00037 Final Report
Page iii
-------
EPA/600/R-16/152
September 2016
Forward
The U.S. Environmental Protection Agency (US EPA) 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. To meet this mandate, US EPA's research program is providing
data and technical support for solving environmental problems today and building a
science knowledge base necessary to manage our ecological resources wisely,
understand how pollutants affect our health, and prevent or reduce environmental
risks in the future.
The National Risk Management Research Laboratory (NRMRL) within the Office of
Research and Development (ORD) is the Agency's center for investigation of
technological and management approaches for preventing and reducing risks from
pollution that threaten human health and the environment. The focus of the
Laboratory's research program is on methods and their cost-effectiveness for
prevention and control of pollution to air, land, water, and subsurface resources;
protection of water quality in public water systems; remediation of contaminated sites,
sediments and ground water; prevention and control of indoor air pollution; and
restoration of ecosystems. NRMRL collaborates with both public and private sector
partners to foster technologies that reduce the cost of compliance and to anticipate
emerging problems. NRMRL's research provides solutions to environmental problems
by: developing and promoting technologies that protect and improve the environment;
advancing scientific and engineering information to support regulatory and policy
decisions; and providing the technical support and information transfer to ensure
implementation of environmental regulations and strategies at the national, state, and
community levels.
Cynthia Sonich-Mullin, Director
National Risk Management Research Laboratory
IA-E12PG00037 Final Report
Page iv
-------
EPA/600/R-16/152
September 2016
Abstract
The 2010 Deepwater Horizon oil spill highlighted the need for better understanding the
interaction of dispersants and crude oil during high-pressure releases. This report
summarizes a study to assess the operational performance of subsurface injection dispersant
use on high-pressure releases within a flume tank. Dispersion experiments were conducted
using South Louisiana Crude, Alaskan North Slope Crude and Intermediate Fuel Oil 120 oils,
with Corexit 9500 and Finasol OSR 52 dispersants and four dispersant-to-oil ratios (DOR 0,
1:20, 1:100, 1:200) at warm and cold temperatures. In situ plume dispersion was monitored
for particle concentration and Droplet Size Distribution (DSD; LISST-100X), and fluorescence
intensity. Samples were collected for Total Petroleum Hydrocarbons and Benzene-Toluene-
Ethylbenzene-Xylene concentrations. Empirical data was subsequently used as input
variables to refine numerical models of droplet size formation (VDROP-J, JETLAG and
Modified Weber Number). This project also generated a fluorescence library of 25 oil types
to expand community knowledge base on optical signatures as a function of oil type. In
general, the addition of dispersant decreased the oil Volume Mean Diameter (VMD), creating
smaller droplets. Dispersions at DOR =1:20 yielded VMD <70 |am and exhibited bimodal DSD,
suggesting that produced droplets would likely remain dispersed in the presence of mixing
energy. Water temperature did not appear to influence the droplets for lighter crude oils.
DSD results suggest a separation of particles within the plume. In situ fluorescence was found
to be a reliable proxy for oil concentration. These findings have implications for the fate and
transport of oil plumes-both for spill response monitoring and numerical modeling.
IA-E12PG00037 Final Report
Page v
-------
EPA/600/R-16/152
September 2016
Acknowledgements
The research presented in this report was funded by the Bureau of Safety and
Environmental Enforcement (BSEE) through Interagency Agreement E12PG00037. Efforts
were partially supported by the U.S. Environmental Protection Agency - Office of Research
and Development (EPA ORD) and the Department of Fisheries Oceans Canada - Bedford
Institute of Oceanography (DFO-BIO). This work was a highly collaborative effort and the
authors would like to thank all of the contributors. The numerical modeling components
are contributions by collaborators Dr. Michel Boufadel and Feng Gao (New Jersey Institute
of Technology) and Dr. Haibo Niu and Linlu Weng (Dalhousie University). The high-
resolution fluorescence component is a contribution of Mary Abercrombie (University of
South Florida). A special thanks to all of the DFO and BDR Contracting staff who made the
tank experiments possible: Patrick Toole, Claire Mclntyre, Cody Sherren, Jennifer Mason,
Peter Thamer, Gary Wohlgeschaffen, Susan Cobanli, and Rod Doane.
IA-E12PG00037 Final Report
Page vi
-------
EPA/600/R-16/152
September 2016
Table of Contents
Notice/Disclaimer iii
Forward iv
Abstract v
Acknowledgements vi
List of Figures ix
List of Tables xv
Acronyms and Abbreviations xvi
Executive Summary xvii
Task A.l Introduction & Relevance 1
TaskA.2 Experimental Methods 5
A.2.1 Flume Tank Description, Flow Calibration, and Operation 5
A.2.2 Waste Water Treatment 8
A.2.3 Subsurface Oil Injection System 8
A.2.4 Submersible Sensor Deployment 12
A.2.5 VOC Air Monitoring 14
A.2.6 Discrete Water Sample Collection 14
A.2.7 Oil and Dispersant Samples 14
A.2.8 Experimental Design - Core and Complimentary Experiments 16
A.2.9 Submersible Sensor Calibration Experiments 17
A.2.10 Submersible Fluorometer and LISST Data Processing 18
A.2.11 Analytical Chemistry Analysis 20
A.2.12 Numerical Modeling Methods 21
TASK A.3 RESULTS 22
A.3.1 ANS Dispersion Effectiveness 22
A.3.2 IFO 120 Dispersion Effectiveness 36
A.3.3 SLC Dispersion Effectiveness 46
A.3.4 Gas Condensate Dispersion Effectiveness 51
A.3.5 Tank Dilution Series Fluorescence Measurements 53
A.3.6 VOC Air Monitoring 58
A.3.7 VDROP-J and JETLAG Numerical Plume Modeling 68
IA-E12PG00037 Final Report Page vii
-------
EPA/600/R-16/152
September 2016
A.3.8 Weber Number Scaling Numerical Plume Modeling 70
Task B.l Introduction & Relevance 72
Task B.2 Experimental Methods 81
B.2.1 Sample Preparation 81
B.2.2 Artificial Seawater Protocol 81
B.2.3 Dispersed Oil in Seawater Protocol 81
B.2.4 Spectrophotometric Analysis 82
Task B.3 Results & Discussion 86
B.3.1 Oil Fluorescence Properties 86
B.3.2 Fluorescence as a Function of Chemistry 99
B.3.3 Flume Tank and Baffled Flask EEM Comparison 110
B.3.4 PARAFAC Modeling 112
DORO 114
DOR 1:100 119
DOR 1:20 123
PARAFAC Summary 126
References 128
Appendices (Separate Document) 132
APPENDIX A - Experiment Logs 132
APPENDIX B - Analytical Chemistry Results 132
APPENDIX C-Jet Release LISST Oil Droplet Size Distribution Histograms 132
APPENDIX D - Jet Release LISST Oil Droplet Size Distribution Time Series Contours 132
APPENDIX E - Submersible Fluorescence Time Series 132
APPENDIX F - Excitation Emission Matrix Contours 239
APPENDIX G - VDROP-J and JETLAG Numerical Plume Modeling Report 265
APPENDIX H - Weber Number Scaling Numerical Plume Modeling Report 293
IA-E12PG00037 Final Report
Page viii
-------
EPA/600/R-16/152
September 2016
List of Figures
Figure 1. Photos of subsurface oil injection at the BIO flume tank showing the formation of the
subsurface oil plume. Note that the background grid size is 1.5 cm x 1.5 cm.
Figure 2. Photo of the DFO BIO flume tank (top) and cross-section of the tank showing the high-flow
manifolds used to generate horizontal water currents (not to scale).
Figure 3. Schematic diagram showing the location of the subsurface injector and in situ instrumentation
submerged within the tank.
Figure 4A. Photo of the pressurized oil vessel used to hold the oil for the subsurface release.
Figure 4B. Schematic diagram of the pressurized oil vessel for subsurface oil release system in the flume
tank.
Figure 5. LISST DSD and VMD (left panels) and time series of concentration and particle size (right
panels) for ANS and Corexit 9500 warm water treatments. From top to bottom, DOR = 0, 1:200,
1:100, 1:20.
Figure 6. LISST DSD and VMD (left panels) and time series of concentration and particle size (right
panels) for ANS and Corexit 9500 cold water treatments. From top to bottom, DOR = 0, 1:200,1:100,
1:20.
Figure 7. In situ submersible fluorescence time series of sub-injection plume of ANS and Corexit 9500
warm water (left panels) and cold water (right panels) treatments. From top to bottom, DOR = 0,
1:200, 1:100, 1:20.
Figure 8. Downstream LISST DSD and VMD (left panels) and time series of concentration and particle
size (right panels) for ANS and Corexit 9500 warm water treatments. From top to bottom, DOR = 0,
1:200, 1:100, 1:20.
Figure 9. Downstream LISST DSD and VMD (left panels) and time series of concentration and particle
size (right panels) for ANS and Corexit 9500 cold water treatments. From top to bottom, DOR = 0,
1:200, 1:100, 1:20.
Figure 10. LISST DSD and VMD (left panels) and time series of concentration and particle size (right
panels) for ANS and Finasol OSR 52 warm water treatments. From top to bottom, DOR = 1:200,1:100,
1:20. Refer back to Figure 5 for ANS DOR = 0.
Figure 11. In situ submersible fluorescence time series of sub-injection plume of ANS and Finasol OSR
52 warm water treatments. From top to bottom, DOR = 1:200, 1:100, 1:20.
Figure 12. Downstream LISST DSD and VMD (left panels) and time series of concentration and particle
size (right panels) for ANS and Finasol OSR 52 warm water treatments. From top to bottom, DOR =
1:200, 1:100, 1:20. Refer back to Figure 6 for ANS DOR = 0.
IA-E12PG00037 Final Report
Page ix
-------
EPA/600/R-16/152
September 2016
Figure 13. LISST DSD with TPC for ANS with Corexit 9500 and Finasol OSR 52 warm water treatments.
DOR = 0 (top panel); DOR = 1:20 experiments are middle and bottom panels.
Figure 14. LISST DSD with TPC (Total Particle Concentration) for DOR = 1:20 experiments of ANS and
Corexit 9500 treatments. Water temperatures increase from top to bottom panels.
Figure 15. LISST TPC (Total Particle Concentration) for DOR = 1:20 experiments of ANS and Corexit
9500 treatments as a function of water temperature.
Figure 16. LISST DSD and VMD (left panels) and time series of concentration and particle size (right
panels) for IFO 120 and Corexit 9500 warm water treatments. From top to bottom, DOR = 0, 1:200,
1:100, 1:20.
Figure 17. LISST DSD and VMD (left panels) and time series of concentration and particle size (right
panels) for IFO 120 and Finasol OSR 52 warm water treatments. From top to bottom, DOR = 1:200,
1:100, 1:20. Refer to Figure 16 for IFO 120 DOR = 0.
Figure 18. Downstream LISST DSD and VMD (left panels) and time series of concentration and particle
size (right panels) for IFO 120 and Corexit 9500 warm water treatments. From top to bottom, DOR =
0, 1:200, 1:100, 1:20.
Figure 19. Downstream LISST DSD and VMD (left panels) and time series of concentration and particle
size (right panels) for IFO 120 and Finasol OSR 52 warm water treatments. From top to bottom, DOR
= 0, 1:200, 1:100, 1:20. Refer to Figure 18 for ANS DOR = 0.
Figure 20. LISST DSD with TPC for IFO 120 with Corexit 9500 and Finasol OSR 52 treatments at warm
temperatures. DOR = 0 (top panel); DOR = 1:20 experiments are middle and bottom panels.
Figure 21. LISST DSD and VMD (left panels) and time series of concentration and particle size (right
panels) for IFO 120 and Corexit 9500 cold water treatments. From top to bottom, DOR = 0, 1:200,
1:100, 1:20.
Figure 22. Downstream LISST DSD and VMD (left panels) and time series of concentration and particle
size (right panels) for IFO 120 and Corexit 9500 cold water treatments. From top to bottom, DOR =
0, 1:200, 1:100, 1:20.
Figure 23. LISST DSD and VMD for IFO 120 (top; DOR = 1:100) and ANS (bottom; DOR = 1:200) with
Corexit 9500 during cold water treatments.
Figure 24. LISST DSD and VMD (left panels) and time series of concentration and particle size (right
panels) for SLC and Corexit 9500 warm water treatments. From top to bottom, DOR = 0, 1:200,1:100,
1:20.
Figure 25. Downstream LISST DSD and VMD (left panels) and time series of concentration and particle
size (right panels) for SLC and Corexit 9500 warm water treatments. From top to bottom, DOR = 0,
1:200, 1:100, 1:20.
Figure 26. LISST DSD with TPC for SLC with Corexit 9500 treatments at warm temperatures. DOR = 0
(top panel); DOR = 1:20 experiments are bottom panels.
IA-E12PG00037 Final Report
Page x
-------
EPA/600/R-16/152
September 2016
Figure 27. In situ submersible fluorescence time series of sub-injection plume of SLC and Corexit 9500
warm water treatments. From top to bottom, DOR = 0, 1:200, 1:100, 1:20.
Figure 28. LISST DSD and VMD (top panels),time series of concentration and particle size (middle
panels), and fluorescence time series (bottom panels) for Gas Condensate and Corexit 9500 warm
water treatments. Left panels are DOR = 0 and right panels are DOR = 1:20.
Figure 29. Calibration lines for fluorometer response vs TPH concentrations.
Figure 30. Calibration lines for fluorometer response vs BTEX concentrations.
Figure 31. Total Particle Concentration and fluorescence time series for ANS crude oil with Corexit
9500 dispersant.
Figure 32. VOC results for subsurface injection experiments (cold water season) using Alaska North
Slope crude oil and four treatment conditions (no dispersant, DOR 1:200, DOR 1:100, DOR 1:20).
Replicate treatments represented by light blue, dark blue and green colored lines.
Figure 33. VOC results for subsurface injection experiments (warm water season) using Alaska North
Slope crude oil and fourtreatment conditions (no dispersant, DOR 1:200, DOR 1:100, DOR 1:20). Corexit
9500 was used as the treating agent. Replicate treatments represented by light blue, dark blue and
green colored lines.
Figure 34. VOC results for subsurface injection experiments (warm water season) using Alaska North
Slope crude oil and fourtreatment conditions (no dispersant, DOR 1:200, DOR 1:100, DOR 1:20). Finasol
OSR 52 was used as the treating agent. Replicate treatments represented by light blue, dark blue and
green colored lines.
Figure 35. VOC results for subsurface injection experiments (cold water season) using IFO 120 and four
treatment conditions (no dispersant, DOR 1:200, DOR 1:100, DOR 1:20). Corexit 9500 was used as the
treating agent. Replicate treatments represented by light blue, dark blue and green colored lines.
Figure 36. VOC results for subsurface injection experiments (warm water season) using IFO 120 and
four treatment conditions (no dispersant, DOR 1:200, DOR 1:100, DOR 1:20). Corexit 9500 was used as
the treating agent. Replicate treatments represented by light blue, dark blue and green colored lines.
Figure 37. VOC results for subsurface injection experiments (warm water season) using IFO 120 and
three treatment conditions (DOR 1:200, DOR 1:100, DOR 1:20). Finasol OSR 52 was used as the treating
agent (note - these treatments were not tested in triplicate).
Figure 38. VOC results for subsurface injection experiments using gas condensate and two treatment
conditions (no dispersant, DOR 1:20). Corexit 9500 was used as the treating agent.
Figure 39. VOC results for subsurface injection experiments using Sweet Louisiana Crude oil and four
treatment conditions (no dispersant, DOR 1:200, DOR 1:100, DOR 1:20). Corexit 9500 was used as the
treating agent.
IA-E12PG00037 Final Report
Page xi
-------
EPA/600/R-16/152
September 2016
Figure 40. Fluorescence peaks of S. Louisiana sweet crude dispersed in ppb QSE (Quinine Sulfate
Equivalents). Symbols represent Fluorescence Intensity Ratio (FIR) locations and the Center
Wavelength (CWL) reported by sensor manufacturers. Bandwidths (BW) are not shown.
Figure 41. Twenty-five oil samples stored in glass bottles.
Figure 42. Trypsinizing baffled flasks containing dispersed oil in artificial seawater (top) and
corresponding samples removed from each flask, ready for spectrofluorometric analysis.
Figure 43. Alaska North Slope dispersed oil in artificial seawater at DOR 1:20 with locations of Fmaxi,
FmaX2, Fmax3 and FmaX4 indicated. Note that maximum fluorescence intensity at FmaX3 is mostly obscured
by masking of second order Rayleigh scattering.
Figure 44. Photographs of pre-analysis samples and corresponding example EEMs of Type I (left) and
II (right) oils; DOR = 1:20 for Arabian Light (light oil, API gravity > 31.1°), Mesa (medium oil, API gravity
22.3 - 31.1°) and heavy oils (IFO 40 and Santa Clara, API gravity < 22.3°).
Figure 45. Fmaxi fluorescence for Light Oils (API gravity > 31°), in order of increasing density: 1. Scotian
Shelf Condensate, 2. Federated, 3. Brent, 4. MC252—Discoverer Enterprise, 5. Hibernia, 6. MC252—
generic, 7. Terra Nova, 8. Gullfaks, 9. Arabian Light. Note discrepancy in Scotian Shelf Condensate
fluorescence pattern (circled) from that of all other Light Oils. It's particularly unusual that
fluorescence intensity at highest DOR is lower than that at DORs 1:200 and 1:100.
Figure 46. Fmaxi fluorescence for Heavy Oils (API gravity < 22.3°), in order of increasing density: 1.
Santa Clara, 2. IFO 40, 3. Cold Lake Dilbit, 4. Access Western Blend Dilbit, 5. Hondo, 6. IFO 120, 7. IFO
180, 8. Belridge Heavy, 9. IFO 300. Note discrepancy in Intermediate Fuel Oils (circled) from that of
all other Heavy Oils.
Figure 47. For all oil types at DOR 0, total concentration of 2-ring, 3-ring, and 4-ring PAHs (ng/L) against
fluorescence intensity (RU) at Fmaxi (top), and against FmaX2 (bottom). Strong linear correlation exists
between 2-ring PAHs and Fmaxifluorescence, but little to no correlation between 3-ring or 4-ring PAHs
and Fmaxi fluorescence intensity (top). Strong linear correlation also exists between 2-ring PAHs and
Fmax2, but no correlation between 3-ring PAHs or 4-ring PAHs and FmaX2 (bottom).
Figure 48. For all oil types at DOR 0, total concentration of 2-ring, 3-ring, and 4-ring PAHs (ng/L) against
fluorescence intensity (RU) at Fmax3 (top), and against Fmax4 (bottom). Strong linear correlation exists
between 3-ring and 4-ring PAHs and both Fmax3 and Fmax4 fluorescence; however, only moderate
correlation exists between 2-ring PAHs and Fmax3 and Fmax4 fluorescence intensity.
Figure 49. For all oil types at DOR 1:20, total concentration of 2-ring, 3-ring, and 4-ring PAHs (ng/L)
against fluorescence intensity (RU) at Fmaxi (top), and against FmaX2 (bottom). A moderate logarithmic
correlation is exhibited between 2-ring PAHs and fluorescence intensity (RU) at Fmaxi and a weaker
correlation between 2-ring PAHs and FmaX2, but no correlation exists between 3-ring or 4-ring PAHs
and fluorescence intensity at either Fmaxi or FmaX2.
IA-E12PG00037 Final Report
Page xii
-------
EPA/600/R-16/152
September 2016
Figure 50. For all oil types at DOR 1:20, total concentration of 2-ring, 3-ring, and 4-ring PAHs (ng/L)
against fluorescence intensity (RU) at FmaX3 (top), and against FmaX4 (bottom). A strong logarithmic
correlation is exhibited between 2-ring PAHs and fluorescence intensity at FmaX3. Moderate
correlations exist between 3-ring PAHs and FmaX3 as well as between 2-ring PAHs and Fmax4. However,
only a weak logarithmic correlation exists between 4-ring PAHs and fluorescence intensity at Fmax3,
and there is no correlation between 3-ring or 4-ring PAHs and Fmax4.
Figure 51. Chemical Dispersibility Ratio (CDR) vs. decreasing oil density (top) and Fluorescence
Dispersibility Ratio (FDR) vs. decreasing oil density (bottom) show only a weak correlation between
chemistry and oil density, and a moderate correlation between fluorescence and oil density. With the
removal of the data point for Scotian Shelf Condensation, correlation between fluorescence and oil
density improves to R2 = 0.71.
Figure 52. Fluorescence Dispersibility Ratio (FDR) vs. Chemical Dispersibility Ratio (CDR) shows weak
correlation between these two ratios.
Figure 53. South Louisiana Crude MC252 EEMS from BFT (left panels) and tank experiments (right
Panels) for DOR = 0, 1:100 and 1:20.
Figure 54. Example of split half validation for the 6-component model of 25 oil types at DOR 0 showing
individual fit of data splits (Set 1, left; and Set 2, right) compared to overall model for Mode 2 (top) and
Mode 3 (bottom) loadings.
Figure 55. Mode 3 Loadings (Excitation) and Mode 2 Loadings (Emission) for all 25 oil types—DORO
using 6-component model. Note difference in x-axis scales. Although components are tightly spaced,
all appear as separate and distinct peaks.
Figure 56. Variation per Component shows Component 1 accounted for >20% to 40% (unique fit and
fit) of the data, while Component 2-contributed 5-10% (unique fit and fit) and Components 3-6
accounted for 5% or less of the data, respectively. While Component 6 accounted for a very low
percentage of the data, the 6-component model was still a better fit to the data than the 5-component
model.
Figure 57. EEM views of the six components of PARAFAC model for 25 oil types at DOR 0. Component
#1: Fmax = Ex 224nm/Em 335nm; Component #2: Fmax = Ex 230nm/Em 340nm; Component #3: Fmax =
Ex 239nm/Em 363nm; Component #4: Fmax = Ex 218nm/Em 290 nm; Component #5: Fmax = Ex
221nm/Em 322nm; Component #6: Fmax = Ex 260nm/Em 474-511nm.
Figure 58. Mode 3 Loadings (Excitation) and Mode 2 Loadings (Emission) for all 25 oil types—DOR 1:100
using 5-component model. Note difference in x-axis scales. Although components are tightly spaced,
all appear as separate and distinct peaks.
Figure 59. Variation per Component shows Component 1 accounted for >35% to almost 50% (unique
fit and fit) of the data, while Components 2-5 accounted for 5% or less of the data, respectively.
IA-E12PG00037 Final Report
Page xiii
-------
EPA/600/R-16/152
September 2016
Figure 60. EEM views of the five components of PARAFAC model for 25 oil types at DOR 1:100.
Component #1: Fmax = Ex 224nm/Em 335nm; Component #2: Fmax = Ex 254-266nm/Em 455-501nm;
Component #3: Fmax= Ex 230nm/Em 344nm; Component #4: Fmax = Ex 242nm/Em 363 nm; Component
#5: Fmax = Ex 218nm/Em 290nm.
Figure 61. Mode 3 Loadings (Excitation) and Mode 2 Loadings (Emission) for all 25 oil types—DOR 1:20
using 5-component model. Note difference in x-axis scales. Effect of full dispersion appears to broaden
and shift emission peaks to longer wavelengths.
Figure 62. Variation per Component shows Component 1 accounted for 25 to 30% of the data (unique
fit and fit) while Component 2 has increased to >10% to 25% (unique fit and fit) of the data.
Contribution from Component 3 and 4 have increased, as well.
Figure 63. EEM views of the five components of PARAFAC model for 25 oil types at DOR 1:20.
Component #1: Fmax = Ex 224nm/Em 335nm; Component #2: Fmax = Ex 233-266nm/Em 432-450nm;
Component #3: Fmax = Ex 230-242nm/Em 501-520nm; Component #4: Fmax = Ex 233nm/Em 349nm;
Component #5: Fmax = Ex 218nm/Em 290nm.
IA-E12PG00037 Final Report
Page xiv
-------
EPA/600/R-16/152
September 2016
List of Tables
Table 1. List of hydrocarbon fluorometers used in this study. QSDE and PAH represent quinine sulfate
dihydrate and petroleum aromatic hydrocarbons, respectively.
Table 2. Water sample collection strategy for the core and complimentary experiments. TPH and BTEX
represent Total Petroleum Hydrocarbons and Benzene-Toluene-Ethylbenzene-Xylene, respectively.
Table 3. Physical and chemical property measurements of the oils used in this study.
Table 4. Step-wise sensor calibration experiment parameters.
Table 5. Calibration equations for the submersible fluorometers. Data in this report have fluorescence
signal in the manufacturer recommended units.
Table 6. Summary of maximum VOC concentrations at the various treatment conditions tested in this
study. Results are for only for warm water experiments.
Table 7. Sensor specifications as listed from manufacturers. Wavelengths listed as Center Wavelengths
(CWL) with Full Width at Half Max (FWHM) and Bandpass (BP). Standards used are QS (Quinine Sulfate
Dihydrate), NDD Salt (Napthalene Disulfonic Disodium) and PTSA Salt (Pyrenetetrasulfonic Acid
Tetrasodium) (From Conmy et al., 2014b).
Table 8. List of oil samples used for EEM analyses. Oils separated by API (American Petroleum Institute)
gravity.
Table 9. EEM fluorescence and chemical characteristics. Refer to Supplemental Table A for full table.
Table 10. Individual hydrocarbon compounds reported as Total Alkanes, Total 2-ring, 3-ring and 4-ring
PAHs.
IA-E12PG00037 Final Report
Page xv
-------
EPA/600/R-16/152
September 2016
Acronyms and Abbreviations
ADV
Acoustic Doppler Velocimetry
ANS
Alaskan North Slope
BIO DFO
Bedford Institute of Oceanography Dept. of Fisheries and Oceans Canada
BSEE
Bureau of Safety and Environmental Enforcement
BTEX
Benzene-Toluene-Ethylbenzene-Xylene
CRRC
Coastal Response Research Center
DCM
Dichloromethane
DE
Dispersion Effectiveness
DOR
Dispersant to Oil Ratio
DSD
Droplet Size Distribution
DWH
Deepwater Horizon Horizon
EEMS
Excitation Emission Matrix Spectroscopy
EPA
U.S. Environmental Protection Agency
GC-FID
Gas Chromatography-Flame Ionization Detector
GC-MS
Gas Chromatography Mass Spectrometry
GoM
Gulf of Mexico
IFO 120
Intermediate Fuel Oil 120
LISST
Laser In Situ Scattering Transmissometry
NEBA
Net Environmental Benefit Analyses
NJ IT
New Jersey Institute of Technology
NRDA
Natural Resource Damage Assessments
NRT
National Response Team
OMA
Oil-Mineral Aggregate
PARAFAC
Parallel Factor Analysis
PSC
Particle Size Concentration
SLC
South Louisiana Crude
STP
Standard Temperatures and Pressures
TPC
Total Particle Concentration
TPH
Total Petroleum Hydrocarbons
UAC
Unified Area Command
VMD
Volume Mean Diameter
VOC
Volatile Organic Compounds
WG-50
Wave Gauges
IA-E12PG00037 Final Report
Page xvi
-------
EPA/600/R-16/152
September 2016
cutive Summary
This report summarizes two projects covered under an Interagency Agreement between the
Bureau of Safety and Environmental Enforcement (BSEE) and the U.S. Environmental
Protection Agency (EPA) in collaboration with the Bedford Institute of Oceanography,
Department of Fisheries and Oceans Canada (BIO DFO), New Jersey Institute of Technology
(NJIT) and Dalhousie University. Both projects dovetail together in addressing the ability to
differentiate physical from chemical dispersion effectiveness using dispersed oil simulations
within a flume tank for improving forensic response monitoring tools. This report is split into
separate Tasks based upon the two projects funded by BSEE:
1) Dispersant Effectiveness, In-Situ Droplet Size Distribution and Numerical Modeling to
Assess Subsurface Dispersant Injection as a Deepwater Blowout Oil Spill Response
Option.
2) Evaluation of Oil Fluorescence Characteristics to Improve Forensic Response Tools.
TASK A: Dispersant Effectiveness, In-Situ Droplet Size Distribution and Numerical Modeling
to Assess Subsurface Dispersant Injection as a Deepwater Blowout Oil Spill Response Option.
The main objectives of work under Task A were to evaluate high velocity subsurface releases
of physically and chemically dispersed oil using a flow-through wave (flume) tank. This project
addressed three issues: (1) performance evaluation of dispersants for subsurface injection into
sub-sea blowouts, (2) tracking, modeling, and predicting the movement and spread of the
deepwater plume and oil surfacing from deepwater blowouts, and (3) evaluating the influence
of dispersant applications in reducing the concentration of volatile organic compounds
emanating from the water surface. Oil dispersion experiments were conducted in the flume
tank at the Department of Fisheries and Oceans Canada, Bedford Institute of Oceanography
(DFO BIO), which is equipped with an underwater oil release system to simulate a high-
pressure release of oil (akin to a deepwater blowout). Subsea plume simulations were
generated with a pressurized underwater oil release system adapted from existing technology
developed by Masutani and Adams (2000). To mitigate wall effects and to generate oil droplets
IA-E12PG00037 Final Report
Page xvii
-------
EPA/600/R-16/152
September 2016
in the size range observed at depth during the Gulf of Mexico Deepwater Horizon (GoM DWH)
oil spill, a high flow-rate of oil (3.8 L/min) was released through a small diameter nozzle (2.4
mm). Although it is impossible to simulate in the tank the extreme hydrostatic pressures that
exist at 1500 m water depth, underwater high-pressure release of crude oil can be simulated
with and without dispersant addition. The researchers also recognize that the shallow nature
of the tank does not allow for investigating the rise velocity of the droplets that would be
observed in a long ("1500m) water column. Rather, the tank allows for gathering data on the
differences in droplet size and distribution during physical and chemical dispersion (akin to
that observed during DWH) and for observing the vertical and horizontal movement of the
droplets. Although results cannot be directly scaled or translated to a deepwater spill in the
ocean, results are still useful for understanding the formation and movement of oil droplets
under varying oil and dispersant type, dispersant amount and water temperature.
A total of 48 core and 24 complimentary flume tank experiments were conducted to evaluate
the effectiveness of dispersant injection and attenuation of the plume as a function of oil type
(US EPA reference oils: Alaskan North Slope (ANS) pipeline blend for a light-medium crude, IFO
120 for a heavy refined product and South Louisiana Crude (SLC) for a light crude, and also a
gas condensate), chemical dispersant type (Corexit 9500 and Finasol OSR 52), dispersant-to-oil
ratio (DOR of 0, 1:20, 1:100, and 1:200; corresponding to DOR concentrations of 0, 5, 1, and
0.5%) and water temperature (< 10 °C for low temperature and > 10 °C for higher
temperature). Experiments were conducted at a fixed horizontal current flow rate of 1 cm/s
(~ l/8th of deep water flow rates in the GoM). Faster current was not permissible as it would
have resulted insufficient time for collection of in situ measurements and discrete samples.
Experiments were conducted using oil at 80 °C, although this is lower than the reservoir
temperatures for the DWH Macondo wellhead (estimated at 130°C), this is as high as the
experimental design would a I low for safety reasons given the limits of the pressurized canister.
Time series dispersion effectiveness was evaluated by measuring dispersed oil concentrations
from samples collected in the flume tank, and via in situ droplet size distribution analysis and
fluorescence measurements. Discrete samples were collected for oil chemical analysis of Total
IA-E12PG00037 Final Report
Page xviii
-------
EPA/600/R-16/152
September 2016
Petroleum Hydrocarbons (TPH) using gas chromatography coupled to a flame ionization
detector (GC-FID) and the analysis of Benzene-Toluene-Ethylbenzene-Xylene (BTEX) via gas
chromatography mass spectrometry (GC-MS), employed to quantify oil concentration and
partitioning of hydrocarbon compounds in seawater.
The produced Droplet Size Distribution (DSD) was determined by using Laser In-Situ Scattering
and Transmissometry instruments (LISST-100X, type C; Sequoia Scientific Inc. Seattle, WA) to
track the full range diameters of chemically and physically dispersed oil droplets. Larger oil
droplets, whether physically or chemically dispersed, may be capable of coalescing and rising
to the surface under less energetic mixing conditions. The LISST measures particle size and
outputs the concentration of particles in 32 logarithmically spaced size bins between 2.5 to
500 |am, thus facilitating a comparison between natural (physical) and chemical dispersion
efficiency of crude oil. All submersible sensors were operated with real-time data acquisition
throughout each experiment. In situ fluorescence was monitored real-time using two Chelsea
Technologies Group AquaTrackas (crude and refined oil types), one Sea Bird - Wet Labs Inc.
ECO (gelbstoff type), two Turner Designs Inc. Cyclops (crude and refined oil types) and one
GmBH Trios (hydrocarbon type) fluorometers. Many of the fluorescence sensors used in this
study are the same models employed to track the subsea plume during the DWH oil spill and
confirm dispersion effectiveness. Sensors used in this work are also ones provided as examples
in the National Response Team (NRT) Subsea Dispersant Monitoring and Assessment Interim
Guidance Document, that states "the Risk Plan should use a properly calibrated oil-specific
fluorometer (e.g., Chelsea UV AQUAtracka, Turner Designs Cyclops, Wet Labs ECO, or
equivalent oil-specific instrument) to enable ongoing improvements in sampling".
Also monitored during experiments was the level of Volatile Organic Compounds (VOC) above
the air-water interface of the tank using a handheld photo-ionization detector based meter to
evaluate concentrations from the perspective of worker safety. Cautioned are the implications
of these shallow water tank results, however as the short vertical water column did not allow
for any stripping or dissolving of volatile compounds into the water column as would be
expected during a deepwater oil release. Correlations between in situ fluorescence data,
IA-E12PG00037 Final Report
Page xix
-------
EPA/600/R-16/152
September 2016
droplet size distribution, total particle concentration, and oil chemistry serve as inputs to the
modeling activities of this project.
Oil droplet size distribution (DSD) data from this study is essential for the improvement of oil
spill trajectory and ocean circulation modeling processes to predict the fate and transport of
subsurface plumes and surface oil slick movement. This has implications for improving the
scientific and response community's understanding on the impacts of dispersant application
at depth, ultimate fate of subsurface dispersed oil plumes and potential natural resource
damages. Recent advancements in the use of numerical modelling have allowed oil droplet
size predictions resulting from a subsurface release. Several different mathematical
approaches have been used to determine how oil would behave flowing out of an orifice at
high pressure. This includes the modified Weber Number technique (Johansen et al., 2013)
and the VDROP-J model (Zhou et al., 2014) to predict oil droplet breakup taking into account
oil viscosity and interfacial tension. However, there is a limited amount of large scale real world
data to help validate the output of these models. This study provided the opportunity to
further test these techniques through the use of several different oil types and treatment
conditions. Additional results from the numerical modelling using data obtained from tank
experiments are presented in Appendix G, with Part 1 using the modified Weber Number and
Part 2 using VDROP.
The premise for this research is that the evaluation and efficacy of chemical dispersants at
depth will differ dramatically from conventional use of chemical dispersants for treating
surface oil slicks. This is due to difference in mixing energy, where for surface slicks is provided
mainly through naturally occurring surface waves and currents, particularly breaking waves.
Monitoring of DSD is essential in differentiating between chemically and physically dispersed
oil. Tank observations using underwater injection experiments provide evidence of stable
dispersion that may be expected during subsea dispersant injection. Larger oil droplets,
whether physically or chemically dispersed, may be capable of coalescing and rising to the
surface under less energetic mixing conditions. The experimental results from this work
demonstrate the chemical dispersion of oil into small droplets and help to predict the
IA-E12PG00037 Final Report
Page xx
-------
EPA/600/R-16/152
September 2016
likelihood of coalescence and resurfacing of oil. Results of the project provide spill responders
with critical information on the utility of subsurface dispersant application as an oil spill
response option and the modeling capabilities that are available to predict oil trajectory during
deep water blowouts. Both assist decision-making regarding countermeasures.
TASK B: Evaluation of Oil Fluorescence Characteristics to Improve Forensic Response Tools.
This project addresses the evaluation of oil fluorescence characteristics and sensor
performance for improving response tools used to inform oil spill countermeasure decision-
making. Fluorescence has long been used as 'one tool in the toolbox' for surface spills and used
to supplement visual confirmation during response efforts. Recent oil and gas production in
extremely remote locations brings an increased risk of spills in under-the-ice and/or deep-sea
environments. For releases in these environs, responders will be evermore reliant on
submersible sensors for plume tracking when the human eye cannot be employed. As such,
the oil spill community has identified the need for better characterization of spilled oil by
fluorometers.
Submersible fluorometers deployed during the 2010 DWH oil spill highlighted the challenges
in ensuring selection of the optimum sensor configuration as fluorescence peaks occur over a
wide nanometer range, vary in shape and wavelength position, are dependent on oil type due
to chemical differences, and are affected by the addition of dispersants. This project addresses
these concerns through the following objectives: (1) Characterization of oil optical properties
as a function of oil type, DOR and concentration; (2) Generation of a comprehensive Excitation
Emission Matrix Spectroscopy, or Matrices (EEMs) library that will be subjected to advanced
statistical analyses for identification of wavelength regions best suited for oil detection; and
(3) Evaluation of sensor performance through a series of experiments in a flume tank capable
of static and flow-through operations, where sensor data will be validated with chemical and
optical analyses.
IA-E12PG00037 Final Report
Page xxi
-------
EPA/600/R-16/152
September 2016
A series of bench-scale dispersed oil-in-seawater experiments were conducted on 25 oils at 4
dispersant-to-oil ratios (DORs) using Corexit 9500 chemical dispersant. Analysis of the
resulting 3D fluorescence EEMs show oil-specific results as well as differing effects of
dispersant and DORs. Results will inform the identification of optimum oil detection
wavelengths in the marine environment as well as confirmation of the chemical effectiveness
of dispersant application. Samples were prepared using baffled flasks to physically disperse
the oil within seawater. The effect of dispersant on oil-specific fluorescence is shown, where
shifts in intensity and peak wavelengths were observed. Results were compared to chemistry
results of oil components.
Results of the laboratory EEMs analysis were compared to EEMs collected under Task A of this
project to compare the applicability of baffled flask fluorescence to large scale mixing
experiments in the flume tank.
Given recent advances with in situ fluorometers, enabling detection at lower UV-wavelengths,
these findings help to discern wavelength regions influenced by dispersed oil within seawater,
improve the interpretation of fluorescence data, and inform decision-making by responders.
Findings from this project will serve to improve confidence in field data, filling operational gaps
and formulating operational guidelines.
Findings: Tasks A and B
Overall findings from both tasks of this project include:
1. Addition of either Corexit 9500 or Finasol OSR 52 chemical dispersants to Alaskan
North Slope (ANS), IFO 120 and South Louisiana Crude (SLC) oils decreased the
Volume Mean Diameter (VMD) and shifted the DSD to smaller droplets. In general,
Corexit 9500 produced smaller droplets compared to Finasol OSR 52.
2. Dispersions created without chemical dispersants or DOR = 1:200 yielded VMD larger
than 70 |am and exhibited unimodal DSD. Dispersions created with DOR = 1:20 yielded
VMD between 2.5 to 70 |am size range with a bimodal distribution. This suggests that
produced droplets from a DOR = 1:20 dispersant injection with ANS would likely
IA-E12PG00037 Final Report
Page xxii
-------
EPA/600/R-16/152
September 2016
remain dispersed in the presence of mixing energy given the larger proportion of
small droplet sizes observed.
3. Particle size analyses near the injection release (LISST Release) exhibited larger VMD
compared to those generated further downstream from release in the tank (LISST
Downstream) indicating a shift from larger to smaller droplets within the plume, with
and without the presence of dispersant during the 12 minute experiments for ANS
and SLC oils. This effect was not always observed with the heavier IFO 120 oil because
small droplets were less predominant for this heavier oil.
4. For ANS, dispersion with < 70 um droplet VMD was observed for the DOR = 1:20
treatments at both cold and warm water temperatures. Water temperature did not
appear to influence the DSD or VMD for this lighter crude oil. However, a
temperature effect was observed on the Total Particle Concentration (TPC), where
lower temperatures were coincident with fewer particles dispersed within the plume
for a given volume of oil injected.
5. The addition of Corexit 9500 or Finasol OSR 52 to IFO 120 during warm temperature
experiments resulted in a shift in DSD and a decrease in VMD; however bimodal
distribution was not achieved and even DOR = 1:20 did not yield VMD less than 70
|am in most cases. At cold water temperatures, lowerdroplet sizes were not observed
with the addition of dispersant, where DOR = 1:20 remained well above 200 |am. This
suggests that dispersant addition to this oil at cold or warm temperatures would not
yield droplet sizes that would likely remain in suspension.
6. For experiments conducted at water temperatures less than 5 °C, The LISST particle
size analyzed yielded unexpected DSD where even a unimodal distribution was not
measured. Chemistry and in situ fluorescence data indicate that the oil was in fact
dispersed adequately. This suggests operational problems with the LISST below 5 °C,
even though it is within the operating temperature of the LISST (manufacturer
manual). Additional testing of the cold water temperature limits of the sensor is
recommended.
IA-E12PG00037 Final Report
Page xxiii
-------
EPA/600/R-16/152
September 2016
7. SLC oil was more dispersible compared to ANS for treatments with and without
chemical dispersant. Bimodal distribution was observed during DOR = 1:20 and some
DOR = 1:100 experiments indicating that the jet release of this particular oil into warm
water produced smaller droplets than the ANS.
8. In situ fluorescence serves as a good proxy for oil concentration during the subsurface
injection experiments. Given the experimental design, fluorescence is better suited
for correlation with particle size analysis and concentration. Heterogeneity of the
produced plumes and the short time scale of experiments (~12 min) led to difficulties
in correlations between the plume particle size analyses and chemistry results. This
is in part due to discrete samples representing 15 second averages as opposed to
instantaneous measures given by fluorometers and particle size analyzers.
9. VOC air monitoring was conducted above the tank at two horizontal locations during
experiments. The gas condensate exhibited the highest surface VOC concentrations,
followed by ANS and SLC which exhibited similar values. Lowest concentrations were
observed for IFO 120 experiments. High VOC concentrations in the air were usually
accompanied by lower BTEX concentrations in the water. For all oils tested, the
addition of chemical dispersants (DOR = 1:20) resulted in a reduction in VOC
concentrations within air compared to experiments without dispersant near the jet
release location above the tank.
10. Computer programs for jet hydrodynamics, droplet size distribution, and movement of
oil droplets within the jet/plume were employed where developed models were
calibrated to experimental data obtained from the oil jet experiments in the flume tank.
The models VDROP-J and JETLAG were used to predict the streamwise velocity and the
holdup along the centerline of the plume, where both models were in agreement,
implying that VDROP-J is capable of predicting the average droplet size distribution in
the plume. In the absence of dispersant, the model VDROP-J predicted the oil DSD
measured by the LISST. In the presence of dispersant, the VDROP-J model captured the
overall trend of the DSD, but was challenged in capturing the peak in droplet
concentration observed for 5 microns. The observed peak is could be due to tip-
IA-E12PG00037 Final Report
Page xxiv
-------
EPA/600/R-16/152
September 2016
streaming (when high DORs oil droplets shed filaments from their edges resulting in
smaller droplets), and VDROP-J does not yet have a module for this component.
11. The Modified Weber Number approach developed by SINTEF is a recent and promising
approach for predicting DSD. Previously, the method has been validated solely by a
light crude oil. For this project, median droplet diameters (dso) and the relative droplet
size (dso/D) were calculated based on the measured droplet sizes obtained from the
tank experiments, and the relations between dso/D and modified Weber number,
Reynolds number, and oil concentration were quantified. Results demonstrate that
chemical dispersants tested here reduced the droplet size of ANS in both cold and
warm temperatures and that dispersants tested here are more effective in reducing
droplet size with ANS compared to IFO 120. A two-step Rosin-Rammler approach was
found to better predict the droplet size distribution in the empirical data as indicated
by higher regression coefficients.
12. Fluorescence EEMs were generated for 25 oil types under varying DOR. Oils could be
separated into two categories based on dispersiblity; where light, medium and heavy
oils were found in each category. Fluorescence peaks are chemistry dependent and
were well correlated with Total Petroleum Hydrocarbon (TPH) and Benzene-Toluene-
Ethylbenzene-Xylene (BTEX) concentrations. EEMs generated from tank and Baffled
Flask Test (BFT) experiments were in agreement with respect to fluorescence peak
position and Fluorescence Intensity Ratio (FIR) values as an indication of dispersion
effectiveness.
IA-E12PG00037 Final Report
Page xxv
-------
EPA/600/R-16/152
September 2016
Tas^ I T.roductirn £ cvance
The 2010 Deepwater Horizon (DWH) oil spill in the Gulf of Mexico has highlighted the pressing
need for a better understanding of the interaction of chemical dispersants and crude oil at
ocean depth. Early in the blowout release, partial emulsification of oil was observed as it rose
to the surface from 1500-m depth, and surface slicks were not continuous (JAG report, 2010).
A decision was made to inject dispersants directly at the release point as a possible means to
increase efficiency of dispersion and to potentially reduce the amount of dispersant needed if
applied at the air-sea interface (CRRC Report, 2010). Large quantities of chemical dispersant
were applied via subsurface injection and traditional spraying from aircraft onto the surface
oil slick (Oil Budget Calculator, 2010). At a Coastal Response Research Center (CRRC) workshop to
discuss the use of subsurface chemical dispersants as an oil spill response option,
recommendations to the RRT (Regional Response Teams) by spill response and research expert
attendees were made on potential advantages of subsurface dispersant injection given the
rate of continuous oil release and preliminary evidence of the dispersant efficacy from the
DWH spill (CRRC, 2010). Potential advantages of this application included the fact that the
fresh (unweathered) oil was considered well suited for dispersion, operators were able to
inject the dispersant directly into the oil stream thereby maximizing dispersant/oil contact,
sufficient control of DOR (Dispersant-to-Oil Ratio) could be maintained, injection may minimize
the need for surface dispersant application because of reduced oil surfacing and optimized
subsurface application would likely promote formation of smaller, more stable droplets of oil,
enhancing biodegradation (Lee et al., 2009).
As recommended by the interagency Unified Area Command (UAC) and on-site emergency spill
response coordinators, a large-scale environmental monitoring program was implemented to
detect and characterize dispersed oil based on field data and plume modeling outputs. This
allowed for tracking the subsurface oil plume emanating from the blowout wellhead. Droplet
Size Distribution (DSD) analysis using the LISST-100X Laser in-situ Scattering and
Transmissometry System (Sequoia Scientific Inc. Seattle, WA) and fluorescence intensity from
IA-E12PG00037 Final Report
Page 1
-------
EPA/600/R-16/152
September 2016
submersible fluorometers were used as an indication of Dispersion Effectiveness onboard the
research vessels, where particle concentrations were monitored to evaluate oil dispersion
(presence of small droplets < 70 |am) based on previous studies for surface dispersant
applications (Li et al., 2009b). Data analysis of the monitoring samples provided sound
evidence of the presence of oil-bearing small particles both in surface waters and in the
subsurface plume (JAG report, 2010). Furthermore, a negative correlation between subsurface
dispersant injection and low molecular weight compounds in surface waters was observed. In
contrast, a strong positive correlation was observed in the subsurface. These results suggest
that subsurface dispersant use may have promoted the formation of small oil droplets in the
deep sea. This would likely enhance the natural weathering and dissolution of oil in the water
column, thus suppressing the presence of oil organic compounds in surface waters.
Although subsurface in situ dispersants were used to counter a deepwater spill blowout, much
uncertainty still exists in terms of the DE (Dispersion Effectiveness) with this type of
application. For example, assumptions of the optimal DOR are based on empirical data mostly
obtained from bench-scale experimental protocols that have been designed for testing at
standard temperatures and pressures (STP), whereas conditions at a wellhead on the ocean
floor or anywhere along a riser beneath the ocean surface could be significantly different.
Hence, DOR for direct injection needs to be better understood. Although theoretical analyses
and experiments suggest that jet breakup of the oil is insensitive to the absolute value of
system hydrostatic pressure for incompressible liquid-liquid systems (Masutani and Adams,
2000), the effects of several ambient environmental factors on subsurface dispersant
effectiveness, including high release pressure, high oil temperature, low water temperature,
and the presence of methane and suspended sediments in the oil plume and/or surrounding
water column remain to be clarified. Improved understanding on the influence of these factors
on DE and the interaction of crude oil and chemical dispersant under a range of turbulent
regimes at depth is required for informed decision-making for future subsurface dispersant
use.
IA-E12PG00037 Final Report
Page 2
-------
EPA/600/R-16/152
September 2016
For evaluating chemical dispersion effectiveness, standard laboratory tests are inherently
limited in simulating real field operational performance due to space constraints that are
critical for transport and dilution efficiency (NRC, 2005). To address the need to evaluate
chemical dispersion effectiveness under more realistic oceanographic and environmental
conditions, a meso-scale wave tank capable of generating breaking and regular non-breaking
wave conditions is currently in operation at the Bedford Institute of Oceanography (BIO),
Dartmouth, Nova Scotia. This tank facility has been used previously to characterize the tank
hydrodynamics and the efficacy of several oil dispersant formulations on dispersion of
different oil types, including fresh and weathered crude oils and heavy fuel oils under breaking
wave conditions (Figure 1) (Lee et al., 2009; Li et al., 2008; Wickley-Olsen et al., 2007).
Mathematical modeling and experimental measurements have been used in the
characterization of the fluid dynamics of the flume tank. In modeling, computerfluid dynamics
software packages have been used to conduct numerical simulation of the fluid field and
transport phenomena of the flume tank under both non-breaking and breaking wave
conditions. Experimentally, wave gauges (WG-50) have been used to monitor wave profiling
throughout the flume tank under various hydrodynamic conditions. Acoustic Doppler
Velocimetry (ADV) has been employed to evaluate the in situ instantaneous three-dimensional
velocity distribution, which is used to compute the velocity gradients and energy dissipation
rates (s) in the tank. Using this facility, previous experiments have assessed chemical
dispersant effectiveness as a function of energy dissipation rate and particle size distribution
(Li et al., 2009a) and demonstrated that the effectiveness of a dispersant is strongly dependent
on wave conditions, dispersant type, and oil type (Lee et al., 2009). A strong correlation has
been established between dispersion effectiveness and in-situ droplet size distribution within
the hydrodynamic regime, particularly energy dissipation rate, under a variety of non-breaking
wave and breaking wave conditions (Li et al., 2008; Li et al., 2009a). The flume tank has also
been operated in flow-through mode to accommodate the effects of underwater currents on
dispersion and dilution of oil (Li et al., 2009b; Li et al., 2010). Experiments have also shown the
reliability of fluorescence measurements as a proxy for oil concentration within physically and
chemically dispersed oil (Conmy et al., 2014). Experimental studies have also been conducted
IA-E12PG00037 Final Report
Page 3
-------
EPA/600/R-16/152
September 2016
to better understand oil-mineral aggregate (OMA) formation and the influence of mineral
fines on the physical and chemical dispersion of oil (Lee et al., 2009).
This report summarizes results from a project that addresses the operational performance of
subsurface injection dispersant use on high pressure releases of oil within the flume tank.
Developed methods were focused on monitoring subsurface oil transport by outfitting a new
high-flow flume tank at the Department of Fisheries and Oceans Canada (DFO) Bedford
Institute of Oceanography (BIO) facility with a new underwater high flow rate oil injection
system. In this way, the efficiency of chemical dispersion during high pressure releases within
the tank can be quantitatively evaluated and compared to experiments with physical
dispersion (without dispersant addition). This work has implications for field response options.
To this end, the objectives of this work were to:
1) Refine existing equipment, technologies, and methodologies for subsurface
dispersant application evaluation and monitoring by measuring dispersed oil
concentration, fluorescence, and in situ oil droplet size distribution,
2) Evaluate effects of water temperature and dispersant type on dispersion efficacy
and dispersed oil droplet size distribution of oil at high temperatures,
3) Evaluate dispersion effectiveness (DE) as a function of oil type and dispersant-to-oil
ratio (DOR) for subsurface dispersant injection,
4) Assess the effect of dispersant application on the VOC concentration in air above the
air-sea interface of the flume tank,
5) Integrate droplet size distribution into deepwater blowout transport/behavior
models to enable prediction of the dispersed oil droplets under high flow subsurface
release velocities.
During the DWH spill, small droplet (d < 70|am) concentrations were monitored to aid in
evaluating oil dispersion efficiency. The particle size and distribution data obtained from the
field monitoring program during the DWH oil spill had a significant role in supporting
emergency oil spill response operations, fate and transport modeling, and impact assessment.
IA-E12PG00037 Final Report
Page 4
-------
EPA/600/R-16/152
September 2016
Findings from this study will have significant implications in further supporting emergency
response operations, spill transport models and assessments for future deepwater spills.
Jas< A r rxoerimental Methods
A.2.1 Flume Tank Description, Flow Calibration, a eration
Oil dispersion experiments were conducted in the flow-through flume tank at BIO. The BIO
flume tank is rectangular shaped with dimensions of 32 m in length x 2 m in height x 0.6 m in
width, with an operational water height of 1.65 m. It was fabricated with carbon steel (3/16")
and the interior and exterior surfaces are coated with a marine epoxy paint finish to reduce
corrosion while operating under marine conditions. Two sets of manifolds consisting of five
inflow and outflow pipes (each constructed of 4" PVC pipe and equipped with a ball valve so
that the flow rate can be controlled) are fixed (1.1 m from the outer edges) at both ends of the
tank (Figure 2). Two high flow centrifugal pumps (Magnatex 3575 Series, 3" suction, 4"
discharge, 600 gpm, Houston, TX), one connected to the inflow manifold and the second
connected to the outflow manifold provide a flow-through system used to generate horizontal
water currents in the tank. A fiberglass holding tank is used to supply seawater for the system
to ensure that a constant flow rate is maintained.
Seawater was obtained from the Bedford Basin, which is directly adjacent to the tank. Two
smaller pumps (5 HP Pacer S Series Centrifugal Pump, 110 gpm, Lancaster, PA) were used to
pull seawater (~50 cm below the surface) through a 3" suction hose from the Basin. A foot
valve was installed at the end of the hose to maintain prime water in the line between fillings.
Prior to entering the tank, the seawater was filtered through high-flow polypropylene bag
filters (5 |am and 25 |am, Atlantic Purification, Dartmouth, NS).
During normal operations, the flume tank (31,500 L) and holding tanks (25, 000 L) were filled
with filtered seawater. A stainless steel baffle was mounted (~0.5 m) in front of the influent
manifold to control current flow. Flow gauges on the influent and effluent lines were
IA-E12PG00037 Final Report
Page 5
-------
EPA/600/R-16/152
September 2016
monitored and valve adjustments were made to obtain a balanced flow rate, and so that the
operational volume was maintained throughout the experiment. Water current velocities
were measured at various depths and locations in the tank using an ADV (Nortek Vectrino+,
Boston, MA) and the flow rates adjusted until the horizontal water current velocities (3,5 cm/s)
were consistent at all measured depths.
Figure 1. Photos of subsurface oil injection at the BIO flume tank showing the formation of the
subsurface oil plume. Note that the background grid size is 1.5 cm x 1.5 cm.
IA-E12PG00037 Final Report
Page 6
-------
EPA/600/R-16/152
September 2016
J
¦
Inflow
Outflow
J
Figure 2. Photo of the DFO BIO flume tank (top) and cross-section of the tank showing the
high-flow manifolds used to generate horizontal water currents (not to scale).
IA-E12PG00037 Final Report
Page 7
-------
EPA/600/R-16/152
September 2016
A.2.2 Waste Water Treatment
Oil absorbent pads (New Pig, Tipton, PA) are used to manually remove oil from the water
surface. The remaining water in the tank is removed by pumping it through an effluent pipe
that discharges the waste water over layers of polypropylene PomPom Oil-Mops (New Pig,
Tipton, PA) that filter the waste water by removing any remaining insoluble oil prior to
discharging it back into the Bedford Basin. Water samples are collected from the treated
effluent and the PomPom's are changed if total petroleum hydrocarbon concentrations exceed
the minimum guidelines (10 ppm) for wastewater discharge in Canada. Pads and Oil-Mops are
discarded as oily waste disposal.
A.2.3 Subsurface Oil Injection System
A custom (engineered in-house) subsurface oil injection system was used to generate
dispersed oil plumes in the tank (Figure 3). Briefly, the system consists of a 2 L stainless steel
pressure vessel that rests in a support rack. A series of valves and pressure gauges are
connected to the pressure vessel. The assembled system is fastened to the outer wall of the
tank by way of a quick connect bulkhead fitting. From the same location inside the tank, the
fitting connects the outer assembly to a nozzle (2.4 mm inner diameter), which extends mid-
width perpendicular to the tank wall (20 cm off the bottom and 9 m downstream from the
inflow manifold) and is angled at the tip, so as to direct the discharge plume downstream.
Given the shallow nature of the tank, this release setup enabled using the horizontal length of
the tank to capture the plume movement.
For each experiment, oil or oil/dispersant premix is added to the pressure vessel (Figure 4A) in
order to reduce the influence of any additional confounding factor of mixing effectiveness.
Inside the pressure vessel is a copper coil that is connected to a water bath to permit the oil to
be heated to 80°C, which takes 30 minutes. Although lower than the estimated oil
temperature during the DWH release (~130°C), this is the highest temperature permissible in
the pressure vessel to avoid risk of explosion. The vessel is then pressurized (40 psi for ANS,
SLC and Condensate; 60 psi for IFO 120) with compressed Nitrogen. A ball valve connected to
IA-E12PG00037 Final Report
Page 8
-------
EPA/600/R-16/152
September 2016
the pressure vessel is manually opened and oil is released through the subsurface nozzle into
the flume tank (Figure 4B). The release time and total volume (determined by mass) of oil
injected are recorded. After each experiment, the entire subsurface injector system was
cleaned by flushing repeatedly with toluene, acetone and fresh water until no visible oil
remained prior to next experiment.
Influent
Manifold
Pressurized
Oil Release
Nozzle
1
Effluent
Manifold
1s
I
4.3 i
12.6 m
32 i
Subsurface
Injector Canister
Fluorometers
1: L1SST-100X
o voc Meter
O Sampling Port
Figure 3. Schematic diagram showing the location of the subsurface injector and in situ
instrumentation submerged within the tank.
IA-E12PG00037 Final Report
Page 9
-------
EPA/600/R-16/152
September 2016
Figure 4A. Photo of the pressurized oil vessel used to hold the oil for the subsurface release.
IA-E12PG00037 Final Report
Page 10
-------
EPA/600/R-16/152
September 2016
80°C
0 0
Legend
esq
Valve
^ Heating Coil
•
Nozzle
—\ 1—
Quick Release
1
Pressure Vessel
2
Water Bath
3
Compressed N2
4
Flume Tank
©
Figure 4B. Schematic diagram of the pressurized oil vessel for subsurface oil release system
in the flume tank.
IA-E12PG00037 Final Report
Page 11
-------
EPA/600/R-16/152
September 2016
A.2.4 Submersible Sensor Deployment
Fluorescence- A total of six hydrocarbon fluorometers that are used worldwide during oil spill
response were evaluated during this study (Table 1). The fluorometers were mounted on an
aluminum frame located 4.3 m from the oil release point with their UV windows and at a depth
of 0.4 m. The instruments were attached to a crosspiece support bar, so that they were all
located the same distance downstream from the oil release point with the UV window pointed
directly down at the bottom of the tank.
Table 1. List of hydrocarbon fluorometers used in this study. QSDE and PAH represent
quinine sulfate dihydrate and petroleum aromatic hydrocarbons, respectively.
Instrument
Excitation/Emission wavelengths and Units
Chelsea UV AQUAtracka (Refined)
239/360nm, |ag/L Perylene
Chelsea UV AQUAtracka (Crude)
239/440nm, |ag/L Carbazole
Turner Designs Cyclops (Fine Oil)
254/350nm, Volts
Turner Designs Cyclops (Crude Oil)
365/510nm, Volts
Sea Bird - WET Labs ECO-FLU
370/460nm, jig/L QSDE
GmbH Trios
254/360nm, jig/L PAH
Several different data acquisition systems were used to control and collect data from the in
situ fluorometers. The GmBH Trios was operated by the manufacturer's power supply and data
acquisition system using the MSDA_DE software, which provided a real-time display of the
signal intensity in calibrated units of |ag/L PAH. The sampling rate was set at one reading every
five seconds and raw data was saved as a comma delimited (.csv) file. The two Turner
instruments were connected to a Databank Handheld Datalogger (Turner Designs, Sunnyvale,
CA), which powered both instruments and recorded data at a sampling rate of 1 reading every
IA-E12PG00037 Final Report
Page 12
-------
EPA/600/R-16/152
September 2016
3 seconds. The datalogger auto-gain feature cycles through settings of lx, lOx, and lOOx
depending on the signal intensity. Raw data was recorded as signal intensity in mV and was
offloaded from the datalogger via USB connection to a laptop and saved as a text (.txt) file. The
Sea Bird - WET Labs and Chelsea instruments were connected to a custom-built power supply
and data acquisition system (Pace Scientific XR5-SE datalogger; Mooresville, NC), which
collected data from the instruments at a sampling rate of one reading per second. The signal
was recorded internally on the datalogger and then sent via wireless connection to a laptop in
real-time display. Raw data was recorded as signal intensity in mV and offloaded as a .txt file.
Particle Size Analysis - Oil droplet size was measured in situ using two LISST-100X particle size
analyzers (Sequoia Scientific, Seattle, WA). The instrument measures particle sizes in the range
of 2.5 - 500 |am in 32 logarithmically spaced bins. The first LISST was located immediately after
the aluminum frame supporting the fluorometer package at a distance of 5.1 m from the oil
release point and the second LISST was located at 16.9 m from the oil release point and both
at a depth of 0.4 m (Figure 3). Placement was informed by the numerical modeling team of
this project to maximize oil droplet detection without saturating the instrument. Both
instruments were connected via a 20 m cable to laptops running the LISST-SOP data acquisition
software (version 5). Prior to the start of each experiment, a background scatter file of the
seawater quality in the tank was generated and used later to subtract from the final
experimental data file. The instruments were operated in real-time mode with a sample
acquisition rate of one measurement every three seconds.
Supplemental Measures - Weather conditions (air temperature, wind speed, wind direction,
humidity, rainfall) for all experiments were recorded using a Vantage VUE Weather station
(Davis Instruments, Hayward, CA). Water temperature and salinity were measured using a YSI
handheld probe. Underwater video of oil droplets and the transport of the plume were
captured using a GoPro Hero4 digital camera, as well as a Sony RX100 III digital camera with
underwater housing.
IA-E12PG00037 Final Report
Page 13
-------
EPA/600/R-16/152
September 2016
A.2.5 VOC Air Monitoring
Surface volatile organic compound (VOC) concentrations were monitored using handheld
ToxiRAE Pro PID portable gas detectors (RAE Systems, San Jose, CA). Two detectors were used
for each experiment, and they were positioned 0.4 m above the water surface at distances of
5.1 and 16.9 m from the oil release point (Figure 3). The detectors were calibrated using a
certified 25 ppm benzene calibration gas (AirLiquide, Dartmouth, NS) according to the
manufacturer's recommended procedure. Instrument drift was checked periodically against
the calibration gas and recalibrated if necessary. During the experiments, the handheld meters
were set to datalogging mode, which recorded VOC concentrations as ppm of benzene every
three seconds. This data was offloaded and saved as a .txt file for processing.
A.2.6 Discrete Water Sample Collection
Water samples for chemical analysis were collected at various time points throughout the
experiments (Table 2). Three stainless steel tubes were attached to the aluminum
fluorometer frame, so that the end of the tube was located at the same depth as the
instrument UV windows (0.4 m). These were attached via peroxide cured silicon tubing (Cole
Parmer, Vernon Hills, IL) to a Masterflex L/S multi-channel digital peristaltic pump (Cole
Parmer, Vernon Hills, IL) which flowed to a three-way valve system. When the valve was set to
bypass mode, the water in the lines was continuously primed and flowing, so it could
instantaneously be switched to sample mode to allow for sample collection. The pump flow
rate was set to approximately 120 mL/min, and all tubing was flushed with clean seawater for
5 minutes prior to the start of any experiment. Tubing was replaced on an as needed basis.
Water samples from the effluent manifolds were also collected through a 1" sampling valve at
the exit of effluent pipe prior to it entering the treatment system.
A.2.7 Oil and Dispersant Samples
Four different hydrocarbon products were tested in this study to cover a range of viscosity and
physico-chemical characteristics: Two crude oils, a fuel oil, and a gas condensate. Samples of
Alaska North Slope crude oil (ANS) and Intermediate Fuel Oil 120 (IFO 120) were obtained from
IA-E12PG00037 Final Report
Page 14
-------
EPA/600/R-16/152
September 2016
BSEE. Sweet Louisiana Crude was obtained from NOAA. Gas Condensate was obtained from
Exxon Mobil and originated from the Sable Offshore Energy Project. Physical properties of the
samples (Table 3) were measured using an Anton Paar SVM 3000 Stabinger Viscometer (Anton
Paar, Saint Laurent, QC). Supplies of chemical dispersants (Corexit 9500 and Finasol OSR 52)
were purchased from the manufacturers.
Table 2. Water sample collection strategy for the core and complimentary experiments. TPH
and BTEX represent Total Petroleum Hydrocarbons and Benzene-Toluene-Ethylbenzene-
Xylene, respectively.
Time (min)
TPH
(Tank)
TPH
(Effluent)
BTEX
(Tank)
BTEX
(Effluent)
Fluorometry
(Tank)
Fluorometry
(Effluent)
Background
X
X
X
T = 0
X
X
X
H
II
O
Ln
X
T= 1.0
X
X
X
T= 1.5
X
T = 2.0
X
X
X
X
T = 2.5
X
T = 3.0
X
X
T = 3.5
X
H
II
b
X
X
X
X
H
II
Ln
X
T = 5.0
X
X
X
H
II
CT)
b
X
X
X
X
H
II
0°
b
X
X
X
X
X
T= 10.0
X
X
X
X
T= 12.0
X
X
X
X
Total #
Samples/Expt
16
7
11
5
3
1
Total # of Samples
Analyzed
TPH-1725
BTEX-1200
Fluorometry - 300
IA-E12PG00037 Final Report
Page 15
-------
EPA/600/R-16/152
September 2016
Table 3. Physical and chemical property measurements of the oils used in this study.
Measurement
Density
Kinematic
BTEX Content
Temperature
Viscosity
Oil Type
cc)
(g/mL)
(centistokes)
(%)
Alaska North Slope
50
0.8529
6.4
2.3
(ANS)
40
0.8600
8.3
25
0.8704
13.1
15
0.8777
18.9
Intermediate Fuel Oil
50
0.9345
134.0
0.2
(IFO 120)
40
0.9411
240.3
25
0.9515
781.4
15
0.9587
2481.5
Gas Condensate (CND)
50
0.7247
0.4
13.4
15
0.7466
0.5
Sweet Louisiana Crude
50
0.8219
3.2
2.4
(SLC)
40
0.8291
4.0
25
0.8733
5.8
15
0.8473
8.2
A.2.8 Experimental Design - Core and Complimentary Experiments
Both the flume tank and holding tanks were filled with filtered seawater as described above.
Seawater temperature and salinity were recorded using a handheld probe (YSI Incorporated,
Yellow Springs, OH). After the flume tank was filled, the in situ instrumentations including the
fluorometers, LISSTs, and VOC meters were positioned in desired locations as indicated
previously. The subsurface oil release system was filled with oil or oil/dispersant premix, which
was heated to operating temperature. The water supply lines leading to the high flow pumps
IA-E12PG00037 Final Report
Page 16
-------
EPA/600/R-16/152
September 2016
were primed and the inflow and outflow pumps were started. The system was run in
recirculation mode for 10 minutes to allow current flow to stabilize in the flume tank. At a set
time point prior to oil injection (5 minutes), data-logging on all instruments was started and
background seawater samples were collected. After the oil was injected into the tank, the real-
time readout of the fluorometer signal was monitored. Once the first spike in signal intensity
was observed (usually after 2 minutes based on the fluorometer signal readout), a stopwatch
was started and the first chemistry samples were collected. At this point the high flow system
was switched from recirculation mode to flow through, which diverted the water flow into the
effluent treatment system instead of returning it to the holding tank. The experiment ran for
12 minutes, at which point the high flow pumps were turned off and the instrument data
acquisition was stopped. The tank was cleaned and drained as described above. Tank and
instruments were cleaned using Big Orange Degreaser (Zep Superior Solutions, Atlanta, GA),
to prevent any potential contamination between experiments. Instrument windows were
cleaned using disposable alcohol wipes (Bausch and Lomb, Vaughan, ON). Water samples were
returned to the lab and stored at 4°C.
A.2.9 Submersible Sensor Calibration Experiments
The calibration experimental setup was similar to the core and complimentary experiments,
except that the oil was added in a step-wise (tank dilution series measurements) fashion to the
flume tank as shown in Table 4. Calibration experiments were conducted in such a way to
create a series of known concentrations of dispersed oil in the flume tank. Predetermined
amounts of oil and dispersant (Corexit 9500) premix were added to the tank (Alaska North
Slope, ANS, crude was used at a DOR of 1:20) using the subsurface injector.
The flume tank was operated in recirculation mode and oil/dispersant premix injections
occurred every 45 minutes, which provided a sufficient time for the dispersed oil
concentrations to stabilize in the tank (previous testing of this system showed that
hydrocarbon concentrations in the tank are homogenous after 45 minutes of recirculation).
The recirculation of water in the tank provided sufficient mixing energy to allow small droplets
IA-E12PG00037 Final Report
Page 17
-------
EPA/600/R-16/152
September 2016
generated by the subsurface injector to remain dispersed in the water column. In situ
instrumentation was located at the same locations as all other experiments. Water samples
were collected at 45 minute time intervals after each oil addition.
Upon reaching homogeneity in the tank (i.e. 45 minutes after each oil addition), the average
fluorometric intensity signal collected over a 4 minute time period was calculated.
Fluorometers were calibrated to manufacturer suggested units using factors provided.
Triplicate water sample analysis results for Total Petroleum Hydrocarbons (TPH) and Benzene-
Toluene-Ethylbenzene-Xylene (BTEX) were averaged that correspond with the same time
points. Fluorescence and chemistry averages were regressed to generate calibration curves of
TPH and BTEX vs signal intensity for oil additions ranging from 1 to 18 ppm. Higher variability
at low concentrations resulted in the exclusion of some data points in the regression
calculation.
A.2.10 Submersible Fluorometer and USS] Data Processing
Raw LISST data files were processed using a statistically-based quality control script written
using the R statistical package (www.r-project.org). In summary, this script identifies and
removes "Over Range" samples (defined as 0 |j.L/L across all particle size bins) and outliers.
Outliers are defined as any reading that is greater than the moving mean (5 data points before
and after the targeted time point) of the dataset multiplied by four times the standard
deviation (over the same interval as the moving mean). Due to the potential for one or more
extreme outliers to skew both the moving mean and standard deviation calculations for points
around them, this outlier detection routine is run iteratively, excluding previously flagged
points, until no more outliers are detected. Once these QC steps have been performed, the
script calculates a number of parameters from the data such as Total Particle Concentration
(TPC), Volume Mean Diameter (VMD), and Particle Size Concentration (PSC). It then goes on to
detect the plume curve (if present) and time-normalizes the data based on that location. Data
are presented as Droplet Size Distribution (DSD). Plots presented include data 2 minutes before
and 8 minutes after the start of the plume curve. Data from the Downstream LISST were
normalized so that the plume began at t = 5 min in order to visually convey that the plume was
IA-E12PG00037 Final Report
Page 18
-------
EPA/600/R-16/152
September 2016
detected in the tank roughly 3 min after detection by the LISST further upstream near point of
injection release.
Similar to the LISST data, a script was used to detect outliers in data collected from the in situ
fluorometers. Curve detection was then performed and the data was time-normalized to
include 2 minutes of data before, and 8 minutes of data after the start of the plume curve. The
baseline of the plume curve was then calculated using data points observed in the first minute
preceding the start of the curve and this baseline was subtracted from the data. Finally, factory
calibration factors were applied to the data values for each instrument before plotting.
Table 4. Step-wise sensor calibration experiment parameters.
Oil Addition #
Mass of Oil Added for
each Addition
(g)
Cumulative Oil
Concentration in Tank
(mg/L)
1
9.45
0.3
2
9.45
0.6
3
12.6
1
4
63
3
5
94.5
6
6
189
12
7
189
18
IA-E12PG00037 Final Report
Page 19
-------
EPA/600/R-16/152
September 2016
A.2.11 Analytical Chemistry Analysis
Total Petroleum Hydrocarbon (TPH) Analysis - The method used for extraction and processing
of TPH samples was developed by DFO in-house (Cole et al., 2007; King et al., 2015). Water
samples were collected in pre-weighed 125 mL amber glass bottles and filled to approximately
90 mL. Sample bottles were weighed and a mass difference was used to determine the total
volume of the collected water sample. The samples were immediately stored at 4°C until ready
for further processing. Within 24 hrs of collection, 10.0 mL of dichloromethane (DCM) was
added to each sample. The samples were shaken by hand for 30 seconds, and then placed on
a Wheaton R2P roller (Wheaton, Millville, NJ) set at 9 rpm. After 18 hours on the roller, a
Pasteur pipette was used to transfer the DCM solvent layer into a pre-weighed 15 mL
graduated centrifuge tube. The solvent was then evaporated under a gentle stream of nitrogen
using an N-Evap (Organomation, Berlin, MA) and topped up with DCM to a final volume of 1.00
mL. The solvent extract was transferred into an auto-sampler vial and stored at -20°C for GC-
FID analysis.
Sample extracts (1 |aL) were injected using an Agilent 7683 auto-sampler into an Agilent 7890B
GC, using splitless injection set to oven track mode (2°C higher than the oven temperature
program). The column used for separations was a Supelco MDN-5s 30 m x 250 |am x 0.25 |am
(length x i.d. x film thickness). Hydrogen was used as a carrier gas with a flow rate of 3.0
mL/min. The GC oven is programmed to an initial oven temperature of 35°C, held for 2 min,
followed by an increase to 320°C at 20°C/min, and held at 320°C for 10 min, with a total run
time of 26.25 min. The GC flame ionization detector (FID) was operated at 320°C with the
hydrogen flow set at 30 mL/min and the air flow set at 400 mL/min. An eight point calibration
was generated using standards prepared from the appropriate crude oil stock that was used
to generate the TPH samples (e.g. ANS, IFO 120, SLC or Gas Condensate). Peak quantification
was performed using relative response factors. Routinely the method of extraction was tested
for efficiency by a spike and recovery study. Typically, a mean percent recovery of >90% was
calculated from filtered seawater spiked with crude oil. Lab and field blanks were incorporated
in the method.
IA-E12PG00037 Final Report
Page 20
-------
EPA/600/R-16/152
September 2016
BTEX Analysis - EPA Method 8240 (purge and trap) was modified by running a gas
chromatograph/mass spectrometer in selected ion monitoring mode to include ethylbenzene
(Cole et al., 2007). To summarize, water samples for BTEX (benzene, toluene, ethylbenzene
and [m,p & o] xylene) analysis were collected in 40 mL purge and trap vials. The vials were
spiked with 40 |aL of 6N HCI to serve as a preservative, so that they can be stored at 4°C for up
to 14 days.
The purge and trap system was a Teledyne Tekmar Stratum PTC purge and trap concentrator
equipped with a Tenax/silica gel/charcoal trap. The auto-sampler was a Teledyne Tekmar
Aquatek 70-vial unit. The auto-sampler transferred a 5 mL aliquot of sample into the purge and
trap chamber, where it was purged with helium for 11 minutes. During this process, the
volatiles were trapped on the Tenax trap and then desorbed at 225°C for 2 min. The desorbed
gases enter a heated transfer line connected to the Agilent 6890 GC injector and subsequently
proceed to the GC column (Supelco MDN-5s 30 m x 250 |am x 0.25 |am length x i.d. x film
thickness).
The GC oven was programmed at an initial oven temperature of 50°C, held for 8 min, followed
by an increase to 280°C at 409C/min, and held at 280°C for 2 min, for a total run time of 18
min. The gases exiting the GC column were detected by an Agilent 5973 mass selective
detector (MS) used in selective ion mode (SIM) monitoring for six ions: 77, 78, 91, 92, 105 and
106 amu. BTEX standards were prepared in 40 mL purge and trap vials. Samples and standards
were analyzed using this method, along with sample blanks and duplicate samples.
A.2.12 Numerical Modeling Methods
Refer to Appendices G and H for numerical modeling components.
IA-E12PG00037 Final Report
Page 21
-------
EPA/600/R-16/152
September 2016
TASK A3 RESULTS
The overarching objective for this project was to evaluate the operational performance of
the subsurface injection of dispersants during deepwater blowouts. Presented here are the
results from a series of flume tank subsurface injection experiments where dispersion
effectiveness was evaluated via response monitoring tools (fluorescence and particle size
analyzers), discrete water sample chemistry analysis and VOC air monitors. The logs for all
experiments conducted can be found in Appendix A. Corresponding chemistry results for
each experiment are tabulated in Appendix B.
A.3.1 ANS Dispersion Effectiveness
Injection experiments were conducted using ANS crude oil, chemically dispersed with Corexit
and Finasol. Regardless of warm (> 11°C) or cold (5.4 - 10.7°C) water temperatures, the
addition of the two tested dispersant lowers the VMD of ANS and shifts the DSD to smaller
droplets within the plume. An example of this trend is shown in Figures 5 and 6. Note that
LISST histogram plots have constrained Y-axes; thus lines that extend slightly above the top
of the plot area represent values that were truncated. Histograms in these figures
correspond to time points at the leading edge of the plume (~2-3 min from oil release).
Contour plot X-axis represents experiment elapsed time. Plots for triplicate experiments for
each treatment are shown in Appendices C and D. All plots represent data from the LISST
positioned closest to the jet release (denoted as Jet Release LISST throughout the document)
and in close proximity to the submersible fluorometers. Histograms represent the particle
concentration for a given size class (Y axes). Contour plots represent the 10 minute time
series of the plume, where colored contours represent the particle concentration
(normalized to max value for comparison purposes), Y axes represent the droplet sizes in |am
and X axes are time in minutes. Time is elapsed time since oil injection into the tank. These
contours allow for ascertaining how the DSD shifts over the duration of the release. A second
LISST positioned further downstream of release (denoted as Downstream LISST throughout
the document) allows for comparing the evolution of the plume in space and time since
release of the plume. For warm temperature experiments, there is a slight decrease in VMD
IA-E12PG00037 Final Report
Page 22
-------
EPA/600/R-16/152
September 2016
for DOR = 1:200 and 1:100 (Corexit) compared to the no-dispersant treatment (DOR = 0), in
this case ~130 |am down to ~80 |am (exact numbers are within text of the figures). A large
shift in DSD is observed for the DOR 1:20 treatment, where VMD is ~ 10 |am. The cold water
treatments exhibit this same trend, where VMD is ~ 10 |am for the DOR = 1:20 treatment
(Figure 6). In situ submersible fluorescence from multiple fluorometers was recorded during
experiments. Example time series for each dispersant and temperature treatment are shown
in Figure 7 and illustrate the impact of dispersant at DOR = 1:20 in the plume. With DORs of
0, 1:200 and 1:100, the plumes tend to exhibit a spike in fluorescence shortly after release
(within 2 min), and then a sharp decline in signal that is brought to extinction by 4 minutes.
For DOR = 1:20, however, the signal remains elevated and with variability for up to 6 minutes.
This indicates that more oil is remaining submerged in the plume for a longer time period.
Time series fluorescence plots for triplicate experiments for each treatment are shown in
Appendix E.
The Downstream LISST positioned further from the jet release and the fluorometers serves
as an indication of plume evolution through the tank. Plots of the Downstream LISST DSD
and VMD for all dispersant treatments for warm and cold water experiments are shown in
Figures 8 and 9, respectively. Comparing these to the LISST results near the jet release
illustrates the decrease in Total Particle Concentration (TPC; represents the maximum
concentration for the entire plume) as the plume disperses through the tank (note the
change of Y axis scale). Also evident is a shift to smaller particles for all DOR treatments as
the plume moves through the tank. Where the decrease in TPC suggests plume dilution in
the tank, the DSD shift to smaller particles suggests that within each experiment larger
droplets were removed from the plume within 6 minutes of the oil release, most likely rising
to the surface of the tank.
Warm water experiments conducted with ANS and Finasol OSR 52 dispersant also yield a
shift in DSD towards smaller VMD for DOR = 1:20 (Figure 10). However the shift is smaller
than that observed with Corexit 9500 (Figure 5), with lowest VMD on the order of ~50-60
IA-E12PG00037 Final Report
Page 23
-------
EPA/600/R-16/152
September 2016
|am. DOR = 1:200 and 1:100 treatments exhibited spikes in fluorescence signal that taper off
within 3 minutes of oil release (Figure 11). Fluorescence for DOR = 1:20 Finasol OSR 52
treatments exhibited a decrease in intensity at ~4 minutes which is faster than that for
treatments with Corexit 9500. The Downstream LISST exhibited a similar shift in DSD and
TPC that was observed with Corexit 9500 treatments (Figures 12 and 13).
Water temperatures for experiments ranged between 5.4 - 20.8 °C. In general there was no
clear trend on the influence of temperature on DSD, VMD fluorescence intensity, or oil
concentrations for the time series for DOR = 0,1:200 or 1:100 treatments. This suggests that
water temperature has little effect on the dispersibility of ANS (80 °C oil temperature) when
released as a jet with little or no exposure to chemical dispersant (in this case the pre-mixing
process prior to release). In contrast, DOR = 1:20 experiments showed a decrease in total
particle concentration (TPC) with decreasing temperature even though no effect was
observed on DSD for the two temperatures. Figure 14 shows three examples of this effect,
where TPC values for each DOR = 1:20 experiment increase as a function of temperature
(Figure 15). It is important to note that for all treatments using ANS, the experiment at the
lowest temperature (SubANS-lOR; 5.4 °C) exhibited anomalous dispersion compared to the
other DOR = 1:200 treatments (Appendices C and D). Because this occurred in only one
experiment out of 33 experiments with ANS, it is difficult to ascribe a cause for this other
than an improper jet release of oil.
IA-E12PG00037 Final Report
Page 24
-------
EPA/600/R-16/152
September 2016
25
2 15
25
20
25
2 15
2.5-3
20
o 10
O
S 5-
2.5-3
TPC = 116.7 i&JL
VMD = 136 um
T = 2.1 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
TPC = 131.5 nUL
VMD = 80.9 um
T = 2.5 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
TPC = 95.5 \i\JL
VMD = 79.5 um
T = 2.3 min.
o 10
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
TPC = 128.7 uL/L
VMD = 9.6 um
T = 3.4 min.
423-499
8-9.5
30-35
113-133
423-499
LISST Particle Size Bins (um)
9—
2 4 6
Time (minutes, normalized)
2 4
Time (minutes, normalized)
t
¦-n
Time (minutes, normalized)
Time (minutes, normalized)
Figure 5. LISST DSD and VMD (left panels) and time series of concentration and particle size
(right panels) for ANS and Corexit 9500 warm water treatments. From top to bottom, DOR
= 0,1:200,1:100,1:20.
IA-E12PG00037 Final Report
Page 25
-------
EPA/600/R-16/152
Particle Size Concentration (Curve TPC Max): SUBANS-9 1387
251
TPC = 147.1 |iL/L
VMD = 83.5 um
20- T = 2.3 min.
£ 15-
o 10-
O
5-
—¦nil III.
30-35
LISST Particle Size Bins (\im)
20
o 10-
O
£ 5-
TPC = 119.3 nL/L
VMD = 86.7 um
T = 2.6 min.
r 10'
i
2.5-3
r—¦¦lllllll lllli—
).5 30-35 113-133 423-499
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
20
o 10
O
£ 5
TPC = 113.1 \iUL
VMD = 62.5 um
T = 2.2 min.
..dllll III.
2.5-3 8-9.5 30-35 113-133 423-499
LISST Particle Size Bins (fim)
TPC = 104.8 |iL/L
VMD = 9.4 um
T = 2.9 min.
c-
2 4 6
Time (minutes, normalized)
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
l
E
3
O _
> J-0.2
u
2 4 6
Time (minutes, normalized)
P"
2 4 6
Time (minutes, normalized)
Q-
2 4 6 8
Time (minutes, normalized)
I
r
i
N
I
Figure 6. LISST DSD and VMD (left panels) and time series of concentration and particle size
(right panels) for ANS and Corexit 9500 cold water treatments. From top to bottom, DOR
= 0,1:200,1:100,1:20.
IA-E12PG00037 Final Report Page 26
-------
EPA/600/R-16/152
September 2016
% 2K,
200
$ 122
I *2
Warm
Cyclop* (Crud« Oi)
rJvf
Chelsea AQUAtracka (Crude Oil)
./V\
Chelsea AQUAtracka (Refined Oil)
•"A.
t *
i o
| |l
i 8:2
J II
* !1
§ II
* 1
5 ~
10
[w^
Chelsea AQUAtracka (Crude Oil)
*J\
V
n
/v
ielsea AQUAtracka (Crude Oil)
\kj" \.y%
Chelsea AQUAtracka (Refined Oil)
j
\ V'\ „
GmbH Trios
r-r"'
..... ... ../
J-=
Cold
1V1
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
vV "v"^
Cyclopa (Cnid» Oil)
Turnof Cyclop* (Raflnod OH)
JS/~v
a
a (Crude Oil)
Isea AQUAtracka (Refined Oil)
Vk
rimf iminutfsi
r Cyclopa (Hannnd OH)
n AQUAtracka (Cmdo oil)
acki (Refined Oil)
GmbH T.toi
T.'"-
Figure 7. In situ submersible fluorescence time series of sub-injection plume of ANS and
Corexit 9500 warm water (left panels) and cold water (right panels) treatments. From top
to bottom, DOR = 0,1:200,1:100,1:20.
IA-E12PG00037 Final Report
Page 27
-------
EPA/600/R-16/152
September 2016
2.5-3
TPC = 39.7 fiL/L
VMD = 86.4 nm
T = 5.3 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
TPC = 41.4 (iL/L
VMD = 63.5 nm
T = 5.3 min.
.¦¦J lib.
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
TPC = 43 \iLIL
VMD = 11.1 fim
T = 6.7 min.
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
TPC = 52 nL/L
VMD = 63.8 um
T = 5.1 min.
—¦¦ill!
i
i...
10
423-499
423-499
C3
2 4 6 8
Time (minutes, normalized)
i
1.0+
CO
i:
¦
i
2 4 6 8
Time (minutes, normalized)
O
2 4 6 8
Time (minutes, normalized)
.r
I
2 4 6 8
Time (minutes, normalized)
Figure 8. Downstream LISST DSD and VMD (left panels) and time series of concentration
and particle size (right panels) for ANS and Corexit 9500 warm water treatments. From top
to bottom, DOR = 0,1:200,1:100,1:20.
IA-E12PG00037 Final Report
Page 28
-------
EPA/600/R-16/152
September 2016
C
o
2 6-
TPC = 23 |iL/L
VMD = 75.1 \im
T = 5.4 min.
r 1.0+
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
TPC = 35.9 liL/L
VMD = 63.9 urn
T = 5.1 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
8-9.5 30-35 113-133
LISST Particle Size Bins (urn)
TPC = 44 nL/L
VMD = 10.1 nm
T = 6.6 min.
¦
UL
2.5-3 8-9.5 30-35
8-9.5 30-35 113-133
LISST Particle Size Bins (urn)
10"
DOC
423-499
g 10^
TPC = 70.3 \xUL
VMD = 59.9 fim |
T = 5.6 min.
—Jl
1
L-
10"
10
10
423-499
2 4 6 8
Time (minutes, normalized)
n
2 4 6 8
Time (minutes, normalized)
:.X>
u
n
2 4 6 8
Time (minutes, normalized)
u
n'
fz
I
2 4 6 8
Time (minutes, normalized)
Figure 9 Downstream LISST DSD and VMD (left panels) and time series of concentration
and particle size (right panels) for ANS and Corexit 9500 cold water treatments. From top
to bottom, DOR = 0,1:200,1:100,1:20.
IA-E12PG00037 Final Report
Page 29
-------
EPA/600/R-16/152
September 2016
20-
25
2.5-3
2.5-3
TPC = 140.2 |iL/L
VMD = 96.1 um
T = 2.5 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
TPC = 157.3 yiL/L
VMD = 85.6 jim
T = 2.8 min.
423-499
o 10
8-9.5 30-35 113-133
LISST Particle Size Bins (jam)
TPC = 152.1 fiL/L
VMD = 51.5 \im
T = 2.1 min.
423-499
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Time (minutes, normalized)
Time (minutes, normalized)
4 6
Time (minutes, normalized)
Figure 10. LISST DSD and VMD (left panels) and time series of concentration and particle
size (right panels) for ANS and Finasol OSR 52 warm water treatments. From top to
bottom, DOR = 1:200,1:100,1:20. Refer back to Figure 5 for ANS DOR = 0.
IA-E12PG00037 Final Report
Page 30
-------
EPA/600/R-16/152
September 2016
SubFIN-02
Turner Cyclops (Crude Oil)
I "SS
— 400
Turner Cyclops (Refined Oil)
t 30-
20 -
Chelsea AQUAtracka (Crude Oil)
I 4-
H 3
^ o -
"§• 6 -
^ o
K'H,,..
I 1 r
rv,
20
/V'v-
Figure 11. In situ submersible fluorescence time series of sub-injection plume of ANS and
Finasol OSR 52 warm water treatments. From top to bottom, DOR = 1:200, 1:100, 1:20.
IA-E12PG00037 Final Report
Page 31
-------
EPA/600/R-16/152
September 2016
TPC = 82.7 liL/L
VMD = 81.5 jim
T = 6.5 min.
II.
2.5-3
2.5-3
10
8-9.5 30-35 113-133
LISST Particle Size Bins (fim)
423-499
O
2 4 6 8
Time (minutes, normalized)
TPC = 43.3 fiL/L
VMD = 72.8 nm
T = 5.9 min.
10'
8-9.5
30-35
113-133
423-499
LISST Particle Size Bins (|im)
2 4 6
Time (minutes, normalized)
TPC = 50.3 \iUL
VMD = 44 um
T = 7.9 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
2 4 6 8
Time (minutes, normalized)
M
1.0+
I
¦- n
Figure 12. Downstream LISST DSD and VMD (left panels) and time series of concentration
and particle size (right panels) for ANS and Finasol OSR 52 warm water treatments. From
top to bottom, DOR = 1:200,1:100,1:20. Refer back to Figure 6 for ANS DOR = 0.
IA-E12PG00037 Final Report
Page 32
-------
EPA/600/R-16/152
September 2016
Jet Release LISST
Downstream LISST
20 -
2 15-
TPC = 116.7 [iLIL
VMD = 136 nm
T = 2.1 min.
25
20
2 15-
o 10-
¦C 5-
25
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
TPC = 128.7 nUL
VMD = 9.6 jim
T = 3.4 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
TPC = 152.1 uL/L
VMD = 51.5 urn
T = 2.1 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
¦d 8
10
•d 8
t 2
10
TPC = 39.7 pUL
VMD = 86.4 \im
T = 5.3 min.
ANS
DOR = 0
...lllll ill.
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
TPC = 43 (iL/L
VMD = 11.1 urn
T = 6.7 min.
ANS-Corexit 9500
DOR = 1:20
LJL
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
TPC = 50.3 \iUL
VMD = 44 urn
T = 7.9 min.
ANS-Finasol OSR 52
DOR = 1:20
8-9.5 30-35 113-133
LISST Particle Size Bins (urn)
Figure 13. LISST DSD with TPC for ANS with Corexit 9500 and Finasol OSR 52 warm water
treatments. DOR = 0 (top panel); DOR = 1:20 experiments are middle and bottom panels.
IA-E12PG00037 Final Report
Page 33
-------
EPA/600/R-16/152
September 2016
20
o 10-
o
£ 5"
TPC = 59.8 \iLIL
VMD = 3.8 urn
T = 4.1 min.
6.8 °C
it
2.5-3
, i
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
TPC = 104.8 \xUL
VMD = 9.4 um
T = 2.9 min
o 10
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
TPC = 128.7 [iUL
VMD = 9.6 um
T = 3.4 min.
14.7 C
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Figure 14. LISST DSD with TPC (Total Particle Concentration) for DOR = 1:20 experiments of
ANS and Corexit 9500 treatments. Water temperatures increase from top to bottom
panels.
IA-E12PG00037 Final Report
Page 34
-------
EPA/600/R-16/152
September 2016
160
120
zr
% 80
40
(_>
CL
0
8 12
Water Temperature (°C)
16
Figure 15. LISSTTPC (Total Particle Concentration) for DOR = 1:20 experiments of ANS and
Corexit 9500 treatments as a function of water temperature.
IA-E12PG00037 Final Report
Page 35
-------
EPA/600/R-16/152
September 2016
A.3.2 1FO 120 Dispersion Effectiveness
Injection experiments were conducted using Intermediate Fuel Oil (IFO 120), chemically
dispersed with Corexit 9500 and Finasol OSR 52. For warm water experiments, temperatures
ranged between 13.5 - 16 °C for treatments with Corexit 9500 and between 17.5 - 20.3 °C for
treatments with Finasol OSR 52. In the DOR = 0, 1:200 and 1:100 treatments using Corexit
9500, VMD typically remained > 200 |am (Figure 16). VMD values were smaller for DOR = 1:20
treatments (~66-120 |am), indicating a shift in DSD, but to a lesser extent than the shift
observed for ANS experiments. Fluorescence data exhibited scatter and noise in the signal for
all but the DOR = 1:20 treatments (Appendix E). A similar trend in DSD, VMD and fluorescence
signal was observed for IFO 120 exposed to Finasol OSR 52 at warm temperatures (Figure 17),
where DOR = 1:200, 1:100 and 1:20 exhibited VMD values of 376.5, 209.5 and 125.8 |am,
respectively. Unlike experiments with ANS, which is less viscous and dense, IFO 120 exposed
to dispersant tended to result in larger oil droplets for a given amount of dispersant added.
Comparing the results of IFO 120 with the two dispersants is challenging because no triplicate
experiments were conducted for Finasol OSR 52 treatments, as the latter treatments were
add-on experiments and not central to the project. In general, from the data collected, Finasol
OSR 52 yielded higher VMD for a given DOR compared to Corexit 9500 at warm temperatures.
As with ANS, the Downstream LISST measured a decrease in TPC and shift to smaller droplet
sizes as the plume moved through the tank for all treatments, but to a lesser extent with DOR
= 1:20 (Figures 18, 19 and 20).
For cold water experiments using IFO 120 exposed to Corexit 9500, temperatures ranged
between (4.9 - 7.5 °C). At these colder temperatures a shift in DSD and VMD was not as
apparent (Figure 21). For DOR = 0, 1:200 and 1:100 VMD typically remained > 223 |am but
was as high as 344 |am. The DOR = 1:20 treatment exhibited VMD of 178-327 |am, suggesting
that this oil was not well dispersed at cold temperatures. Fluorescence time series data were
noisy for all experiments except the DOR = 1:20 (Appendix E). The Downstream LISST
recorded extremely low particle concentrations, further suggesting poor dispersion (Figure
22). During the IFO 120 cold water treatments, one experiment resulted in an anomalous
IA-E12PG00037 Final Report
Page 36
-------
EPA/600/R-16/152
September 2016
DSD histogram that was similar to an anomalous one observed during one of the ANS
experiments (Figure 23). In both cases, the experiments were conducted at the coldest
temperatures during the course of this study (4.9 and 5 °C). Suspected as a possible cause
may be the LISST instrument itself. The manual reports that the lower operating
temperature for the LISTT-100X is -10 °C. However, the data suggests that our particular unit
may have experienced some complications at low temperatures. This is supported by the
fact that the fluorescence signal and chemistry data for these experiments indicate no
anomalies. Further testing would be needed to confirm the effect of low temperatures on
particle size analysis results using our instrument to rule out any potential issues with
operating at temperatures between 5 and -10 °C.
One aspect to note with the IFO 120 cold water experiments is that a few of the treatments
were conducted at water temperatures of ~12 °C, which overlaps with the temperatures of
the warm water group. This was the result of erratic weather patterns that at times were
difficult to work around. Thus, when interpreting the temperature data, caution must be
exercised for these particular experiments (refer to Appendix A for temperature log), and for
the interpretation in this section, they were excluded as they do not represent cold
conditions.
IA-E12PG00037 Final Report
Page 37
-------
EPA/600/R-16/152
Seotember 2016
25
£ 15-
¦£ 5-
2.5-3
2.5-3
25 -
20-
2 15-
o 10-
% 5-
25
£ 15
TPC = 74.4 \iUL
VMD = 279.7 urn
T = 2.5 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
TPC = 164.6 \iUL
VMD = 196 um
T = 2.1 min.
-¦¦¦¦¦HI
8-9.5 30-35 113-133
LISST Particle Size Bins (jam)
TPC = 111.5 nL/L
VMD = 66.4 um
T = 2.6 min.
423-499
TPC = 75.8 uL/L
VMD = 234.5 urn
T = 2.4 min.
423-499
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
&
2 4 6
Time (minutes, normalized)
r
2 4 6 8
Time (minutes, normalized)
2 4 6 8
Time (minutes, normalized)
i
¦o
r
I
r
423-499
r
¦- n
n'
I
™ n
2 4 6 8
Time (minutes, normalized)
Figure 16. LISST DSD and VMD (left panels) and time series of concentration and particle
size (right panels) for IFO 120 and Corexit 9500 warm water treatments. From top to
bottom, DOR = 0,1:200,1:100,1:20.
IA-E12PG00037 Final Report
Page 38
-------
EPA/600/R-16/152
September 2016
20 -
£ 15-
o 10-
S s-
2.5-3
25 -
£ 15-
o 10-
O
t 5-
2.5-3
20
£ 15-
2.5-3
TPC = 18.7 nL/L
VMD = 376.5 |im
T = 2.1 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
-.¦¦J
423-499
TPC = 72.6 |iL/L
VMD = 209.5 urn
T = 2.3 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
TPC = 78.6 \iLIL
VMD = 125.8 urn
T = 3.1 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
423-499
2 4 6
Time (minutes, normalized)
9
423-499
2 4 6
Time (minutes, normalized)
V
rr
r
¦ n
t
¦ n
2 4 6 8
Time (minutes, normalized)
Figure 17. LISST DSD and VMD (left panels) and time series of concentration and particle
size (right panels) for IFO 120 and Finasol OSR 52 warm water treatments. From top to
bottom, DOR = 1:200,1:100,1:20. Refer to Figure 16 for IFO 120 DOR = 0.
IA-E12PG00037 Final Report
Page 39
-------
EPA/600/R-16/152
September 2016
TPC = 20.4 \iLIL
VMD = 197.3 um
T = 6.2 min.
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins ((jm)
423-499
TPC = 62.8 fiL/L
VMD = 211.8 ^m
T = 5.5 min.
-
f 102
-------
EPA/600/R-16/152
September 2016
TPC = 3.8 |iL/L
VMD = 151.2 |im
T = 5.2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
TPC = 16.7 |iL/L
VMD = 146.2 jim
T = 5.8 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
TPC = 49.9 \iLJL
VMD = 122.1 urn
T = 7.1 min.
8-9.5 30-35 113-133
LISST Particle Size Bins ((.im)
10'
423-499
10'
—.illlll-
5 113-133 423-499
10
4 6
Time (minutes, normalized)
Time (minutes, normalized)
4 6
Time (minutes, normalized)
n1
I
¦ n
n1
-
o
<->
n1
o
-------
EPA/600/R-16/152
September 2016
Jet Release LISST
Downstream LISST
2.5-3
TPC = 74.4 jiL/L
VMD = 279.7 jim
T = 2.5 min.
¦d 8 -
o 4
O
£ 2-
TPC = 20.4 \iLIL
VMD = 197.3 jim
T = 6.2 min.
IFO 120
DOR = 0
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
TPC = 111.5 \iLIL
VMD = 66.4 jim
T = 2.6 min.
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins (urn)
423-499
TPC = 68.2 nL/L |FO 120-Corexit 9500
VMD = 56 fim
DOR = 1:20
T = 8.8 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
423-499
8-9.5 30-35 113-133
LISST Particle Size Bins (fim)
TPC = 78.6 \iLIL
VMD = 125.8 nm
T = 3.1 min.
10 -
2 6-
o 4 -
O
% 2-
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
TPC = 49.9 uL/L
vmd = 122.1 jim IFO 120-Finasol OSR 52
T = 7-1min DOR = 1:20
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
Figure 20. LISST DSD with TPC for IFO 120 with Corexit 9500 and Finasol OSR 52 treatments
at warm temperatures. DOR = 0 (top panel); DOR = 1:20 experiments are middle and
bottom panels.
IA-E12PG00037 Final Report
Page 42
-------
EPA/600/R-16/152
September 2016
TPC = 30.8 jiL/L
VMD = 296.3 |im
T = 2.6 min.
-.«¦
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (urn)
TPC = 27.9 fiL/L
VMD = 301.1 nm
T = 2 min.
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
423-499
Time (minutes, normalized)
423-499
TPC = 25.8 |iL/L
VMD = 344 urn
T = 2.5 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (|jm)
M
TPC = 164.9 \±lL
VMD = 280.2 |im
T = 2.2 min.
2.5-3
...llllllllll
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
— 102"
|
N
<35
a>
o
t
CL
101 "
423-499 il
Time (minutes, normalized)
Hf,r
2 4 6 8
Time (minutes, normalized)
2 4
Time (minutes, normalized)
I
¦-n
Figure 21. LISST DSD and VMD (left panels) and time series of concentration and particle
size (right panels) for IFO 120 and Corexit 9500 cold water treatments. From top to bottom,
DOR = 0,1:200,1:100,1:20.
IA-E12PG00037 Final Report
Page 43
-------
EPA/600/R-16/152
September 2016
2.5-3
2.5-3
2.5-3
TPC = 14.3 \iLIL
VMD = 276.2 jim
T = 5.8 min.
-¦¦ll
8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
423-499
TPC = 1.5 jiL/L
VMD = 157.5 nm
T = 5.6 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
TPC = 4.6 |iL/L
VMD = 186.3 \im
T = 7.6 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
423-499
TPC = 20.5 \iUL
VMD = 145.2 urn
T = 5.2 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (urn)
423-499
10"
CO
a>
o
¦¦5
10'
101
10
« 1
2 4 6 8
Time (minutes, normalized)
2 4 6 8
Time (minutes, normalized)
2 4 6 8
Time (minutes, normalized)
2 4 6 8
Time (minutes, normalized)
'
J
J
c
o
ro
c
0)
o
c
o
o
0)
E
3
o
>
J
1.0+
0.2
f
¦-n
n1
U
Figure 22. Downstream LISST DSD arid VMD (left panels) and time series of concentration
and particle size (right panels) for IFO 120 and Corexit 9500 cold water treatments. From
top to bottom, DOR = 0,1:200,1:100,1:20.
IA-E12PG00037 Final Report
Page 44
-------
EPA/600/R-16/152
September 2016
TPC = 194.5 |iL/L
VMD = 365.4 fim
20 -I T = 2.9 min.
15 -
o 10
O
2.5-3 8-9.5 30-35 113-133 423-499
LISST Particle Size Bins (^m)
TPC = 260.5 nL/L
VMD = 318.2 nm
T = 4.7 min.
J
0 J—i 1 — 1—
2.5-3 8-9.5 30-35 113-133 423-499
LISST Particle Size Bins (|im)
Figure 23. LISST DSD and VMD for IFO 120 (top; DOR = 1:100) and ANS (bottom; DOR =
1:200) with Corexit 9500 during cold water treatments.
IA-E12PG00037 Final Report
Page 45
-------
EPA/600/R-16/152
September 2016
A.3.3 SLC Dispersion Effectiveness
Experiments involving South Louisiana Crude (SLC) oil treated with Corexit 9500 were
conducted for warm water conditions (16.6 - 19.6 °C) to compare dispersion between ANS
and SLC. Experiments with SLC yielded higher TPC values compared to ANS results, most
likely the result of slightly larger amounts of oil added to the pressure canister (~25-50 g) due
to the lower viscosity of SLC resulting in more oil injected by the injector, so comparisons
shouldn't be made regarding TPC. The observed VMD of physically-dispersed SLC oil (neat;
DOR = 0; ~123-148 |am) was found to be less than that of ANS (>200 |am). The addition of
dispersant yielded a shift in DSD and VMD to smaller particles, where DOR = 1:200 and 1:100
exhibited diameters of ~91-108 |am, and DOR = 1:20 ranged between ~15 -21 |am, as depicted
in Figure 24. The Downstream LISST results indicate smaller droplet size as the plume moves
through the tank (size fractionation) and a decrease in TPC (plume dilution), further
demonstrating this trend for all oils (Figures 25 and 26). The fluorescence data indicates a
strong signal with little scatter for up to 4 min in these treatments (Figure 27; Appendix E).
Using these results, comparisons can be made to results of SLC with Corexit 9500 from
surface plume simulations (oil released into tank via pour in from flask) from Conmy et al.,
2014a (and unpublished data) from those experiments. No apparent differences between
DSD and VMD for DOR = 0 treatments were found. For DOR = 1:20 VMD values are similar,
however, the range of droplet diameters for surface simulations is larger with particles up to
200 |am. In subsurface injection jet experiments the range of diameters is narrower, where
particles > 100 |am were not observed. This suggests that the combination of the chemical
dispersant tested here, elevated turbulent mixing from the jet release and higher oil
temperature of 80 °C yielded smaller droplets. To discern the dominant factor controlling
the difference, additional testing would need to be conducted.
IA-E12PG00037 Final Report
Page 46
-------
2.5-3
2.5-3
2.5-3
TPC = 110.8 nL/L
VMD = 121.3 um
T = 2.9 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (^im)
TPC = 142.1 |iL/L
VMD = 103.3 \im
1 = 2 min.
.iillll
liiii...
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
TPC = 180.7 |iL/L
VMD = 91.7 um
T = 2.3 min.
423-499
llll
llllllia.
113-133 423-499
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
TPC = 132.5 \iUL
VMD = 18.4 |im
T = 2.5 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
EPA/600/R-16/152
September 2016
o
Time (minutes, normalized)
4 6 8
Time (minutes, normalized)
c
o
10
n
4 6 8
Time (minutes, normalized)
I
I
i
4 6 8
Time (minutes, normalized)
Figure 24. LISST DSD and VMD (left panels) and time series of concentration and particle size
(right panels) for SLC and Corexit 9500 warm water treatments. From top to bottom, DOR
= 0,1:200,1:100,1:20.
IA-E12PG00037 Final Report
Page 47
-------
10
10
£ 6-
2 6-
TPC = 68.8 fiL/L
VMD = 104.5 um
T = 5.6 min.
1.0+
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
TPC = 51.7 uL/L
VMD = 81.2 pm
T = 5.4 min.
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
TPC = 33.8 \iLIL
VMD = 66.8 |im
T = 6.5 min.
8-9.5 30-35 113-133
LISST Particle Size Bins fiimt
TPC = 57.7 \iLIL
VMD = 22.6 um
T = 5.7 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
10
10'
10
10'
10'
10
03
CL-
IO'
423-499
4 6
Time (minutes, normalized)
Time (minutes, normalized)
CD
4 6
Time (minutes, normalized)
4 6
Time (minutes, normalized)
0.4
0.2
r 1.0+
0.8
0.6
0.4
I 0.2
n
0
r1.0+
0.6
0.4
o i 0.2
Figure 25. Downstream LISST DSD and VMD (left panels) and time series of concentration
and particle size (right panels) for SLC and Corexit 9500 warm water treatments. From top
to bottom, DOR = 0,1:200,1:100,1:20.
IA-E12PG00037 Final Report Page 48
-------
EPA/600/R-16/152
September 2016
Jet Release LISST
TPC = 110.8 |iL/L
VMD = 121.3 nm
T = 2.9 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (|im)
TPC = 132.5 |iL/L
VMD = 18.4 um
T = 2.5 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
Downstream LISST
TPC = 68.8 |iL/L
VMD = 104.5 ^im
T = 5
DOR = 0
8-9.5 30-35 113-133 423-499
LISST Particle Size Bins (um)
TPC = 57.7 uL/L
VMD = 22.6 um
T = 5.7 min.
SLC - Corexit 9500
DOR = 1:20
8-9.5 30-35 113-133
LISST Particle Size Bins (urn)
Figure 26. LISST DSD with TPC for SLC with Corexit 9500 treatments at warm temperatures.
DOR = 0 (top panel); DOR = 1:20 experiments are bottom panels.
IA-E12PG00037 Final Report
Page 49
-------
SubSLC-05
Turner Cyclops (Crude Oil)
n tensity |mV
A-
1" ns
=§¦ 10
Turner Cyclops (Refined Oil)
..... f - _
Chelsea AQUAtracka (Refined Oil)
r—----- {/A'' T
SubSLC-02
Turner Cyclops (Crude Oil)
f 150
IT 100
IX
1 - - \'"W
Turner Cyclops (Refined Oil)
S 15 -
S- 10
1 o
Av, -^¦VA./V\
Chelsea AQUAtracka (Crude Oil)
Chelsea AOUAtracka (Refined Oil)
1" 4 -
iS
r\
oooooc
iasoi/61
/ "v _/v
1 1
...
024 Time (minutes) 6810
SubSLC—03
Turner Cyclops (Crude Oil)
f Tsl -
# 100 -
rv. ....
5 30
s- 20 -
1 1o
A- - ..._
5 0
,A.
1" :
/~a.
~ 100 -
A...
I f
/ " v -
0 2 4 Time (minutest 6810
300 .
£¦ 200 -
SubSLC-OSR
Turner Cyclops (Crude Oil)
fK.
1" 40
§• 20
Turner Cyclops (Refined Oil)
!rJsiJ\x
Chelsea AQUAtracka (Crude Oil)
! I
11:
. . ._ __
Chelsea AOUAtracka (Refined Oil)
firit.
1 V
5:=l
- _...
EPA/600/R-16/152
September 2016
Figure 27. In situ submersible fluorescence time series of sub-injection plume of SLC and
Corexit 9500 warm water treatments. From top to bottom, DOR = 0,1:200,1:100,1:20.
IA-E12PG00037 Final Report
Page 50
-------
EPA/600/R-16/152
September 2016
A.3.4 Gas Condensate Dispersion Effectiveness
Injection experiments were conducted with Gas Condensate and Corexit 9500 for warm
water conditions (10 - 12 °C temperature range) and DOR = 0 and 1:20 only. The Gas
Condensate consisted of mostly C15 alkanes and lower PAHs (napthalene and alkylated
derivatives. The VMD for Gas Condensate with no dispersant added ranged between ~150
- 215 |am (Figure 28). With the addition of dispersant, VMD for the triplicates were 60.4, 68.2
and 170.4 |am suggesting that dispersant at DOR = 1:20 shifts the DSD to smaller particles for
most experiments. Large variability in the triplicates was observed, however at this time
there is no clear explanation as to the cause. The corresponding fluorescence data for these
treatments indicate a strong signal with little scatter for up to 3 min in both treatments
(Figure 28; Appendix E).
IA-E12PG00037 Final Report
Page 51
-------
EPA/600/R-16/152
September 2016
TPC = 50.1 |iL/L
VMD = 183.3 [im
T = 2.6 min.
DOR = 0
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (|.im)
Time (minutes, normalized)
2 15
£ 5
TPC = 138.3 [xUL
VMD = 68.2 nm
T = 2.2 min.
DOR = 1:20
8-9.5 30-35 113-133
LISST Particle Size Bins (|am)
Bum = lao
H
i
2 4 6
Time (minutes, normalized)
SubCND-03
subcnd-04
Tumor Cyclops (Crutto Oil)
I I
1 I
DOR = 0
¥ ,0H
1 :
DOR = 1:20
f
1 1
Tuinoi Cyclops Oil)
f so
f 2
' "" J
i H
® o"o
Cholsoa AQUAdacko (Crude Oil)
1 I
% *
* I
4 §
fh __
f:::
% O.O
__)V "
i
a-
i '
B. 1
I '«~
* 11
i
* 'li
i i i » i
T'"— 1
Figure 28. LISST DSD and VMD (top panels),time series of concentration and particle size
(middle panels), and fluorescence time series (bottom panels) for Gas Condensate and
Corexit 9500 warm water treatments. Left panels are DOR = 0 and right panels are DOR =
1:20.
IA-E12PG00037 Final Report
Page 52
-------
EPA/600/R-16/152
September 2016
A.3.5 Tank Dilution Series Fluorescence Measurements
Submersible fluorescence results are presented in units recommended by manufacturers and
using calibration factors provided by the manufacturers. Efforts were made to correlate the
fluorescence intensity with TPH and / or BTEX concentration but were not possible due to
issues inherent with the discrete sample collection. In order to fill bottles for chemical analysis,
a 30 second time period was needed. Due to the short time period of the experiments and the
heterogeneity of the plume concentration through time (evident from the fluorescence time
series), oil concentrations within the bottles represent an average over a 30 second time
period that cannot be aligned with the time series data, which are generated on the time scale
of seconds. Given this fact, a dilution series within the tank using ANS was conducted to
provide a calibration curve for fluorometers to a known concentration of oil in a homogeneous
tank akin to Conmy et al., 2014a. Calibration regression results for all submersible
fluorometers can be found in Figures 29 and 30 for TPH and BTEX, respectively and regression
equations are tabulated in Table 5. Strong correlations between oil concentration and
fluorescence intensity were observed, suggesting that fluorescence signal may serve as a proxy
for TPH or BTEX at specific time points within the tank. This is an advantage as fluorescence
intensity and oil droplet concentrations time series can therefore be calibrated and employed
to provide for chemistry estimates that can be correlated with particle / oil droplet
concentrations at fine time scales within the tank during experiments. For example, comparing
the TPC and fluorescence signature for ANS with and without dispersant illustrates the
differences in the oil droplet concentration and dissolved oil during injection experiments and
the utility of monitoring both to understanding plume dynamics (Figure 31).
IA-E12PG00037 Final Report
Page 53
-------
EPA/600/R-16/152
September 2016
GmbH Trios
60
r«
i)
-30 -
10
TPH(ppm)
Turner Cyclops-Crude Optics
9000 i
8000
Chelsea AQUAtracka- Refined
Optics
10
TPH (ppm)
WetLabs ECO
Turner Cyclops-Refined Optics
1400
1200
7000
_ 6000 -
£ 5000 -
4000 -
.E» 3000 -
w
2000 -
1000
0 «
-1000 Q
10
TPH (ppm)
R2 = 0.99871
>
J. 800
I 600
55 400
0
0 2 4 6 8 10 12 14
TPH (ppm)
Chelsea AQUAtracka- Crude
Optics
^ 1.2
? 1
I 0.8
4>
0.6
I 0.4
0.2
0
10
TPH (ppm)
R2 = 0.96941
Figure 29. Calibration lines for fluorometer response vsTPH concentrations.
TPH (ppm)
IA-E12PG00037 Final Report
Page 54
-------
EPA/600/R-16/152
September 2016
Chelsea AQUAtracka- Crude
Optics
R2 = 0.986348
P 0.4
100 150
Btex(ug/L)
Chelsea AQUAtracka- Refined
Optics
R2 = 0.985245
100 150
Btex(ug/L)
WetLabs ECO
GmbH Trios
R2 = 0.99437
Btex(ug/L)
Btex(ug/L)
Turner Cyclops- Crude Optics
Turner Cyclops- Refined Optics
100
Btex(ug/L)
Btex(ug/L)
Figure 30. Calibration lines for fluorometer response vs BTEX concentrations.
IA-E12PG00037 Final Report
Page 55
-------
EPA/600/R-16/152
September 2016
~100
DOR = 0
1* 0.0
0 2
|
4 6 8 10
Time (minutes, normalized)
1
lHv_
Chelsea AQUAtracka (Refined Oil)
1
lv
WetLabs ECO
!
Vi
v
J100 -
80
Si 60-
O 40
20
o-
uj 150
OT 100
DOR = 1:20
0 2
4 6 8 10
Time (minutes, normalized)
Chelsea AQUAtracka (Crude Oil)
'vt\h
Chelsea AQUAtracka (Refined Oil)
i VUu
WetLabs ECO
VA"W\(
Time (minutes)
Time (minutes)
Figure 31. Total Particle Concentration and fluorescence time series for ANS crude oil with
Corexit 9500 dispersant.
IA-E12PG00037 Final Report
Page 56
-------
EPA/600/R-16/152
September 2016
Table 5. Calibration equations for the submersible fluorometers. Data in this report have
fluorescence signal in the manufacturer recommended units.
Instrument
Factory
Calibration
Standard
(units)
TPH Calibration
Equation
BTEX Calibration
Equation
Chelsea
Aquatracka
(Crude Optics)
Perylene
(ug/L)
[TPH] = ([Perylene]-
0.3834)/0.06051
[BTEX] = ([Perylene]-
0.3165)/0.004922
Chelsea
Aquatracka
(Refined
Optics)
Carbazole
(ug/L)
[TPH] = ([Carbazole]-
0.1804)/0.09575
[BTEX] = ([Carbazole]-
0.08487)/0.007584
GmbH Trios
PAH (ug/L)
[TPH] = ([PAH]-
12.288)/2.2733
[BTEX] = ([PAH]-
9.559)/0.1871
Turner Cyclops
(Crude Optics)
Signal (mV)
[TPH] =
(Signal+320.26)/503.94
[BTEX] =
(Signal+1152.2)/42.429
Turner Cyclops
(Refined
Optics)
Signal (mV)
[TPH] = (Signal-
299.29)/73.339
[BTEX] = (Signal-
212.05)/6.0593
Wetlabs ECO
QSDE (uM/L)
[TPH] = (QSDE-
0.2102)/0.4362
[BTEX] =
(QSDE+0.5403)/0.03697
IA-E12PG00037 Final Report
Page 57
-------
EPA/600/R-16/152
September 2016
A.3.6 VOC Air Monitoring
For all experiments, the Volatile Organic Compounds (VOC) measurements exhibited higher
variability compared to the in water sensor measurements. The installation of a wind curtain
along the western side of the tank helped to reduce the prevailing winds coming directly off
the water, however the effects of wind were not completely eliminated. The observed
variability is likely caused by differences in wind speed and direction both among the triplicate
experiments (typically run on the same day), and among the different treatments (which were
run over days/weeks). The VOC meters were installed with the air intakes pointing down and
were 0.4 m above the water surface at the top edge of the tank. This positioning helped to
reduce the effects of wind, given that the tank walls acted as an additional wind blocker.
Two VOC meters were deployed above the tank, but only results from the VOC meter closest
to the oil release are presented here (Jet Release VOC meter; the VOC meter directly above
the fluorometer rack). Results from the second VOC meter (Downstream VOC meter) installed
11.8 m farther downstream are more variable, both in concentrations between triplicate runs
and in the time it takes for airborne VOC concentrations to reach the meter. In general,
readings from the second meter showed a broader plume with a lower peak VOC
concentration. Due to an instrument malfunction, approximately 17 experiments are missing
data from the Downstream VOC meter. All results from the Jet Release VOC meter are
presented in Figures 32-39. Note that the Y-axis scale differs depending on the oil type (20 ppm
for IFO, 45 ppm for ANS & SLC, 250 ppm for Gas Condensate).
Of the four different hydrocarbon products tested, experiments using the gas condensate
exhibited the highest surface VOC concentrations, followed by ANS and SLC which exhibited
similar values. The lowest concentrations were observed for IFO 120 experiments. Higher
concentrations of VOC in the air were usually accompanied by lower BTEX concentrations in
the water for each oil type (analytical chemistry results in Appendix B). Chemistry results from
the water column (effluent, listed in Appendix B tables) samples help to verify the findings
from the VOC meters. In general, the measured concentrations of BTEX in the effluent water
IA-E12PG00037 Final Report
Page 58
-------
EPA/600/R-16/152
September 2016
samples were higher for experiments using dispersant compared to the untreated
experiments. The effluent port in the flume tank during normal operation produces a depth
integrated water sample which does not draw off the water surface. Therefore, oil that rises
to the surface is not drawn into the effluent, and so the tank effluent can be used as a measure
of how much oil was dispersed into the water column. Regardless of the oil product tested,
the use of chemical dispersants resulted in a reduction in VOC concentrations in the air above
the water compared to corresponding experiments without dispersant. These results
comparing the mean maximum VOC concentrations (30 second before/after peak readings)
measured during each experiment are summarized in Table 6. A general trend was also
observed where increasing the DOR resulted in lower surface VOC concentrations near the jet
release location. Statistical analysis using ANOVA followed by confidence interval test (Tukey's
test) to compare the means found that there were significant differences between VOC
readings for ANS at a DOR of 1:20 versus no dispersant (both Corexit and Finasol), as well as
significant differences for SLC at a DOR of 1:20 versus no dispersant and DOR 1:100 and 1:200.
Caution should be used when extrapolating these results to other spill scenarios, given that
this was a shallow water tank so the effects of dissolution of VOCs from oil droplets in a
deepwater blowout would not be accounted for in these experiments. Due to wind effects
mentioned previously, trends in VOC concentrations above the plume further down the tank
could not be established. Further, wind conditions may have contributed to the observed
variability in the measurements. The effects of wind on the dilution and transport of VOCs
should also be considered during a real world spill scenario, and so the absolute values of VOC
concentrations measured in this study should only be used to compare the relative differences
between treatments, and should not be used as a guide for worker exposure. Caution must be
exercised however in that these results merely represent VOCs that make it to the air-sea
interface from a very shallow wave tank. They cannot simulate the dissolution of VOCs into
water that would be expected in a deep water column.
IA-E12PG00037 Final Report
Page 59
-------
EPA/600/R-16/152
September 2016
Table 6. Summary of maximum VOC concentrations at the various treatment conditions
tested in this study. Results are for only for warm water experiments.
Avg. Peak VOC Concentration (ppm), n = 3
ANOVA
Oil Type
No Dispersant
DOR 1:200
DOR 1:100
DOR 1:20
p-value, a =
0.05
ANS (Corexit 9500)
23.07
13.27
12.43
0.13
0.023
ANS (Finasol OSR52)
23.07
16.56
7.17
2.9
0.024
IFO 120
1.00
0.90
7.37
0.17
0.133
Condensate
121.23
-
-
19.73
0.152
SLC
28.53
27.5
16.75
1.53
0.001
IA-E12PG00037 Final Report
Page 60
-------
EPA/600/R-16/152
September 2016
No Dispersant
DOR 1:200
45
—SubANS-1
SubANS-5
SubANS-9
SubANS-2R
—SubANS-6R
SubANS-1 OR
40
40
To 25
§20
8 20
o
c
o
o 15
o
§10
o 15
> 10
100 200 300 400
Time (seconds)
5
0
0
100
200
300
Time (seconds)
400
500
600
0
500
600
DOR 1:100
DOR 1:20
—SubANS-4R
SubANS-8R
SubANS-12R
—SubANS-3
— SubANS-7
SubANS-11
-30
-30
« 25
8 20
8 20
> 10
> 10
0
100
200
300
Time (seconds)
400
500
600
100
200
300
Time (seconds)
400
500
600
Figure 32. VOC results for subsurface injection experiments (cold water season) using Alaska
North Slope crude oil and four treatment conditions (no dispersant, DOR 1:200, DOR 1:100,
DOR 1:20). Replicate treatments represented by light blue, dark blue and green colored
lines.
IA-E12PG00037 Final Report
Page 61
-------
EPA/600/R-16/152
September 2016
No Dispersant
DOR 1:200
—SubANS-14
—SubANS-18
SubANS-22
— SubANS-13
SubANS-17
SubANS-21
40
« 25
To 25
8 20
o 15
o 15
> 10
> 10
0
100
200
300
Time (seconds)
400
500
600
0
100
200
300
Time (seconds)
400
500
600
DOR 1:100
DOR 1:20
SubANS-16
SubANS-20
SubANS-24
—SubANS-15R
— SubANS-19
SubANS-23
40
« 25
3 25
8 20
o 15
o 15
> 10
> 10
100
200
300
Time (seconds)
400
500
600
100
200
300
Time (seconds)
400
500
600
Figure 33. VOC results for subsurface injection experiments (warm water season) using
Alaska North Slope crude oil and four treatment conditions (no dispersant, DOR 1:200, DOR
1:100, DOR 1:20). Corexit 9500 was used as the treating agent. Replicate treatments
represented by light blue, dark blue and green colored lines.
IA-E12PG00037 Final Report
Page 62
-------
EPA/600/R-16/152
September 2016
No Dispersant
DOR 1:200
—SubFIN-07
—SubFIN-08
SubFIN-09
—SubANS-13
— SubANS-17
SubANS-21
40
40
% 25
"S 25
§20
C
o
o 15
o
§10
8 20
o 15
> 10
5
0
0
100
200
300
Time (seconds)
400
500
600
0
100
200
300
Time (seconds)
400
500
600
DOR 1:100
DOR 1:20
—SubFIN-04
—SubFIN-05
SubFIN-06
—SubFIN-02
SubFIN-03
— 30
•S30
3 25
® 20
g 20
> 10
> 10
5
100 200
0
0
100
200
300
Time (seconds)
400
500
600
300
Time (seconds)
400
500
600
Figure 34. VOC results for subsurface injection experiments (warm water season) using
Alaska North Slope crude oil and four treatment conditions (no dispersant, DOR 1:200, DOR
1:100, DOR 1:20). Finasol OSR 52 was used as the treating agent. Replicate treatments
represented by light blue, dark blue and green colored lines.
IA-E12PG00037 Final Report
Page 63
-------
EPA/600/R-16/152
September 2016
20
18
No Dispersant
20
18
DOR 1:200
SublFO-5
—SublFO-2R
—SublFO-6R
SublFO-10R
_16
E
•I14
_16
E
s14
c
.212
c
512
c 10
©
cS
c 10
9
1 8
1 8
o 6
§<
S 6
2
0
0.00
2
100.00 200.00 300.00 400.00
500.00 600.00
0
0
100
200 300
400
500 600
Time (seconds)
Time (seconds)
20
DOR 1:100
20
DOR 1:20
—SublFO-3
—SublFO-4R
18
SublFO-11R
18
SublFO-8R
_16
E
I14
_16
£
c
.212
1 12
TO
c 10
a>
|10
®
I 8
i 8
O
o 6
O
> 4
o
U 6
O
> 4
2
2
0
0.00
100.00 200.00 300.00 400.00
500.00 600.00
0 "
0
100
200 300
400
500 600
Time (seconds)
Time (seconds)
Figure 35. VOC results for subsurface injection experiments (cold water season) using IFO
120 and four treatment conditions (no dispersant, DOR 1:200, DOR 1:100, DOR 1:20). Corexit
9500 was used as the treating agent. Replicate treatments represented by light blue, dark
blue and green colored lines.
IA-E12PG00037 Final Report
Page 64
-------
EPA/600/R-16/152
September 2016
20
18
_16
E
a14
! 12
<3
I 10
®
1 8
S 6
2
0
No Dispersant
—SublFO-13
SublFO-17
SublFO-21
100
200 300 400
Time (seconds)
500
600
DOR 1:200
20
18
_16
£
t14
c
.2 12
<3
$10
S 6
2
0
—SublFO-14
—SublFO-18
SublFO-22
100
200 300 400
Time (seconds)
500
600
20
18
DOR 1:100
DOR 1:20
-SublFO-15
SublFO-19
SublFO-23
20
18
— 16
£
I14
J 12
re
Sio
o
o 6
§4
2
0
200 300 400
Time (seconds)
500
600
100
200 300 400
Time (seconds)
—SublFO-16
SublFO-20
SublFO-24
500
600
Figure 36. VOC results for subsurface injection experiments (warm water season) using IFO
120 and four treatment conditions (no dispersant, DOR 1:200, DOR 1:100, DOR 1:20). Corexit
9500 was used as the treating agent. Replicate treatments represented by light blue, dark
blue and green colored lines.
IA-E12PG00037 Final Report
Page 65
-------
EPA/600/R-16/152
September 2016
IFO-120 & Finasol @ DOR 1:200,1:100,1:20
20 | —SUbFIN-10|
18 —SubFIN-11
SubFIN-12
_16
E
I14
| 12
To
c 10
1 8
g 6
;l IAaaJLa
0 100 200 300 400 500 600
Time (seconds)
Figure 37. VOC results for subsurface injection experiments (warm water season) using IFO
120 and three treatment conditions (DOR 1:200, DOR 1:100, DOR 1:20). Finasol OSR 52 was
used as the treating agent (note - these treatments were not tested in triplicate).
No Dispersant
DOR 1:20
250
250
.2 150
.2 150
£ 100
£ 100
A
100
200
100
300
Time (seconds)
400
500
600
200
300
Time (seconds)
400
500
600
Figure 38. VOC results for subsurface injection experiments using gas condensate and two
treatment conditions (no dispersant, DOR 1:20). Corexit 9500 was used as the treating agent.
IA-E12PG00037 Final Report
Page 66
-------
EPA/600/R-16/152
September 2016
No Dispersant
—SubSLC-01
SubSLC-05
SubSLC-09
100 200 300 400 500 600
Time (seconds)
DOR 1:200
—SubSLC-02
—SubSLC-06
SubSLC-10
100
200 300 400
Time (seconds)
500
600
45
40
DOR 1:100
DOR 1:20
—SubSLC-03
SubSLC-07
45
40
j 35
a
S30
c
o
2 25
§20
o 15
o
§10
100 200 300 400 500 600
Time (seconds)
—SubSLC-04R
SubSLC-08R
SubSLC-12R
1
100
200 300 400
Time (seconds)
500
600
Figure 39. VOC results for subsurface injection experiments using Sweet Louisiana Crude oil
and four treatment conditions (no dispersant, DOR 1:200, DOR 1:100, DOR 1:20). Corexit
9500 was used as the treating agent.
IA-E12PG00037 Final Report
Page 67
-------
EPA/600/R-16/152
September 2016
A.3.7 VDROP-J and JETLAG Numerical Plume Modeling
Refer to Appendix G for detailed summary of the VROP-J and JETLAG numerical modeling
component along with figures. Modeling the movement of oil released underwater is a
challenging task due to limitations in measuring hydrodynamics in an oil-water system.
Computational Fluid Dynamics (CFD) models are capable of reproducing the hydrodynamics
provided they have sufficient resolution. However, current CFD models cannot predict the
droplet size distribution. For this reason, we used a suite of programs to understand jet
hydrodynamics, the droplet size distribution, and the movement of oil droplets within the
jet/plume. The developed models were calibrated to experimental data of oil jet released
underwater in the BIO tank. Based on the properties of the jet (mass flow rate 3.8 L/min
through a 2.4 mm orifice), the regime of the jet is atomization, which indicates that the jet
would break into small droplets. The models VDROP-J and JETLAG were used to predict the
streamwise centerline velocity and the holdup (volume of oil divided by the total volume of
fluids in a control volume) along the centerline of the plume, where both models were in
agreement. This implies that VDROP-J is adequate to predict the average droplet size
distribution in the plume. In the absence of dispersant, the model VDROP-J predicted oil DSD
that is very close to that measured by the LISST instrument. However, In the presence of
dispersant premixed with the oil, the VDROP-J model captured the overall trend of the DSD,
but could not capture the peak in droplet concentration observed at 5 |am. The observed peak
is most likely due to tip-streaming (when at high DORs, oil droplets shed filaments from their
edges resulting in smaller droplets), and VDROP-J does not have such a module at this time but
is considered for future development.
The computational fluid dynamics (CFD) program Fluent (www.ansys.com) was used to model
the hydrodynamics of the horizontal jet experiments. The standard k-s model was used to
model turbulence, and the Volume of Fluid (VOF) was used to model the two phases (oil and
water). The profiles of the holdup (ratio of oil volume to total volume), velocity magnitude,
eddy diffusivity and turbulent dissipation rate were presented. Findings indicate that the
holdup drops sharply with distance from the source to a few percent within 0.50 m from the
source, suggesting the occurrence of water entrainment into the plume. A significant reduction
IA-E12PG00037 Final Report
Page 68
-------
EPA/600/R-16/152
September 2016
in the energy dissipation rate was also observed, by orders of magnitude along the centerline,
starting from 104 watt/kg to 10"4 watt/kg. Both holdup and energy dissipation values have
important consequences on oil droplet breakup and coalescence. The plume exhibited a core
of high velocity and high mixing, while the edge of the plume had more or less violent
conditions, which is probably due to the entrained water squishing the edges of the plume.
The velocity and eddy diffusivity are needed to predict the movement of individual oil droplets.
The shape of the plume was circular near the orifice, but became oblate horizontally at a
centerline distance of 2.0 m, which is due to both the buoyancy of the whole plume and its
inertia. This suggests that the narrow width of the tank (0.60 m) did not affect the jet
hydrodynamics (otherwise the jet would be elongated in the vertical). The width of the tank
had an effect on the jet dynamics only near the surface as the plume became elongated along
the tank near the surface.
The CFD approach has its limitations as it smooths out the edge of the oil jet/plume, and thus
does not allow for the formation of large eddies around the plume. Here, large eddy
simulations (LES) were used to capture the large eddies where the movement of individual oil
droplets employed a lagrangian approach. Water velocity and the eddy diffusivity were used
to transport oil droplets, and the effect of individual oil droplet buoyancy and inertia were
accounted for. Accounting for the inertia of oil droplets has not been done previously in the
oil spill literature. Neglecting the inertia of the droplets results in overestimates of their rise
rate as the inertia from a horizontal jet tends to propel the droplet more horizontally, and thus
their rise gets delayed also by turbulent mixing. Results suggest that oil droplets with a
diameter less than 100 |am would mix uniformly in the plume, while those close to 500 |am
would tend to be above the centerline of the plume. This indicates that, when measuring the
droplet size distribution using the LISST, the placement of the LISST would not affect the
reading of droplets that are less than 100 microns. But the LISST needs to be placed judicially
to capture particles that are 300 to 500 |am, otherwise LISST placement below the centerline
would underestimate the actual droplets in that range. In contrast, LISST placement above the
centerline does not allow for determining that the concentration values represent the whole
cross section of the plume.
IA-E12PG00037 Final Report
Page 69
-------
EPA/600/R-16/152
September 2016
A.3.8 Weber Number Scaling Numerical Plume Modeling
Refer to Appendix H for detailed summary of the Weber Number Scaling numerical modeling
component. During the Deepwater Horizon oil spill, modeling activities for predicting oil
droplet size distribution formed in subsea oil blowouts was critical given their direct
influence on the fate and transport of oil in the marine environment. The scientific
community's knowledge on droplet size distributions and our capability to predict the
distributions are still limited. A recent and promising approach for predicting DSD is the
Modified Weber Number approach developed by SINTEF. Thus far, this method has been
based on experimental results, validated by a light crude oil (Oseberg Blend crude oil). Here,
this approach is validated over a range of oil types (IFO 120 and ANS) using a series of
experiments conducted with a subsurface release of oil within the DFO horizontal flow tank.
Based on the measured droplet sizes obtained from the tank experiments, corresponding
median droplet diameters (dso) and the relative droplet size (dso/D) were calculated, where D
is the nozzle diameter. Accordingly, the relations between dso/D and modified Weber number,
Reynolds number, and oil concentration were quantified. With regression analyses, the
empirical coefficients for the prediction of droplets size distribution based on the modified
Weber number were determined for a certain type of oil (e.g., IFO 120 and ANS). The results
indicated that chemical dispersants play an important role in reducing the droplet size of ANS
in both cold and warm temperatures. The effectiveness of dispersant in reducing droplet size
is higher for ANS compared to IFO 120. There may be thresholds for the dose of chemical
dispersant to some oils (e.g., IFO 120) but further data analyses are needed to confirm this.
There may also be over dose of dispersant to some oils (e.g., ANS) when the DOR is high,
eventually affecting the droplet size distribution. Furthermore, the data indicate that the
distributions of the data with d/dso<= 1 and d/dso> 1 are significantly varied. Therefore, a two-
step Rosin-Rammler approach was introduced to more accurately predict the droplet size
distribution. The regression coefficients for the two-step Rosin-Rammler are higher compared
to the single step in most cases (Appendix H), indicating the advantage of the proposed two-
step Rosin-Rammler approach. It should also be noted that the measured interfacial tension
(IFT) for the IFO 120 and ANS with different DORs appear to be significantly different compared
IA-E12PG00037 Final Report
Page 70
-------
EPA/600/R-16/152
September 2016
to the measured results from SINTEF for the modified Weber number approach, possibly due
to the characteristics of different oils.
IA-E12PG00037 Final Report
Page 71
-------
Task B.l Introduction & Relevance
BSEE's Remote Sensing & Surveillance of Oil Spills broad agency announcement that funded this
work states that "In remote sensing, a sensor other than human vision or conventional
photography is used to detect or map oil spills." Thus, although certain remote sensing of oil
spills is traditionally linked to detection of oil on the sea surface from above, the scope of the
technology can be extended to include the detection of oil in the deep-sea and/or under-the-ice
conditions using various sensors, as responders cannot use vision within the water column. As
demonstrated during the 2010 Gulf of Mexico Deepwater Horizon (DWH) oil spill, oil detection
by fluorescence can enable responders to discern trajectory of plumes and assess effectiveness
of dispersant countermeasures (ACT, 2008; Joint Analysis Group Report, 2010). The information
gained from such technologies was used to track oil in the water column and inform response
strategies to protect natural resources potentially at risk; thus supporting both Net
Environmental Benefit Analyses (NEBA) and Natural Resource Damage Assessments (NRDA). To
advance the application of this methodology, this project evaluated fluorescence characteristics
of various oils with and without dispersants to aid in the selection and refinement of in situ
sensors for use in oil spill response operations.
The overall objective of this work was to translate oil fluorescence R&D into operational tools for
oil spill response. Tabulating information on the optimum fluorescence wavelengths for oil
detection as a function of oil type and DOR assists responders selecting sensors and establishing
Best Practices for rapid decision making during spill response. The results of this project are
timely and can be used in conjunction with the National Response Team (NRT) guidance
document, Environmental Monitoring for Atypical Dispersant Operations: Including Guidance for
Subsea Application and Prolonged Surface Application, for incident-specific decisions concerning
monitoring subsea dispersant use (www.nrt.org). It specifically calls upon using oil-specific
submersible fluorometers with laboratory and on-board ship analyses using fixed wavelength
and scanning spectrofluorometers to enable improvements to monitoring sampling during
dispersant application. Findings from this project provide additional scientific information in
support of implementing guidance recommendations.
-------
EPA/600/R-16/152
September 2016
Fluorescence characteristics - All fluorophores (molecules that fluoresce) have characteristic
wavelengths for maximum absorption of light and characteristic wavelengths at which they emit
light as fluorescence. Absorption and fluorescence can occur at either narrow or wide wavelength
ranges depending on the chemistry and complexity of the fluorophores. A variety of naturally
occurring fluorescent compounds occur in the ocean, from ones with narrow wavelength ranges
with sharp fluorescence peak maxima (pigments, proteins) to complex compounds with wide
diffuse peaks over long wavelength ranges, such as the ubiquitous Colored Dissolved Organic
Matter (CDOM) or petroleum oils.
Fluorescence characteristics of complex mixtures can overlap if structurally similar compounds
are shared. Such is the case with CDOM and the aromatic fraction of crude oils. Both are
comprised of a variety of organic molecules and both exhibit complex, three-dimensional EEM
spectra. In general, crude oils have a broad excitation peak centered in the ultraviolet spectrum
(< 300 nm) and two emission peaks, one centered in the ultraviolet spectrum around 340 nm and
a much larger and broader peak in the visible around 445 nm (Bugden et al., 2008). These peaks
result from the single ring benzene derivatives and the "polynuclear aromatic" fraction that are
particularly susceptible to UV excitation wavelengths. EEMs exhibit distinct fingerprints for
different oils as illustrated by previous studies (Bugden et al., 2008; Kepkay et al., 2008).
DWH in situ oil fluorescence - Deployment of submersible fluorometers during the DWH oil spill
response illustrated the utility of this forensic tool that enabled large-scale monitoring of oil
concentrations to a depth of approximately 1600 m. Co-deployment of the fluorometers
alongside other response sensors [Conductivity-Temperature-Depth (CTD), Dissolved Oxygen
(DO), Laser In-Situ Scattering and Transmissometry (LISST)] from multiple platforms (e.g.
profilers) with real-time capabilities improved our understanding of the processes influencing the
fate and behavior of the oil in the presence and absence of chemical dispersants. Added to this,
extensive water column sampling also involved discrete sample collection for oil particle
concentration and size, Total Petroleum Hydrocarbons (TPH), Volatile Organic Carbon (VOC) and
other physical, chemical, biological factors. To date, the in-depth reviews by the Joint Analysis
Group (JAG) charged with data analysis have found that of all the variables measured, the most
IA-E12PG00037 Final Report
Page 73
-------
EPA/600/R-16/152
September 2016
highly correlated in the subsea plume are in situ DO and oil fluorescence intensity (Joint Analysis
Group Report, 2010). Such a correlation is not unexpected as laboratory tests show that
enhanced oxygen utilization can result from microbial respiration in the presence of oil
compounds (Venosa et al., 2002b). Beyond the underlying biochemical mechanisms however,
likelihood of correlation is increased based on the fact that variables measured in situ at high
sampling rates are better to capture plume heterogeneity. Hence, the utility of in situ
fluorescence as a tool was ascertained early in the response due to such correlations, the high
temporal and spatial resolution provided by the sensors, and also the advantages afforded by
real-time capability compared to discrete analyses.
However, the multitude of submersible fluorometers used in the DWH response called to
attention differences in the sensitivity and analytical capability of the instruments used due to
differences in configuration of excitation and emission wavelengths, methods of calibration,
sensitivity, and correlation to oil concentration (Figure 40, Table7). Many are not customized to
capture oil fluorescence peak maxima, rather only a fraction of the signal (Fuller et al., 2003;
Conmy et al., 2004 Conmy et al., 2014b). Furthermore, the ability of any fluorescence sensor
(laboratory or field submersible) to detect oil is a function of (1) how well the sensor matches the
excitation and emission wavelengths of the oil (including bandwidth of the wavelength filters or
bandpasses from gratings, (2) the power of the light source, and (3) the sensitivity of the detector.
When tracking in the subsea became necessary early in the response, fluorometers used for
detection of CDOM (i.e., WET Labs ECO series) were deployed on the vertical profilers as they
were widely available, were capable of full ocean depth deployment and had been previously
shown to detect oil in water (Wet Labs, Inc. website, www.wetlabs.com). These sensors typically
have light sources that excite at wavelengths slightly longer than peak absorption by
hydrocarbons and detect emission in the visible. They employ filters centered on excitation (Ex)
and emission (Em) wavelengths at 370 and 460 nm (ExEm37o/460nm). Although the center
wavelength of the filters does not capture the peak of the oil fluorescence signal, the wide
bandwidth of the emission filters (120 nm Full Width at Half Max) and the broad nature of the
fluorescence peaks means that CDOM sensors are capable of detecting a large portion of the
IA-E12PG00037 Final Report
Page 74
-------
EPA/600/R-16/152
September 2016
visible fluorescence signal. CDOM fluorometers were used to detect oil during the response in
part because of their accessibility, but also because these sensors capture some portion of the
oil fluorescence peak that occurs at the longer UV wavelengths where CDOM peaks also exist.
To quell questions regarding the ability of the ECO CDOM fluorometer to detect oil in the subsea
plume, calibration tests were conducted at Louisiana State University (LSU) using Mississippi
Canyon 252 (MC252) source oil. They provided a means to convert raw fluorescence data to
Quinine Sulfate Dihydrate Equivalents (QSDE, the standard typically used for CDOM) to ppm of
oil (JAG report, 2010). The calibrations were conducted in flasks on orbital shakers at 90
revolutions per minute (rpm), where oil concentrations ranged between 1-50 ppm. Dispersant
(Corexit 9500) was added at a DOR of 1:2.5 and 1:25. The response of the fluorometer was linear
with respect to oil but varied as a function of DOR, with a quenching of fluorescence in the
presence of more dispersant per unit oil. Results of this test indicated that the ECO sensor was
a sufficient proxy for oil concentrations greater than 1 ppm (NOAA, 2010). However, as the
response continued and after the well was capped, oil concentrations in the subsea plume
decreased as well as the magnitude of the fluorescence anomaly due to dilution and degradation
of the oil, particularly at further distances from the wellhead. Concern was raised that a
fluorometer with higher sensitivity for oil (one with a hydrocarbon-specific configuration) was
needed. At that time, Chelsea UV Aquatrackas (ExEm239/360nm) were deployed to track the plume
in the far field of the response geographic region with the expectation (and subsequent
confirmation) that it would detect fluorescence signal at lower oil concentrations.
IA-E12PG00037 Final Report
Page 75
-------
EPA/600/R-16/152
September 2016
ppb QSE
400 450
Emu von Inm)
1 F275/325 nm = 553 ppb
2 265/502 nm = 532 ppb
Q FIR = F280/340 : 280/445 r,m
~ Chelsea UV Aquatracka
F23S. 360 = 74 ppb; 14% of Peak 1
F23a-440= 86 ppb; 16% ofPeak2
A Turner Cyclops
F254/350 = 281 ppb; 51 % of Peak 1
(Fine oil)
Fs2
-------
EPA/600/R-16/152
September 2016
Table 7. Sensor specifications as listed from manufacturers. Wavelengths listed as Center
Wavelengths (CWL) with Full Width at Half Max (FWHM) and Bandpass (BP). Standards used
are QS (Quinine Sulfate Dihydrate), NDD Salt (Napthalene Disulfonic Disodium) and PTSA Salt
(Pyrenetetrasulfonic Acid Tetrasodium) (From Conmy et al., 2014b).
Manufacturer
Instrument
Light source
Excitation X (nm)
Emission "k (nm)
Detector
Dynamic Range
Chelsea
UV AQUAtracka
Xenon lamp
239 CWL
360 CWL
PMT
0.001- 10 (jg/L Carbazole
Technologies Group
UV AQUAtracka
Xenon lamp
239 CWL
440 CWL
PMT
0.001 - 10 (jg/L Perylene
Seapoint Sensors
SUVF
LED
370, 12 FWHM
440, 40 FWHM
Photodiode
0.1 - 1500 (jg/LQS
TriOS, GmbH
EnviroFLU-HC, I)S
Xenon lamp
254, 25 FWHM
360, 50 FWHM
Photodiode
0 - 5000 ppb Phenantiiren
Cyclops (Fine oil)
LED
254, 40 nmBP
350, 50 nmBP
Photodiode
0 - 10,000 ppb NDD Salt
Turner Designs
Cyclops (Crude oil)
LED
320, 130 nmBP
510, 180 nmBP
Photodiode
0 - 2700 ppb PTSA Salt
Cyclops (CDOM)
LED
320, 130 nmBP
470, 60 nmBP
Photodiode
0 - 2500 ppb QS
WetLabs
WetS tar
LED
370, 10 FWHM
460, 120 FWHM
Photodiode
0.100 - 1000 ppb QS
ECO-FLU, triplet, puck LED
370, 10 FWHM
460, 120 FWHM
Photodiode
0.01 - 500 ppb QS
Post-DWH response sensor tank testing -To address persisting uncertainties regarding sensor
performance in the subsea, a team of scientists conducted experiments in May 2011 to study the
dynamic range, sensitivity, and response of in situ fluorometers to changing excitation or
emission properties of fresh and weathered MC252 oil (NOAA Science Box Award, PI: Michelle
Wood; Co-PI's from EPA, NOAA, University of South Florida). The experiment was conducted
within the flow-through flume tank at the BIO in Dartmouth, Nova Scotia, taking into
consideration environmental factors such as wave energy and ocean currents. Experiments
included the stepwise addition of oil and dispersant (DOR of 1:25; 0.3 -12 ppm of MC252 SLC oil)
to the flume tank while collecting in situ fluorescence and droplet-size distribution data, as well
as coincident discrete samples for chemistry and EEM analyses. The flume tank was operated in
static mode and each addition of oil and dispersant was allowed to homogenize priorto collecting
discrete samples and coincident sensor measurements to calculate the least linear squares
regressions. Results indicated that all sensors tested were responsive to changes in MC252 oil
concentration regardless of wavelength configuration. Linear response of the WET Labs ECO,
IA-E12PG00037 Final Report
Page 77
-------
EPA/600/R-16/152
September 2016
Turner Designs Cyclops and the Chelsea Technologies Group AQUAtrackas sensors as a function
of oil concentration was observed, where lowest concentrations were not below the detection
limit of any sensor tested (Conmy et al., 2014a). Results demonstrated that all sensors exhibited
a wide dynamic range of detection for MC 252 oil and were capable of detecting oil at the lowest
concentration (approximately 300 ppb oil), which is significantly lower than the LSU calibration
study (1 ppm) and a common misconception during the response (Conmy et al., 2014a).
Differences in the detection limit between the studies may be explained by differences in the
design, scale and the amount of physical dispersion of the tests, where the tank can provide
mixing energies similar to those found in the field.
The 2011 study findings answered critical questions about sensor performance to detecting
MC252 oil. However, the experiment highlighted the need for future studies to evaluate sensor
performance using a variety of DORs and for multiple oil types. Evident from the DWH spill and
post-spill research was that further R&D is needed to transfer knowledge gained through
laboratory 3-D Excitation Emission Matrix (EEM) Spectroscopy into practical information for
fluorescence tools used during spill response. Fluorescent properties are oil specific and
investigating variations in EEMs as function of oil type and dispersant-to-oil ratios better prepares
the community in identifying sensors for response options. To that end, the objectives of this
project were to:
I. Generate a comprehensive EEMs database, building upon existing data at the
Department of Fisheries and Oceans Canada, to provide fluorescence peak
information as a function of oil type, weathering state, concentration and Dispersant-
to-Oil Ratios (DORs).
II. Critically examine the database using advanced statistical methods and models to
identify wavelengths best suited for oil monitoring during dispersant application and
degradation.
III. Conduct flume tank experiments to determine submersible sensors capable of
providing data comparable to scanning and/or fixed wavelength laboratory
fluorometers for rapid deployment during response efforts.
IA-E12PG00037 Final Report
Page 78
-------
EPA/600/R-16/152
September 2016
Through this project, a comprehensive EEMs library database was generated covering a wide
variety of oils from light to heavy fuel and crude oils and diluted bitumen. Varying DORs (1:20,
1:100, 1:200, 0) and oil concentrations were evaluated as the presence of dispersant alters EEM
fingerprints. EEMs were subjected to advanced statistical analyses and models to identify
wavelengths best suited for oil monitoring during dispersant application and subsequent
tracking. Fluorescence is a non-destructive characterization tool that is routinely used to examine
complex organic mixtures (foods, wine, medical compounds, aquatic organic matter, oils). Unlike
single compound solutions, they exhibit broad, diffuse peaks that result from overlapping smaller
peaks with similar chemistry. Although EEMs can be a substantial source of information on
chemical composition and variability amongst samples, the high-dimensionality (intensity by
emission by excitation) and nonlinearity of the data equates to difficulties in data interpretation
and extraction of practical information as a characterization tool (Bieroza et al., 2010). Therefore,
it is difficult to determine which underlying chemical components are responsible for which
portion of the fluorescence fingerprint. Combining standard techniques for EEM analysis such as
assessment of particular fluorescence peak features including peak height and wavelength
position via 'peak picking' with Parallel Factor Analysis (PARAFAC) modeling results in a more
comprehensive understanding of the chemical constituents. The use of advanced multivariate
analyses such as PARAFAC has gained popularity as an effective means to deconvolve complex,
broad peaks into their underlying smaller components (Stedmon et al., 2003; Boehme et al.,
2004; Christensen et al., 2005; Stedmon and Bro, 2008). Here, we processed EEMs data with
scripts in the N-way toolbox for Matlab (Andersen and Bro, 2000) and SOLO software
(Eigenvector, Inc) and used the algorithms to isolate wavelengths to best characterize an oil type.
An excellent review of these chemometric techniques and applications is provided in Bieroza et
al., 2010. This approach will allow for comparing oil in water mixtures for similarities and
contrasting features.
Results were evaluated for the Fluorescence Intensity Ratio (FIR) technique (Bugden et al. 2008;
Kepkay et al. 2008). The latter calculates the ratio at ExEm28o/340nm to ExEm28o/445nm as an indicator
IA-E12PG00037 Final Report
Page 79
-------
EPA/600/R-16/152
September 2016
of oil dispersion. Previous studies at DFO COOGER have shown that dispersed oil fluoresces over
two peaks centered on emission wavelengths of 340 nm and 445 nm, at excitation wavelength
280 nm, and that chemical dispersion enhances the emission intensity at 445 nm (Bugden et al.
2008; Kepkay et al. 2008). Postulated is that the fluorescence intensity at ExEm28o/340nm
represents the dispersion of lower molecular weight aromatic hydrocarbons, while intensity at
ExEm28o/445nm corresponds to higher molecular weight aromatic compounds.
Finally our work addresses the disconnect that exists between fluorescence research conducted
in laboratories and the collection of fluorescence data from submersible sensors. By conducting
laboratory-based and tank-based experiments on the same oil type and DOR, comparisons
between EEMs can be made across scales. This helps to determine how well the in situ sensors
are aligned in detecting dispersed oil.
IA-E12PG00037 Final Report
Page 80
-------
Task B.2 Experimental Methods
EPA/600/R-16/152
September 2016
B.2.1 Sample Preparation - Twenty-five oil samples from the DFO and EPA stockpiles (covering
a wide range of viscosity and oil type) were used for spectrofluorometric testing, where oil
characteristics were tabulated for the test oils based on an extensive literature search (Table 8;
Supplemental Material A). All glassware used in this study was cleaned to ensure highest
analytical integrity including solvent rinsing, deionized water rinsing and baking in a muffle
furnace at 450°C where appropriate. Samples were stored in 125mL amber glass bottles with
PTFE-lined caps (Figure 41).
B.2.2 Artificial Seawater Protocol - Artificial seawater was used for DOR mixing to avoid
interference of fluorophores found in natural seawater with oil fluorescence signal. Fresh
artificial seawater was made to salinity of 28 ppt and was prepared in 1 L quantity at the
beginning of each experiment by adding Tropic Marin® salts (Appendix A) to 1 L ultrapure water
dispensed from a Millipore Milli-Q unit (< 4 ppb DOM) into a 1.5 L glass beaker, covering the
beaker with aluminum foil, and stirring with a magnetic stir-bar on electric stir plate for 20
minutes at room temperature (~24°C).
B.2.3 Dispersed Oil in Seawater Protocol - A series of dispersed-oil-in-sea water experiments
were performed using baffled trypsinizing flasks (baffled flasks) with artificial seawater, MC252
oil and Corexit 9500 chemical dispersant (Venosa et al., 2002a). Four petroleum oil / dispersant
solutions were prepared for each oil type at the following DORs: 0, 1:20, 1:100, and 1:200. Oil
was pipetted into an 8.6 ml ambervial, followed by addition of the appropriate amount of Corexit
9500 chemical dispersant into the vial. Teflon-lined capped vials were shaken by hand for 60
seconds and 10 |aL of dispersant / oil mixture was pipetted (Eppendorf positive displacement
micropipettes, 1-20 |aL) into 100 mL artificial seawater contained in each of three replicate flasks.
Flasks were covered with parafilm and placed on a New Brunswick Scientific Innova 2100
platform shaker (orbit = 1.9 cm) for 12 minutes at 200 rpm. Approximately 3.5mL of the resulting
dispersed-oil-in-seawater was immediately dispensed through a spigot near the bottom of each
flask into three 4.0-mL UV-grade quartz cuvettes, which were immediately covered with Teflon
IA-E12PG00037 Final Report
Page 81
-------
EPA/600/R-16/152
September 2016
stoppers to prevent evasion of volatile components during fluorescence analyses (Figure 42).
After removal of spectrophotometric samples, additional volumes of sample were removed from
the baffled flasks for extraction of total petroleum hydrocarbons (TPH) by dimethylene chloride
(DCM). TPH analysis follows the same Gas Chromatography-Mass Spectrometry (GC-MS) method
as in Task A of this project.
B.2.4 Spectrophotometric Analysis,- A Horiba Scientific Aqua Log spectrofluorometer was used
to analyze the 25 oil types with varying DOR. A series of analyses were initially performed while
varying the instrument's settings (excitation and emission increments, gain setting, integration
time) in order to determine optimal settings for the entire experimental protocol. Excitation-
Emission Matrices (EEMs) were generated using the following instrument parameters: 200 - 800
nm excitation (3 nm increments), 249 - 828 nm emission range (CCD detector at 534 nm 8 pixel
increments), medium gain setting and integration time of 0.1 sec. A quinine sulfate dihydrate
dilution series was created consisting of: 0.5N H2SO4 solvent; 100 ppm 1° (primary stock)
solution; 100 ppb 2° (secondary stock) solution; 1,3,5,10 and 20ppb quinine sulfate solutions.
Dilutions were analyzed for fluorescence and used for cross-calibration with instrument software
built-in quinine sulfate tool to convert results into Quinine Sulfate Equivalents (QSE) and
demonstrate linearity of fluorescence in a dilution series. All data processing and spectral
corrections follow the manufacturer's manual. Dilution series with oil concentrations between 1
- 500 ppb were also generated to determine lower detection limits for oils. EEMs are presented
in Raman Units (RU).
IA-E12PG00037 Final Report
Page 82
-------
EPA/600/R-16/152
September 2016
Figure 41. Twenty-five oil samples stored in glass bottles.
IA-E12PG0Q037 Final Report
Page 83
-------
EPA/600/R-16/152
September 2016
Table 8. List of oil samples used for EEM analyses. Oils separated by API (American Petroleum
Institute) gravity.
Light (API >31.1°)
Medium (API 22.3-31.1°)
Heavy (API <22.3°)
Arabian Light (32.2°)
Alaska North Slope (29.7°)
Access Western Blend Dilbit
(21.3°)
Brent (38.2°)
Alaskan North Slope (10%
weathered)
Belridge Heavy (13.6°)
Federated (39.4°)
Heidrun (28.6°)
Cold Lake Dilbit (21.5°)
GuIIfaks (32.7°)
Lago (25.0°)
Hondo (19.5°)
Hibernia (35.6°)
Mesa (30.3°)
IFO 40 (21.9°)
MC252—Discoverer
Enterprise (37.2°)
Sea Rose (29.8°)
IFO 120 (18.4°)
MC252—generic (35.2°)
Vasconia (26.3°)
IFO 180 (14.1°)
Scotian Shelf Condensate
(53.2°)
IFO 300
Terra Nova (33.8°)
Santa Clara(22.1°)
IA-E12PG00037 Final Report
Page 84
-------
EPA/600/R-16/152
September 2016
Figure 42. Trypsinizing baffled flasks containing dispersed oil in artificial seawater (top) and
corresponding samples removed from each flask, ready for spectrofluorometric analysis.
IA-E12PG0Q037 Final Report
Page 85
-------
EPA/600/R-16/152
September 2016
Task B.3 Rev.ils 6. « scussion
B.3.1 Oil Fluorescence Properties
Four characteristic excitation/emission (Ex/Em) peak locations were identified • F maxl F max4
(Figure 43) for all 25 oil types at four DORs (Figure 43). The highest intensity peak (Fmaxi)
occurred, without exception, at Ex 221-239 nm/Em 335-344 nm and was paired with a blue-
shifted, lower intensity peak (Fmax2) at Ex 215-221 nm/Em 285-308 nm in all oil types. A third
broad, low-intensity peak (Fmax3) was observed at Ex 215-305 nm/Em 418-571 nm, to varying
degrees across oil types, corresponding with oil categories determined by API gravity (Table 8).
Light crude oils exhibited Fmax3 fluorescence at all DORs with the exception of Scotian Shelf
Condensate. Of note is that Scotian Shelf Condensate appeared physically unlike any of the other
light oils: clear in color with apparent very low viscosity. Since viscosity is largely determined by
the size and relative weight of component hydrocarbons (Fingas 2011), fewer complex
fluorophores would likely be present in this oil type. Fluorescence in the Fmax3 region was
identified at all DORs in only one medium weight oil (Heidrun), and was not present at any DOR
in one medium oil (Vasconia). Two medium-weight oils emitted measurable fluorescence in the
Fmax3 region only with full dispersion (Lago and Mesa), while Sea Rose showed fluorescence at
DORs 1:100 and 1:20. One medium weight oil, Alaska North Slope (both fresh and 10%
weathered), exhibited unusual Fmax3 behavior, with measureable fluorescence at DOR 0, 1:100
and 1:20, but not at DOR 1:200. Finally, for the heavy weight oils, Fmax3 was almost completely
absent at all DORs, with the exception of fluorescence at DOR 1:20 for Cold Lake Dilbit (Diluted
Bitumen) and IFO 40, and across all DORs for one anomalous member of this group—Access
Western Blend Dilbit. Dilbit is a mixture of bitumen—essentially a heavy crude oil with API
gravity < 10.0°—and a diluent—either a light condensate or naptha (Priaro 2016). The
combination of characteristics from both oil types may account for the unusual Fmax3 fluorescence
observed in this oil type. Additionally, Intermediate Fuel Oils (IFOs) are not true crude oils, but
marine fuels consisting of a mixture of post-refinery heavy residual oil and refined diesel fuel,
which may also help to explain the appearance of Fmax3 fluorescence in IFO 40. Clearly, the
presence of fluorescence in the Fmax3 region, especially at DOR 1:20, appears to be related to API
gravity, and thus to density as well as kinematic viscosity since API gravity = (141.5/Specific
IA-E12PG00037 Final Report
Page 86
-------
EPA/600/R-16/152
September 2016
Gravity) - 131.5 (Fingas, 2011). The absence of Fmax3 region fluorescence in heavy weight oils
may be due to retention of energy within the large, complex hydrocarbons which make up the
highest density oils. Additionally, the appearance of fluorescence in the Fmax3 region at highest
DORs for the medium weight oils (Lago, Mesa, and Sea Rose) suggests that smaller droplet sizes
were created via the dispersion which could lead to a decrease in reabsorption of fluorescence
within the oil - water mixture. A fourth region of broad, low-intensity fluorescence (Fmax4) was
identified at Ex 269-291 nm/Em 326-353 nm for all oil types at all DORs. Fmaxi and Fmax4 oil-in-
water fluorescence regions appear to be analogous to the characteristic colored dissolved
organic matter (CDOM) fluorescence regions 'Ac' at Ex 260/Em 400-460 and 'C' at Ex 320-365/Em
420-470 (Coble, 2014).
In addition to maximum intensity for each fluorescence peak (in RU), full width at half maximum
(FWHM) was also recorded. Further, fluorescence intensity at Ex/Em 281/340 and 281/456 nm
was recorded to enable calculation of the FIR for all samples. Optimum settings for signal
collection on the HORIBA AquaLog necessitated excitation at 3 nm intervals, which accounts for
the 1 nm discrepancy from the published FIR wavelengths (Bugden, et al. 2008). Fluorescence
intensity at the specified Ex/Em wavelength settings of five off-the-shelf in situ fluorometers
(Conmy et al., 2014a & b), which were all employed in the response to the DWH spill, was
recorded. Those wavelengths were also adjusted slightly to compensate for signal collection
intervals on the HORIBA AquaLog. Selected results are presented in Table 9 along with results of
chemical analyses, and complete fluorescence results are included as a Supplemental Table A.
EEM contour 'fingerprint' plots for all oils, which characterize each oil type and illustrate the
effect of dispersant on the fluorescence properties, are presented in Appendix F. The ability to
identify oil source can be useful in the prevention and abatement of oil spill pollution. To that
end, efforts to determine characteristic fluorescence fingerprints have existed since the 1970s
(Frank, 1978) and have received renewed attention with the advent of improved fluorescence
detection systems (Bugden, 2008).
Intensity of Fmaxi was consistently strong across oil types, with no ambiguity in peak location. The
observed Ex/Em range of significant fluorescence intensity was fairly narrow with FWHM of only
IA-E12PG00037 Final Report
Page 87
-------
EPA/600/R-16/152
September 2016
37-50 nm, and little to no change in peak location with increasing DOR. However, Six of the nine
light oil types, but just one of the seven medium oil types and one of the nine heavy oil types1
displayed this slight increase (approximately 4.5 nm) in FWHM with maximum dispersion (DOR
1:20). One medium weight oil (Lago) and one heavy oil (Access Western Blend Dilbit) showed
the same slight increase in FWHM at both DORs 1:100 and 1:20. The impact of applying the Inner
Filter Effect correction tool (IFE) to fluorescence intensity was also calculated for Fmaxi. This
correction utilizes the measured absorbance of the sample to correct for fluorescence emitted
by fluorophores within the sample, but re-absorbed within the sample itself. Of note is that
application of the IFE resulted in only a small magnification of the fluorescence signal at DORs 0,
1:200 and 1:100 for all oil types; however, there was a clear delineation between two categories
of oil types at DOR 1:20: Oil Type I, with IFE effect > 2.5 and Oil Type II, with IFE effect < 2.5 (Table
9). This appears to be due to the increase in optical density, and thus absorbance, possibly caused
by interaction between Corexit 9500 and well dispersed Type I oils. Photographs of four
representative pre-analysis samples, along with the resulting EEMs of oil type are shown in Figure
44 to illustrate the difference in fluorescence between the types regardless of being a light,
medium or heavy crude oil.
Due to variation from laboratory to laboratory, and even differences in instrument to instrument
performance from the same manufacturer, it is necessary to convert fluorescence intensity "raw
counts" to a standardized unit for useful reporting purposes. Traditionally, the fluorescence
community has utilized a dilution series of quinine sulfate dihydrate in weak acid to convert
instrument output to Quinine Sulfate Equivalents (QSE) (Coble, 1996). However, in recent years
the alternate method of reporting in Raman Units (RU) has gained favor (Murphy et al., 2010).
Due to inherent properties of water molecules, the Raman scatter peak is a reliable feature which
can be used through collecting a scan of ultra-pure water at the beginning of each day, and then
using the ratio of raw counts to the area under the curve of the Raman peak (approximately 381-
426 nm) to convert fluorescence to RU. As the Quinine Sulfate SRM is no longer available from
1 Increase of approximately 4.5 nm in FWHM in Fmaxi seen in light oils Arabian Light, Brent, Federated, Gullfaks,
Hibernia, and Terra Nova; in medium oil Mesa; and in heavy oil IFO 120 at DOR 1:20.
IA-E12PG00037 Final Report
Page 88
-------
EPA/600/R-16/152
September 2016
NIST (National Institute of Standards and Technology), we have reported results in RU and offer
a conversion factor to QSE using the highest quality quinine sulfate dihydrate readily available.
Overall, Fmaxi intensity ranged from a minimum of 39.58 RU (Access Western Blend Dilbit DOR 0)
to 3090.23 RU (IFO 120 DOR 1:100). Since all of the Intermediate Fuel Oils and the Scotian Shelf
Condensates showed unusual fluorescence profiles which tended to skew the results for the
aforementioned reasons (Figures 45 and 46), these will be eliminated from the remaining
discussion. Fmaxi intensity within Type I oils ranged from 357.62 RU (Arabian Light DOR 1:200) to
1998.60 RU (MC252 Discoverer Enterprise DOR 1:20), while the range in Type II Oils was the
overall low of 39.58 previously mentioned to a high of 1098.90 (Heidrun DOR 1:20).
While the excitation wavelength of maximum intensity for Fmax2 remained relatively consistent,
the emission wavelength varied within, as well as among, oil types. The occurrence of double
and triple peaks, as well as minor sub-peaks, within the Fmax2 region was fairly common. It was
sometimes difficult to distinguish the Fmax2 peak from the shoulder of a very strong Fmaxi peak,
especially at higher DORs. For this reason, determination of the true FWHM was sometimes
problematic. For Fmax2 intensity, Type I Oils ranged from 63.95 RU (Brent DOR 1:200) to 437.32
RU (MC252 Discoverer Enterprise DOR 1:20), and Type II Oils ranged from 25.07 RU (Belridge
Heavy DOR 0) to 164.07 RU (Heidrun DOR 1:20).
For oil types exhibiting an Fmax3 peak, it was most apparent at the highest DOR (1:20) and some
oils exhibited a strong Fmax3 peak across all DORs (e.g., Brent, Federated). However, for those oils
the Fmax3 peak at DOR 1:20 was significantly blue shifted (peak moved to lower wavelengths) from
the Fmax3 location observed at lower DORs. FWHM of the Fmax3 peak was much greater than that
of any other peak (145-283 nm), with the exception of the three lower DORs of Access Western
Blend Dilbit (52-56 nm). Identification of highest Fmax3 intensity proved somewhat problematic
as it tended to lay within the second order Rayleigh region, a band of high intensity light resulting
from scattering by water molecules. The edge of highest intensity could also lie in this region, so
determination of the true FWHM was also problematic for many oil types. Traditionally, second
order Rayleigh is eliminated by simply masking this region (10-12 nm). Although algorithms have
been developed to model the character of fluorescence peaks lying within (Zepp, et al. 2004;
IA-E12PG00037 Final Report
Page 89
-------
EPA/600/R-16/152
September 2016
Bahram, et al. 2006), assumptions about the linearity of fluorescence must be made, and the
true signal behavior cannot be known. For this reason, as our goal was to identify signals which
could also be detected by in situ instruments, the decision was made to identify the maximum
fluorescence intensity lying outside of the second order Rayleigh region rather than to try to
interpolate the data.
As previously mentioned, Fmax3 intensity was not always present, and it was observed far more
often in Type I Oils with a range of 2.64 RU (Arabian Light DOR 1:200) to 744.69 (MC252
Discoverer Enterprise DOR 1:20). Only four of the Type II Oils exhibited Fmax3 peaks and these
ranged from 2.45 RU (Access Western Blend Dilbit DOR 0) to 174.93 RU (Heidrun DOR 1:20).
As with Fmax2, the Fmax4 region sometimes contained double peaks. Unique spectral shapes for
this region were also observed, especially in higher-density oils such as Access Western Blend
Dilbit, Belridge Heavy, and Cold Lake Dilbit. FWHM ranged from 27 nm to 73 nm, for all oil types
but one. The exception was Access Western Blend Dilbit, with FWHM of 77-110. Intensity at
Fmax4ranged from 33.53 RU (Arabian Light DOR 1:200) to 231.86 RU (MC252 Discoverer Enterprise
DOR 1:20) in Type I Oils and from 4.93 RU (Access Western Blend Dilbit DOR 0) to 116.97 RU
(Heidrun DOR 1:20) in Type II Oils.
Results of the concentration dilution series showed that the HORIBA AquaLog was consistently
capable of detecting dispersed oil in artificial seawater in the three oil types tested (Alaska North
Slope, IFO 120, and MC252 Discoverer Enterprise) at all four DORs down to at least 50 ppb.
However, detecting dispersed oil below 100 ppb necessitated increasing the integration time to
10 sec. per scan in order to collect sufficient data, which resulted in a total analysis time of
approximately 30 minutes for each sample. Since the HORIBA AquaLog scans from high to low
wavelengths and much of the fluorescence signal from petroleum resides in the UV region,
photobleaching of the sample as well as temperature effects certainly may have impacted these
results.
IA-E12PG00037 Final Report
Page 90
-------
EPA/600/R-16/152
September 2016
600
E
£
00
£
_S»
«u
>
CB
500 -I
max3
400 -I
300 -I
max4
max2
250
300
350
400
EX Wavelength (nm)
Figure 43. Alaska North Slope dispersed oil in artificial seawater at DOR 1:20 with locations of
Fmaxi, Fmax2, Fmax3 and Fmax4 indicated. Note that maximum fluorescence intensity at Fmaxs is
mostly obscured by masking of second order Rayleigh scattering.
IA-E12PG0Q037 Final Report
Page 91
-------
EPA/600/R-16/152
September 2016
Table 9. EEM fluorescence and chemical characteristics. Refer to Supplemental Table A for full table.
2-ring
3-ring
4-ring
Type 1
F 1
r max
Alkanes
PAHs
PAHs
PAHs
Oils
DOR
(RU)
IFE
FIR
(Hg/L)
(Hg/L)
(Hg/L)
(Hg/L)
Alaska North
0
697.07
1.16
21.59
375
145
15
8
Slope
1:200
715.01
1.15
35.41
1:100
839.60
1.32
6.59
1:20
1171.63
3.33
0.88
3019
477
89
65
Alaska North
0
812.97
1.19
21.70
545
182
19
14
Slope (10%
1:200
831.70
1.21
21.70
weathered)
1:100
828.06
1.28
9.08
"'-.v- - - . -
1:20
1109.51
3.01
0.91
3312
499
99
74
Arabian Light
0
400.42
1.18
7.29
733
113
11
12
1:200
357.62
1.07
11.61
1:100
426.82
1.26
2.70
1:20
701.75
4.19
0.39
6004
571
71
103
Brent
0
646.18
1.15
7.59
1068
162
21
12
HHH
1:200
660.37
1.16
6.65
1:100
708.16
1.37
1.97
1:20
1098.42
3.05
0.68
5954
456
97
58
Federated
0
574.35
1.14
3.70
1921
197
41
30
1:200
607.97
1.21
2.01
1:100
645.28
1.37
0.94
1:20
1223.17
4.37
0.36
6501
488
129
87
Gullfaks
0
937.00
1.24
5.79
762
326
47
26
1:200
934.42
1.27
5.57
1:100
933.08
1.35
3.24
"A"
1:20
1524.21
3.73
0.71
1943
642
107
60
Hibernia
0
938.08
1.24
6.91
2289
296
39
24
1:200
951.49
1.34
3.22
1:100
978.62
1.47
1.69
1:20
1812.41
4.02
0.49
6095
541
94
59
MC252
0
998.50
1.27
5.01
1578
300
50
30
(Discoverer
1:200
1009.18
1.37
3.06
Enterprise)
1:100
1085.54
1.48
1.22
*
1:20
1998.60
4.76
0.39
3992
482
101
68
MC252
0
857.35
1.24
4.42
1468
231
36
21
(generic)
1:200
877.78
1.25
3.56
1:100
964.02
1.57
0.95
1:20
1795.13
4.73
0.40
5093
511
113
67
IA-E12PG00037 Final Report
Page 92
-------
EPA/600/R-16/152
September 2016
2-ring
3-ring
4-ring
Type 1
F 1
r max
Alkanes
PAHs
PAHs
PAHs
Oils
DOR
(RU)
IFE
FIR
(Hg/L)
(M-g/L)
(M-g/L)
(Mg/L)
MESA
0
757.84
1.19
18.14
1388
234
34
24
HH
1:200
806.76
1.22
11.97
1:100
745.17
1.26
6.44
¦¦¦
1:20
1107.09
2.77
1.04
4088
439
85
62
Sea Rose
0
1145.29
1.28
9.80
1583
285
35
19
1:200
1223.98
1.33
7.59
1:100
1236.63
1.55
2.18
1:20
1973.55
3.56
0.71
5903
601
120
70
Terra Nova
0
665.50
1.16
6.93
1038
168
18
11
¦¦¦
1:200
719.72
1.22
3.79
¦HM!
1:100
821.24
1.37
1.47
1:20
1380.34
3.76
0.40
5608
463
77
50
IA-E12PG00037 Final Report
Page 93
-------
EPA/600/R-16/152
September 2016
2-ring
3-ring
4-ring
Type II
F 1
r max
Alkanes
PAHs
PAHs
PAHs
Oils
DOR
(RU)
IFE
FIR
(M-g/L)
(M-g/L)
(Mg/L)
(Mg/U
Access
0
39.58
1.02
11.41
93
15
4
10
Western
1:200
46.52
1.02
11.62
Blend Dilbit
1:100
49.84
1.03
8.54
;
1:20
60.19
1.08
1.13
258
40
17
37
Belridge
0
118.69
1.07
4.90
42
31
21
30
Heavy
1:200
161.75
1.08
5.22
1:100
140.96
1.07
4.62
1:20
147.09
1.12
3.30
44
49
35
52
Cold Lake
0
120.61
1.05
18.92
155
74
20
22
Dilbit
1:200
120.65
1.05
17.33
1:100
125.85
1.06
11.50
1:20
133.15
1.17
1.92
368
160
50
57
Heidrun
0
902.69
1.28
4.50
382
337
55
43
1:200
909.47
1.26
6.22
1:100
964.31
1.33
3.74
1:20
1098.90
2.19
0.77
684
524
96
79
Hondo
0
283.04
1.09
18.38
412
76
8
2
1:200
312.27
1.08
15.38
1:100
274.80
1.07
17.44
1:20
288.01
1.07
16.33
319
67
7
1
IFO-40
0
1173.91
1.20
27.94
1324
408
152
168
1:200
1246.63
1.22
47.73
1:100
1338.56
1.25
35.74
1:20
1458.79
2.29
4.57
4354
1033
475
561
IFO-120
0
3030.69
1.68
109.67
343
607
88
49
1:200
2903.21
1.61
101.83
¦¦1
1:100
3090.23
1.65
85.76
1:20
2527.73
1.67
34.32
840
885
212
118
IFO-180
0
1263.05
1.22
33.67
866
410
235
320
1:200
1394.42
1.23
36.08
1:100
1703.55
2.37
36.31
1:20
1532.99
1.63
12.19
2933
1189
797
1095
IFO-300
0
720.55
1.11
61.26
446
216
79
188
1:200
443.51
1.14
25.85
1:100
465.91
1.07
45.70
1:20
661.50
1.10
39.81
366
183
65
157
Lago
0
352.22
1.07
12.13
1289
114
32
18
IA-E12PG00037 Final Report
Page 94
-------
EPA/600/R-16/152
September 2016
2-ring
3-ring
4-ring
Type II
F 1
r max
Alkanes
PAHs
PAHs
PAHs
Oils
DOR
(RU)
IFE
FIR
(Hg/L)
(Hg/L)
(Hg/L)
(Hg/L)
1:200
398.40
1.13
11.73
1:100
367.75
1.12
8.88
1:20
453.10
1.75
0.92
4221
258
88
51
,
Santa CI a-a
0
157.30
1.05
25.93
209
27
1
0
Ml
1:200
147.55
1.05
22.73
1:100
154.98
1.08
18.26
1:20
169.39
1.12
6.97
1196
72
9
3
Scotian
0
946.52
1.15
40.62
447
125
2
0
Shelf
Condensate
1:200
1:100
1408.59
1487.16
1.44
1.53
53.68
58.06
1:20
1337.98
1.50
47.65
1057
216
4
0
Vasconia
0
844.93
1.15
34.56
2550
317
98
59
1:200
1:100
828.37
835.62
1.16
1.17
44.43
34.09
1:20
935.79
1.67
3.41
4402
467
164
95
IA-E12PG00037 Final Report
Page 95
-------
EPA/600/R-16/152
September 2016
-'3W/
Cor&uk tSoa ~D<& 1 <50
/vie^A
V-3£>
!>D
Arabian Light
DOR 1:20
250 30O 350
EX Wavelength (nm)
IFO-40
DOR 1:20
O
Mesa
DOR 1:20
250 300 350
EX Wavelength (nm)
Santa Clara
DOR 1:20
?
250 300 350
EX Wavelength (nm)
2 SO 300 350
EX Wavelength (nm)
Figure 44. Photographs of pre-analysis samples and corresponding example EEMs of Type I
(left) and II (right) oils; DOR = 1:20 for Arabian Light (light oil, API gravity > 31.1°), Mesa
(medium oil, API gravity 22.3 - 31.1°) and heavy oils (IFO 40 and Santa Clara, API gravity <
22.3°).
IA-E12PG0Q037 Final Report
Page 96
-------
EPA/600/R-16/152
September 2016
c
31°), in order of increasing density:
1. Scotian Shelf Condensate, 2. Federated, 3. Brent, 4. MC252—Discoverer Enterprise, 5.
Hibernia, 6. MC252—generic, 7. Terra Nova, 8. Gullfaks, 9. Arabian Light. Note discrepancy
in Scotian Shelf Condensate fluorescence pattern (circled) from that of all other Light Oils.
It's particularly unusual that fluorescence intensity at highest DOR is lower than that at DORs
1:200 and 1:100.
IA-E12PG00037 Final Report
Page 97
-------
EPA/600/R-16/152
September 2016
3500 n
3000 -
2500 -
~ DORO
¦ DOR 1:200
a DOR 1:100
• DOR 1:20
Fmaxi Fluorescence Intensity (RU)
Heavy Oils
C
-------
EPA/600/R-16/152
September 2016
B.3.2 Fluorescence as a Function of Chemistry
Samples of dispersed oil in artificial seawater (DOR 0 and DOR 1:20 for each oil type), extracted
into methylene chloride were analyzed via GC-MS. Total alkanes, 2-ring, 3-ring, and 4-ring PAHs
(see Table 10 for list of hydrocarbons in each class) were each plotted against F maxl, Fmax2, Fmax3,
and Fmax4 (Figures 47-50). Results showed highest correlation at DOR 0 between total 3-ring PAHs
and fluorescence intensity at Fmax3 and Fmax4 (Figure 48) followed by that of 2-ring PAHs and
fluorescence intensity at Fmaxi and Fmax2 (Figure 47) and between 4-ring PAHs and fluorescence
intensity at Fmax3 and Fmax4 (Figure 48). It is important to note, however, that only 12 of the 25 oil
types exhibited any Fmax3 fluorescence at DOR 0.2 These correlations support the fact that larger,
more complex PAHs fluoresce at longer emission wavelengths.
For all oils at DOR 1:20, logarithmic relationships rather than linear relationships best modeled
all correlations; however, overall these were weaker than those found at DOR 0. Highest
correlation was observed between 2-ring PAHs and Fmax3 intensity (Figure 50), with moderate
correlations observed between 2-ring PAHs and fluorescence at Fmaxi and between 2-ring PAHs
and Fmax2 fluorescence (Figure 49), and between 3-ring PAHs and Fmax3 fluorescence (Figure 50).
Only weak correlations were observed between 2-ring PAHs and fluorescence at Fmax2 (Figure 49)
and between 4-ring PAHs and Fmax3 fluorescence (Figure 50). Clearly, full dispersion at DOR 1:20
results in widely varying changes in fluorescence intensity across all oil types.
2 Oil types exhibiting Fmax3 fluorescence at DOR 0: Access Western Blend Dilbit, Alaska North Slope (both fresh
and 10% weathered), Arabian Light, Brent, Federated, Gullfaks, Heidrun, Hibernia, MC252 (both Discoverer
Enterprise and generic), and Terra Nova.
IA-E12PG00037 Final Report
Page 99
-------
EPA/600/R-16/152
September 2016
Table 10. Individual hydrocarbon compounds reported as Total Alkanes, Total 2-ring, 3-ring and 4-ring PAHs.
Total Alkanes:
Total 2-ring PAHs:
Total 3-ring PAHs
Total 4-ring PAHs:
n-decane
Naphthalene
phenanthrene
pyrene
undecane
Methylnaphthalene
anthracene
methylpyrene
dodecane
Dimethylnaphthalene
methylphenanthrene
dimethylpyrene
tridecane
Trimethylnaphthalene
dimethylphenanthrene
trimethylpyrene
tetradecane
tetramethylnaphthalene
trimethylphenanthrene
tetramethylpyrene
pentadecane
Acenaphthene
tetramethylphenanthrene
naphthobenzothiophene
hexadecane
Acenaphthylene
fluoranthene
methylnaphthobenzothiophene
heptadecane
Fluorene
dimethylNBenzothiophene
2,6,10,14-TMPdecane
(pristine)
Methylfluorene
trimethylNbenzothiophene
octadecane
Dimethylfluorene
tetramethylNbenzothiophene
2,6,10,14-TMHdecane
(phytane)
Trimethylfluorene
benz[a]anthracene
nonadecane
Dibenzothiophene
chrysene
eicosane
methyldibenzothiophene
methylchrysene
heneicosane
dimethyldibenzothiophene
dimethylchrysene
docosane
trimethyldibenzothiophene
trimethylchrysene
tricosane
tetramethyldibenzothiophene
tetramethylchrysene
tetracosane
benzo[b]fluoranthene
pentacosane
benzo[k]fluoranthene
IA-E12PG00037 Final Report
Page 100
-------
EPA/600/R-16/152
September 2016
Total Alkanes:
Total 2-ring PAHs:
Total 3-ring PAHs
Total 4-ring PAHs:
hexacosane
benzo[e]pyrene
heptacosane
perylene
octacosane
n-nonacosane
tricontane
n-heneicontane
dotriacontane
tritriacontane
tetratriacontane
n-pentatriacontane
17a(H), 21(B (H)-hopane
17(B(H), 21a(H)-hopane
IA-E12PG00037 Final Report
Page 101
-------
The effect of DOR 1:20 on dissolved hydrocarbons can also be investigated by taking the ratio
of total alkanes + PAHs at DOR 1:20 to total alkanes + PAHs at DOR 0 to yield the Chemical
Dispersibility Ratio (CDR). The ratio ranges from between 0.8 for two heavy oils (Hondo and
IFO 300) and 7.8 for Arabian Light. Although heavy oils tended to have lower CDRs and light
oils tended to have higher ratios, oil density was not correlated with chemical dispersion For
example, the heavy oil Santa Clara (API Gravity 22.1°) had the third highest CDR (5.4), while
Scotian Shelf Condensate, by far the lightest oil (API Gravity 46.6°), had a CDR of only 2.2
(Figure 51 The effect of dispersion on fluorescence intensity can be similarly investigated by
taking the ratio of Fmaxi fluorescence intensity at DOR 1:20 to that at DOR 0, resulting in the
Fluorescence Dispersibility Ratio (FDR). The FDR also shows a general increasing trend with
increasing API Gravity, but only a moderate linear correlation (R2 = 0.55). The relationship
between CDR and FDR exhibited weak linear correlation (R2 = 0.17) (Figure 52).
All four Intermediate Fuel Oils (IFO 40, IFO 120, IFO 180, and IFO 300) as well as Scotian Shelf
Condensate (SSC), showed fluorescence and chemistry anomalies that tended to skew overall
results. With respect to SSC, all other light oils (API Gravity < 22.3°) exhibited increasing
fluorescence intensity with increasing DOR, culminating in an increase at DOR 1:20; however,
SSC showed a decrease in fluorescence intensity at DOR 1:20, dropping to below the level
exhibited at DOR 1:200 (Figure 45). Additionally, SSC was the only light oil which exhibited no
Fmax3 fluorescence at any DOR. Chemically, SSC is unique, containing a high proportion of 2-
ring to 3-ring PAHs—52.2 for DOR 0 and 58.6 for DOR 1:20. With the exception of Santa Clara,
with a 2-ring to 3-ring ratio of 31.9 at DOR 0, all other oil types had a ratio of 10 or less at both
DOR 0 and DOR 1:20. SSC also contained no 4-ring or 5-ring PAHs, unlike all other oils with the
exception of DOR 0 Santa Clara. All Intermediate Fuel Oils fell into the heavy oil group (API
Gravity > 31°), in which all other oils showed little to no increase in fluorescence intensity with
increasing DOR as well as maximum Fmaxi intensity of just 60-288 RU. The IFOs, however,
showed far greater Fmaxi intensity across the board (721-3031 RU) along with clear separation
with increasing DOR. Like Scotian Shelf Condensate, IFO 120, IFO 180, and IFO 300 also
exhibited a drop in Fmaxi intensity at DOR 1:20; in fact, IFO 120 Fmaxi at DOR 1:20 was actually
-------
EPA/600/R-16/152
September 2016
17% lower than at DOR 0. These same three IFOs also had the highest overall concentration of
PAHs, and all four IFOs were the only oils to contain any anthracene. For all oil types, total
alkanes as a function of fluorescence intensity was found to be only loosely correlated, as total
concentration increased overall in relation to fluorescence intensity with no clear relationship.
IA-E12PG00037 Final Report
Page 103
-------
EPA/600/R-16/152
September 2016
800
350000
2, 3 and 4-Ring PAHs vs. F
DOR = 0
maxl
700
300000
600
250000
y = 0.3352x+ 6.4568
R2 = 0.8203
500
200000 i
~ ~
^ 400
150000 =
~~
.= 300
100000
y = 35.602X+ 18013
R2 = 0.0809
~
i
1 I ' ¦
500
50000
100
-''km
y = 0.0437x+ 12.588 A
R2 = 0.237
1000
1500
Intensity (RU)
2000
2500
3000
~ 2 Ring PAHs
Linear (2 Ring PAHs)
3 Ring PAHs
¦ 4 Ring PAHs
Linear (3 Ring PAHs) Linear (4 Ring PAHs)
800
350000
2, 3 and 4-Ring PAHs vs. F,
DOR = 0
max2
700
300000
600
250000
_ 500
200000 2
y = 0.3352x+ 6.4568
R2 = 0.8203
i 400
150000 .£
300
100000
200
50000
100
~ ¦
50
100
150
200
Intensity (RU)
250
300
350
400
Linear (2 Ring PAHs)
~ 2 Ring PAHs
3 Ring PAHs ¦ 4 Ring PAHs
Figure 47. For all oil types at DOR 0, total concentration of 2-ring, 3-ring, and 4-ring PAHs (|ig/L) against
fluorescence intensity (RU) at Fmaxi (top), and against Fmax2 (bottom). Strong linear correlation exists
between 2-ring PAHs and Fmaxi fluorescence, but little to no correlation between 3-ring or 4-ring PAHs
and Fmaxi fluorescence intensity (top). Strong linear correlation also exists between 2-ring PAHs and
Fmax2, but no correlation between 3-ring PAHs or 4-ring PAHs and Fmax2 (bottom).
IA-E12PG00037 Final Report
Page 104
-------
EPA/600/R-16/152
September 2016
400
50000
2, 3 and 4-Ring PAHs vs. F,
DORO
max3
45000
350
y = 11.926X+ 75.883
4. R2 = 0.6661
40000
300
35000
250
_ -' y = 1438.3x + 4939.!
R2 = 0.8118
30000
200
25000
20000 2
c 150
15000
100
10000
50
Jl-
y = 2.3946x+3.5402 5000
R2 = 0.8673
A A
10
15
Intensity (RU)
20
25
~ 2 Ring PAHs
Linear (2 Ring PAHs)
~ 3 Ring PAHs
¦ 4 Ring PAHs
Linear (3 Ring PAHs) Linear (4 Ring PAHs)
800
350000
2, 3 and 4-Ring PAHs vs. F,
DORO
max4
700
300000
600
250000
y = 11.926x+ 75.883
R2 = 0.6661
500
200000 3
~~
400
150000 c
~ ~
c 300
100000
"" y = 1438.3x+4939.9
R2 = 0.8118
200
50000
100
y = 2.3946x+3.5402 ±
R2 = 0.8673
100
150
Intensity (RU)
200
250
300
~ 2 Ring PAHs
Linear (2 Ring PAHs)
~ 3 Ring PAHs
¦ 4 Ring PAHs
Linear (3 Ring PAHs) Linear (4 Ring PAHs)
Figure 48. For all oil types at DOR 0, total concentration of 2-ring, 3-ring, and 4-ring PAHs (|xg/L) against
fluorescence intensity (RU) at Fmax3 (top), and against Fma*4 (bottom). Strong linear correlation exists
between 3-ring and 4-ring PAHs and both FmaX3 and FmaX4 fluorescence; however, only moderate
correlation exists between 2-ring PAHs and FmaX3 and FmaX4 fluorescence intensity.
IA-E12PG0Q037 Final Report
Page 105
-------
EPA/600/R-16/152
September 2016
2500
2, 3 and 4-Ring PAHs vs. F,
DOR = 1:20
1200000
maxl
1000000
2000
800000
ZT 1500
600000
1000
400000
y = 216.8ln(x) - 999
R2 = 0.5402
500
200000
500
1000
1500
Intensity (RU)
2000
2500
3000
~ 2 Ring PAHs A 3 Ring PAHs ¦ 4 Ring PAHs —Log. (2 Ring PAHs)
2, 3 and 4-Ring PAHs vs. F,
DOR = 1:20
2500
1200000
max2
1000000
2000
800000
3- 1500
600000
1000
400000
y = 204.38ln(x)-566.55
* R2 = 0.3315
500
200000
100
300
Intensity (RU)
200
400
500
600
Log. (2 Ring PAHs)
~ 2 Ring PAHs A 3 Ring PAHs ¦ 4 Ring PAHs
Figure 49. For all oil types at DOR 1:20, total concentration of 2-ring, 3-ring, and 4-ring PAHs (ng/L)
against fluorescence intensity (RU) at Fmaxi(top), and against Fma*2 (bottom). A moderate logarithmic
correlation is exhibited between 2-ring PAHs and fluorescence intensity (RU) at Fmaxi and a weaker
correlation between 2-ring PAHs and FmaX2, but no correlation exists between 3-ring or 4-ring PAHs
and fluorescence intensity at either Fmaxl or Fmax2"
IA-E12PG0Q037 Final Report
Page 106
-------
EPA/600/R-16/152
September 2016
2, 3 and 4-Ring PAHs vs. F,
DOR 1:20
700
120000
max3
y = 111.78ln(x)-145.61
R2 = 0.7329
600
100000
500
80000
400
y = 9362.3ln(x) +12699 B ¦
R2 = 0.2893
60000
2 300
40000
¦o 200
20000
100
y = 18.032ln(x)- 6.0361
R2 = 0.6687
100
200
300
400 !
Intensity (RU)
500
600
700
800
~ 2 Ring PAHs
Log. (2 Ring PAHs)
a 3 Ring PAHs
¦ 4 Ring PAHs
Log. (3 Ring PAHs) Log. (4 Ring PAHs)
2500
1200000
2, 3 and 4-Ring PAHs vs. F,
DOR = 1:20
maxl
1000000
2000
800000
CT 1500
600000
1000
y = 221.74ln(x)-544.41
R2 = 0.525
400000
500
200000
50
100
150
Intensity (RU)
200
250
Log. (2 Ring PAHs)
~ 2 Ring PAHs ~ 3 Ring PAHs ¦ 4 Ring PAHs
Figure 50. For all oil types at DOR 1:20, total concentration of 2-ring, 3-ring, and 4-ring PAHs (ng/L)
against fluorescence intensity (RU) at Fmax3 (top), and against Fma*4 (bottom). A strong logarithmic
correlation is exhibited between 2-ring PAHs and fluorescence intensity at FmaX3. Moderate
correlations exist between 3-ring PAHs and FmaX3 as well as between 2-ring PAHs and Fmax4. However,
only a weak logarithmic correlation exists between 4-ring PAHs and fluorescence intensity at FmaX3,
and there is no correlation between 3-ring or 4-ring PAHs and Fmax4.
IA-E12PG0Q037 Final Report
Page 107
-------
EPA/600/R-16/152
September 2016
Chemical Dispersibility Ratio vs. API Gravity
y = 0.0641X+ 1.4558
R2 = 0.1026
API Gravity (degrees)
Fluorescence Dispersibility Ratio vs. API Gravity
API Gravity (degrees)
Figure 51. Chemical Dispersibility Ratio (CDR) vs. decreasing oil density (top) and Fluorescence
Dispersibility Ratio (FDR) vs. decreasing oil density (bottom) show only a weak correlation between
chemistry and oil density, and a moderate correlation between fluorescence and oil density. With the
removal of the data point for Scotian Shelf Condensation, correlation between fluorescence and oil
density improves to R2 = 0.71.
IA-E12PG00037 Final Report
Page 108
-------
EPA/600/R-16/152
September 2016
2.5
~ 2.0
D
1.5
1.0
>¦ 0.5
0.0
Fluorescence Dispersibility Ratio vs.
Chemical Dispersibility Ratio
y = 0.091x+ 1.1749
R2 = 0.1721
H-
H-
H-
H-
H-
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0
Total PAHs+Alkanes in DOR 1:20 Oils (ng/L)/Total PAHs+Alkanes in DOR 0 Oils (ng/L)
9.0
Figure 52. Fluorescence Dispersibility Ratio (FDR) vs. Chemical Dispersibility Ratio (CDR) shows weak
correlation between these two ratios.
IA-E12PG00037 Final Report
Page 109
-------
EPA/600/R-16/152
September 2016
B33 Flume Tank and Baffled Flask EEM Comparison
In addition to the EEMs generated from the BFT of 25 oil types, EEMS were also generated from
the discrete sample collection during the flume tank experiments using South Louisiana Crude oil
(SLC) in Task A of this project report. Samples for EEM analysis were collected and immediately
analyzed on the same Horiba Aqualog at DFO using identical analysis protocols and data
processing. A comparison of SLC MC252 EEMs for varying DOR from the BFT (left) and the flume
tank (right) experiments are illustrated in Figure 53. Note that that the contour coloring for the
peaks is identical between experiments, but the baseline color varied, where black was used for
the BFT EEMs and blue used for tank EEMs, but the appearance of the blue color this is not to be
confused with the presence of higher fluorescence in regions away from the peak fluorescence.
Fluorescence Intensity Ratios (FIR) were calculated for the tank EEMs and found to be between
7.1 and 9.1 for DOR = 0, 1.3 and 4.3 for DOR = 1:100 and 0.6 and 0.8 for DOR = 1:20. This is
follows the findings of Bugden et al., 2008 where a decrease in FIR is observed with the addition
of dispersant. It is also consistent with the BFT EEMs which show a 4.9 for DOR = 0 and 0.4 for
DOR = 1:20 (Supplemental Table A). These results indicate that FIR can be an indicator of
dispersion effectiveness for SLC oil.
IA-E12PG00037 Final Report
Page 110
-------
EPA/600/R-16/152
September 2016
MS252 BFT
a
DOR = 0
MC252 Tank
DOR = 0
EX Wavelength (nm)
EX Wavelength (nm)
P-
?¦
I
DOR = 1:20 ¦
250 300 350
EX Wavelength (nm)
DOR = 1:100
DOR = 1:100
250 300 350
EX Wavelength (nm)
rjk
/
6 »
DOR = 1:20
250 300 350
EX Wavelength (nm)
250 300 350
EX Wavelength (nm)
Figure 53. South Louisiana Crude MC252 EEMS from BFT (left panels) and tank experiments
(right Panels) for DOR = 0,1:100 and 1:20.
IA-E12PG0Q037 Final Report
Page 111
-------
EPA/600/R-16/152
September 2016
B3.4 PARAFAC Modeling
Originally designed to model complexity in the field of psychometrics (Carroll and Chang, 1970;
Harshman, 1970), parallel factor analysis (PARAFAC, also known as canonical decomposition or
CANDECOMP) was first employed in the analysis of fluorescence data within the next ten years
(Appellof and Davidson, 1981). Over the past twenty years, PARAFAC has been widely embraced
by chemometricians and used to tease apart the overlapping fluorescence components of
complex chemical mixtures containing fluorescent substances ranging from proteins and
pigments to pesticides and PAHs (Andersen and Bro, 2003). More recently, PARAFAC analysis
has been used in the analysis of the fate and transport of dispersed oil from the Deepwater
Horizon Oil Spill (Mendoza, et al., 2013; Zhou, et al., 2003).
Presented with hundreds of complex fluorescence EEM data sets containing samples, excitation,
and emissions; PARAFAC analysis can reduce this to data sets containing samples and intensity
at a few important wavelength pairs (Murphy et al., 2014). In the past, this information gathering
was often done via time-consuming "peak-picking" whereby EEMs were visually inspected for
apparent Fmax location, then fluorescence intensity data at that excitation/emission point was
copied and pasted into a spreadsheet for further analysis. PARAFAC provides the capability to
turn that somewhat qualitative task into a more quantitative exercise; however, careful
preparation of the data is critical in order to obtain a meaningful outcome. PARAFAC analysis
also allows the consideration of minor fluorescence peaks, which may have been overwhelmed
by high-intensity major peaks, but may be no less important in the analysis of EEM results. More
importantly, PARAFAC analysis allows for direct comparison to chemical composition upon
successful modelling of an EEM data set (Murphy, et al., 2014). The steps that must be
undertaken for successful PARAFAC analysis are: (1) assembling the dataset; (2) preprocessing to
correct biases, remove scatter and normalize; (3) exploring the dataset to remove possible
outliers and develop preliminary models; (4) validating the model by determining the proper
number of components and evaluating model fit; (5) interpreting results (Murphy, et al., 2013).
In order to identify connections between the fluorescence profiles and underlying chemical
complexity of the 25 oil types in the BFT analysis (Figure 53), PARAFAC analysis was performed
IA-E12PG00037 Final Report
Page 112
-------
EPA/600/R-16/152
September 2016
on the fluorescence data. The PLS Toolbox (Eigenvector, Inc.) was used within MATLAB
(MathWorks, Inc. 2014b) to accomplish this task. After importing raw data and assembling
datasets, three constraints were applied to all samples: normalization, EEM filtering, and non-
negativity. Normalization was conducted to compensate for the wide variation in fluorescence
intensity across oil types (Fmaxi = 39.6 RU for Access Western Blend Dilbit to Fmaxi = 3090.2 RU for
IFO 120) in order to prevent samples with high fluorescence intensity values from skewing the
model. Further, normalization of maximum intensity to 1 (inf-Norm) was chosen rather than
normalization of the entire area of fluorescence (1-Norm) to preserve differences in spectral
shape. EEM filtering was applied in order to remove artifacts of the fluorescence analysis process
known as first and second order Rayleigh scatter. This was accomplished by interpolating data
across those regions (12 nm for first order Rayleigh and 24 nm for second order Rayleigh); zero
values were also assigned to sub-Rayleigh wavelengths since fluorescence emission takes place
at wavelengths above excitation due to Stokes shift. Raman scatter, the other light-related
artifact which must be removed before PARAFAC analysis can be performed, was accomplished
as sample analysis was done by subtracting that day's sample blank from each sample. Upon
running several PARAFAC test models using 4, 5, 6 and 7-components on a dataset containing the
DOR 0 sample from flask #1 of all 25 oil types, data between excitation at 200 nm and 212 nm
was excluded. The inherent "noise" typically found at excitation < 240 nm, related to the low
intensity of xenon lamps in that region, led to this decision. Excluding data at excitation and
emission wavelengths above 680 nm was also employed in order to improve processing results
since no fluorescence information of value was contained in that region.
The biggest challenge in PARAFAC modelling is in determining the most appropriate number of
component factors. While it is important to ensure separation of all individual factors, it is also
critical not to select too many components in order to avoid over-fitting the data. Several ways
of doing this are suggested in the PARAFAC tutorial (Bro, 1997): comparison of the resulting factor
profiles with background knowledge of expected components, consideration of the residuals,
and split half validation of the model. The latter has also been recommended by other
researchers (Harshman and Lundy, 1994; Murphy, et al., 2013). Split half analysis is accomplished
by dividing the data into two independent subsets and applying the model to each of the subsets.
IA-E12PG00037 Final Report
Page 113
-------
EPA/600/R-16/152
September 2016
In theory, if the correct number of components has been selected, the two halves of the data
should each fit the model well; however, Murphy cautions that a relatively large data set is
necessary in order for this to hold true (2013). Smilde, et al., (2004) also caution that some
phenomenon observed within a subset of data may not match the overall model, but instead
may just be present in that particular random half of the dataset. Thus, it could be anticipated
that split half validation will work better with samples within oil weight subdivisions than with
the dataset containing all 25 oil types as a whole. Bro and Kiers (2003) have also advised using
core consistency of the model to validate that the correct number of components have been
selected. All of these methods were employed for the following analyses by first noting the
percentage of data fit by the model, next checking the core consistency of the model, then
inspecting residuals, inspecting the loadings for Mode 3 (excitation) and Mode 2 (emission), and
inspecting EEMs of each component. Finally, split half analysis was done. In all cases, several
models were run with different numbers of components to ensure selection of the most
appropriate model.
DORO
Initially, a five-component model was fit to the dataset, followed by 4-, 6- and 7-component
models. Best overall fit was obtained with the six-component model, which explained 99.504%
of the data. Core consistency was 52%, and split half validation was 56.4% (Figure 54). Review
of residuals showed they were minimal with random distribution, inspection of plots of Mode 2
and Mode 3 loadings (Figure 55), variation per component (Figure 56), as well as EEMs of
individual components (Figure 57) all supported choice of the 6-factor model for best fit.
IA-E12PG00037 Final Report
Page 114
-------
EPA/600/R-16/152
September 2016
Similarity measure of splits and overall model 56.4%
eraII model
Set 1
Set 2
40 50 60
Mode 2
rail mode!
40 50
Mode 2
eraII model
Set 1
Set 2
Mode 3
eraII model
Set 1
Set 2
Mode 3
Figure 54. Example of split half validation for the 6-component model of 25 oil types at DOR 0
showing individual fit of data splits (Set 1, left; and Set 2, right) compared to overall model for
Mode 2 (top) and Mode 3 (bottom) loadings.
IA-E12PG0Q037 Final Report
Page 115
-------
EPA/600/R-16/152
September 2016
Mode 3 Loadings
Mode 2 Loadings
Q , Vsi/S r I I ¦ I ¦ i. n
250 300 350 400 450 500 550 600 650
Emission (nm)
350
Excitation (nm)
Figure 55. Mode 3 Loadings (Excitation) and Mode 2 Loadings (Emission) for all 25 oil types—
DORO using 6-component model. Note difference in x-axis scales. Although components are
tightly spaced, all appear as separate and distinct peaks.
IA-E12PG00037 Final Report
Page 116
-------
EPA/600/R-16/152
September 2016
Variation per Component
40
X 35
Fit (% X)
Unique Fit (% X)
£ 30
OJ
X 20
ir 15
1 .
iT 5
0
2
3
4
5
6
Component number
Figure 56. Variation per Component shows Component 1 accounted for >20% to 40% (unique
fit and fit) of the data, while Component 2-contributed 5-10% (unique fit and fit) and
Components 3-6 accounted for 5% or less of the data, respectively. While Component 6
accounted for a very low percentage of the data, the 6-component model was still a better fit
to the data than the 5-component model.
IA-E12PG00037 Final Report Page 117
-------
EPA/600/R-16/152
September 2016
Component #1
400.
Component #2
2 300
2 300
Emission nm)
400
400
Component #3
.2 300
"
0.09
400.
0.08
0.07
350
0.06
0.05
300
0.04
0 03
0.02
250
0.01
1
c
200
400 450 500
Emission (nm)
Component #4
400 450
Emission (nm)
Component #5
Ann
400 450
Emission (nm)
Component #6
2 300
Emission (nm)
400 450 500
Emission (nm)
Figure 57, EEM views of the six components of PARAFAC model for 25 oil types at DOR 0.
Component #1: Fmax = Ex 224nm/Em 335nm; Component #2: Fmax = Ex 230nm/Em 340nm;
Component #3: Fmax = Ex 239nm/Em 363nm; Component #4: Fmax = Ex 21Snm/Em 290 nm;
Component #5: Fmax = Ex 221nm/Em 322nm; Component #6: Fmax = Ex 260nm/Em 474-
511nm.
600
IA-E12PG0Q037 Final Report
Page 118
-------
EPA/600/R-16/152
September 2016
DOR 1:100
A six-component model was fit to the dataset containing all 25 oil types at DOR 1:00 since that
was the best fit for the DOR 0 dataset, followed by a 7-component model, which returned an
error warning that two or more components may be fitting the same feature, as well as core
consistency <0%. Finally, a 5-component model was fit to the dataset. Interestingly, for the DOR
1:100 dataset, the 5-component model proved to be the best fit, explaining 99.353% of the data
with core consistency of 72% and split half validation of 75.8%. Residuals were minimal and
randomly distributed, and visual inspection of loadings (Figure 58), variation per component
(Figure 59) and component EEMs (Figure 60) led to acceptance of the 5-component model.
IA-E12PG00037 Final Report
Page 119
-------
EPA/600/R-16/152
September 2016
Mode 3 Loadings
Mode 2 Loadings
350
Excitation (nm)
250 300 350 400 450 500 550 600 650
Emission (nm)
Figure 58. Mode 3 Loadings (Excitation) and Mode 2 Loadings (Emission) for all 25 oil types—
DOR 1:100 using 5-component model. Note difference in x-axis scales. Although components
are tightly spaced, all appear as separate and distinct peaks.
IA-E12PG00037 Final Report
Page 120
-------
EPA/600/R-16/152
September 2016
Variation per Component
I Fit (% X)
Unique Fit (% X)
Component number
Figure 59. Variation per Component shows Component 1 accounted for >35% to almost 50%
(unique fit and fit) of the data, while Components 2-5 accounted for 5% or less of the data,
respectively.
IA-E12PG00037 Final Report
Page 121
-------
EPA/600/R-16/152
September 2016
Component #1
Component #2
2 300
Emission (nm)
400
Component #3
2 300
Emission {nm)
0.2
0.18
0.16
0.14
E
0.12
a
5
0.1
Is
"O
0.08
i2
0.06
0.04
0.02
0
400
Component #5
.2 300
2 3U0
Emission (nm)
Component #4
Emission (nm)
250 300 350 400 450 500 550 600
Emission (nm)
Figure 60. EEM views of the five components of PARAFAC model for 25 oil types at DOR 1:100.
Component #1: Fmax = Ex 224nm/Em 335nm; Component #2: Fmax = Ex 254-266nm/Em 455-
501nm; Component #3: Fmax= Ex 230nm/Em 344nm; Component #4: Fmax = Ex 242nm/Em 363
nm; Component #5: Fmax = Ex 218nm/Em 290nm,
0.08
0.07
J 0.06
10.05
I 0.04
I 0.03
I 0.02
10.01
'o
IA-E12PG00037 Final Report
Page 122
-------
EPA/600/R-16/152
September 2016
DOR 1:20
Once again, a six-component was fit to the dataset containing all 25 oil types at DOR 1:20;
however, an error message warning that two or more components may be fitting the same data
was displayed, and the core consistency was <0%. Fitting a 5-component model to the data,
however, resulted in 98.891% of the data explained by the model as well as core consistency of
84% and a split half validation of 84%. Overall, residuals were minimal and randomly distributed;
however, residuals appeared to occur at somewhat higher wavelengths than at other DORs.
Visual inspection of loadings (Figure 60), variation per component (Figure 62), and component
EEMS (Figure 63) led to final acceptance of the 5-component model.
IA-E12PG00037 Final Report
Page 123
-------
EPA/600/R-16/152
September 2016
250 300 350 400 450
Excitation (nm)
Mode 3 Loadings
Mode 2 Loadings
400 450 500
Emission (nm)
Figure 61. Mode 3 Loadings (Excitation) and Mode 2 Loadings (Emission) for all 25 oil types—
DOR 1:20 using 5-component model. Note difference in x-axis scales. Effect of full dispersion
appears to broaden and shift emission peaks to longer wavelengths.
Variation per Component
Fit (% X)
Unique Fit (% X)
Component number
Figure 62. Variation per Component shows Component 1 accounted for 25 to 30% of the data
(unique fit and fit) while Component 2 has increased to >10% to 25% (unique fit and fit) of the
data. Contribution from Component 3 and 4 have increased, as well.
IA-E12PG00037 Final Report
Page 124
-------
Component #1
Emission (nm)
Component #3
250 300 350 400 450 500 550 600
Emission (nm)
Component #5
Emission {nm)
EPA/600/R-16/152
September 2016
Component #2
3501
E*
E
-I 300 H
cn
*"5
,2
2501
2oo 1 i i i i i i 1
250 300 350 400 450 500 550 600
Emission (nm)
Component #4
Emission (nm)
Figure 63. EEM views of the five components of PARAFAC model for 25 oil types at DOR 1:20.
Component #1: Fmax = Ex 224nm/Em 335nm; Component #2: Fmax = Ex 233-266nm/Em 432-
450nm; Component #3: Fmax = Ex 230-242nm/Em 501-520nm; Component #4: Fmax = Ex
233nm/Em 349nm; Component #5: Fmax = Ex 218nm/Em 290nm.
IA-E12PG00037 Final Report
Page 125
-------
EPA/600/R-16/152
September 2016
PARAFAC Summary
PARAFAC analysis of EEM datasets for the 25 oil types at DOR 0, DOR 1:100 and DOR 1:20 show
interesting changes in fluorescence intensity with increasing dispersion. However, we see a
decrease in distinct components from six at DOR 0 to five at DOR 1:100 and 1:20. From analysis
of plots of Mode 3 (Excitation) and Mode 2 (Emission) Loadings, it appears that increased
dispersion results in a broadening and shift to longer emission wavelengths as well as in a larger
contribution of fluorescence intensity at higher wavelengths. Upon examination of the EEMs of
each component, several other patterns emerge. Even with the minimal dispersion at DOR 1:100,
contribution to the overall model from a broad fluorescence peak which provided the least
contribution to the overall model at DOR 0~Component #6—became second in importance at
DOR 1:100, albeit with a contribution to the model of only about 5%. Upon full dispersion at DOR
1:20, this broad, high-wavelength peak retained importance to the model of approximately 5-
7%; however, another broad, but slightly lower wavelength peak appeared as Component #2
with 12-25% contribution to the overall model. Throughout the entire analysis, Component #1
at Ex 224nm/Em 335nm remained the most important contribution to the model, which confirms
this fluorescence region as the best target for detecting oil in the marine environment. However,
since the region represented by Component #2 in the DOR 1:20 dataset becomes a major
contribution to the model only upon effective dispersion, the FIR ratio (Bugden et al., 2008) can
be used to track this important parameter.
The MC252 oil samples used for these analyses, both ones collected onboard the Discoverer
Enterprise during DWH and the generic version provided by BP, are classified as light, sweet crude
based on density and sulfur content. Overall, oil types range from light to heavy due to the
proportion of n-alkanes and cyclo-alkanes vs. aromatic hydrocarbon compounds, while sulfur
content determines the rank of sweet (<1%) vs. sour (>1%). These characteristics arise from
kerogen source and reservoir maturity (Tissot and Welty, 1978). The 25 oils analyzed in the BFT
cover a wide range of light to heavy oil types, as well as a range of sulfur content. Oil fluorescence
phenomena arise from the presence of Tt-bonding in C=C bonds, leading to highest fluorescence
intensity from polycyclic aromatic hydrocarbons (Ryder, 2005), with fluorescence intensity
tending to increase with increasing molecular weight (Mendoza, et al., 2013). However, the
IA-E12PG00037 Final Report
Page 126
-------
EPA/600/R-16/152
September 2016
presence of fluorescence quenching species, as well as energy transfer between complex
molecules, complicates the isolation of compound-specific fluorescence in crude oil analysis.
Fluorescence research has shown that heavy oils generally have broad, weak fluorescence while
lighter oils have narrower, more intense emission bands (Steffens and Landulfo, 2010). Due to
the hundreds, if not thousands, of complex hydrocarbons present in crude oils, characterization
of fluorescence arising from specific PAH molecules would not be useful. However, PARAFAC
analysis of these 25 oil types has shown that it is possible to use fluorescence characterization in
specific wavelength regions for detect ion of non-dispersed vs. dispersed oil across a wide variety
of oil types.
The well depth of the MC252 oil source is by far the deepest of all our 25 oil type sources
(approximately 1600 m); however, a number of other oil types were sourced from offshore well
locations. These include the light oils Brent and Gullfaks from the North Sea (140-230 m water
depth) as well as Hibernia, Scotian Shelf Condensate and Terra Nova from offshore eastern
Canada (12-100 m water depth). Intermediate weight oils Heidrun from the Norwegian Sea (350
m water depth) and Sea Rose from off the coast of Newfoundland, Canada (100 m water depth)
as well as the heavy oil Hondo from offshore California (260 m water depth) were also included
in this study. The intermediate weight Alaskan North Shore, both fresh and 10% weathered,
would be representative of oil which may be sourced from offshore Alaska in the future.
Additionally, with the presence of approximately 3,000 platforms in the U.S. Gulf of Mexico
(BOEM, 2016), understanding the characterization of non-dispersed and dispersed MC252 oil will
certainly aid in preparedness for the possibility of future oil spill events in that region.
IA-E12PG00037 Final Report
Page 127
-------
EPA/600/R-16/152
September 2016
References
Alliance for Coastal Technologies Report (2008) Hydrocarbon sensors for oil spill prevention and response.
Ref # [UMCES]CBL08-095, 28 pp.
Andersen, C. M., R. Bro (2003) Practical aspects of PARAFAC modeling of fluorescence excitation-emission
data. J. Chemom., 17:200-217.
Appellof, C.J., and E.R. Davidson (1981) Strategies for Analyzing Data from Video Fluorometric Monitoring
of Liquid Chromatographic Effluents. J. Anal. Chem. 53:2053-2056.
Bahram, M., et al. (2006). "Handling of Rayleigh and Raman scatter for PARAFAC modeling of fluorescence
data using interpolation." Journal of Chemometrics 20(3-4): 99-105.
Bieroza, M., A. Baker, J. Bridgeman (2010) Classification and calibration of organic matter fluorescence
data with multiway analysis methods and artificial neural networks: an operational tool for improved
drinking water treatment. Envirometrics, 22: 256-270.
Boehme, J.R., PG. Coble, R.N. Conmy (2004) Examining CDOM fluorescence variability using principal
component analysis: seasonal and regional modeling of three-dimensional fluorescence in the Gulf of
Mexico. Marine Chemistry, 89:3-14.
Bro, R. (1997) PARAFAC. Tutorial and applications. Chemometrics and Intelligent Laboratory Systems.
38:149-171.
Bro, R., and H.A.L. Kiers (2003) A new efficient method for determining the number of components in
PARAFAC models. J. Chemometrics 17:274-286.
Bugden, J.B.C., C.W. Yeung, P.E. Kepkay, K. Lee (2008) Application of ultraviolet fluorometry and EEMS to
fingerprint oil and chemically dispersed oil in seawater. Marine Pollution Bulletin, 56(4): 677-85.
CRRC Report (2010) Coastal Response Research Center Deepwater Horizon Dispersant USE Meeting
Report, May 26-27, 2010, 108 pp.
Carroll, J.D. and J.-J. Chang (1970) Analysis of Individual Differences in Multidimensional Scaling via an N-
Way Generalization of "Eckart-Young" Decomposition. Psychometrika 35:283-319.
Christensen, A.B. Hansen, J. Mortensen, O. Andersen (2005) Characterization and matching of oil samples
using fluorescence and PARAFAC. Anal. Chem., 77:2210-17.
Coble, P.G. (1996) Characterization of marine and terrestrial DOM in seawater using EEM spectroscopy.
Mar. Chem., 51: 325-346.
Coble, P.G. (2014) Aquatic Organic Matter Fluorescence. In Aquatic Organic Matter Fluorescence. Coble,
P.G., Lead, L., Baker, A., Reynolds, D.M. and Spencer, R.G.M., Eds., Cambridge University Press, pp. 75-
122.
Cole, M.G., T. King, K. Lee (2007) Analytical technique for extracting hydrocarbons from water using
sample container as extraction vessel in combination with a roller apparatus. Anal. Chem., 77: 2210-17.
IA-E12PG00037 Final Report
Page 128
-------
EPA/600/R-16/152
September 2016
Conmy, R.N., P.G. Coble and C.E. Del Castillo (2004) Calibration and performance of the WetLabs' SAFIre
in situ fluorometer using seawater. Contin. Shelf Res., 24: 431-42.
Conmy, R. N., et al. (2014a) Submersible Optical Sensors Exposed to Chemically Dispersed Crude Oil: Wave
Tank Simulations for Improved Oil Spill Monitoring. Environ Sci Technol 48(3): 1803-1810.
Conmy, R.N., C.E. Del Castillo, B. Downing, R.F. Chen (2014b) Calibrations and Quality Control of In situ
Sensors. In: Fluorescence Applications in Aquatic Science. Eds: A. Baker & P.G. Coble, Cambridge Univ.
Press, 31 pp.
Fingas, M. (2011) Introduction to Oil Chemistry and Properties. In Oil Spill Science and Technology; Fingas,
M., Eds; Elsevier, pp. 51-59.
Frank, U. (1978) A Review of Fluorescence Spectroscopic Methods for Oil Spill Source Identification.
Toxicological and Environment Chemistry Rev., 2:163-185.
Fuller, C.B., J.S. Bonner, C.A. Page, G. Arrambide, M.C. Sterling, Jr., T. Ojo (2003) Field instruments for real
time in-situ crude oil concentration measurements. A Conrad Blucher Institute, Texas A&M Corpus Christi
Report, 10 pp.
Harshman, R.A. (1970) Foundations of the PARAFAC procedure: Models and conditions for an
"explanatory" multimodal factor analysis. UCLA Working Papers in Phonetics 16:1-84.
Harshman, R.A. and M.E. Lundy (1994) PARAFAC: Parallel factor analysis. Computational Statistics & Data
Analysis 18:39-72.
Johansen, O., et al. (2013) Droplet breakup in subsea oil releases - Part 2: Predictions of droplet size
distributions with and without injection of chemical dispersants. Marine Pollution Bulletin 73(1): 327-335.
Joint Analysis Group Report (2010) Review of Data to Examine Subsurface Oil in the Vicinity of DWH MC-
252, May 19-June 19, 2010; 29 pp.
Kepkay, P., W. Young, J.B.C. Bugden, Z. Li, K. Lee (2008) Ultraviolet fluorescence spectroscopy (UVFS): A
new means of determining the effectiveness of chemical dispersants on oil spills. International Oil Spill
Conference. API, Washington D.C., Savannah, GA, pp. 639-44.
King, T.L., B. Robinson, C. Mclntyre, P. Toole, S. Ryan, F. Saleh, M.C. Boufadel, K. Lee (2015) Fate of surface
spills of cold lake blend diluted bitumen treated with dispersant and mineral fines in a wave tank. In:
Environmental Engineering Science, Vol. 32, No. 3, 01.03.2015, p. 250-261.
Lee, K., Li, Z., Boufadel, M.C., Venosa, A.D. and Scott Miles, M. (2009) Wave tank studies on dispersant
effectiveness as a function of energy dissipation rate and particle size distribution. Final report submitted
to NOAA/CRRC/UNH. Pp67+ appendix.
Li, Z., Lee, K., King, T., Boufadel, M.C. and Venosa, A.D., (2008) Oil droplet size distribution as a function of
energy dissipation rate in an experimental wave tank 2008 International Oil Spill Conference. American
Petroleum Institute, Washington D.C., Savannah, GA, pp. 621-626.
Li, Z., Lee, K., King, T., Boufadel, M.C. and Venosa, A.D., (2009a) Evaluating Chemical Dispersant Efficacy
in an Experimental Wave Tank: 2, Significant Factors Determining In Situ Oil Droplet Size Distribution.
Environmental Engineering Science, 26(9): 1407-1418.
IA-E12PG00037 Final Report
Page 129
-------
EPA/600/R-16/152
September 2016
Li, Z., Lee, K., King, T., Boufadel, M.C. and Venosa, A.D., (2009b) Evaluating crude oil chemical dispersion
efficacy in a flow-through wave tank under regular non-breaking wave and breaking wave conditions.
Marine Pollution Bulletin, 58: 735-744.
Li, Z., Lee, K., King, T., Boufadel, M.C. and Venosa, A.D., (2010) Effects of temperature and wave conditions
on chemical dispersion efficacy of heavy fuel oil in an experimental flow-through wave tank. Marine
Pollution Bulletin, In Press, Corrected Proof.
Li, Z., K. Lee, T. King, P.E. Kepkay, O. Mikkelsen, C. Pottsmith (2011) Monitoring dispersed oil droplet size
distribution at the GOM Deepwater Horizon spill site. International Oil spill Conference Proceedings,
March 2011.
Masutani, S.M., E.E. Adams (2000) Experimental Study of Multiphase Plumes with Application to Deep
Ocean Oil Spills," Final Report to U.S. Dept. of the Interior, Minerals Management Service, Contract No.,
1435-01-98-CT-30946.
Mendoza, W.G., D.D. Riemer, R.G. Zika (2013) Application of fluorescence and PARAFACto assess vertical
distribution of subsurface hydrocarbons and dispersant during the Deepwater Horizon oil spill.
Environmental Science Processes & Impacts 15:1017-1030.
Murphy, K. R., et al. (2010). "Measurement of Dissolved Organic Matter Fluorescence in Aquatic
Environments: An Interlaboratory Comparison." Environ Sci Technol 44(24): 9405-9412.
Murphy, K. R., C.A. Stedmon, D. Graeber, R. Bro (2013) Fluorescence Spectroscopy and mutli-way
techniques, PARAFAC. Anal. Methods, 5: 6557-6566.
Murphy, K.R., Bro, R., Stedmon, C.A. (2014) Chemometric Analysis of Organic Matter Fluorescence. In:
Aquatic Organic Matter Fluorescence. Eds: Coble, P.G., Lead, J., Baker, A., Reynolds, D.M., Spencer, R.G.M.
Cambridge University Press, pp.339-375.
NOAA Technical Report NOS OR&R 25 (2010): Joint Analysis Group, Deepwater Horizon Oil Spill: Review
of Preliminary Data to Examine Subsurface Oil in the Vicinity of MC252, May 19 - June 19, 2010, 168 pp.
Oil Budget Calculator (2010) NOAA Report Oil Budget Calculator: Deepwater Horizon, Federal
Interagency Solutions Group, 217 pp.
NRC Report (2005) National Research Council Oil Spill Dispersants: Efficacy and Effects, Washington, DC,
400 pp.
Priaro, M. (2014) A Bitumen and Dilbit Primer. https://www.behance.net/gallery/16654901/A-BITUMEN-
AND-DILBIT-PRIMER (accessed June 10, 2016).
Rosin, P., E. Rammler (1933) The Laws Governing the Fineness of Powdered Coal. J. Institute of Fuel, 7,
29-36.
Ryder, A.G. (2005) Analysis of Crude Petroleum Oils Using Fluorescence Spectroscopy. Reviews in
Fluorescence 2005 pp. 169-198.
Smilde, A., Bro, R., Geladi, P. (2004) Validation and Diagnostics. In: Multi-way Analysis with Applications
in the Chemical Sciences. John Wiley & Sons, Ltd. Pp. 145-173
IA-E12PG00037 Final Report
Page 130
-------
EPA/600/R-16/152
September 2016
Stedmon, C.A., S. Markager, R. Bro (2003) Tracing dissolved organic matter in aquatic environments using
a new approach to fluorescence spectroscopy. Mar. Chem., 82: 239-54.
Stedmon, C.A., R. Bro (2003) Characterizing dissolved organic matter fluorescence with parallel factor
analysis: a tutorial. Limnol. Oceanogr. Methods, 6: 572-579.
Steffens, J., Landulfo, E., Courrol, L.C., Guardani, R. (2011) Application of Fluorescence to the Study of
Crude Petroleum. J. Fluoresc. 21:859-864.
Tissot, B.P. and D.H. Welty (1978) Classification of Crude Oils. In: Petroleum Formation and Occurrence.
Springer-Verlag, pp. 369-377.
Venosa, A.D., D.W. King, G.A. Sorial (2002a) The Baffled Flask Test for Dispersant Effectiveness: A Round
Robin Evaluation of Reproducibility and Repeatability. Spill Science &Technology Bulletin, 7(5-6): 299-308.
Venosa, A.D., K. Lee, M.T. Suidan, S. Garcia-Blanco, S. Cobanli, M. Moteleb, J.R. Haines, G. Tremblay, M.
Hazelwood (2002b) Bioremediation and Biorestoration of a Crude Oil-Contaminated Freshwater Wetland
on the St. Lawrence River. Bioremediation Journal,6(3):261-281.
Wickley-Olsen, E., Boufadel, M.C., King, T., Li, Z., Lee, K., Venosa, A.D. (2007) Regular and breaking waves
in wave tank for dispersion effectiveness testing. In: Proceedings of the 30th Arctic and Marine Oil spill
Program (AMOP) Technical Seminar, Edmonton, AB, Canada. Environment Canada, Ottawa, Ontario,
Canada, pp. 161-187.
Zepp, R. G., et al. (2004). "Dissolved organic fluorophores in southeastern US coastal waters: correction
method for eliminating Rayleigh and Raman scattering peaks in excitation-emission matrices." Marine
Chemistry 89(1-4): 15-3.
Zhao, L., et al. (2014). "Evolution of droplets in subsea oil and gas blowouts: Development and validation
of the numerical model VDROP-J." Marine Pollution Bulletin 83(1): 58-69.
Zhou, Z, Guo, L., Shiller, A.M., Lohrenz, S.E., Asper, V.L., Osburn, C.L. (2003) Characterization of oil
components from the Deepwater Horizon oil spill in the Gulf of Mexico using fluorescence EEM and
PARAFAC techniques. Marine Chemistry 148:10-21.
IA-E12PG00037 Final Report
Page 131
-------
EPA/600/R-16/152
September 2016
Appendices
APPENDIX A - Experiment Logs
APPENDIX B-Analytical Chemistry Results
APPENDIX C-Jet Release LISST Oil Droplet Size Distribution Histograms
APPENDIX D - Jet Release LI! Droplet Size Distribution Time Series Contours
APPENDIX E - Submersible Fluorescence Time Series
APPENDIX F - Excitation Emission Matrix Contours
APPENDIX G - VDROP-J and JET LAG Numerical Plume Modeling Report
APPENC ber Number Scaling Numerical Plume Modeling Report
IA-E12PG00037 Final Report
Page 132
-------
EPA/600/R-16/152
Appendix A
APPENDIX A - Experiment Logs for subsurface injection experiments with Alaskan North Slope crude, IFO-120, South Louisiana
crude and Gas Condensate.
Table Al. Log sheet for subsurface injection experiments with ANS and Corexit 9500 for cold water temperatures.
Experiment ID
Date
Oil Type
Oil
Amount
(g)
Dispersant-
to-Oil Ratio
DOR
Oil
Temp.
°C
Seawater
Temp.
°C
Air
Temp.
°C
Salinity
ppt
Oil
Injection
Time
seconds
Injection
Pressure
psi
Weather
Conditions
SubANS-1
22-May-14
ANS
208.0
0
80
11.4
11.0
28.1
4
40
Calm
SubANS-2R
02-Dec-14
ANS
290.5
1:200
80
6.4
0.0
27.7
5
40
clear/windy
SubANS-3
23-May-14
ANS
284.5
1:100
80
11.2
14.0
28.0
5
40
Clear, calm
SubANS-4R
03-Dec-14
ANS
287.2
1:20
80
6.8
0.5
28.6
5
40
breezy
SubANS-5
26-May-14
ANS
279.3
0
80
8.4
9.0
28.6
5
40
Overcast
SubANS-6R
02-Dec-14
ANS
335.0
1:200
80
6.1
-2.0
27.7
5
40
light wind
SubANS-7
30-May-14
ANS
276.3
1:100
80
8.5
17.0
29.1
5
40
Clear, calm
SubANS-8R
03-Dec-14
ANS
297.2
1:20
80
7.0
4.5
28.8
5
40
rainy/breezy
SubANS-9
02-Jun-14
ANS
281.4
0
80
9.7
17
29.1
5
40
Clear, calm
SubANS-lOR
17-Dec-14
ANS
344.5
1:200
80
5.4
2.0
26.5
5
40
calm
SubANS-11
06-Jun-14
ANS
276.8
1:100
80
10.7
14.0
29.0
5
40
Overcast
SubANS-12R
03-Dec-14
ANS
295.7
1:20
80
7.3
6.5
28.8
5
40
rain
Averages
288.0
8.2
7.8
28.3
IA-E12PG00037 Final Report Appendices
Page 133
-------
EPA/600/R-16/152
Appendix A
Table A2. Log sheet for subsurface injection experiments with IFO-120 and Corexit 9500 for cold water temperatures.
Experiment ID
Date
Oil Type
Oil
Amount
(g)
Dispersant-
to-Oil Ratio
DOR
Oil
Temp.
°C
Seawater
Temp.
°C
Air
Temp.
°C
Salinity
ppt
Oil
Injection
Time
seconds
Injection
Pressure
psi
Weather
Conditions
SublFO-lR
17-Dec-14
IFO-120
253.6
0
80
4.9
1.0
25.9
10
60
calm
SublFO-2R
04-Dec-14
IFO-120
208.2
1:200
80
6.7
6.0
28.1
7
60
windy
SublFO-3
20-Jun-14
IFO-120
213.9
1:100
80
13.2
12.4
27.9
7
62
Partly cloudy, windy
SublFO-4R
05-Dec-14
IFO-120
219.6
1:20
80
5.6
-3.5
29.3
10
30
sunny calm
SublFO-5
17-Jun-14
IFO-120
275.1
0
80
12.0
20.0
28.7
7
62
Clear, calm
SublFO-6R
04-Dec-14
IFO-120
215.6
1:200
80
6.6
5
27.4
8
60
windy
SublFO-7R
10-Dec-14
IFO-120
239.3
1:100
80
7.5
9.0
28.4
8
60
heavy rain/wind
SublFO-8R
05-Dec-14
IFO-120
243.3
1:20
80
5.4
-3.5
28.4
10
60
sunny breezy
SublFO-9
17-Jun-14
IFO-120
359.6
0
80
12.7
20.3
29.0
7
62
Clear, sunny
SublFO-lOR
04-Dec-14
IFO-120
221.7
1:200
80
6.6
4.5
27.2
8
60
windy
SublFO-llR
17-Dec-14
IFO-120
N/A
1:100
80
4.9
0.5
25.8
10
60
calm
SublFO-12R
10-Dec-14
IFO-120
204.8
1:20
80
6.8
7.0
29.5
9
60
rain/wind
Averages
241.3
7.7
6.6
28.0
IA-E12PG00037 Final Report Appendices
Page 134
-------
EPA/600/R-16/152
Appendix A
Table A3. Log sheet for subsurface injection experiments with ANS and Corexit 9500 for warm water temperatures.
Experiment ID
Date
Oil Type
Oil
Amount
(g)
Dispersant-
to-Oil Ratio
DOR
Oil
Temp.
°C
Seawater
Temp.
°C
Air
Temp.
°C
Salinity
ppt
Oil
Injection
Time
seconds
Injection
Pressure
psi
Weather
Conditions
SubANS-13
05-Sep-14
ANS
303.7
0
80
17.7
19.0
29.4
5
40
sunny
SubANS-14
08-Sep-14
ANS
295.2
1:200
80
16.0
18.0
29.8
5
40
sunny
SubANS-15R
10-Sep-14
ANS
304.3
1:100
80
13.8
16.5
30.0
5
40
cloudy, calm
SubANS-16
10-Sep-14
ANS
291.9
1:20
80
14.7
19.0
30.0
5
40
partly sunny
SubANS-17
05-Sep-14
ANS
299.6
0
80
18.1
23.5
29.5
5
40
sunny, calm
SubANS-18
08-Sep-14
ANS
297.7
1:200
80
16.2
19.0
29.7
5
40
sunny
SubANS-19
09-Sep-14
ANS
283.4
1:100
80
15.3
18.5
29.9
5
40
sunny, calm
SubANS-20
ll-Sep-14
ANS
289.6
1:20
80
14.1
16.0
30.0
5
40
overcast, calm
SubANS-21
08-Sep-14
ANS
297.1
0
80
15.1
14.5
29.8
5
40
sunny
SubANS-22
09-Sep-14
ANS
281.8
1:200
80
14.2
14.0
30.0
5
40
cloudy, calm
SubANS-23
10-Sep-14
ANS
284.4
1:100
80
13.4
16.0
30.1
5
40
partly cloudy
SubANS-24
ll-Sep-14
ANS
285.8
1:20
80
13.6
17.5
30.0
5
40
sunny, calm
Average
292.9
15.2
17.6
29.9
IA-E12PG00037 Final Report Appendices
Page 135
-------
EPA/600/R-16/152
Appendix A
Table A4. Log sheet for subsurface injection experiments with IFO-120 and Corexit 9500 for warm water temperatures.
Experiment ID
Date
Oil Type
Oil
Amount
(g)
Dispersant-
to-Oil Ratio
DOR
Oil
Temp.
°C
Seawater
Temp.
°C
Air
Temp.
°C
Salinity
ppt
Oil
Injection
Time
seconds
Injection
Pressure
psi
Weather
Conditions
SublFO-13
12-Sep-14
IFO-120
256.8
0
80
14.9
18.5
29.9
7
60
Sunny, breezy
SublFO-14
15-Sep-14
IFO-120
279.0
1:200
80
13.5
14.0
30.0
8
60
Sunny, breezy
SublFO-15
16-Sep-14
IFO-120
336.2
1:100
80
14.0
12.0
30.0
8
60
sunny, calm
SublFO-16
17-Sep-14
IFO-120
315.9
1:20
80
14.7
14.0
29.9
7
60
Clear, calm
SublFO-17
12-Sep-14
IFO-120
293.3
0
80
14.7
20.0
30.0
8
60
sunny, windy
SublFO-18
15-Sep-14
IFO-120
331.8
1:200
80
13.8
16.5
30.0
8
60
cloudy, breezy
SublFO-19
16-Sep-14
IFO-120
353.8
1:100
80
14.7
14.0
30.0
7
60
overcast, calm
SublFO-20
17-Sep-14
IFO-120
345.6
1:20
80
15.2
18.0
29.9
7
60
Sunny, breezy
SublFO-21
12-Sep-14
IFO-120
303.6
0
80
15.2
20.0
30.0
8
60
sunny, windy
SublFO-22
15-Sep-14
IFO-120
363.3
1:200
80
14.0
16.5
30.0
8
60
cloudy
SublFO-23
16-Sep-14
IFO-120
352.6
1:100
80
14.7
16.5
30.0
7
60
overcast, calm
SublFO-24
17-Sep-14
IFO-120
380.0
1:20
80
16.0
20.0
30.0
7
60
sunny, calm
Average
326.0
14.6
16.7
30.0
IA-E12PG00037 Final Report Appendices
Page 136
-------
EPA/600/R-16/152
Appendix A
Table A5. Log sheet for subsurface injection experiments with Gas Condensate and Corexit 9500 for warm water temperatures.
Experiment ID
Date
Oil Type
Oil
Amount
(g)
Dispersant-
to-Oil Ratio
DOR
Oil
Temp.
°C
Seawater
Temp.
°C
Air
Temp.
°C
Salinity
ppt
Oil
Injection
Time
seconds
Injection
Pressure
psi
Weather
Conditions
SubCND-01
15-Jun-15
Condensate
169.3
None
ambient
10.0
20.5
27.4
4
40psi
Clear, calm
SubCND-02R
17-Jun-15
Condensate
299.8
None
ambient
11.1
17.2
27.8
6
40psi
Windy, partly sunn^
SubCND-03
16-Jun-15
Condensate
328.6
None
ambient
10.0
14.5
28.5
6
40psi
Overcast
SubCND-04
16-Jun-15
Condensate
308.9
1:20
ambient
10.8
17.0
28.4
6
40psi
Overcast, breezy
SubCND-05
17-Jun-15
Condensate
308.7
1:20
ambient
11.9
16.5
26.7
6
40psi
Windy
SubCND-06
17-Jun-15
Condensate
301.8
1:20
ambient
11.6
17.0
27.2
6
40psi
Windy, partly sunn^
SubCND-07
15-Jun-15
None
-
None
-
10.0
20.5
27.4
5
40psi
Clear, calm
Average
288.4
11.5
18.4
27.5
IA-E12PG00037 Final Report Appendices
Page 137
-------
EPA/600/R-16/152
Appendix A
Table A6. Log sheet for subsurface injection experiments with ANS and Finasol OSR 52 for warm water temperatures.
Experiment ID
Date
Oil Type
Oil
Amount
(g)
Dispersant-
to-Oil Ratio
DOR
Oil
Temp.
°C
Seawater
Temp.
°C
Air
Temp.
°C
Salinity
ppt
Oil
Injection
Time
seconds
Injection
Pressure
psi
Weather
Conditions
SubFIN-01
09-Jul-15
ANS
275.6
1
:20
80
16.6
17.5
26.9
5
40psi
Sunny
SubFIN-02
09-Jul-15
ANS
279.7
1
:20
80
17.4
23.5
27.2
5
40psi
Sunny
SubFIN-03
17-J u 1-15
ANS
268.4
1
:20
80
16.8
24.0
28.3
5
40psi
Sunny, breezy
SubFIN-04
13-J u 1-15
ANS
251.4
1
100
80
17.0
24.0
27.8
5
40psi
Sunny, calm
SubFIN-05
13-J u 1-15
ANS
261.7
1
100
80
18.6
27.5
27.6
5
40psi
Sunny, calm
SubFIN-06
13-J u 1-15
ANS
258.0
1
100
80
20.8
30.0
27.5
5
40psi
Sunny, calm
SubFIN-07
15-J u 1-15
ANS
283.0
1
200
80
14.2
19.0
28.7
5
40psi
Foggy
SubFIN-08
15-J u 1-15
ANS
264.8
1
200
80
17.5
23.5
28.0
5
40psi
Sunny
SubFIN-09
17-J u 1-15
ANS
267.9
1
200
80
17.0
22.0
28.0
5
40psi
Sunny
Average
267.8
17.3
23.4
27.8
IA-E12PG00037 Final Report Appendices
Page 138
-------
EPA/600/R-16/152
Appendix A
Table A7. Log sheet for subsurface injection experiments with IFO-120 and Finasol OSR 52 for warm water temperatures.
Experiment ID
Date
Oil Type
Oil
Amount
(g)
Dispersant-
to-Oil Ratio
DOR
Oil
Temp.
°C
Seawater
Temp.
°C
Air
Temp.
°C
Salinity
ppt
Oil
Injection
Time
seconds
Injection
Pressure
psi
Weather
Conditions
SubFIN-10
30-Jul-15
IFO-120
131.6
1:200
80
17.5
15.0
28.0
6
60psi
Light fog
SubFIN-11
30-Jul-15
IFO-120
264.9
1:100
80
18.7
26.2
6
60psi
Sunny, calm
SubFIN-12
30-Jul-15
IFO-120
297.5
1:20
80
20.3
27.5
25.9
6
60psi
Sunny
Average
231.3
18.8
21.3
26.7
IA-E12PG00037 Final Report Appendices
Page 139
-------
EPA/600/R-16/152
Appendix A
Table A8. Log sheet for subsurface injection experiments with SLC and Corexit 9500 for warm water temperatures.
Experiment ID
Date
Oil Type
Oil
Amount
(g)
Dispersant-
to-Oil Ratio
DOR
Oil
Temp.
°C
Seawater
Temp.
°C
Air
Temp.
°C
Salinity
ppt
Oil
Injection
Time
seconds
Injection
Pressure
psi
Weather
Conditions
SubSLC-01
23-J u 1-15
MC252
317.3
0
80
16.6
18.0
27.7
6
40psi
Calm, overcast
SubSLC-02
24-Jul-15
MC252
315.2
1:200
80
17.0
19.5
28.1
6
40psi
Sunny, calm
SubSLC-03
27-J u 1-15
MC252
317.5
1:100
80
16.9
17.5
28.5
6
40psi
Light rain
SubSLC-04
28-Jul-15
MC252
317.1
1:20
80
17.4
18.5
27.6
6
40psi
Rain
SubSLC-05
23-J u 1-15
MC252
322.0
0
80
17.0
24.5
27.8
6
40psi
Sunny, calm
SubSLC-06
24-Jul-15
MC252
319.5
1:200
80
17.1
23.0
28.1
6
40psi
Sunny
SubSLC-07
27-J u 1-15
MC252
322.9
1:100
80
17.0
18.5
28.5
6
40psi
Rain
SubSLC-08
28-Jul-15
MC252
318.0
1:20
80
17.8
18.5
27.2
6
40psi
Drizzle, breezy
SubSLC-09
23-J u 1-15
MC252
329.4
0
80
17.2
24.5
27.9
6
40psi
Sunny
SubSLC-10
24-Jul-15
MC252
325.2
1:200
80
17.1
23.5
28.4
6
40psi
Overcast
SubSLC-11
27-J u 1-15
MC252
330.6
1:100
80
17.0
18.5
28.0
6
40psi
Rain
SubSLC-12
28-Jul-15
MC252
321.1
1:20
80
18.1
21.0
27.3
6
40psi
Windy, drizzle
Average
321.3
17.2
20.5
27.9
IA-E12PG00037 Final Report Appendices
Page 140
-------
EPA/600/R-16/152
Appendix B
APPENDIX B - Analytical TPH and BTEX analytical chemistry values for each experiment time
point.
Table Bl. Total Petroleum Hydrocarbon (TPH) and Benzene-Toluene-Ethylbenzene-Xylene
(BTEX) values for subsurface injection experiments of ANS with Corexit 9500 for warm water
experiments. TPH method detection limit is < 1 ppm, represented by 0 in top table. Effluent
samples are represented with # E in time column.
Time (min)
TPH (ppm) DOR
= 0
TPH (ppm) DOR =
1:200
TPH (ppm) DOR =
1:100
TPH (ppm) DOR =
1:20
SUBANS13 SUBANS17
SUBANS21
SUBANS14
SUBANS18
SUBANS22
SUBANS15R
SUBANS19
SUBANS23
SUBANS16 SUBANS20
SUBANS24
0
0
0
0
0
1
0
Data missing
0
0
0
0
1
2
37
0
127
72
84
67
32
46
34
8
41
2.5
174
253
359
237
132
128
159
65
72
60
113
3
57
149
60
98
177
97
103
159
82
77
44
3.5
24
35
60
11
17
40
25
31
27
62
49
4
2
8
146
4
2
15
1
5
7
5
13
4.5
0
1
14
1
2
3
0
2
20
1
3
5
0
0
34
0
0
1
0
0
4
0
1
5.5
0
0
36
0
0
0
0
0
1
0
0
6
0
0
7
0
0
0
0
0
0
1
2
6.5
0
0
3
0
0
1
1
0
0
1
0
7
0
0
4
0
0
0
1
0
1
1
0
8
0
0
3
0
0
0
0
0
0
0
0
10
0
0
1
0
0
0
0
0
0
1
0
12
0
0
1
0
0
0
0
0
1
0
0
14
0
0
1
0
0
0
0
0
0
0
0
2 E
0
0
0
0
0
0
0
0
0
0
0
4 E
0
0
0
0
0
0
0
0
0
0
0
6 E
0
0
2
0
0
0
0
0
0
0
0
8 E
7
37
12
19
6
10
15
16
32
18
17
10 E
12
13
5
11
23
13
21
25
19
18
29
12 E
8
5
4
8
18
10
5
10
6
7
5
14 E
3
3
3
4
11
4
2
3
3
3
2
Time (min)
BTEX(ppb) DOR
= 0
BTEX(ppb) DOR =
1:200
BTEX(ppb) DOR =
1:100
BTEX(ppb) DOR =
1:20
SUBANS13 SUBANS17
SUBANS21
SUBANS14
SUBANS18
SUBANS22
SUBANS15R
SUBANS19
SUBANS23
SUBANS16
SUBANS20
SUBANS24
0
0
0
0
2
3
1
Data missing
0
0
2
0
2
2
0
0
744
106
16
312
16
124
88
0
13
3
161
1333
920
1218
2063
1447
791
1161
1529
1445
636
4
14
248
1883
94
65
346
45
127
140
262
209
5
2
15
514
10
15
21
7
14
72
9
22
6
6
6
317
8
7
5
6
4
15
19
23
7
2
2
102
6
6
4
8
3
7
12
3
8
4
1
92
5
6
3
3
2
4
9
3
10
1
1
59
4
5
2
3
2
4
11
2
12
3
4
39
3
5
2
3
2
4
4
2
14
3
1
31
3
4
2
3
1
4
3
2
6 E
0
0
71
1
1
3
2
0
5
0
1
8 E
37
107
195
255
145
178
331
312
517
408
393
10 E
37
50
122
193
278
349
379
322
232
267
298
12 E
27
77
88
152
232
168
74
163
107
108
76
14 E
75
10
59
87
109
62
32
44
48
49
27
IA-E12PG00037 Final Report Appendices
Page 141
-------
EPA/600/R-16/152
Appendix B
Table B2. Total Petroleum Hydrocarbon (TPH) and Benzene-Toluene- Ethylbenzene -Xylene
(BTEX) values for subsurface injection experiments of ANS with Corexit 9500 for cold water
experiments. TPH method detection limit is < 1 ppm, represented by 0 in top table. Effluent
samples are represented with # E in time column.
Time (min)
TPH (ppm) DOR
= 0
TPH (ppm) DOR =
1:200
TPH (ppm) DOR =
:100
TPH (ppm) DOR =
1:20
SUBANS1
SUBANS5
SUBANS9
SUBANS2R
SUBANS6R
SUBANS10R
SUBANS3
SUBANS7
SUBANS11
SUBANS4R SUBANS8R
SUBANS12R
0
0
0
0
0
0
0
1
0
0
0
1
2
214
17
132
70
109
201
5
189
106
2
25
301
2.5
66
0
48
150
192
110
269
87
88
79
160
114
3
7
0
60
131
83
3
18
18
37
71
46
42
3.5
0
4
35
27
5
34
2
7
44
5
29
4
0
1
3
10
4
1
2
11
6
34
4.5
0
0
2
13
2
0
0
15
3
10
5
0
0
2
4
1
0
0
0
4
3
4
5.5
0
0
0
2
1
0
0
0
1
2
1
6
0
0
1
0
0
0
0
0
5
1
6.5
0
0
0
0
0
1
0
0
0
0
3
1
7
0
0
0
0
0
1
0
0
0
0
1
0
8
0
0
0
0
0
1
0
0
0
0
1
0
10
0
0
0
0
0
1
0
0
0
0
0
0
12
0
0
0
0
0
0
0
0
0
0
0
0
14
0
0
0
0
0
1
0
0
0
0
0
0
2 E
0
0
0
0
0
1
0
0
0
0
0
0
4 E
0
0
0
0
0
1
0
0
0
0
0
0
6 E
0
0
0
0
0
20
0
0
1
0
0
1
8 E
7
8
4
16
19
32
13
4
12
4
4
12
10 E
6
0
4
5
34
8
5
9
11
23
13
23
12 E
6
1
5
7
17
5
10
6
4
12
10
12
14 E
2
1
2
13
4
5
5
1
2
5
14
6
Time (min)
BTEX(ppb) DOR
= 0
BTEX(ppb) DOR =
1:200
BTEX(ppb) DOR =
1:100
BTEX(ppb) DOR =
1:20
SUBANS1
SUBANS5
SUBANS9
SUBANS2R
SUBANS6R
SUBANS10R
SUBANS3
SUBANS7
SUBANS11
SUBANS4R
SUBANS8R
SUBANS12R
0
1
3
0
1
0
1
1
1
0
1
5
2
940
1303
2604
0
638
774
2488
1937
19
18
52
3
123
341
1437
1599
1143
274
350
566
1034
1046
470
570
4
5
11
36
109
133
200
52
21
69
163
63
469
5
3
5
13
52
71
18
11
7
13
145
66
72
6
3
4
8
6
16
7
7
6
7
11
100
21
7
2
3
7
3
6
14
4
5
6
8
16
12
8
2
3
7
3
4
8
4
4
5
5
8
7
10
3
5
3
3
3
3
3
4
4
3
5
12
3
5
2
2
2
4
3
3
2
3
5
14
2
5
2
2
3
2
3
3
2
2
4
6 E
no sample
no sample
no sample
0
0
898
no sample
no sample
no sample
5
17
73
8 E
no sample
no sample
no sample
295
348
418
no sample
no sample
no sample
150
121
232
10 E
no sample
no sample
no sample
338
417
185
no sample
no sample
no sample
360
216
336
12 E
no sample
no sample
no sample
170
215
128
no sample
no sample
no sample
205
186
163
14 E
no sample
no sample
no sample
95
no sample
78
no sample
no sample
no sample
78
197
58
IA-E12PG00037 Final Report Appendices
Page 142
-------
EPA/600/R-16/152
Appendix B
Table B3. Total Petroleum Hydrocarbon (TPH) and Benzene-Toluene- Ethylbenzene -Xylene (BTEX)
values for subsurface injection experiments of ANS with Finasol OSR 52 for warm water experiments.
TPH method detection limit is < 1 ppm, represented by 0 in top table. Effluent samples are
represented with # E in time column.
Time (min)
TPH (ppm) DOR =
SUBFIN-07 SUBFIN-08
1:200
SUBFIN-09
TPH (ppm) DOR =
SUBFIN-04 SUBFIN-05
1:100
SUBFIN-06
TPH (ppm) DOR =
SUBFIN-01 SUBFIN-02
1:20
SUBFIN-03
0
0
1
16
7
1
0
0
1
0
2
68
42
13
84
9
24
84
38
17
2.5
147
164
111
214
123
66
113
150
206
3
78
28
106
116
109
67
97
122
197
3.5
35
24
59
113
99
48
22
113
109
4
22
2
5
6
16
26
8
17
6
4.5
4
2
13
15
17
9
9
18
31
5
2
0
4
7
13
6
3
10
18
5.5
0
0
4
4
8
4
1
4
12
6
2
0
1
2
3
3
1
4
4
6.5
0
0
1
2
2
2
1
3
2
7
0
0
0
1
2
0
1
2
1
8
1
0
0
1
2
2
1
2
1
10
0
0
0
2
2
0
1
1
0
12
0
0
0
0
1
1
1
1
1
14
0
0
0
0
1
0
1
0
2 E
0
0
0
0
1
0
1
2
0
4 E
0
0
0
1
0
0
1
1
0
6 E
0
0
0
2
1
1
1
1
1
8 E
18
25
15
22
17
24
20
12
21
10 E
23
9
11
20
23
18
21
19
16
12 E
6
5
6
6
10
6
7
13
11
14 E
2
2
4
3
3
4
3
5
5
Time (min)
BTEX(ppb) DOR =
SUBFIN-07 SUBFIN-08
1:200
SUBFIN-09
BTEX(ppb) DOR =
SUBFIN-04 SUBFIN-05
1:100
SUBFIN-06
BTEX(ppb) DOR =
SUBFIN-01 SUBFIN-02
1:20
SUBFIN-03
0
0
1
0
0
2
1
1
6
2
2
81
187
0
28
1
1
0
28
1
3
894
908
1056
1063
766
535
585
1203
1937
4
467
238
304
527
337
305
439
530
589
5
201
45
36
144
224
78
95
143
184
6
11
13
21
37
50
22
24
47
59
7
16
10
15
16
17
10
9
16
19
8
9
8
8
8
9
6
4
8
12
10
6
6
5
1
6
4
3
5
8
12
3
5
4
3
5
3
2
4
5
14
4
5
2
5
3
2
2
3
5
6 E
0
3
0
0
0
0
0
0
1
8 E
390
444
287
464
312
413
380
229
287
10 E
385
192
253
331
373
308
365
327
261
12 E
119
123
132
105
148
129
115
183
188
14 E
58
66
68
51
51
65
52
68
87
IA-E12PG00037 Final Report Appendices
Page 143
-------
EPA/600/R-16/152
Appendix B
Table B4. Total Petroleum Hydrocarbon (TPH) and Benzene-Toluene- Ethylbenzene -Xylene (BTEX)
values for subsurface injection experiments of IFO 120 with Corexit 9500 for warm water experiments.
TPH method detection limit is < 1 ppm, represented by 0 in top table. Effluent samples are
represented with # E in time column.
Time (min)
TPH (ppm) DOR
= 0
TPH (ppm) DOR =
1:200
TPH (ppm) DOR =
1:100
TPH (ppm) DOR =
1:20
SUBIF013
SUBIF017
SUBIF021
SUBIF014
SUBIF018
SUBIF022
SUBIF015
SUBIF019
SUBIF023
SUBIF016
SUBIFO20
SUBIF024
0
6
4
0
0
0
0
3
2
1
0
0
1
2
16
69
151
87
20
108
187
162
202
27
159
103
2.5
56
155
117
87
55
113
81
414
344
129
171
127
3
35
43
32
14
16
411
58
30
101
110
4
107
3.5
31
9
16
10
11
8
6
1
4
45
1
4
4
4
8
2
4
18
5
2
0
1
15
1
2
4.5
3
3
0
2
10
5
8
1
0
4
0
0
5
8
4
1
2
5
0
1
1
0
2
1
1
5.5
4
7
0
4
5
0
1
0
0
0
0
0
6
5
2
0
2
2
0
1
0
1
1
1
1
6.5
0
1
1
2
1
0
1
1
0
0
0
0
7
0
2
1
2
0
0
1
1
0
1
0
0
8
0
2
1
1
0
0
0
0
1
0
0
0
10
0
1
0
1
0
0
1
0
0
0
0
0
12
0
1
1
1
0
0
0
0
0
0
1
0
14
0
0
1
0
0
0
0
0
1
0
0
1
2 E
0
0
0
1
0
0
0
0
0
0
0
0
4 E
0
0
0
0
lost
0
0
0
0
0
0
0
6 E
0
0
0
0
0
0
0
1
0
0
0
0
8 E
3
15
9
4
2
13
3
17
15
2
40
51
10 E
5
39
21
5
13
22
10
15
46
13
25
31
12 E
2
15
4
6
8
7
7
5
14
30
7
7
14 E
1
4
2
1
3
2
2
3
3
1
3
4
BTEX(ppb) DOR
= 0
BTEX(ppb) DOR =
1:200
BTEX(ppb) DOR =
1:100
BTEX(ppb) DOR =
1:20
Time (min)
SUBIF013 SUBIF017
SUBIF021
SUBIF014
SUBIF018
SUBIF022
SUBIF015
SUBIF019
SUBIF023
SUBIF016
SUBIFO20
SUBIF024
0
0
0
3
0
0
2
0
1
0
2
2
2
11
9
68
98
0
138
166
6
89
4
123
15
3
40
111
42
40
32
75
144
102
277
64
22
163
4
5
37
1
3
26
5
3
2
4
5
1
5
5
1
1
1
8
9
1
0
2
2
0
1
2
6
1
1
1
3
1
0
2
2
0
1
2
7
0
0
1
3
1
0
2
0
1
2
8
0
0
1
0
1
0
2
0
1
2
10
0
0
1
0
1
0
2
2
0
1
2
12
1
0
1
0
1
0
2
0
1
2
14
0
0
1
0
1
0
2
0
1
2
6 E
0
0
1
0
1
1
0
17
1
0
1
2
8 E
8
5
20
8
7
31
11
28
40
20
62
69
10 E
7
18
27
10
27
32
19
16
33
51
25
36
12 E
2
6
8
4
11
11
10
11
19
11
8
15
14 E
0
1
1
1
6
4
2
4
6
3
5
7
IA-E12PG00037 Final Report Appendices
Page 144
-------
EPA/600/R-16/152
Appendix B
Table B5. Total Petroleum Hydrocarbon (TPH) and Benzene-Toluene- Ethylbenzene -Xylene (BTEX)
values for subsurface injection experiments of IFO 120 with Corexit 9500 for cold water experiments.
TPH method detection limit is < 1 ppm, represented by 0 in top table. Effluent samples are
represented with # E in time column.
Time (min)
TPH (ppm) DOR
= 0
TPH (ppm)
DOR =
1:200
TPH (ppm) DOR =
1:100
TPH (ppm) DOR =
1:20
SUBIF01R
SUBIF05
SUBIF09
SUBIF02R SUBIF06R
SUBIFO10R
SUBIF03
SUBIF07R
SUBI F011R
SUBIF04R SUBIF08R
SUBIF012R
0
0
0
2
1
2
1
5
0
1
1
0
1
2
111
118
210
30
12
1
110
1
36
7
2
28
2.5
50
165
55
15
14
30
135
28
29
11
107
69
3
3
32
7
4
1
35
10
19
2
2
81
39
3.5
3
2
1
1
2
11
3
5
1
6
36
10
4
1
1
1
1
1
2
1
5
1
4
2
2
4.5
1
1
1
1
0
1
1
10
1
2
2
2
5
1
1
1
0
1
1
1
7
1
1
2
1
5.5
1
0
1
0
0
1
1
4
0
1
1
1
6
0
1
0
0
0
1
3
1
1
1
1
6.5
1
0
1
1
0
0
1
3
0
2
1
0
7
1
0
2
1
0
1
1
4
1
1
1
1
8
1
0
1
0
0
0
1
1
1
1
1
0
10
1
0
1
1
0
1
1
1
1
0
1
1
12
0
1
1
0
1
0
1
1
1
1
1
1
14
1
1
1
1
0
0
1
1
0
1
1
0
2 E
0
0
0
0
0
1
0
0
1
0
0
0
4 E
0
0
0
0
1
1
1
1
1
0
1
1
6 E
2
0
0
1
1
1
1
1
0
0
1
0
8 E
4
4
13
0
2
0
2
0
2
5
2
1
10 E
1
6
15
1
3
3
5
2
1
2
10
9
12 E
1
5
5
1
1
1
2
2
1
2
7
4
14 E
0
2
3
0
1
1
1
2
1
1
1
2
BTEX(ppb) DOR
= 0
BTEX(ppb) DOR =
1:200
BTEX(ppb) DOR =
1:100
BTEX(ppb) DOR =
1:20
Time (min)
SUBIF01R SUBIF05
SUBIF09
SUBIF02R
SUBIF06R
SUBI FO10R
SUBIF03
SUBIF07R
SUBI F011R
SUBIF04R SUBIF08R
SUBIF012R
0
1 0
0
0
0
1
0
0
0
0
2
0
2
3 73
277
20
30
1
22
1
7
0
225
7
3
6 53
10
1
32
20
25
4
4
164
94
4
1 1
1
0
0
2
4
0
5
14
3
5
1 0
1
0
0
2
0
15
0
0
4
1
6
1 1
1
0
0
2
0
2
0
0
2
0
7
1 1
0
0
0
2
0
8
0
0
2
0
8
1 1
0
0
0
2
0
1
0
0
2
0
10
1 0
0
0
0
2
0
1
0
0
2
0
12
1 0
0
0
0
2
0
0
0
0
2
0
14
1 0
0
0
0
2
0
0
0
0
2
0
6 E
14 no sample
no sample
0
0
1
no sample
0
0
0
2
0
8 E
5 no sample
no sample
1
4
1
no sample
1
5
5
8
1
10 E
1 no sample
no sample
2
1
5
no sample
3
1
2
27
16
12 E
1 no sample
no sample
1
0
2
no sample
6
0
1
11
6
14 E
1 no sample
no sample
0
0
2
no sample
4
0
1
3
0
IA-E12PG00037 Final Report Appendices
Page 145
-------
EPA/600/R-16/152
Appendix B
Table B6. Total Petroleum Hydrocarbon (TPH) and Benzene-Toluene- Ethylbenzene -Xylene (BTEX)
values for subsurface injection experiments of IFO 120 with Finasol OSR 52 for warm water
experiments. TPH method detection limit is < 1 ppm, represented by 0 in top table. Effluent samples
are represented with # E in time column.
Time (min)
TPH (ppm) DOR =1:200
TPH (ppm) DOR =1:100
TPH (ppm) DOR =1:20
SUBFIN-10
SUBFIN-11
SUBFIN-12
0
1
2
0
2
45
55
3
2.5
14
170
181
3
10
14
117
3.5
3
19
46
4
2
4
9
4.5
2
3
2
5
1
3
3
5.5
0
2
2
6
2
1
2
6.5
1
2
2
7
1
1
0
8
1
2
1
10
1
2
0
12
1
0
0
14
1
0
1
2 E
1
2
2
4 E
1
1
1
6 E
0
0
0
8 E
1
8
11
10 E
0
5
28
12 E
0
3
9
14 E
0
2
4
Time (min)
BTEX(ppb) DOR =1:200
BTEX(ppb) DOR =1:100
BTEX(ppb) DOR =1:20
SUBFIN-10
SUBFIN-11
SUBFIN-12
0
0
1
0
2
17
22
10
3
24
59
180
4
1
11
6
5
1
2
1
6
0
1
1
7
0
1
1
8
0
1
0
10
0
1
0
12
0
1
0
14
0
1
0
6 E
0
1
0
8 E
3
20
46
10 E
4
9
53
12 E
1
4
16
14 E
0
2
8
IA-E12PG00037 Final Report Appendices
Page 146
-------
EPA/600/R-16/152
Appendix B
Table B7. Total Petroleum Hydrocarbon (TPH) and Benzene-Toluene- Ethylbenzene -Xylene (BTEX)
values for subsurface injection experiments of SLC with Corexit 9500 for warm water experiments.
TPH method detection limit is < 1 ppm, represented by 0 in top table. Effluent samples are
represented with # E in time column.
Time (min)
TPH (ppm) DOR
SUBSLC-01 SUBSLC-05
= 0
SUBSLC-09
TPH (ppm) DOR =
SUBSLC-02 SUBSLC-06
1:200
SUBSLC-10
TPH (ppm) DOR =
SUBSLC-03 SUBSLC-07
1:100
SUBSLC-11
TPH (ppm) DOR =
SUBSLC-04 SUBSLC-08
1:20
SUBSLC-12
0
1
3
2
3
0
1
0
0
2
2
3
1
2
48
13
100
6
3
4
8
26
11
2
4
1
2.5
156
140
130
130
110
195
262
119
171
22
56
32
3
105
120
106
75
29
111
164
100
60
120
130
131
3.5
52
109
104
87
20
38
122
45
22
85
93
115
4
48
111
40
45
28
8
30
16
8
69
55
42
4.5
9
29
39
56
10
4
13
7
6
57
23
18
5
2
10
22
17
9
3
3
4
13
28
12
9
5.5
4
3
10
14
4
2
3
2
6
10
6
4
6
3
3
3
7
3
2
3
7
6
3
2
6.5
2
3
4
3
0
1
2
4
3
2
2
7
1
0
4
3
2
2
2
2
2
2
2
2
8
1
1
2
3
0
1
2
2
1
3
2
1
10
1
1
2
3
0
1
2
2
1
1
12
1
0
1
0
1
0
1
2
1
1
14
0
0
2
3
2
1
1
2
1
1
2 E
2
1
1
0
1
0
0
1
2
1
1
4 E
2
2
1
2
0
1
0
1
2
1
1
6 E
1
1
1
0
1
0
1
2
1
2
8 E
13
12
11
16
12
14
4
5
7
8
11
17
10 E
11
9
5
10
7
12
13
17
17
12
13
17
12 E
10
12
8
11
10
7
9
9
12
15
11
13
14 E
8
5
6
9
4
2
7
4
6
9
7
6
Time (min)
BTEX(ppb) DOR
SUBSLC-01 SUBSLC-05
= 0
SUBSLC-09
BTEX(ppb) DOR =
SUBSLC-02 SUBSLC-06
1:200
SUBSLC-10
BTEX(ppb) DOR =
SUBSLC-03 SUBSLC-07
1:100
SUBSLC-11
BTEX(ppb) DOR =
SUBSLC-04 SUBSLC-08
1:20
SUBSLC-12
0
2
6
3
2
2
4
0
1
8
2
141
1
523
0
2
12
54
359
6
0
1
3
3
1082
1268
1161
1046
578
1794
2058
1518
590
2455
1038
708
4
316
1507
721
664
457
464
923
668
77
1128
494
890
5
58
74
548
380
317
87
124
91
152
118
110
65
6
17
59
44
199
129
23
60
40
136
34
42
35
7
17
24
31
49
32
14
25
14
55
17
20
16
8
10
19
21
28
15
8
21
9
15
7
9
10
10
8
13
15
9
7
7
7
6
19
6
8
8
12
6
12
13
6
6
6
6
4
0
4
5
7
14
6
12
11
5
6
5
4
4
5
3
5
6
6 E
0
0
0
0
1
0
6
0
1
0
5
1
8 E
337
335
212
252
396
344
140
249
224
371
337
220
10 E
263
371
174
240
199
323
265
444
394
6
392
197
12 E
207
296
276
211
335
190
210
225
264
371
258
343
14 E
136
130
185
184
113
64
162
110
147
136
146
222
IA-E12PG00037 Final Report Appendices
Page 147
-------
EPA/600/R-16/152
Appendix B
Table B8. Total Petroleum Hydrocarbon (TPH) and Benzene-Toluene- Ethylbenzene -Xylene (BTEX)
values for subsurface injection experiments of Gas Condensate with Corexit 9500 for warm water
experiments. TPH method detection limit is < 1 ppm, represented by 0 in top table. Effluent samples
are represented with # E in time column.
Time (min)
TPH (ppm) DOR
SUBCND-01 SUBCND-02R
= 0
SUBCND-03
TPH (ppm) DOR =
SUBCND-04 SUBCND-05
1:20
SUBCND-06
0
0
0
0
0
0
0
2
4
41
29
76
50
25
2.5
4
12
12
32
102
42
3
2
1
3
17
35
5
3.5
0
2
0
9
2
26
4
0
2
0
2
3
2
4.5
0
1
0
4
5
0
5
0
1
0
0
4
0
5.5
0
1
0
1
9
0
6
0
0
0
0
7
0
6.5
0
1
0
0
2
0
7
0
0
0
0
0
0
8
0
0
0
0
3
0
10
0
0
0
0
2
1
12
0
1
0
0
2
0
14
0
0
0
0
1
0
2 E
0
0
0
0
0
0
4 E
0
0
0
0
1
0
6 E
0
0
0
0
0
0
8 E
1
1
2
5
3
4
10 E
0
1
1
9
3
5
12 E
0
1
0
8
4
7
14 E
0
1
0
3
5
3
Time (min)
BTEX(ppb) DOR
SUBCND-01 SUBCND-02R
= 0
SUBCND-03
BTEX(ppb) DOR =
SUBCND-04 SUBCND-05
1:20
SUBCND-06
0
1
4
4
5
1
13
2
998
2326
4041
6345
4418
1298
3
706
186
772
3665
2618
1261
4
41
741
72
323
1104
532
5
19
28
36
46
1195
252
6
18
99
25
28
962
35
7
15
27
22
18
355
40
8
11
29
17
15
863
26
10
11
29
12
11
186
27
12
6
18
15
9
269
24
14
6
12
7
7
371
17
6 E
1
3
324
336
61
8
8 E
322
228
579
834
462
683
10 E
282
174
205
1393
553
853
12 E
90
135
86
1012
633
848
14 E
43
84
34
459
636
520
IA-E12PG00037 Final Report Appendices
Page 148
-------
EPA/600/R-16/152
Appendix C
APPENDIX C-Jet Release LISST Oil Droplet Size Distribution Histograms*
*Note that LISST histogram plots have constrained Y-axes, thus lines which extend slightly above
the top of the plot area represent values which were truncated.
IA-E12PG00037 Final Report Appendices
Page 149
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SUBANS-1 1387
25
20
15-
TPC = 108.8 fiL/L
VMD = 101.4 um
T = 2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max); SUBANS-5 1387
25
20
15
10
TPC = 48 uL/L
VMD = 146.5 fim
T = 2.2 min.
2.5-3
¦¦
¦lllllh.—¦!
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SUBANS-9 1387
TPC = 147.1 jiL/L
VMD = 83.5 |im
T = 2,3 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Figure CI. LISST DSD plot for AIMS, no dispersant, cold water experiments.
IA-E12PG00037 Final Report Appendices
Page 150
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubANS-2R 1174
25
20
I 15-
TPC = 105.6 fiL/L
VMD = 93,5 um
T = 2.8 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubANS-6R 1174
25
20
TPC = 119.3 (.iL/L
VMD = 86,7 um
T = 2.8 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubANS-10R 117.
25
20
0
1
E
m
o
c
o
0
u
1
0.
15
10
TPC = 260.5 nL/L
VMD = 318.2 um
T = 4.7 min.
Jl
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-489
Figure C2. LISST DSD plot for AIMS, DOR 1:200 (Corexit 9500), cold water experiments.
IA-E12PG00037 Final Report Appendices
Page 151
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SUBANS-3 1387
25
20
I 15-
TPC = 79,7 nL/L
VMD = 66.9 urn
T = 2.6 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max); SUBANS-7 1387
25
20
TPC = 113.1 nL/L
VMD = 82.5 um
T = 2.2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): subans-11 1387
25
TPC = 207.1 jiL/L
VMD = 87.5 um
T = 2.1 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
FigureC3. LISST DSD plot for AIMS, DOR 1:100 (Corexit 9500), cold water experiments.
IA-E12PG00037 Final Report Appendices
Page 152
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubANS-4R 1174
25
20
I 15-
o
c
o
o
®
0
1
10
TPC = 59.8 (iUL
VMD = 3.8 jim
T = 4.1 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (urn)
423-499
Particle Size Concentration (Curve TPC Max); SubANS-8R 1174
25
20
15
0
1
CL-
IO
TPC = 72.5 |iUL
VMD = 12.7 urn
T = 2 min.
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
423-499
Particle Size Concentration (Curve TPC Max): SubANS-12R 117.
25
20
0
1
E
m
o
c
o
0
u
1
0.
15
10
TPC = 104.8 jiUL
VilD = 9.4 jxm
T = 2.9 min.
1 A
2.5-3 8-9.5 3
8-9.5 30-35 113-133
LISST Particle Size Bins (fim)
423-499
Figure C4. LISST DSD plot for AIMS, DOR 1:20 (Corexit 9500), cold water experiments.
IA-E12PG00037 Final Report Appendices
Page 153
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubANS-13 1174
25
TPC = 116.7 jiL/L
VMD = 136 jim
T = 2.1 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max); SubANS-17 1174
25
20
15
TPC = 90.7 fiUL
VMD = 161.7 jim
T = 2.8 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
423-499
Particle Size Concentration (Curve TPC Max): SubANS-21 1174
25
20
0
1
E
m
o
c
o
u
0
1
0.
15
10
TPC = 121.8 jiL/L
VMD = 138.6 jim
T = 2.5 min.
—¦¦ill
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
423-499
Figure C5. LISST DSD plot for AIMS, DOR 0 (Corexit 9500), warm water experiments.
IA-E12PG00037 Final Report Appendices
Page 154
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubANS-14 1174
25
TPC = 172 jiUL
VMD = 97.5
T = 2 min.
2,5-3 8-9,5 30-35 113-133
LISST Particle Size Bins (umj
423-499
Particle Size Concentration (Curve TPC Max); SubANS-18 1174
25
20
TPC = 131.5 nL/L
VMD = 80.8 (im
T = 2.5 min.
2,5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
423-499
Particle Size Concentration (Curve TPC Max): SubANS-22 1174
25
TPC = 115.4 jiL/L
VMD = 79.5 p.m
T = 2.3 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
423-499
Figure C6. LISST DSD plot for AIMS, DOR 1:200 (Corexit 9500), warm water experiments.
IA-E12PG00037 Final Report Appendices
Page 155
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubANS-15R 117.
25
TPC = 111.7 ,uUL
VMD = 94,8 fim
T = 2.2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubANS-19 1174
25
20
15
TPC = 95.5 uL/L
VMD = 78.5 fim
T = 2.3 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubANS-23 1174
25
TPC = 97.4 jiL/L
VMD = 79.5 pm
T = 2.6 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (pm)
423-499
Figure C7. LISST DSD plot for AIMS, DOR 1:100 (Corexit 9500), warm water experiments.
IA-E12PG00037 Final Report Appendices
Page 156
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubANS-16 1174
TPC = 128.7 (iL/L
VMD = 9.8 jim
T = 3.4 min
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubANS-20 1174
25
20
15
TPC = 118.7 nL/L
VMD = 8.6 fim
T = 3.2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubANS-24 1174
25
TPC = 113.9 jiL/L
VMD = 9.7 um
T = 2.6 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (fim)
423-499
Figure C8. LISST DSD plot for AIMS, DOR 1:20 (Corexit 9500), warm water experiments.
IA-E12PG00037 Final Report Appendices
Page 157
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SublF0-1R 1174
25
20
15-
TPC = 30,8 |iL/L
VMD = 296.3 jim
T = 2.6 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SublFO-5 1387
25
20
TPC = 82.8 nL/L
VMD = 284.7 um
T = 2.1 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SublFO-9 1387
TPC = 124.8 jiL/L
VMD = 114.4 um
T = 2.8 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Figure C9. LISST DSD plot for IFO-120, no dispersant, cold water experiments.
IA-E12PG00037 Final Report Appendices
Page 158
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SublF0-2R 1174
25
20
15-
TPC = 12.6 fiL/L
VMD = 390.5 um
T = 2.5 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SublFQ-SR 1174
25
20
15
10
TPC = 27.9 §iUL
VlflD = 301.1 jim
T = 2 min.
>ll
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
423-499
Particle Size Concentration (Curve TPC Max): SublFO-10R 117*
25-
20
15-
10
TPC = 46.4 jiL/L
VMD = 430.4 jim
T = 3.2 min.
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
423-489
Figure C10. LISST DSD plot for IFO-120, DOR 1:200 (Corexit 9500), cold experiments.
IA-E12PG00037 Final Report Appendices
Page 159
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SUBIFO-3 1387
25
20
TPC = 46.1 fiL/L
VMD = 223 pm
T = 2.5 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (urn)
423-499
Particle Size Concentration (Curve TPC Max): SubiFO-7R 1174
25
20
15
10
TPC = 25.8 fiLIL
VMD = 344 fim
T = 2.5 min.
2.5-3
-T—
8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
M
423-499
Particle Size Concentration (Curve TPC Max): SublFO-11R 117*
25-
20
15
10
TPC = 194.5 jiL/L
VMD = 385.4 nm
T = 2.9 min.
-
I
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (pm)
423-489
Figure Cll. LISST DSD plot for IFO-120, DOR 1:100 (Corexit 9500), cold experiments.
IA-E12PG00037 Final Report Appendices
Page 160
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SublF0-4R 1174
25
20
I 15-
TPC = 25.3 fiL/L
VMD = 327.4 fim
T = 2.2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubiFO-8R 1174
25
20
15
0
1
CL-
IO
2.5-3
TPC = 164.9 nL/L
VMD = 280.2 um
T = 2.2 min.
¦'
iillill
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubiFG-12R 117*
25
20
0
1
E
m
o
c
o
0
J5
u
1
0.
15
10
TPC = 37.1 jiL/L
VMD = 178.3 jim
T = 3.1 min.
2.5-3
illlllll
lllll.
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-489
Figure C12. LISST DSD plot for IFO-120, DOR 1:20 (Corexit 9500), cold water experiments.
IA-E12PG00037 Final Report Appendices
Page 161
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SublFO-13 1174
25
20
15-
10
TPC = 28.5 fiL/L
VMD = 247.6 fim
T = 2.6 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max); SublFO-17 1174
25
20
15
10
TPC = 39.3 \xUL
VMD = 223.7 jim
T = 3 min.
i
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
.mill!
423-499
Particle Size Concentration (Curve TPC Max): SublFO-21 1174
25
20
15
10
TPC = 74.4 jiL/L
VMD = 279.7 Jim
T = 2.5 min.
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
423-489
Figure C13. LISST DSD plot for IFO-120, no dispersant, warm water experiments.
IA-E12PG00037 Final Report Appendices
Page 162
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SublFO-14 1174
25
20
I 15-
o
c
o
o
®
0
1
10
TPC = 39,9 fiL/L
VMD = 258 pm
T = 2.4 min.
2,5-3 8-9,5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max); SublFO-18 1174
25
20
15
0
1
Q.
10
2,5-3
TPC = 37,9 nUL
VMD = 340.4 )im
T = 3.2 min.
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
I
423-499
Particle Size Concentration (Curve TPC Max): SublFO-22 1174
25
20
0
1
E
m
o
c
o
0
J5
u
1
0.
15
10
TPC = 75.8 fiL/L
VMD = 234.5 um
T = 2,4 min.
2.5-3
.¦¦¦¦111
iill
8-9.5 30-35 113-133
LISST Particle Size Bins (pm)
423-499
Figure C14. LISST DSD plot for IFO-120, DOR 1:200 (Corexit 9500), warm experiments.
IA-E12PG00037 Final Report Appendices
Page 163
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SublFO-15 1174
25
TPC = 84,8 nL/L
VMD = 204.5 (im
T = 2.2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max); SublFO-19 1174
25
TPC = 161.5 nL/L
VMD = 260.5 (im
T = 2.4 min
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SublFO-23 1174
25
TPC = 164.6 jiL/L
VMD = 196 um
T = 2.1 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Figure C15. LISST DSD plot for IFO-120, DOR 1:100 (Corexit 9500), warm experiments.
IA-E12PG00037 Final Report Appendices
Page 164
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubIFO-16 1174
25
20
I 15-
o
c
o
o
®
0
1
10
TPC = 78 fiL/L
VMD = 119.8 iim
T = 2.5 min.
2,5-3 8-9,5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max); SublFO-20 1174
25
TPC = 241.1 nL/L
VMD = 86.4 (im
T = 2.1 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SublFO-24 1174
25
TPC = 111.5 jiL/L
VMD = 66,4 um
T = 2.6 min.
2,5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (fim)
423-499
Figure C16. LISST DSD plot for IFO-120, DOR 1:20 (Corexit 9500), warm experiments.
IA-E12PG00037 Final Report Appendices
Page 165
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubCND-01 1174
25
20
I 15-
o
c
o
o
®
0
1
10
TPC = 51,2 pL/L
VMD = 215.7 um
T = 2.2 min.
2,5-3 8-9,5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration {Curve TPC Max): SubCND-02R 117-
25
20
15
0
1
CL-
IO
2,5-3
TPC = 22.9 fiL/L
VMD = 153.3 pm
T = 2.3 min.
iilllll
hi
8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubCND-03 1174
25
20
0
1
E
m
o
c
o
0
u
1
0.
15
10
TPC = 50,1 jiL/L
VMD = 183.3 jim
T = 2,6 min.
2,5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-489
Figure C17. LISST DSD plot for Gas Condensate, no dispersant.
IA-E12PG00037 Final Report Appendices
Page 166
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubCND-04 1174
25
20
15-
TPC = 138.3 fiL/L
VMD = 68.2 fim
T = 2.2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max); SubCND-05 1174
TPC = 445.9 (.iL/L
VMD = 170.4 um
T = 2.9 min
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubCND-06 1174
25
20
15
TPC = 181.2 jiL/L
VMD = 60.4 um
T = 2.2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Figure C18. LISST DSD plot for Gas Condensate, DOR 1:20 (Corexit 9500).
IA-E12PG00037 Final Report Appendices
Page 167
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubCND-07 1174
TPC = 2,7 |iL/L
VMD = 258.9 pm
T = 2.2 min.
2.5-3
8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
423-499
Figure C19. LISST DSD plot for air injection, no dispersant.
IA-E12PG00037 Final Report Appendices
Page 168
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubFIN-07 1174
25
20
I 15-
TPC = 105.3 jiL/L
VMD = 95.5 urn
T = 2.2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubFIN-08 1174
25
20
TPC = 140.2 nL/L
VMD = 96.1 um
T = 2.5 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubFIN-09 1174
25
TPC = 103.9 nL/L
VMD = 92.2 um
T = 2.2 min.
¦¦¦
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Figure C20. LISST DSD plot for AIMS, DOR 1:200 (Finasol OSR 52), warm experiments.
IA-E12PG00037 Final Report Appendices
Page 169
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubFIN-04 1174
25
TPC = 157.3 |iUL
VMD = 85,6 urn
T = 2.8 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubFIN-05 1174
25
20
TPC = 111.8 nL/L
VMD = 78.2 um
T = 2.5 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubFIN-06 1174
25
TPC = 109.7 fiL/L
VMD = 75.5 fim
T = 2.2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Figure C21. LISST DSD plot for AIMS, DOR 1:100 (Finasol OSR 52), warm experiments.
IA-E12PG00037 Final Report Appendices
Page 170
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubFIN-01 1174
25
TPC = 138.8 |iUL
VMD = 61,4 |im
T = 2.4 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubFIN-02 1174
25
20
TPC = 152.1 nL/L
VMD = 51.5 um
T = 2.1 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubFIN-03 1174
25
20
0
1
E
m
o
c
o
0
u
1
0.
15
10
TPC = 155.3 (iL/L
VMD = 48.8 um
T = 2.9 min.
¦!
m
llh.—
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Figure C22. LISST DSD plot for AIMS, DOR 1:20 (Finasol OSR 52), warm water experiments.
IA-E12PG00037 Final Report Appendices
Page 171
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubFIN-10 1174
25
TPC = 18,7 nL/L
VMD = 376.5 jim
T = 2.1 min.
20
15
10
5
0
2.5-3
8-9.5
30-35
113-133
423-499
LISST Particle Size Bins (jim)
Figure C23. LISST DSD plot for IFO-120, DOR 1:200 (Finasol OSR 52), warm water experiments.
IA-E12PG00037 Final Report Appendices
Page 172
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubFIN-11 1174
25
TPC = 72,6 nL/L
VMD = 209.5 jim
T = 2.3 min.
o 10
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (fim)
423-499
Figure C24. LISST DSD plot for IFO-120, DOR 1:100 (Finasol OSR 52), warm experiments.
IA-E12PG00037 Final Report Appendices
Page 173
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubFIN-12 1174
25
TPC = 78,6 nL/L
VMD = 125.8 jim
T = 3.1 min.
15
10
5
0
2.5-3
8-9.5
30-35
113-133
423-499
LISST Particle Size Bins (jim)
Figure C25. LISST DSD plot for IFO-120, DOR 1:20 (Finasol OSR 52), warm experiments.
IA-E12PG00037 Final Report Appendices
Page 174
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubSLC-01 1174
25
20
15-
TPC = 93.9 fiL/L
VMD = 148 jim
T = 2 min.
2,5-3 8-9,5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubSLC-05 1174
25
20
TPC = 110.8 nL/L
VMD = 121.3 tim
T = 2.9 min.
2,5-3 8-9,5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubSLC-09 1174
25
20
TPC = 126.8 jiL/L
VMD = 140.4 um
T = 2,2 min.
2,5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Figure C26. LISST DSD plot for SLC, no dispersant, warm water experiments.
IA-E12PG00037 Final Report Appendices
Page 175
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubSLC-02 1174
25
TPC = 142.1 jiL/L
VMD = 103.3 um
T = 2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubSLC-06 1174
25
TPC = 164.1 nL/L
VMD = 99.9 pm
T = 2.3 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (um)
423-499
Particle Size Concentration (Curve TPC Max): SubSLC-10 1174
TPC = 300.3 (iLIL
VMD = 103 jim
T = 2.1 min
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (pm)
423-499
Figure C27. LISST DSD plot for SLC, DOR 1:200 (Corexit 9500), warm water experiments.
IA-E12PG00037 Final Report Appendices
Page 176
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubSLC-03 1174
25
TPC = 180,7 jiL/L
VMD = 91.7 nm
T = 2.3 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (jim)
423-499
Particle Size Concentration (Curve TPC Max): SubSLC-07 1174
25
TPC = 104 uL/L
VMD = 91 nm
T = 2.2 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
423-499
Particle Size Concentration (Curve TPC Max): SubSLC-11 1174
25
TPC = 161.4 fiL/L
VlflD = 108.2 nm
T = 2.5 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
423-499
Figure C28. LISST DSD plot for SLC, DOR 1:100 (Corexit 9500), warm water experiments.
IA-E12PG00037 Final Report Appendices
Page 177
-------
EPA/600/R-16/152
Appendix C
Particle Size Concentration (Curve TPC Max): SubSLC-04R 117'
25
20
I 15-
o
c
o
o
®
0
1
10
TPC = 63.8 nL/L
VMD = 21.2 urn
T = 2.3 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (urn)
423-499
Particle Size Concentration (Curve TPC Max): SubSLC-08R 117-
25
20
15
TPC = 132.5 (jL/L
VMD = 18.4 urn
T = 2.5 min.
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (nm)
423-499
Particle Size Concentration (Curve TPC Max): SubSLC-12R 117-
25
20
0
1
E
m
o
c
0
u
J5
u
1
0.
15
10
TPC = 116.7 fiL/L
VMD = 14.8 tim
T = 2.8 min.
!¦- —¦
A
lli
2.5-3 8-9.5 30-35 113-133
LISST Particle Size Bins (tim)
423-499
Figure C29. LISST DSD plot for SLC, DOR 1:20 (Corexit 9500), warm water experiments.
IA-E12PG00037 Final Report Appendices
Page 178
-------
EPA/600/R-16/152
Appendix D
APPENDIX D - Jet Release LISST Oil Droplet Size Distribution Time Series Contours
IA-E12PG00037 Final Report Appendices
Page 179
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SUBANS-1 1387
1.0+
0,8
0.6
0.4
0.2
0
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SUBANS-5 1387
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SUBANS-9 1387
1.0+
0.8
0.6
0.4
0.2
0
2 4 6 8
Time (minutes, normalized)
Figure Dl. LISST contour plot for ANS, no dispersant, cold water experiments.
— 10
t:
CD
0-
\?—
10'
IA-E12PG00037 Final Report Appendices
Page 180
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubANS-2R 1174
— 10
t:
CD
0-
10'
— 10
~ 10
re
Q-
>
1.0+
0.8
0.6
0.4
> 0.2
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubANS-6R 1174
1.0+
0.8
0.6
0.4
0.2
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubANS-10R 1174
1.0+
0.8
0.6
0.4
0.2
4 6 8
Time (minutes, normalized)
FigureD2. LISST contour plot for ANS, DOR 1:200 (Corexit 9500), cold water experiment.
IA-E12PG0Q037 Final Report Appendices
Page 181
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SUBANS-3 1387
— 10
t:
CD
0-
10'
o-
— 10z
10
— 10
¦c
re
0.
101 1
1.0+
0.8
0.6
0.4
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SUBANS-7 1387
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): subans-11 1387
2 4 6 8
Time (minutes, normalized)
0.2
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
Figure D3. LISST contour plot for ANS, DOR 1:100 (Corexit 9500), cold water experiment.
IA-E12PG0Q037 Final Report Appendices
Page 182
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubANS-4R 1174
— 10
t:
CD
0-
10'
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubANS-8R 1174
— 10'
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubANS-12R 1174
— 10
1
re
Q.
101 I
2 4 6 8
Time (minutes, normalized)
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
Figure D4. LISST contour plot for ANS, DOR 1:20 (Corexit 9500), cold water experiment.
IA-E12PG0Q037 Final Report Appendices
Page 183
-------
EPA/600/R-16/152
Appendix D
— 10
t:
CD
0-
Contour (z-axis constrained): SubANS-13 1174
r>m
10'
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubANS-17 1174
— 10'
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubANS-21 1174
— 10
1
cs
0.
2 4 6 8
Time (minutes, normalized)
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
Figure D5, LISST contour plot for ANS, no dispersant (Corexit 9500), warm experiment.
IA-E12PG0Q037 Final Report Appendices
Page 184
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubANS-14 1174
— 10
t:
re
Q.
— 10'
— 10
1
re
0.
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubANS-18 1174
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubANS-22 1174
1011
0-
2 4 6 8
Time (minutes, normalized)
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
Figure D6. LISST contour plot for ANS, DOR 1:200 (Corexit 9500), warm experiment.
IA-E12PG0Q037 Final Report Appendices
Page 185
-------
EPA/600/R-16/152
Appendix D
— 10
t:
CD
0-
10'
Contour (z-axis constrained): SubANS-15R 1174
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubANS-19 1174
1.0+
0.8
0.6
0.4
— 10z
10
0
0.2
1.0+
0.8
0.6
0.4
0.2
— 10
1
cs
0.
101 1
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubANS-23 1174
0"
2 4 6 8
Time (minutes, normalized)
1.0+
0.8
0.6
0.4
0.2
Figure D7. LISST contour plot for ANS, DOR 1:100 (Corexit 9500), warm experiment.
IA-E12PG0Q037 Final Report Appendices
Page 186
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubANS-16 1174
— 10
t:
re
Q.
— 10z
10
— 10
1
re
0.
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubANS-20 1174
ft
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubANS-24 1174
1011
2 4 6 8
Time (minutes, normalized)
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
Figure DS. LISST contour plot for ANS, DOR 1:20 (Corexit 9500), warm experiment.
IA-E12PG0Q037 Final Report Appendices
Page 187
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SublFO-1R 1174
— 10
t:
re
Q.
— 10'
— 10
1
re
0.
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SublFO-5 1387
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SublFO-9 1387
1011
5
2 4 6 8
Time (minutes, normalized)
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
Figure D9, LISST contour plot for IFO-120, no dispersant, cold water experiment
IA-E12PG0Q037 Final Report Appendices
Page 188
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SublFO-2R 1174
— 10
t:
re
Q.
— 10'
t:
«
Q.
— 10
1
cs
0.
101
4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SublFO-6R 1174
4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SublFO-10R 1174
2 4 6 8
Time (minutes, normalized)
1.0+
1.0+
J*
j
c
o
*3
ra
i—
4-1
c
0)
c
O
O
E
3
1
1.0+
0.8
0.6
0.4
Figure D10. LISST contour plot for IFO-120, DOR 1:200 (Corexit 9500), cold experiment.
IA-E12PG0Q037 Final Report Appendices
Page 189
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SUBIFO-3 1387
— 10
t:
CD
0-
10'
— i
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SublFO-4R 1174
— 10
t:
CD
0-
10'
— i
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SublFO-13 1174
— 10
t:
CD
0-
10'
1.0+
0.8
0.6
0.4
— 10z
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SublFO-17 1174
ry—
0.2
1.0+
10
0.8
0.6
0.4
0.2
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SublFO-21 1174
— 10
1
re
0.
101 1
1.0+
0.8
0.6
0.4
0.2
2 4 6 8
Time (minutes, normalized)
Figure D13. LISST contour plot for IFO-120, no dispersant, warm water experiment.
IA-E12PG0Q037 Final Report Appendices
Page 192
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SublFO-14 1174
— 10
t:
CD
0-
10'
— i
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SublFO-15 1174
— 10
t:
CD
0-
10'
1.0+
0.8
0.6
0.4
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SublFO-19 1174
— 10z
10
0.2
1.0+
0.8
0.6
0.4
0.2
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SublFO-23 1174
— 10
1
re
0.
101 1
1.0+
0.8
0.6
0.4
0.2
2 4 6 8
Time (minutes, normalized)
Figure D15. LISST contour plot for IFO-120, DOR 1:100 (Corexit 9500), warm experiment.
IA-E12PG0Q037 Final Report Appendices
Page 194
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SublFO-16 1174
— 10
t:
re
Q.
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SublFO-20 1174
— 10z
10
9
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SublFO-24 1174
— 10
1
re
0.
101 1
0
2 4 6 8
Time (minutes, normalized)
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
Figure D16. LISST contour plot for IFO-120, DOR 1:20 (Corexit 9500), warm experiment.
IA-E12PG0Q037 Final Report Appendices
Page 195
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubCND-01 1174
— 10
t:
re
Q.
— 10'
— 10
1
re
0.
101
1.0+
0.8
0.6
0.4
0.2
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubCND-02R 1174
-1.0+
0.8
0.6
0.4
0.2
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubCND-03 1174
&
1.0+
0.8
0.6
0.4
0.2
2 4 6 8
Time (minutes, normalized)
Figure D17, LISST contour plot for Gas Condensate, no dispersant.
IA-E12PG0Q037 Final Report Appendices
Page 196
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubCND-04 1174
— 10
t:
CD
0-
10'
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubCND-05 1174
— 10'
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubCND-06 1174
— 10
1
re
Q.
101 I
k
2 4 6 8
Time (minutes, normalized)
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
Figure D18. LISST contour plot for Gas Condensate, DOR 1:20 (Corexit 9500).
IA-E12PG0Q037 Final Report Appendices
Page 197
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubCND-07 1174
Time (minutes, normalized)
Figure D19. LISST contour plot for air injection, no dispersant.
IA-E12PG00037 Final Report Appendices
Page 198
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubFIN-07 1174
— 10
t:
CD
0-
10'
p
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubFIN-08 1174
— i
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubFIN-04 1174
— 10
t:
CD
0-
10'
D
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubFIN-05 1174
— i
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubFIN-01 1174
— 10
t;
CD
0-
10'
o
— 10z
10
Cr
— 10
1
re
0.
101
1.0+
0.8
0.6
0.4
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubFIN-02 1174
0.2
1.0+
0.8
0.6
0.4
0.2
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubFIN-03 1174
0r
2 4 6 8
Time (minutes, normalized)
1.0+
0.8
0.6
0.4
0.2
Figure D22. LISST contour plot for ANS, DOR 1:20 (Finasoi OSR 52), warm experiment.
IA-E12PG0Q037 Final Report Appendices
Page 201
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubFIN-10 1174
— 10
t
re
0.
2 4 6 8
Time (minutes, normalized)
-1.0+
0.8
0.6
0.4
0.2
Figure D23. LISST contour plot for IFO-120, DOR 1:200 (Finasol OSR 52), warm experiment.
IA-E12PG00037 Final Report Appendices
Page 202
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubFIN-11 1174
— 10
t
(8
Q_
10'
7
-1.0+
0.8
0.6
0.4
0.2
2 4 6 8
Time (minutes, normalized)
Figure D24. LISST contour plot for IFO-120, DOR 1:100 (Finasol OSR 52), warm experiment.
IA-E12PG00037 Final Report Appendices
Page 203
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubFIN-12 1174
— 10
t
(8
Q_
10
-1.0+
0.8
0.6
0.4
0.2
2 4 6 8
Time (minutes, normalized)
Figure D25. LISST contour plot for IFO-120, DOR 1:20 (Finasol OSR 52), warm experiment.
IA-E12PG00037 Final Report Appendices
Page 204
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubSLC-01 1174
— 10
t:
CD
0-
10'
Q-
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubSLC-05 1174
— i
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubSLC-02 1174
— 10
t:
CD
0-
o
10'
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubSLC-06 1174
— 10
1
re
a.
10
>
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubSLC-10 1174
10
t
re
a
10'
P
2 4 6 8
Time (minutes, normalized)
1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
-1.0+
0.8
0.6
0.4
0.2
Figure D27. LISST contour plot for SLC, DOR 1:200 (Corexit 9500), warm water experiment.
IA-E12PG0Q037 Final Report Appendices
Page 206
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubSLC-03 1174
— 10
t:
re
Q.
10
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubSLC-07 1174
— 10'
2 4 6 8
Time (minutes, normalized)
Contour (z-axis constrained): SubSLC-11 1174
— 10
E
re
Cl
10
2 4 6 8
Time (minutes, normalized)
n
1.0+
0.8
0.6
0.4
f
¦ n
-1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
Figure D28. LISST contour plot for SLC, DOR 1:100 (Corexit 9500), warm experiment.
IA-E12PG0Q037 Final Report Appendices
Page 207
-------
EPA/600/R-16/152
Appendix D
Contour (z-axis constrained): SubSLC-04R 1174
— 10
t:
re
Q.
10'
— i
2 4 6 8
Time (minutes, normalized)
1.0+
0.8
0.6
0.4
0.2
-1.0+
0.8
0.6
0.4
0.2
1.0+
0.8
0.6
0.4
0.2
n
Figure D29. LISST contour plot for SLC, DOR 1:20 (Corexit 9500), warm water experiment.
IA-E12PG0Q037 Final Report Appendices
Page 208
-------
EPA/600/R-16/152
Appendix E
APPENDIX E - Submersible Fluorescence Time Series
IA-E12PG00037 Final Report Appendices
Page 209
-------
EPA/600/R-16/152
Appendix E
SUBANS-1
Turner Cyclops (Crude Oil)
200
1 so
100
50
o
2.0
1.5
1 .o
0.5
O.O
2.0
1 .5
1 .O
O.S
O.O
40
30
20
10
30
25
20
1 5
10
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
"YV,
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
GmbH Trios
J
o
2
4 6
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
8
10
SUBANS-
Turner Cyclops (Refined Oil)
300
200
100
3.0
2.5
2.0
1.5
1 .0
0.5
O.O
2.5
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
j
lArt
Av _ _ . .
"S. 2.0
S" 1-5
Q. "I .O
^ 05
S> O.O
60
50
40
30
20
10
WetLabs ECO
GmbH Trios
j
^ _
O 2 4 6 8
Time (miri utes)
Turner Cyclops (Crude Oil)
10
SUBANS-
Turner Cyclops (Refined Oil)
_J
^
Chelsea AQUAtracka (Crude Oil)
I
Chelsea AQUAtracka (Refined Oil)
I
|
WetLabs ECO
i\
GmbH Trios
fx
/
40
30
20
10
250
200
1 50
1 oo
50
2.5
2.0
1.5
-I .O
0.5
O.O
1.5
1.0
0.5
O.O
50
40
30
20
10
25
20
15
10
Time (minutes)
Figure El. Raw in-situ fluorometer signal for ANS, no dispersant, cold experiment.
IA-E12PG00037 Final Report Appendices
Page 210
-------
EPA/600/R-16/152
Appendix E
SubANS-2R
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
"o -1.2
3 4:8
s 8:5
0.2
=^, O-O
-0.2
J 15
1 .o
Q- O.S
~B) O.O
UJ 25
CO 20
S -15
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
A--/-a.-.
GmbH Trios
150
3r 1 oo
S SO
8:§
c 1.2
® 1 .o
O.S
<5 0.6
Q- 0.4
_i 0.2
^ -8:2
s as
to 20
30
20
10
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SubANS-6R
\
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea
AQUAtracka (Refined Oil)
.......... Pwx.,
/^v
Wet Labs ECO
GmbH Trios
_
^ ^
~-A/v"-
M. 150
^ 1 oo
j= so
8:1
8:3
O.O
g 1 o
_5> 0.8
o.e
2 0.4
0.2
g) O.O
li_j 30
q 25
co 20
30
20
10
Time (minutes)
Turner Cyclops (Crude Oil)
SubANS-1 OR
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
GmbH Trios
Time (minutes)
Figure E2. Raw in-situ fluorometer signal for ANS, DOR 1:200 (Corexit 9500), cold expt.
IA-E12PG00037 Final Report Appendices
Page 211
-------
EPA/600/R-16/152
Appendix E
SUBANS-3
Turner Cyclops (Crude Oil)
350
300
250
200
1 50
1§8
O
1.5
1.O
0.5
O.O
2.5
2.0
1 .5
1 .O
0.5
O.O
30
25
20
1 5
10
20
15
10
500
400
300
200
100
2.5
2.0
1 .5
1 .O
0.5
O.O
-0.5
— 3
60
50
40
30
20
1 O
35
in 30
25
a- 20
—¦ 15
1
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
J
VvJL -
Chelsea AQUAtracka (Refined Oil)
J
WetLabs ECO
)
CSmbH Trios
O 2
4 6 8 IO
Time (minutes)
SUBANS-
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
1
V
Chelsea AQUAtracka (Crude Oil)
f>
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
Time (minutes)
Turner Cyclops (Crude Oil)
SUBANS-1 1
> 250
200
150
'5j 100
a> 50
"o 3.0
a 2-5
_S 2.0
(5 "J
t> 1 .o
0.5
g> o.o
g 3.0
® 2.5
"E- 2.0
te 1.5
1 .o
0.5
^ O.O
S 58
co 40
o 30
20
s> 10
40
30
20
10
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
~x
WetLabs ECO
GmbH Trios
Time (minutes)
Figure E3. Raw in-situ fluorometer signal for ANS, DOR 1:100 (Corexit 9500), cold expt.
IA-E12PG00037 Final Report Appendices
Page 212
-------
EPA/600/R-16/152
Appendix E
1000
- 800
r 6oo
400
200
30
20
10
£ 150
o 100
80
GO
40
20
1000
' 800
j* 600
400
200
35
30
25
20
18
UJ 200
CO 1 SO
^ 100
O) 50
80
60
40
20
: 2000
; 1500
i 1000
i 500
o
* 35
^ is
r
10
LU 1 50
Q
W 1 OO
80
GO
40
20
SufoANS-4R
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
.A ,
Wet Labs ECO
GmbH Trios
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
S u b A N S -8 R
Turner Cyclops (Refined Oil)
J1 W-v-
' '', /V.. ¦ - A
v/v
Chelsea ftQUAtracka (Crude Oil)
VA
w
kA
AaJ l
Chelsea AQUATmcka (Refined Oil)
/vA^/^'uV
WetLabs ECO
GmbH Trios
f
Time (minutes)
Turner Cyclops (Crude Oil)
SubANS-1 2R
Turner Cyclops (Refined Oil)
Vv\
Chelsea AQUAtracka (Crude Oil)
_
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
Time (minutes)
Figure E4. Raw in-situ fluorometer signal for ANS, DOR 1:20 (Corexit 9500), cold expt.
IA-E12PG00037 Final Report Appendices
Page 213
-------
EPA/600/R-16/152
Appendix E
SUBANS-1 3
250
200
1 SO
1 OO
50
O
¦S o
"5 2.0
« 1.5
H 1.0
° 0.5
00
c 2.0
-j|. 1.5
a> 1 .O
^ 0.5
3) O.O
30
20
10
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
H
Wet Labs ECO
GmbH Trios
250
200
1 50
1 OO
50
1.5
1 .O
0.5
O.O
t .5
1.0
0.5
O.O
20
15
10
— 30
3? 25
< 20
15
d 10
a §
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SUBANS-17
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
^V,
Chelsea AQUAtracka (Refined Oil)
vV
r
WetLabs ECO
GmbH Trios
150
1 OO
50
2.0
1 .5
1 .O
0.5
O.O
2.0
1 .5
-I .O
O.S
O.O
40
30
20
10
40
30
20
10
Time (minutes)
Turner Cyclops (Crude Oil)
SU BANS-21
A'
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
~ -A.
Time (minutes)
Figure E5. Raw in-situ fluorometer signal for ANS, DOR 0 (Corexit 9500), warm expt.
IA-E12PG00037 Final Report Appendices
Page 214
-------
EPA/600/R-16/152
Appendix E
SUBANS-14
20
15
-IO
1.5
1.0
0.5
O.O
2.5
2.0
1.5
1.0
0.5
O.O
50
40
30
20
10
O
50
40
30
20
1 o
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
GmbH Trios
~"V
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SUBANS-18
15
10
2.0
1 .5
1 .O
0.5
O.O
2.0
1 .5
1 .O
0.5
O.O
30
20
10
40
30
20
10
Turner Cyclops (Refined Oil)
)
rMa
W-A
Chelsea AQUAtracka (Refined Oil)
_._J
WetLabs ECO
GmbH Trios
x~\
--
' —
Time (minutes)
Turner Cyclops (Crude Oil)
S U BANS-22
Turner Cyclops (Refined Oil)
15 -
/ 7\
10 -
J
5 ^
/
v\
O -
5> 0.0
Time (minutes)
Figure E6. Raw in-situ fluorometer signal for ANS, DOR 1:200 (Corexit 9500), warm expt.
IA-E12PG00037 Final Report Appendices
Page 215
-------
EPA/600/R-16/152
Appendix E
SUBANS-1 5R
IS
10
2.0
1 .5
1 .O
0.5
O.O
3.0
2.5
2.0
1 .5
1 .O
0.5
O.O
30
20
10
40
30
20
-IO
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
V
Chelsea AQUAtracka (Refined Oil)
V.
Wet Labs ECO
GmbH Trios
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SUBANS-19
1 s
10
t .5
1.0
O.S
O.O
1.5
1.0
0.5
O.O
30
20
1 O
O
40
30
20
10
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
__A-V
Chelsea
AQUAtracka (Refined Oil)
nrx
WetLabs ECO
GmbH Trios
/VA'-
Time (minutes)
Turner Cyclops (Crude Oil)
S U BANS-23
3= 0
-5 2.5
TO 2.0
-e 1-5
s 10
_i 0.5
0.0
o>
® 1 .5
IT 1.0
^ 0.5
s> 0.0
40
30
20
10
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
Time (minutes)
Figure E7. Raw in-situ fluorometer signal for ANS, DOR 1:100 (Corexit 9500), warm expt.
IA-E12PG00037 Final Report Appendices
Page 216
-------
EPA/600/R-16/152
Appendix E
SUBANS-1 6
Turner Cyclops (Crude Oil)
30
20
10
Turner Cyclops (Refined Oil)
\
Chelsea AQUAtracka (Crude Oil)
-Av_
Chelsea AQUAtracka (Refined Oil)
r\i,
140
-120
100
so
so
40
20
so
60
40
20
_r~ \ a.a^
Wet Labs ECO
\
GmbH Trios
"'V,
30
20
10
<-> 2
® 6
120
1 OO
SO
60
40
20
o
80
60
40
20
Time (minutes)
Turner Cyclops (Crude Oil)
SUBANS-20
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
y/yww^j
IL __
Chelse
a AQUAtracka (Refined Oil)
WetLabs ECO
vA^l
GmbH Trios
/
£ 25
" 20
.•ir 15
g 10
® 5
"5 5
co 10O
a
=" 50
60
40
20
Time (minutes)
Turner Cyclops (Crude Oil)
S U BANS-24
Turner Cyclops (Refined Oil)
_
Chels
ea AQUAtracka (Crude Oil)
Aa
Chelse
a AQUAtracka (Refined Oil)
j V [
_
WetLabs ECO
GmbH Trios
Time (minutes)
Figure E8. Raw in-situ fluorometer signal for ANS, DOR 1:20 (Corexit 9500), warm expt.
IA-E12PG00037 Final Report Appendices
Page 217
-------
EPA/600/R-16/152
Appendix E
SublFO-1 R
Turner Cyclops (Crude Oil)
-/V
1 .O
0.8
o.e
0.4
0.2
o.o
-0.2
-0.4
O.S
0.4
0.3
0.2
0.1
O.O
-0.1
O.OS
O.OO
-O.OS
O.S
o.o
-0.5
-1.0
-1.5
-2.0
Turner Cyclops (Refined Oil)
VAA^
V-wvV
VvA"
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
j^\yJJywVMv^
[/
Wet Labs ECO
10
GmbH Trios
120
1 oo
80
GO
40
20
-I .O
0.5
O.O
6
30
20
10
^ O
80
60
40
20
O
t .5
1.0
O.S
o.o
1 o
30
25
20
15
10
\l ^
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SUBIFO-5
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
V.
-¦¦'-yt/v}
\
WetLabs ECO
wv-^. /V^
GmbH Trios
,r v
Aav,
Time (minutes)
Turner Cyclops (Crude Oil)
SUBIFO-9
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
JvAx
A
Time (minutes)
Figure E9. Raw in-situ fluorometer signal, IFO-120, no dispersant, cold water expt.
IA-E12PG00037 Final Report Appendices
Page 218
-------
EPA/600/R-16/152
Appendix E
SublFO-2R
= -2
1 .o
8:1
# 81
c O.O
.22 -0.2
c -0.4
0.3
0.2
0.1
O.O
-O.l
0.05
O.OO
-0.05
-O.l O
o.a
O.O
-0.5
Turner Cyclops (Crude Oil)
Vwwvv-—-v.-
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
1 V vv\A-AA-JV-'-'v'V 'Wv'~
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
GmbH Trios
MAaA^V
J\f\j\j\^ J\j\y~ -JV
-1.5
1.0
O.S
O.O
2.5
2.0
1.5
1.0
O.S
O.O
-0.5
0.20
0.1 5
O.l O
0.05
O.OO
-0.05
0.08
0.06
8:83
O.OO
-0.02
-0.04
S 05
C* O.O
~°-5
^ -1 .o
8
3? 6
< 4
, 2
3 -2
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SublFO-6R
^vyvAAMA/1
Turner Cyclops (Refined Oil)
- —-A- ^
Chelsea AQUAtracka (Crude Oil)
yy^A/A
VvVyfyJ
l/*Vv
WetLabs ECO
vy v
GmbH Trios
rA A--- j\ /v''^A\/A-x /
Time (minutes)
Turner Cyclops (Crude Oil)
SublFO-1 OR
o
1 .5
t .o
O.S
O.O
O.O©
0.04
0.02
O.OO
-0.02
-0.04
-0.06
0.10
0.05
O.OO
-0.05
O.S
O.O
-0.5
-t .o
10
>u
Turner Cyclops (Refined Oil)
-A -AA/^VAAvA "wVi/nJV^/^-^'\/v
Chelsea AQUAtracka (Crude Oil)
v\y-vAv-v
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
JVWAATV
-V- V,/^^/V^WV/A^JlYsy
Time (minutes)
Figure E10. Raw in-situ fluorometer signal, IFO-120, DOR 1:200 (Corexit 9500), cold expt.
IA-E12PG00037 Final Report Appendices
Page 219
-------
EPA/600/R-16/152
Appendix E
SUBIFO-3
Turner Cyclops (Crude Oil)
20
1 o
o
-10
1.5
-1.0
0.5
0.0
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
-1.5
1.0
o.s
0.0
-0.5
-1 -O
-1 .5
20
1 5
IO
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
GmbH Trios
A-
^ ^
.£. 1.0
S- o.a
CL O.O
J -0.5
S 0.1s
s 0.10
1= 0.05
2 0.00
_| -0.05
1)0.10
J 0.10
£~ 0.05
o- O.OO
%-o.os
g II
CO 1.5
O 1 .0
i 0-5
Sj O.O
=i- -0.5
S -2
Time (minutes)
Turner Cyclops (Crude Oil)
Su bl FO-7R
A,
UaA'a-
-aA,A-/v^
-V'
V
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
wVw
y-V\
",^VAjvv'*vAv
w p-^-vvJ W»r-Aj
WetLabs EGO
JrV
VP
/ *"Vi
l^kAWOAV/
GmbH Trios
\ / \f
-6
1 .5
1 .O
0.5
O.O
0.5
0.4
0.3
0.2
O.I
O.O
-0.-1
0.10
0.05
O.OO
-0.05
1 .O
0.5
O.O
-0.5
-1 .O
-t .5
1 O
3
Time (minutes)
Turner Cyclops (Crude Oil)
SublFO-1 1 R
Turner Cyclops (Refined Oil)
Av^/v^vAV^ VV'VV^YVVVV/V^- V\/V^xA-'Aa>vvv\/v\/Vv A/vy\r-"
Chelsea AQUAtracka (Crude Oil)
v U^fW uavw
W/J Ham w ^
\^"vi /VxA 'v'
-vA A^A/\/V^VVvyVVs^-V^'
Time (minutes)
Figure Ell. Raw in-situ fluorometer signal, IFO-120, DOR 1:100 (Corexit 9500), cold expt.
IA-E12PG00037 Final Report Appendices
Page 220
-------
EPA/600/R-16/152
Appendix E
SublFO-4R
=»
4
-2
-6
6
-S*
4
c
2 -
"H
O
o>
s
O.G
rc
0.4
,
0.2 -
cx>
o.o -
a>
S
0.1 -
o.o -
-0.1
-0.2 -
-0.3 -
-0.4 -
¦ ¦ ¦
o -
-1 -
1 o
£
5 -
=d
o -
-5 -
—
80
GO
40
c
20 -
=
O
c£-
i=
5
3
e
2 -
~c=
o -
2.5
2.0 -
1.5-
1 .o -
~S>
o.o -
a>
Q)
0.8
r-
0.6
0.4 -
¦
0.2 -
g>
o.o -
LU
4 -
CO
G»
2 -
o -
-2 -
40 -
«=C
30
20 -
s»
10 -
o -
>"
o -
5»~
-£r
8
e
<5
2 -
c
a>
N
6 -
-e
4 -
o
5*
o
a>
4 -i
a>
1
1 -
a>
o -
LU
4 -
2
1
5»
O
-1 -1
GO -
30 -
18
A
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
^/AAAiA,
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
-WYV'lVV'
GmbH Trios
Time (minutes)
Figure E12. Raw in-situ fluorometer signal, IFO-120, DOR 1:20 (Corexit 9500), cold expt.
IA-E12PG00037 Final Report Appendices
Page 221
-------
EPA/600/R-16/152
Appendix E
SUBIFO-1 3
1 .o
0-8
0.6
0.4
0.2
O.O
0.25
0.20
O.l 5
0.10
0.05
O.OO
-0.05
2.0
1 .5
1 -O
0.5
O.O
-0.5
-1 .O
25
20
1 5
10
O
1 .5
1 .O
0.5
O.O
0.5
O.4.
0.3
0.2
O.l
O.O
-O.l
1 .O
0.5
O.O
-0.5
-1 .O
-1 .5
-2.0
-2.5
30
20
10
3= °
~S 1-5
¦S 1 -O
« o.s
O.O
j= 0.6
« 0.5
"5* 0.4
0.3
S_ O 2
i 0.1
a -o:i
30
25
20
1 5
1 O
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
GmbH Trios
4 6
Time (minutes)
Turner Cyclops (Crude Oil)
SUBIFO-1 Y
Turner Cyclops (Refined Oil)
/ \rr\
Chelsea AQUAtracka (Crude Oil)
\^j\a
\m,
Chelsea AQUAtracka (Refined Oil)
J^X
Vv-,
WetLabs ECO
GmbH Trios
/'V.-A
Time (minutes)
Turner Cyclops (Crude Oil)
to
SU BIFO-21
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
_
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
..aaY'- -
Time (minutes)
Figure E13. Raw in-situ fluorometer signal for IFO-120, no dispersant, warm water expt.
IA-E12PG00037 Final Report Appendices
Page 222
-------
EPA/600/R-16/152
Appendix E
SUBIFO-14
Turner Cyclops (Crude Oil)
°
~§5 2.0
« 1.5
ts 1-0
° 0.5
o.o
£ O.S
o 0.4
0.3
« 0.2
o.-i
=! O.O
§ -o.-i
25
20
15
10
Turner Cyclops (Refined Oil)
Chelse
a AQUAtracka (Crude Oil)
_
Chelsea
AQUAtracka (Refined Oil)
_*/v_
^
WetLabs ECO
A
GmbH Trios
-v- - ay vyv/"v/V V
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SUBIFO-18
Turner Cyclops (Refined Oil)
1.5
1 .o
0.5
O.O
0.4
0.3
0.2
O.I
O.O
-0.1
1 .5
1 .O
O.S
O.O
-O.S
-1 .O
25
20
15
10
40
30
20
lO
O
15
lO
O
2.5
2.0
1 .5
1.0
0.5
O.O
0.6
0.4
0.2
O.O
30
20
10
Chelsea AQUAtracka (Crude Oil)
/V
U.A^~"~ W „
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
yvA
I V-
Time (minutes)
Turner Cyclops (Crude Oil)
SUBIFO-22
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
vv
WetLabs ECO
GmbH Trios
Time (minutes)
Figure E14. Raw in-situ fluorometer signal, IFO-120, DOR 1:200 (Corexit 9500) warm expt.
IA-E12PG00037 Final Report Appendices
Page 223
-------
EPA/600/R-16/152
Appendix E
SUBIFO-1 5
50
40
30
20
IO
20
15
-IO
0.8
0.6
0.4
0.2
O.O
-0.2
50
40
30
20
1 O
120
1 OO
80
60
40
20
30
25
20
1 5
-IO
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
AL.
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
GmbH Trios
Time (minutes)
Turner Cyclops (Crude Oil)
SUBIFO-1 9
Turner Cyclops (Refined Oil)
<3 2
-1.0
0.5
O.O
Chelsea AQUAtracka (Crude Oil)
\TM
Vll
A
Chelsea AQUAtracka (Refined Oil)
U
WetLabs ECO
"A,
GmbH Trios
so
60
40
20
1 OO
80
60
40
20
30
20
IO
S 2
1
O
1 15
1 .O
^ 0.5
O.O
V,
Time (minutes)
Turner Cyclops (Crude Oil)
SUBIFO-23
A/v
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
GmbH Trios
Time (minutes)
Figure E15. Raw in-situ fluorometer signal, IFO-120, DOR 1:100 (Corexit 9500) warm expt.
IA-E12PG00037 Final Report Appendices
Page 224
-------
EPA/600/R-16/152
Appendix E
SUBIFO-1 6
1 oo
80
60
40
20
O
30
20
10
o 2
1 .2
J:8
o.e
o_4
0.2
0.0
10
so
60
40
20
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
s
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
J#V'
Wet Labs ECO
GmbH Trios
30
25
20
1 5
10
2.0
1.5
1.0
0.5
O.O
L1J 20
<0 15
o 10
1 50
1 OO
50
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SUBIFO-20
Turner Cyclops (Refined Oil)
J
¦a
Chelsea AQUAtracka (Refined Oil)
1
Wv
;
WetLabs ECO
w
GmbH Trios
E. 200
>s -I 50
"55 too
S so
C O
g 40
so
w 20
o 10
-I .5
IT -i .0
^ 0.5
5^ 0.0
LU 15
» 10
-150
< IOO
Time (minutes)
Turner Cyclops (Crude Oil)
SUBIFO-24
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
K
WetLabs ECO
GmbH Trios
Time (minutes)
Figure E16. Raw in-situ fluorometer signal, IFO-120, DOR 1:20 (Corexit 9500), warm expt.
IA-E12PG00037 Final Report Appendices
Page 225
-------
EPA/600/R-16/152
Appendix E
SubCND-01
25
20
15
10
1.2
1 -O
0.8
0.6
0.4
0.2
O.O
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
g 0.6
ET 0.4
s. 0.2
d 0.0
s -0 2
1 so
100
so
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
GmbH Trios
=s-
E
8
6 -
CO
O -
-E
—4 -
E
30
20 -
o>
c.
-lO -
O -
d>
0
1.5 -
Jq
1.0 -
0.5 -
O.O -
0.3 -
0.2 -
o_
O.O -
gj>
Q
CO
Cf
2.5 -
2.0 -
1:S
0.5 -
=L
-0.5 -
80 -
60 -
20
O -
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SubCN D-02R
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
V
Chelsea AQUAtracka (Refined Oil)
\
WetLabs ECO
GmbH Trios
Time (minutes)
Turner Cyclops (Crude Oil)
SubCND-03
20
is
10
1 .5
t .O
0.5
O.O
0.4
0.3
0.2
0.1
O.O
-O.I
2.0
1 .5
1 .O
0.5
O.O
-0.5
120
1 OO
80
GO
40
20
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
Time (minutes)
Figure E17. Raw in-situ fluorometer signal for Gas Condensate, no dispersant.
IA-E12PG00037 Final Report Appendices
Page 226
-------
EPA/600/R-16/152
Appendix E
SubCND-04
"B °
E 80
60
to 40
v 20
1 .O
0.5
O.O
I 250
< 200
ISO
100
S 50
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
GmbH Trios
I iA^^VvV-*Vr^rv~A'vV|Vy'vwA/^, .^y.AA-W^V—
Tim e ( m i n utes )
SubCND-05
15
10
1 OO
80
60
40
20
j5 1 -5
IT 10
_2" 0.5
O.O
S O
400
5 300
^ 200
g> 10O
Turner Cyclops (Refined Oil)
^¦x _.. _ —
Chelsea AQUAtracka (Crude Oil)
A._ _
Chelsea AQUAtracka (Refined Oil)
1
V
\^h^-^^AvvWW\7U1„. —— A-
-w^-. -wV""AavvJ*V"<. .. y*-M^
WetLabs ECO
rx^
GmbH Trios
i
1 5
10
-S o
80
60
40
20
10
^ 0.5
g> O.O
-1
250
200
150
100
50
Time (minutes)
Turner Cyclops (Crude Oil)
SubCND-06
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
1
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
y\
Time (minutes)
Figure E18. Raw in-situ fluorometer signal for Gas Condensate, DOR 1:20 (Corexit 9500).
IA-E12PG00037 Final Report Appendices
Page 227
-------
EPA/600/R-16/152
Appendix E
SubCND-07
23.5
23.0
22.5
22.0
S 1-10
"jS, 1 .08
g. 1 .06
1 .04
1.02
.20
.16
.14
.1 2
.10
.08
<3 5
Turner Cyclops (Crude Oil)
Vv-v-vVaay^W* ¦T"-AvV^V aJ
Y v'Wi
Turner Cyclops (Refined Oil)
*y» —V\^v »v~vvVA^'
Time (minutes)
Figure E19. Raw in-situ fluorometer signal for air injection, no dispersant.
IA-E12PG00037 Final Report Appendices
Page 228
-------
EPA/600/R-16/152
Appendix E
SubFIN -07
250
200
1 50
100
50
Turner Cyclops (Crude Oil)
1 250
200
150
1 oo
<» SO
15
1 o
"o 1 -2
§ 1 -O
S 0.8
m 0.6
2 0.4
i 0.2
o.o
J 1-5
Er 1.0
0.5
O.O
30
20
10
40
30
20
10
400
300
200
100
O
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SubFIN -08
\
Turner Cyclops (Refined Oil)
J
Chelsea AQUAtracka (Refined Oil)
|
"fVV\
_
-------
EPA/600/R-16/152
Appendix E
SutoFIN-04
3SO
300
250
200
g 2.0
_S 1.5
c5 1 .O
" 0-5
g) O.O
« 2.5
-S, 2.0
-I .5
«=»- 1 .O
d 0.5
O.O
uj 40
£ 30
O 20
10
50
40
30
20
1 O
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
i/\jW
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
V
GmbH Trios
1 50
1 oo
50
S 1.5
-£ 1 -O
jj 0.5
O.O
I 15
"E- 1 .o
S- 0.5
O.O
uj 30
o 25
to 20
30
20
10
yx
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SubFIN-05
v .
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
jVY
J >
Chelsea AQUAtracka (Refined Oil)
V
Wet Labs ECO
GmbH Trios
200
1 50
100
50
O
JL -15
-£r -i o
1 .o
0.8
0.6
0.4
0.2
O.O
Time (minutes)
Turner Cyclops (Crude Oil)
SubFI N-06
Turner Cyclops (Refined Oil)
A~
Chelsea AQUAtracka (Crude Oil)
'V,
VL„
8ig
. °-4
0.2
S> O.O
30
20
10
Chelsea AQUAtracka (Refined Oil)
yv-A - w
WetLabs ECO
GmbH Trios
Time (minutes)
Figure E21. Raw in-situ fluorometer signal ANS, DOR 1:100 (Finasol OSR 52) warm expt.
IA-E12PG00037 Final Report Appendices
Page 230
-------
EPA/600/R-16/152
Appendix E
600
;£=r 400
| 200
^ O
^ 40
~ 3°
~ 20
g -IO
J±J IOO
co 80
G* 60
40
5> 20
60
40
20
SubFIN-01
Turner Cyclops (Crude Oil)
~\d V.
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
V-
GmbH Trios
"V
1 000
i 800
> 600
400
200
o
40
i 30
i* 20
10
5 o
uj 100
c=> 80
6 60
i 40
O) 20
_ 100
80
Q- SO
=d 40
5 20
o
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SubFIN-02
Va-
W
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
snrf/l
Mv/Vv
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
<4-
GmbH Trios
/wv,
300
£ 200
® 100
= 30
•=• 25
20
15
c 10
® 5
"o 3.0
J i.o
s q:g
0.5
g> O-O
o
111 80
S SO
a 40
% 20
80
^ 6°
_j 40
gi 20
Time (minutes)
Turner Cyclops (Crude Oil)
Su foFI N-03
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
K/A
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
Time (minutes)
Figure E22. Raw in-situ fluorometer signal for ANS, DOR 1:20 (Finasol OSR 52) warm expt.
IA-E12PG00037 Final Report Appendices
Page 231
-------
EPA/600/R-16/152
Appendix E
Su bFI N-1 O
0.6
0.4
0.10
o.os
_J -0.05
gi-O.IO
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
—A/-ajW-
\,
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
1
UK
/JiaHJi
w ^
Time (minutes)
Figure E23. Raw in-situ fluorometer signal for IFO-120, DOR 1:200 (Finasol OSR 52), warm water
experiment.
IA-E12PG00037 Final Report Appendices
Page 232
-------
EPA/600/R-16/152
Appendix E
SubFIN-11
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
v-Jr
2 1 -2
to l-O
-S 0.8
(5 0.6
S 0.4
i 0.2
O.O
Chelsea AQUAtracka (Crude Oil)
8:f
8:1
"=*• 0.1
=d O.O
-O.-l
-2
50
5 40
2 30
i 20
g) 10
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
f\
GmbH Trios
W" v/ — ^
V
^
,
Time (minutes)
Figure E24. Raw in-situ fluorometer signal for IFO-120, DOR 1:100 (Finasol OSR 52), warm water
experiment.
IA-E12PG00037 Final Report Appendices
Page 233
-------
EPA/600/R-16/152
Appendix E
SubFIN-12
Turner Cyclops (Crude Oil)
60
40
20
as -4
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
s 15
<2 10
2 100
SO
GmbH Trios
Tim e ( m i n utes )
Figure E25. Raw in-situ fluorometer signal IFO-120, DOR 1:20 (Finasol OSR 52) warm expt.
IA-E12PG00037 Final Report Appendices
Page 234
-------
EPA/600/R-16/152
Appendix E
SutoSLC-01
250
200
1 50
1 OO
50
O
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
50
40
30
20
10
GmbH Trios
100
50
20
is
10
O 2.0
_2 1 -5
ro 1 O
<-> 0.5
^ O.O
— s
-
UJ 250
S 200
C* 1SO
60
50
40
30
20
IO
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
SubSLC-05
"W AAA/- -
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Refined Oil)
v
WetLabs ECO
GmbH Trios
-A" - — - —-
120
1 00
so
so
§8
-20
1.5
1.0
0.5
O.O
^ 150
22 100
50
40
30
20
-IO
Time (minutes)
Turner Cyclops (Crude Oil)
SubSLC-09
^wvx/V"
Turner Cyclops (Refined Oil)
A>.
..A-
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
uA/\/va,a_
V
WetLabs ECO
Wa
GmbH Trios
_ _/V- ~J V "x7
Time (minutes)
Figure E26. Raw in-situ fluorometer signal for SLC, no dispersant, warm water expt.
IA-E12PG00037 Final Report Appendices
Page 235
-------
EPA/600/R-16/152
Appendix E
SutoSLC-02
¦g -ISO -
5- 1 oo
1 50
= o
f 2»
— 15
•"£ 10
Jl °
"5 3.5
« 2l5
1 ?:8
1 .o
O.S
gi O.O
llj 250
£ 200
C* 150
50
40
30
20
io
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
xF
\.
Chelsea AQUAtracka (Refined Oil)
%u7V
Wet Labs ECO
rV\.
GmbH Trios
500
400
300
200
1 OO
30
25
20
1 5
10
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
S u bS LC-06
y---Uai
Turner Cyclops (Refined Oil)
® 8
uj 300
q 250
co 200
so
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
60
40
20
> 500
400
300
"w 200
5 100
c O
¥ 25
— 20
-is
'<» 10
uj 300
^ 25°
22 200
^ 1 50
^ 10O
80
60
40
20
Time (minutes)
Turner Cyclops (Crude Oil)
SubSLC-1 O
Turner Cyclops (Refined Oil)
; v
Chelsea AQUAtracka (Crude Oil)
w
Chelsea AQUAtracka (Refined Oil)
V-Ai yy
WetLabs ECO
GmbH Trios
Time (minutes)
Figure E27. Raw in-situ fluorometer signal for SLC, DOR 1:200 (Corexit 9500), warm expt.
IA-E12PG00037 Final Report Appendices
Page 236
-------
EPA/600/R-16/152
Appendix E
SutoSLC-03
200
150
100 -
50
O
40
30 -
20
10
£ 300
« 200
1 OO
80
60
40
20
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
V
Wet Labs ECO
GmbH Trios
M- 400
3^ 300
•5 200
S 100
^ 20
15
s 10
Lij 250
Q 200 "
"2 150
80
SO
40
20
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
S u bS LC-07
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
W\a
WetLabs ECO
GmbH Trios
200
1 50
IOO
SO
25
20
15
IO
m 300
q 250
co 200
5> 50
60
40
20
Time (minutes)
Turner Cyclops (Crude Oil)
SubSLC-1 1
-V- ^\A^WWWVV-
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
Time (minutes)
Figure E28. Raw in-situ fluorometer signal for SLC, DOR 1:100 (Corexit 9500), warm expt.
IA-E12PG00037 Final Report Appendices
Page 237
-------
EPA/600/R-16/152
Appendix E
500
400
300
200
100
o
40
30
20
IO
Su bSLC-04R
Turner Cyclops (Crude Oil)
Turner Cyclops (Refined Oil)
•v..
Chelsea AQUAtracka (Crude Oil)
500
400
300
200
100
1 so
100
50
Chelsea AQUAtracka (Refined Oil)
ha
Wet Labs ECO
GmbH Trios
^ 400
— 300
~ 200
S 100
= o
E. 40
30
"5 20
5 10
500
400
300
200
100
o
150
-lOO
50
Tim e ( m i n utes )
Turner Cyclops (Crude Oil)
S u bS LC-08 R
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
Chelsea AQUAtracka (Refined Oil)
Wet Labs ECO
GmbH Trios
> 500
400
300
"w 200
100
= o
g 40
so
w 20
o 10
500
400
300
200
100
O
150
100
50
Time (minutes)
Turner Cyclops (Crude Oil)
Su foSLC-1 2R
Turner Cyclops (Refined Oil)
Chelsea AQUAtracka (Crude Oil)
An.
Chelsea AQUAtracka (Refined Oil)
WetLabs ECO
GmbH Trios
Time (minutes)
Figure E29. Raw in-situ fluorometer signal for SLC, DOR 1:20 (Corexit 9500), warm expt.
IA-E12PG00037 Final Report Appendices
Page 238
-------
EPA/600/R-16/152
Appendix F
Appendix F - Excitation Emission Matrix Contours
IA-E12PG00037 Final Report
Page 239
-------
EPA/600/R-16/152
Appendix F
DOR = 0
400 -
DOR = 1:200
250 300 350
EX Wavelength (nm)
EX Wavelength (nm)
DOR = 1:100
600
400
DOR = 1:20
o
200
250
300
350
400
200
250
300
350
400
EX Wavelength (nm) EX Wavelength (nm)
Figure Fl. Light Oil Category - IFO-40 oil with dispersant EEMs. Colored contours represent intensity, scaled to maximum
fluorescence peak (red).
IA-E12PG00037 Final Report
Page 240
-------
EPA/600/R-16/152
Appendix F
DOR = 0
<5)
m
200
250 300 350
EX Wavelength (nm))
DOR, = 1:200
200
250 300 350
EX Wavelength (nm))
400
600
400
DOR = 1:100
O)
•
400
200
250 300
EX Wavelength (nm))
350
200
250
300 350
EX Wavelength (nm))
Figure F2. Light Oil Category - Arabian Light crude oil with dispersant EEMs. Colored contours represent intensity, scaled to
maximum fluorescence peak (red).
400
IA-E12PG00037 Final Report
Page 241
-------
EPA/600/R-16/152
Appendix F
600 -I
¦
600 -
¦
DOR = 0
DOR = 1:200
I
f
^ 400 -
1
1
400 "
I
1
o
>
o
200
250 300 350
EX Wavelength (nm)
400
250 300 350
EX Wavelength (nm)
DOR = 1:100
a
200
250 300 350
EX Wavelength (nm)
400
250 300 350
EX Wavelength (nm)
DOB = 1:20
Figure F3. Light Oil Category - Brent crude oil with dispersant EEMs. Colored contours represent intensity, scaled to maximum
fluorescence peak (red).
IA-E12PG00037 Final Report Page 242
-------
EPA/600/R-16/152
Appendix F
600 -h
DOR = 0
600
DOR = 1:200
400 -|
200
250 300
EX Wavelength (nm)
350
400
DOR = 1:100
EX Wavelength (nm)
Figure F4, Light Oil Category - Federated crude oil with
maximum fluorescence peak (red).
200 250 300 350 400
EX Wavelength (nm)
dispersant EEMs. Colored contours represent intensity, scaled to
250 300 350
EX Wavelength (nm)
200
IA-E12PG00037 Final Report
Page 243
-------
EPA/600/R-16/152
Appendix F
600 -
p
600
DOR = 0 1
DOR = 1:200
o
I
I °
200
250 300
EX Wavelength (nm)
350
400
200
250
300 350
EX Wavelength (nm)
400
600 -i
400
DOR = 1:100
600
DOR = 1:20
200 250 300 350 400 200 250 300 350 400
EX Wavelength (nm) EX Wavelength (nm)
Figure F5. Light Oil Category - Gullfaks crude oil with dispersant EEMs. Colored contours represent intensity, scaled to maximum
fluorescence peak (red).
IA-E12PG00037 Final Report
Page 244
-------
600 -i
600 -
EPA/600/R-16/152
Appendix F
DOR = 0
400 -|
DOR = 1:200
200
250 300
EX Wavelenath (nm)
350
400
300
EX Wavelength (nm)
600
400 -
DOR a 1:100
:«d
600
I
DOR = 1:20
400
6
200
250
350
EX Wavelength (nm) EX Wavelength (nm)
Figure F6. Light Oil Category - Hibernia crude oil with dispersant EEMs. Colored contours represent intensity, scaled to maximum
fluorescence peak (red).
IA-E12PG00037 Final Report
Page 245
-------
EPA/600/R-16/152
Appendix F
600 -¦
600
DOR = 0
I
400 -
400 -
250 300 350
EX Wavelength (nm)
DOR = 1:200
250 300 350
EX Wavelength (nm)
600
400
DOR = 1:100
400
200
250
300
350
400
EX Wavelength (nm) EX Wavelength (nm)
Figure F7. Light Oil Category - MC252 (Discoverer Enterprise) crude oil with dispersant EEMs. Colored contours represent
intensity, scaled to maximum fluorescence peak (red).
IA-E12PG00037 Final Report
Page 246
-------
DOR = 0
EPA/600/R-16/152
Appendix F
400 h
250 300 350
EX Wavelength (nm)
DOR = 1:200
DOR = 1:100
A
300
EX Wavelength (nm)
DOR = 1:20
200
250
300
350
400
EX Wavelength (nm) EX Wavelength (nm)
Figure F8. Light Oil Category - MC252 (Generic) crude oil with dispersant EEMs. Colored contours represent intensity, scaled to
maximum fluorescence peak (red).
IA-E12PG00037 Final Report
Page 247
-------
EPA/600/R-16/152
Appendix F
600
DOR = 0
DOR = 1:200
*
250 300 350
EX Wavelength (nm))
250 300 350
EX Wavelength (nm))
DOR = 1:20
DOR = 1:100
I
t
°
250 300 350
EX Wavelength (nm))
300
EX Wavelength (nm)
Figure F9. Light Oil Category - Scotian Shelf Condensate crude oil with dispersant EEMs. Colored contours represent intensity,
scaled to maximum fluorescence peak (red).
IA-E12PG00037 Final Report
Page 248
-------
EPA/600/R-16/152
Appendix F
400 -
DOR = 0
600 -
DOR = 1:200
400 -
200 250
300
EX Wavelength (nm)
350 400 200
600 -
300
EX Wavelength (nm)
350 400
DOR = 1:100
DOR = 1:20
400
%
200
250
350
400
200
250
300
350
400
EX Wavelength (nm) EX Wavelength (nm)
Figure F10. Light Oil Category - Sea Rose crude oil with dispersant EEMs. Colored contours represent intensity, scaled to maximum
fluorescence peak (red).
IA-E12PG00037 Final Report
Page 249
-------
EPA/600/R-16/152
Appendix F
DOR = 0
400
DOR = 1:200
a
200
250 300 350
EX Wavelength (nm)
400
250 300 350
EX Wavelength (nm)
DOR = 1:100
a
400 •
K
&
DOR = 1:20
200
400
200
250 300 350
EX Wavelength (nm)
400
250 300 350
EX Wavelength (nm)
Figure Fll. Light Oil Category - Terra Nova crude oil with dispersant EEMs. Colored contours represent intensity, scaled to
maximum fluorescence peak (red).
IA-E12PG00037 Final Report
Page 250
-------
600 -
DOR = 0
O
200
250 300 350
EX Wavelength (nm)
600
DOR = 1:100
400
©
200
250 300 350
EX Wavelength (nm))
EPA/600/R-16/152
Appendix F
DOR = 1:200
<0
400 200
600
250 300 350
EX Wavelength (nm))
400 -
DOR = 1:20
oi
400
203
250 300 350
EX Wavelength (nm))
I
400
Figure F12. Medium Oil Category - ANS (Alaskan North Slope) crude oil with dispersant EEMs. Colored contours represent
intensity, scaled to maximum fluorescence peak (red).
IA-E12PG00037 Final Report Page 251
-------
EPA/600/R-16/152
Appendix F
DOR = 0
DOR = 1:200
250 300 350
EX Wavelength (nm))
250 300 350
EX Wavelength (nm)
DOR = 1:100
a
400 -
DOR = 1:20
6
200
250
300
20O
EX Wavelength (nm) EX Wavelength (nm))
Figure F13. Medium Oil Category -10% Weathered ANS (Alaskan North Slope) crude oil with dispersant EEMs. Colored contours
represent intensity, scaled to maximum fluorescence peak (red).
IA-E12PG00037 Final Report
Page 252
-------
600 ¦
600 —|
DOR = 0
400
EPA/600/R-16/152
Appendix F
DOR = 1:200
300
EX Wavelength (nm)
250 300 350
EX Wavelength (nm)
600
DOR = 1:100
400
DOR = 1:20
EX Wavelength (nm) EX Wavelength (nm)
Figure F14. Medium Oil Category - Heavy IFO-120 oil with dispersant EEMs. Colored contours represent intensity, scaled to
maximum fluorescence peak (red).
IA-E12PG00037 Final Report
Page 253
-------
400
DOR = 0
I
400
&
250 300 350
EX Wavelength (nm)
200
DOR = 1:100
400 H
EPA/600/R-16/152
Appendix F
DOR = 1:200
o
I
250 300 350
EX Wavelength (nm)
400
DOR = 1:20
200 250 300 350 400 200 250 300 350 400
EX Wavelength (nm) EX Wavelength (nm)
Figure F15. Medium Oil Category - Heavy IFO-180 oil with dispersant EEMs. Colored contours represent intensity, scaled to
maximum fluorescence peak (red).
IA-E12PG00037 Final Report
Page 254
-------
EPA/600/R-16/152
Appendix F
600
DOR = 0
600
DOR = 1:200
400 -
250 300 350
EX Wavelength (nm)
200
250 300 350
EX Wavelength (nm)
DOR = 1:100
<3>
600 -
DOR = 1:20
400 -|
200
250 300 350
EX Wavelength (nm)
400
200 250 300 350 400
EX Wavelength (nm)
Figure F16, Medium Oil Category - Heidrun crude oil with dispersant EEMs. Colored contours represent intensity, scaled to
maximum fluorescence peak (red).
IA-E12PG00037 Final Report Page 255
-------
EPA/600/R-16/152
Appendix F
DOR = 0
600 -h
f
400 -
DOR = 1:200
o
200
250 300 350
EX Wavelength (nm)
400
250 300 350
EX Wavelength (nm)
600
DOR = 1:100
400 -
w
1
DOR = 1:20
; Jft
250
400
EX Wavelength (nm) EX Wavelength (nm)
Figure F17. Medium Oil Category - Lago crude oil with dispersant EEMs. Colored contours represent intensity, scaled to maximum
fluorescence peak (red).
IA-E12PG00037 Final Report
Page 256
-------
600
EPA/600/R-16/152
Appendix F
DOR = 0
DOR = 1:200
400 -|
200
250 300 350
EX Wavelength (nm)
600
250 300 350
EX Wavelength (nm)
400
DOR = 1:100
600
DOR = 1:20
250
200
250
300
350
400
EX Wavelength (nm) EX Wavelength (nm)
Figure FIB. Medium Oil Category - Mesa crude oil with dispersant EEMs. Colored contours represent intensity, scaled to maximum
fluorescence peak (red).
IA-E12PG00037 Final Report
Page 257
-------
600
400
DOR = 0
600
400
EPA/600/R-16/152
Appendix F
DOR = 1:200
o
*
200
250 300 350
EX Wavelength (nm))
400
250 300 350
EX Wavelength (nm))
DOR = 1:100
DOR = 1:20
400 ¦
A
EX Wavelength (nm)) Wavelength (nm))
Figure F19. Medium Oil Category - Santa Clara crude oil with dispersant EEMs. Colored contours represent intensity, scaled to
maximum fluorescence peak (red).
IA-E12PG00037 Final Report
Page 258
-------
EPA/600/R-16/152
Appendix F
600 H
DOR, = 0
600 h
DOR = 1:200
200
250 300 350
EX Wavelength (nm)
200
600
250 300 350
EX Wavelength (nm)
400
400
DOR = 1:100
600 -
DOR = 1:20
400 -
200
250
300
350
400
200
250
300
350
400
EX Wavelength (nm) EX Wavelength (nm)
Figure F20. Medium Oil Category - Vasconia crude oil with dispersant EEMs. Colored contours represent intensity, scaled to
maximum fluorescence peak (red).
IA-E12PG00037 Final Report
Page 259
-------
EPA/600/R-16/152
Appendix F
600
500
400 ¦
300 -
DOR = 0
600 -h
400 -
200
250 300 350
EX Wavelength (nm»
400
DOR = 1:200
200
250 300 350
EX Wavelenath
400
600 -i
DOR = 1:20
DOR = 1:100
400 -\
200
250 300
EX Wavelength (nm))
350
400
EX Wavelength (nm))
Figure F21, Heavy Oil Category - Belridge crude oil with dispersant EEMs. Colored contours represent intensity, scaled to
maximum fluorescence peak (red).
IA-E12PG00037 Final Report
Page 260
-------
EPA/600/R-16/152
Appendix F
600
DOR = 0
400 -|
400
DOR = 1:200
0
300
EX Wavelength (nm)
350
200
250 300
EX Wavelength (nm)
350
400
400
600
DOR = 1:100
DOR = 1:20
400 -|
a
200
250
300
350
400
EX Wavelength (nm) EX Wavelength (nm)
Figure F22. Heavy Oil Category - Hondo crude oil with dispersant EEMs. Colored contours represent intensity, scaled to maximum
fluorescence peak (red).
IA-E12PG00037 Final Report
Page 261
-------
EPA/600/R-16/152
Appendix F
600
DOR = 0
600 -
¦
DOR = 1:200
f
If
5 400 -
o
i
o
250 300 350
EX Wavelength (nm)
DOR = 1:100
6
•i
200
1 i ¦ i • i
250 300 350
1 i
400
EX Wavelength (nm)
400 -
250 300 350
EX Wavelength (nm)
DOR = 1:20
4>
200
250 300
EX Wavelength (nm)
350
400
Figure F23. Heavy Oil Category - IFO-300 crude oil with dispersant EEMs. Colored contours represent intensity, scaled to
maximum fluorescence peak (red).
IA-E12PG00037 Final Report
Page 262
-------
600
400
DOR = 0
200
250 300
EX Wavelength (nm»
350
600
DOR = 1:100
200
250
300
350
400
400
600
400
400
200
EPA/600/R-16/152
Appendix F
DOR = 1:200
i.0>
250 300 350
EX Wavelength (nm)
DOR = 1:20
4
EX Wavelength (nm)) EX Wavelength (nm))
Figure F24. Dilbit Oil Category - Access Western Blend oil with dispersant EEMs. Colored contours represent intensity, scaled to
maximum fluorescence peak (red).
IA-E12PG00037 Final Report
Page 263
-------
600 •
600 -
DOR = 0
EPA/600/R-16/152
Appendix F
DOR = 1:200
400
250 300 350
EX Wavelength (nm)
400
t
200
250 300 350
EX Wavelength (nm)
400
600
DOR = 1:100
DOR = 1:20
200
250
300
350
400
?
400
250
300
350
400
EX Wavelength (nm) Wavelength (nm)
Figure F25. Dilbit Oil Category - Cold Lake oil with dispersant EEMs. Colored contours represent intensity, scaled to maximum
fluorescence peak (red).
IA-E12PG00037 Final Report
Page 264
-------
EPA/600/R-16/152
Appendix G
IAA No. E12PG00037
Final Report
Dispersant Effectiveness, In-Situ Droplet Size Distribution and Numerical Modeling to Assess
Subsurface Dispersant Injection as a Deepwater Blowout Oil Spill Response Option
and
Evaluation of Oil Fluorescence Characteristics to Improve Forensic Response Tools
APPENDIX G VDROP-J and JETLAG Numerical Plume Modeling
Report on Simulated Oil Plume Advanced Numerical Modeling using VDROP-J and JETLAG
Prepared by:
Michel C. Boufadel, PhD, PE and Feng Gao
New Jersey Institute of Technology
Civil and Environmental Engineering
University Heights, Newark, NJ 07102
973-596-5657
Michel.boufadel@njit.edu
IA-E12PG00037 Final Report
Page 265
-------
EPA/600/R-16/152
Appendix G
Introduction
The goal was to develop the needed tools to simulate the horizontal release of oil from a
horizontal orifice. The models were to be calibrated (i.e., trained) by simulating the release of oil
from a 2.4 mm orifice in the BIO tank (32 m long, 0.6 m wide, and 2.0 m tall, but contains water
at a depth of 1.5 m). The BIO tank contains also currents at an approximate speed of 5.0 cm/s
along the duration of the jet.
The orifice had an elevation of 0.25 m from the inside bottom of the tank, and thus the distance
between it and the water surface is 1.25 m. The diameter of the orifice was 2.4 mm. The oil was
Alaskan North Slope (ANS) whose density p = 866 kg/m3, dynamic viscosity n=11.5 cp, and
interfacial tension with water a =0.02 N/m. The oil mass flow rate was 58g (approximately 0.06
liter per second, 3.6 liter per minute, around 1.0 gpm). The oil temperature in the canister was 80
°C while the water temperature in the tank was 15 °C. Due to the short duration of the
experiments and the relatively small volume of released oil, it is unlikely that the oil temperature
raised the water temperature by a measurable amount.
Based on the volumetric oil flow rate and the diameter of the orifice, the average oil exit velocity
is 13.3 m/s. In the absence of dispersant, the Reynolds and Weber numbers are:
Rc pV.D (866)(13.3)(0.0024)^2 500
pi 0.0113
(1)
We = py^D = (866)(13.3)2 (0.0024) =
cr 0.02
(2)
Another important number is the Ohnsorge number given as:
0, = ^L = —!L- = 0.055 ,3)
Re (poD)°5 (866 x 0.02 x 0.0024)°5
Based on Figure 1 of Johansen et al. (2013), the resulting jet is in the atomization regime (the blue
dot in the graph).
IA-E12PG00037 Final Report
Page 266
-------
EPA/600/R-16/152
Appendix G
1
r
0
c
0
1 01
3
C
•
0>
0
(A
1 0.01
jC
o
0.001
1 10 100 1000 10000 100 000
Reynolds number, Re
Figure 1: Experimental conditions plotted in the diagram of the Ohnesorge vs. Reynolds number as
obtained byJohansen et al. (2013).The injection rate varied from 0.1 to 20 L/min, with nozzle diameters
varied from 0.5 to 20 mm. The oil viscosity is presumed to be 5 cP. The thick line in the diagram shows
the boundary between transitional and atomization breakup. The blue disk represents the experimental
conditions of the BIO tank (this study).
Technical Approach
There are two major challenges with modeling oil jets. The first is the hydrodynamics of the jet
and the other is the formation of droplets. However, we approach this problem first from an
engineering point of view where we attempt to understand the engineering properties of the jet,
and then by zooming in on the hydrodynamics of the jet and the movement of oil droplets. For
this purpose, we use first the models VDROP-J and JETLAG, which provide the average
hydrodynamics properties along the centerline of the jet. The model VDROP-J provides also the
centerline droplet size distribution. In the second step, we use the models Fluent a Computational
Fluid Dynamics (CFD) model and the model NEM03D for tracking the individual oil droplets.
The layout of the document is as follows: The next Section addresses the modeling using the
engineering approach. The following section addresses the detailed hydrodynamics of the plume
and the movement of individual oil droplets as they interact with their surroundings. In that
Section, we address the droplet size distribution within the plume, which would help to design
future experiments.
~ 01 L/M
-•-02 Un*
i I
" K
A Al 1 1—.
-VOlMw
2.0 Ur*
-SO Urn*
100
1 I ^
\
\
\
\
95
mm
in I
;
tvti
\ 1 m
y
>V-»
—
(— af\
-
-*i
' '4 "
-*-200 Umi
Si
m ~
*
10 mr
* *
0
Jr
If?! *
~ ¦ V'
M
-»
—*
20 mm +
4
DwSpi
•
••••
IA-E12PG00037 Final Report
Page 267
-------
EPA/600/R-16/152
Appendix G
Engineering Properties of Jet Hydrodynamics
For jet hydrodynamics, we use herein the models VDROP and JETLAG to capture the behavior of
the plume. While the model VDROP provides the centerline velocity based on correlations from
the literature, the model JETLAG provides the average velocity within the plume. To allow
comparison, we used the Gaussian approximation for the velocity profile across the jet, and we
computed the peak (centerline) velocity from JETLAG. The resulting velocity along the centerline
of the plume is reported in Figure 2 for VDROP and for JETLAG in the presence of a current Ua=3.0
cm/s and in the absence of current (Ua=0 cm/s). The agreement is very good between the two
cases lending further credence that VDROP-J is compatible with numerical models of plumes such
as JETLAG.
10 -
I
£•
8
1 i
O
e
H
3
c
O
U
0.1
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
Distance to the jet exit along the plume trajectory (m)
Figure 2: The velocity along the centerline of the jet/plume using VDROP-J and JETLAG. The velocity
along the centerline of the jet/plume using VDROP-J and JETLAG. Note that the agreement was
particularly good considering that no fitting was conducted, rather each model was run with the
parameters stated in the Problem Statement Section.
The holdup is defined as the ratio of the volume of oil at a particular location in the plume to the
total volume (oil+water) at that location. The value of the holdup affects the droplet size
distribution, and reflects the intensity of water entrainment into the jet/plume; a large holdup
value results in a high rate of coalescence between droplets. Figure 3 reports the holdup along
the centerline using VDROP-J and JETLAG, and one clearly notes that the agreement is good
further reflecting the compatibility of VDROP-J with JETLAG for evaluating the mass of oil in the
jet/plume. The holdup decreased sharply from around 10% at few centimeters from the pipe exit
to almost 0.5% after 2.5 m. Thus at 2.5 m, 99.50% of the fluid in the plume is made up of water.
BIO Tank Experiment
Centerline velocity from JETLAG (Ua=0 cm/s)
I
Centerline velocity from JETLAG (Ua=3 cm/s)
\.
\\
— — Centerline velocity from VDROP-J
\V
\V\
\V.
\\.
\V ..
vX
Tu..
IA-E12PG00037 Final Report
Page 268
-------
EPA/600/R-16/152
Appendix G
l
o.i -
~ 0.01
o
X
0.001
0.0001
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
Distance to the jet exit along the plume trajectory (m)
Figure 3: Holdup (ratio of oil volume to total volume of fluid) as function of distance from the pipe exit.
Note the rapid decrease, which required a logarithmic scale for the holdup.
BIO Tank Experiment
\
Centerline velocity from JETLAG (Ua=0 cm/s)
\
A
Centerline velocity from JETLAG (Ua=3 cm/s)
Centerline velocity from VDROP-J
s
Figure 4 shows the simulation results from JETLAG for the shape of the plume. It shows the
centerline along with the lower and upper bound of the plume along the vertical plan (note that
the plume is circular, and thus three dimensional). The location of the LISST used in the
experiments is also reported, and its results will be discussed later in this document.
1.5
1
|
a. 0.5
Q
0
-0.5
0 0.5 1 1.5 2 2.5 3
Distance to the jet exit along the plume trajectory (m)
Figure 4: Plume centerline and boundaries based on the model JETLAG. The LISST was located within
the plume near the lower edge of the plume.
X LISST location
Plume centerline from JETLAG
— -Plume boundary from JETLAG
IA-E12PG00037 Final Report
Page 269
-------
EPA/600/R-16/152
Appendix G
Droplet Size Distribution
Using only the parameters from the Problem Statement, we obtained the breakage constant
Kb=0.1 from the correlation in Zhao et al. (2014). We then used VDROP-J to predict the droplet
size distribution at the location of the LISST (given in Figure 5), and we compared it to the
observed data from the LISST in Figure 4, which shows an exceptional agreement. The only
adjustable parameter was the initial droplet size, selected at 500 microns herein. The initial size
of the droplets results from the so-called primary breakup, and it depends on shear flow near the
orifice and thus cannot be predicted by VDROP-J, which relies on turbulence away from
boundaries.
0.9
0.6
~ Experimental data
> 0.4
VDROP-J modeling results
U 0.2
0
100
200
300
500
400
600
Droplet diameter (micron)
Figure 5: Droplet size distribution obtained from VDROP-J and from the LISST (location reported in Figure
3) in the absence of dispersant.
Based on the good agreement between the model VDROP-J and experimental data, one can
predict the DSD at various locations in the plume, as illustrated in Figure 5; it is clear that the DSD
changes drastically in the first meter of its trajectory, and the DSD does not change much
afterward. This is reasonable as the mixing energy decreases rapidly with distance from the
orifice (to the power "-4").
IA-E12PG00037 Final Report
Page 270
-------
EPA/600/R-16/152
Appendix G
¦LISST
~ Experimental data
LISST location (2.6 m downstream distance)
downstream distance = 0.2 m
downstream distance = 0.4 m
downstream distance = 0.6 m
0
100 200 300 400 500 600 700 800
Droplet diameter (micron)
Figure 6: The droplet size distribution at various locations from the orifice along the centerline. The DSD
essentially reached the steady state distribution within less than a meter.
The impact of dispersant was evaluated by premixing the oil with dispersant at the DOR of 1:20,
and releasing oil at the same rate as before (see Problem Statement). The experimental and
modeling results are reported in Figure 7. In the VDROP-J model, the same parameters used
earlier were used again with the exception of the interfacial tension, which was calibrated to be
0.0013 N/m, a reduction of 15 folds. The results in Figure 6 show a reasonable fit of the model to
the data. The fit is good for larger sizes, but it is relatively poor at the smaller sizes. To better
illustrate this discrepancy, we used a different scale in the lower panel of Figure 7. The
discrepancy is due to the fact that the oil and dispersants were premixed resulting in the so-called
tip-streaming, whereby the oil peels off from the droplet without the action of mixing. It is a pure
chemical process that VDROP-J is not designed to handle.
~
~
~ Experimental data
VDROP-J modeling results
0
100
200
300
400
500
600
Droplet diameter (mieron)
IA-E12PG00037 Final Report
Page 271
-------
EPA/600/R-16/152
Appendix G
i
0.9 -
§ 0-8 -
I 0.7 -
h
| 0.6
5
© 0.5 -
o
> 0.4
J5
| 0.3-
U 0.2 -
0.1 -
0
1 10 100
Droplet diameter (micron)
Figure 7: Droplet size distribution in the presence of dispersant obtained at the location of the LISST in
Figure 4. The difference between the two graphs is that the lower one has a logarithmic scale for the
size axis.
Computational Fluid Dynamics
Evaluating the hydrodynamic properties of an oil jet (e.g., a blowout) is very difficult for a variety
of reasons including: 1) The oil jet velocity is very large at the orifice (e.g., around 15 m/s for our
experiment), and drops sharply with distance, and 2) there are multiple phases present, namely
oil, water, and oil droplets in water, and 3) the turbulent energy and the presence of eddies
prevents accurate evaluation of average speeds. For example, when using an acoustic Doppler
velocimeter (ADV), one obtains different readings depending on the phase passing in front of the
ADV. Hence, one needs a robust method to study the oil jet underwater numerically to
complement experimental measurement.
We used computational fluid dynamics (CFD) to simulate the experiment of oil jet carried out at
Bedford Institute of Oceanography in Canada. In the CFD simulation, the computational domain
consisted of 2,777,029 nodes. The turbulence model k-s was selected, as it is most appropriate
for such problems because of its usefulness of application in free-shear layer flows, calculation
stability, and relative easiness of convergence (Aronson et al., 2000; Pope, 2000; Wilcox, 1998).
The model relies on solving two equations to model the turbulent kinetic energy and the
turbulence dissipation rate. The k-s model belongs to the family of RANS (Reynolds Average
Navier Stokes) models, which aim to solve for the average flow field of turbulent flow (Eaton and
Johnston, 1981).
In the present simulation, the achievement of steady state was observed based on monitoring
the conservation of mass (e.g. the inflow rate of the oil equates the rate of the outflow of oil) with
a maximum of 1% difference. A mesh independent study was performed to ensure that the
simulation results does not change with the further refinement of mesh. This was done by
~ Experimental data
VDROP-J modeling results
IA-E12PG00037 Final Report
Page 272
-------
EPA/600/R-16/152
Appendix G
monitoring the average pressure at the outlet to see whether it changes with mesh refinement.
After approximately 10,000 iterations, the simulation reached steady state with first order of
accuracy as the mass conservation information was monitored. All simulations were run in
parallel on 20 processors located on the NRDP Computational Laboratory.
Figure 8 shows results of CFD where the profile of the plume that was modelled based on an oil
flow rate of 1.0 L/second. The purple lines represent the contours of the holdup, which is equal
to the volume of oil divided by the volume of fluids in a given control volume. The holdup at the
exit is 1.0, and one notes that it decreases to a few percent and even lower with 0.5 m. The
contours of the velocity magnitude show velocities larger than 0.2 to 0.3 m/s within the plume
but a sharp decrease at the edge of the plume. One also shows the velocity vectors outside of
the plume reflecting the entrainment of fluids (i.e., water) to the plume, especially near the exit
to the tank.
Figure 9 shows the edge of the plume (delineated using 10% of the velocity) along with three
locations for obtaining cross sections. Figures 10, 11, and 12 show cross sections of the holdup
(lines) and velocity magnitude.
Figure 10 shows that the cross section is more or less circular as the exit point to the tank was
circular. The lack of perfect circularity is due mesh discretization. It appears that the bottom of
the tank, which was at 25 cm below the exit, has an effect on the general flow circulation, as the
flow vectors below the jet are different from those above it. But it is not sure if the bottom has
any measurable effects on the hydrodynamics within the plume.
IA-E12PG00037 Final Report
Page 273
-------
EPA/600/R-16/152
Appendix G
VELOCITY (MS):
1 1.5 2
Horizontal distance (m)
Figure 8: The profile of the plume through a vertical plane passing through the center axis. The figure
shows the contour of the velocity magnitude (flooded colors) and the holdup (volume of oil to total
liquid volume), using purple lines. The arrows indicate the velocity component in the plan of the figure.
The length of the velocity vector does not represent the magnitude but only represents the direction.
IA-E12PG00037 Final Report
Page 274
-------
EPA/600/R-16/152
Appendix G
c
O,
Ci
0.5 -
B surface 2
/
~
surface 3
A
surface 1
-0.5
0
0.5
1.5
2
2.5
Distance to the jet exit along the plume trajectory (m)
Figure 9: The locations of cut cross sections along the plume trajectory. The cross sections are numbered
as surface-1, surface-2 and surface-3 as they are distant away from the orifice. "Surface-1" is 0.5 m away
from the orifice and perpendicular to the horizontal direction. "Surface-2" is centered at 0.81 m above
the jet orifice and 45°to the tank bottom. "Surface-3" is parallel to the horizontal, and 10 cm below the
water surface. The green curve denotes the center line of the jet plume and the dotted orange lines
denote the edge of the plume (which is defined by 10 % of the centerline velocity magnitude). The cross
sections are cut while being at downstream locations (i.e., one sees the jet coming toward them).
IA-E12PG00037 Final Report Page 275
-------
EPA/600/R-16/152
Appendix G
VELOCITY (M/S): 0,04 0.1 0.18 0.26
Horizontal distance (m)
Figure 10: Surface-1 (Purple lines indicate the contour of the holdup, continuous flood contour indicates
the velocity magnitude. The arrow vectors indicate the velocity vectors that parallel to the present
surface)
Figure 11 shows the cross section B-B' (see Figure 9). One notes that the plume is no longer
circular in comparison to Figure 10. Also, the center of the plume is closer to the top portion of
the plume than it is to the bottom of the plume. First, note that the velocity in this cross section
is going in the direction of B-B', which means it is upward. We believe the non-circular shape of
the plume in Figure 11 is due to two complementary processes: 1) The close distance to the top
portion of the plume is most likely due to buoyancy which has more effects on the center of the
plume than on the edge of the plume because the center of the plume has more oil in it.
Therefore, the buoyancy of the center of the plume is larger than that of the top, causing the
center to be closer to the top of the plume. The converse occurs for the bottom of the plume. 2)
As buoyancy pulls the plume upward, we believe that the lower edge of the plume is carried
further out due to inertia.
IA-E12PG00037 Final Report
Page 276
-------
EPA/600/R-16/152
Appendix G
0.5
0.4
?
i 0.3
O
o
0
>- 0.2
0.1
0 0.1 0.2 0.3 0.4 0.5
X dicrection (m)
Figure 11: Surface-2 (Purple lines indicate the contour of the holdup, continuous flood contour indicates
the velocity magnitude. The arrow vectors indicate the velocity vectors that parallel to the present
surface)
Figure 12 shows the cross section C-C' where the velocity vectors are outward (i.e., away from
the plume). Notice that the width of the plume is determined by the width of the tank (i.e., 0.60
m). Notice also that the velocity magnitude did not drop a lot from Figure 11, which is probably
due to the fact that the plume reached the water surface and is thus confined to move
horizontally. In this figure, the effect of inertia discussed for Figure 11 is more prevalent, as one
notes that the distance between the center of the plume and the downstream edge of the plume
(i.e., toward B) is much larger than in the opposite direction. Note that the downstream edge of
the plume here is the lower edge of the plume of Figure 11.
VELOCITY (MS): 0.02 0.04 0.06 0.08 0.1 0.12 0.14
IA-E12PG00037 Final Report
Page 277
-------
EPA/600/R-16/152
Appendix G
0.6
0.5
0.4
£
C
o
z 0.3
u
¦5
>
0.2
0.1
0 0.1 0.2 0.3 0.4 0.5 0.6
X direction (m)
Figure 12: Surface-3 (Purple curve indicates the contour of the holdup, continuous flood contour
indicates the velocity magnitude. The arrow vectors indicate the velocity vectors that parallel to the
present surface)
The eddy diffusivity is defined as;
k2
D = 0.9 —
£
(4)
where k is the kinetic energy due to turbulence per unit mass and s is energy dissipation rate per
unit mass (watt/kg), and it reflects the intensity of turbulent mixing. The unit of D is m2/s.
Figure 13 shows the eddy diffusivity profile in a vertical plane passing through the center axis of
the plume. The eddy diffusivity is important for the mixing of oil and water within the plume, and
it was used later to predict the movement of individual oil droplets. One notes that the plume
has a core of high mixing surrounded by a layer of relatively small mixing. This is probably due to
the fact that the edge of the plume is "constrained" by the entrainment of water into it, a process
VELOCITY (MS):
IA-E12PG00037 Final Report
Page 278
-------
EPA/600/R-16/152
Appendix G
that tends to reduce mixing. However, the entrainment does not seem to affect the core of the
plume.
EDDY DIFFUSIVITY (rrf-2/s): 0.00015 0.0002 0.0004
Figure 13: The profile of eddy diffusivity of the plume (Purple lines indicates the contour of the holdup,
flood contour indicates eddy diffusivity. The arrow vectors indicate the velocity vectors that parallel to
the present surface. The length of the velocity vector does not represent the magnitude but only
represent the direction.
1 1.5 2
Horizontal distance (m)
Figures 14, 15, and 16 report the eddy diffusivity at the cross sections used earlier (see Figure 9).
One notes that the eddy diffusivity magnitude in Figure 14 (cross section 1) and Figure 15 (cross
section 2) is comparable. However, one notes a large decrease for Figure 16 (cross section 3).
This is probably because the eddy diffusivity is due to the k-s turbulence model, and turbulence
was unhindered between cross section 1 and cross section 2. However, cross section 3 was only
0.10 m below the water surface where turbulent eddies were quashed by the water surface,
which acted as a boundary.
IA-E12PG00037 Final Report
Page 279
-------
EPA/600/R-16/152
Appendix G
EDDY DIFFUSIVITY (MA2yS): 0.0001 0.0002 0.0003 0.00045
Horizontal distance (m)
Figure 14: Surface-1, purple lines indicate the contour of the holdup, continuous flood contour indicates
eddy diffusivity. The arrow vectors indicate the velocity vectors that parallel to the present surface
EDDY DIFFUStVITY (MA2/S): 5E-05 0.0001 0.0003 0.0005
X dicrection (m)
Figure 15: Surface-2 (Purple lines indicate the contour of the holdup, continuous flood contour indicates
eddy diffusivity and the arrow vectors indicate the velocity vectors that are parallel to the present
surface)
IA-E12PG00037 Final Report
Page 280
-------
EPA/600/R-16/152
Appendix G
0.6
0.5
0.4
I
o
¦o
>-
0.2
0.1
°0 0.1 0.3 0.4 0.5 0.6
X direction (m)
Figure 16: Surface-3 (Purple lines indicates the contour of the holdup, flood contour indicates eddy
diffusivity and the arrow vectors indicate the velocity vectors that parallel to the present surface)
The value of the energy dissipation rate e (watt/kg) is commonly needed to determine the
breakup of oil droplets and their coalescence. Although, we have not conducted the breakup of
oil droplets within the current framework, we provide herein the values of s for future
(imminent) application. Figure 17 shows the profile contour of s , where the values drop sharply
with the centerline distance (note the logarithmic scale of the contours).
EDDY DIFFUSIVITY (MA2/S) : 5E-05 0.0001 0.0003 0.0005 0.0007
IA-E12PG00037 Final Report
Page 281
-------
EPA/600/R-16/152
Appendix G
TURBULENT DISSIPATION RATE J/KG.S : 1 E-05
0.5 1 1.5 2 2.5 3
Horizontal distance (m)
Figure 17: The profile of turbulent dissipation rate in the plume. Purple lines indicates the contour of
the holdup, continuous flood contour indicates turbulent dissipation rate. The arrows indicate velocity
vectors components that are parallel to the page surface. The length of the velocity vector does not
represent the magnitude but only represent the direction.
Figure 18,19, and 20 show the contours of s in cross sections 1, 2, and 3, respectively (see Figure
9 for reference). While the maximum value is at or near the center, one notes a sharp decrease
of s between cross sections. The decrease is much sharper than that of the eddy diffusivity, which
is probably because the turbulent kinetic energy k behaved more or less similar to s, and thus the
ratio in Eq. 4 remained more or less uniform with distance from the source.
IA-E12PG00037 Final Report
Page 282
-------
EPA/600/R-16/152
Appendix G
TURBULENT DISSIPATION RATE (J/KG.S): 0.001 0.002 0.005 0.006
Horizontal distance (m)
Figure 18: Surface-1 (Purple lines indicates the contour of the holdup, continuous flood contour indicates
eddy diffusivity. The arrow vectors indicate the velocity vectors that parallel to the present surface)
I I
TU RBU LE N T DIS SIP ATI ON RATE (J/KG S):
1E-05
9E-05 0.0001
0.00018
X direction (m)
Figure 19: Surface-2 (Purple lines indicates the contour of the holdup, continuous flood contour indicates
turbulent dissipation rate. The arrow vectors indicate the velocity vectors that parallel to the present
surface)
IA-E12PG00037 Final Report
Page 283
-------
EPA/600/R-16/152
Appendix G
TURBULENT DISSIPATION RATE (J/KG.S): 1E-05 2E-05 5E-05 0.0001
0.6
0.5
0.4
B
c
>-
0.2
0.1
2.3
^ i 11 i tin 11 id^M
0 0.1 0.2 0.3 0.4 0.5 0.6
X direction (m)
Figure 20: Surface-3 (Purple lines indicate the contour of the holdup, continuous flood contour indicates
turbulent dissipation rate. The arrow vectors indicate the velocity vectors that parallel to the present
surface)
Figure 21 reports the variation of s along the centerline of the plume, where one notes a sharp
decrease with distance (note the logarithmic scale on the vertical). Theoretical arguments in
water jets revealed that e decreases proportional to x 4, where x is the centerline distance from
the source.
IA-E12PG00037 Final Report Page 284
-------
EPA/600/R-16/152
Appendix G
1. 00E+03
1. 00E+02
1. 00E+01
1. OOE+OO
1. 00E-01
1. 00E-03
1. 00E-04
1. 00E-05 L_
0.00 0.25 0.50 0.75
1. 00 1. 25
1. 50
1.75 2.00 2.25 2.50
Horizontal disntance (m)
Figure 21: Turbulent dissipation rate along the centerline of the plume (The turbulent dissipation rate
decreases sharply when exiting the orifice.)
Particle tracking using Lagrangian method coupled with CFD
In the experiment of oil jet underwater, we observed that some oil droplets exit the plume and
rise up individually. This interesting phenomenon gives rise to how the oil droplets behave when
exiting the jet orifice. In the present model, we aim to understand how oil droplets behave after
exiting the orifice (e.g., how the large oil droplets exit the plume and oil droplets of which
diameters remain within the plume. However, CFD (Computational Fluid Dynamics) does not
consider the behavior of single droplets. For example, RANS (Reynold Averaged Navier Stokes
Equations) in CFD considers the oil and water fully mixed, which is different from the real
situation, where they are in different phases: oil droplets in water.
We modelled herein the trajectories of oil droplets using a Lagrangian method coupled with CFD
data. The flow field was obtained from the CFD simulations (Section I), and used as input to our
particle tracking model NEM03D.
The velocity of water is:
U = u.x + v. v + w.z
(5)
The velocity of the particle is:
IA-E12PG00037 Final Report
Page 285
-------
EPA/600/R-16/152
Appendix G
Up = up.x + vp.y + wp.z
(6)
For particle tracking:
^ = UP+R^2DAt
dt
(7)
Where
~rp=xp:x+Y~y+zp:z
(8)
Equation (3) is thus written in each of the coordinate directions:
dX„
—~ = u„ +
dt
RpDxAt
(9a)
D I
\-f = Vp+R^
(9b)
dZ.
—L = w +
dt
RpD^At
(9c)
Where Dx, Dy, and Dz are the eddy diffusivities computed in Section I (Eq. 4). The velocities
up,vp and wp are given based on the momentum equation. In the x direction:
—- = (—)*(Ucos/?-t/ cos a) = (—)*(u-u )
dt zd Td
(10)
Momentum equation in the y direction:
dv f f
—- = (—) * (U sin P — U sin a) = (—) * (v- v )
dt Td zd
(11)
IA-E12PG00037 Final Report
Page 286
-------
EPA/600/R-16/152
Appendix G
Momentum equation in the z direction:
dw f p — p
^L = (A)*(W_W ) + J^_^L*g
dt Pw
(12)
U sin p
Ccos fi
U sine;
U; cos a
Figure 22: The angles of the velocities in cylindrical coordinates for the water velocity U (left panel) and
for the oil droplet velocity U (right panel).
The parameter Td is the Stokes drag coefficient given as:
r Pq*D2
d I** Mo
(13)
The parameter fx is a correction for the Stokes coefficient to account for situations where the
p ^ vi ^ D
flow is not laminar, but when the droplet Reynolds number Re = —— is less than 100. It
Mw
is calculated based on the following equations (Miller et al., 1998):
i
fx = 1 + 0.0545 Re+ 0.1 * Re^ (1 - 0.03 * Re)
(14)
The slip velocity for the calculation of the Reynolds number is given by the Euclidean norm:
IA-E12PG00037 Final Report
Page 287
-------
EPA/600/R-16/152
Appendix G
f#, =11 u-up 11=
(w - up f + (¦v - wp f + (w - wp )2
1/2
(15)
By putting equation (5)-(14) into the NEMO 3D code, one can obtain the trajectories of the oil
droplets with different diameters.
The oil droplets with different diameters rise up not only because of the flow filed velocity, but
also due to the individual velocities of its own. As shown in Figure 23, oil droplets with larger
diameters rise ahead of the centerline of the plume, and oil droplets with a 1000 microns exist
the plume at around 0.5 m from the orifice in the horizontal direction. The present method can
predict at which point the oil droplets will exit the plume. In addition, when combined with LISST
data, one can also estimate how much percentage of oil droplets with large diameters will exit
the plume after released.
5
1. 25
¦¦¦¦ Edge of the plume
—1000 microns
1
—500 microns
0. 75
—250 microns
—100 microns
5
50 microns
0. 25
0
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75 3
Horizontal distance (m)
Figure 23: The trajectories of oil droplets with different diameters. The edge of the plume is defined by
10 % of the centerline velocity.
When the inertia terms are not estimated for oil droplets, the buoyancy effect is estimated by
terminal velocity directly (e.g., the right hand side of Equation (9a), (9b), (9c) is zero) while the
present method uses equation (5)-(14) to consider the effect of inertia and buoyancy.
Figure 24 shows the evaluation of the combined effect of inertia and buoyancy effect on oil
droplets when rising; the oil droplet trajectories rise faster when the inertial effects are not
accounted for. This is because the jet is horizontal and tends to propel the droplets horizontally.
IA-E12PG00037 Final Report
Page 288
-------
EPA/600/R-16/152
Appendix G
Thus, not accounting for the inertia of droplets overestimate their rise rate. The difference
becomes more pronounced for the large oil droplets, such as 500 and 1000 microns, which is
because the large droplets have a higher buoyancy (due to their large volume) and have larger
inertia due to their large mass. The results in Figure 24 demonstrate that one needs to account
for both the effects of inertia and buoyancy when considering the movement of oil droplets in
jets.
.5
1. 25
1
75
>.5
0. 25
0
0
Horizontal distance (m)
¦¦¦¦ Edge of the plume
—0 micron (no inertia)
—100 microns (no inertia)
^—500 microns (no inertia)
—1000 microns (no inertia)
—1000 microns
(with inertia and buoyancy)
—500 microns
(with inertia and buoyancy)
—100 microns
(with inertia and buoyancy)
—50 microns
(with inertia and buoyancy)
^—50 microns (no inertia)
Figure 24: Evaluation of the effects of inertia and buoyancy on oil droplets trajectories. In the absence
of inertia, oil droplets rise faster.
Figures 25 through 28 show the trajectories of oil droplets of various sizes. Figures 25 and 26
show identical results, which suggests that the mixing due to eddy diffusivity was large enough to
minimize the effect of buoyancy between 50 and 100 microns. However, the effect of buoyancy
seems to become important for droplets of size 500 microns (Figure 27) and 1,000 microns (Figure
28). For the latter, the droplet trajectories were outside of the 10% boundary of the plume.
IA-E12PG00037 Final Report
Page 289
-------
EPA/600/R-16/152
Appendix G
1. 5
1.25
1
0. 75
g 0.5
0. 25
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75
Hori zonta1 dis tance (m)
Figure 25: The trajectories of oil droplets when the diameter is 50 microns (with inertia and turbulent
diffusion)
8 0. 75
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75
Horizontal distance (m)
Figure 26: The trajectories of oil droplets when the diameter is 100 microns (with inertia and turbulent
diffusion)
IA-E12PG00037 Final Report
Page 290
-------
EPA/600/R-16/152
Appendix G
i
0 ' ' 1 ' 1 1 1 1 1 1 1—
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75
Horizontal distance (m)
Figure 27: The trajectories of oil droplets when the oil droplet diameter is 500 microns with inertia and
turbulent diffusion.
Horizontal distance (m)
Figure 28: The trajectories of oil droplets when the diameter is 1000 microns, with inertia and turbulent
diffusion.
IA-E12PG00037 Final Report Page 291
-------
EPA/600/R-16/152
Appendix G
References
Aronson, D., Chroneer, Z., Elofsson, P., Fellbom, H., 2000. Comparison between CFD and PIV
measurements in a passenger compartment. SAE Technical Paper.
Eaton, J., Johnston, J., 1981. A review of research on subsonic turbulent flow reattachment.
AIAA journal 19, 1093-1100.
Johansen, 0., Brandvik, P.J., Farooq, U., 2013. Droplet breakup in subsea oil releases - Part 2:
Predictions of droplet size distributions with and without injection of chemical
dispersants. Mar Pollut Bull 73, 327-335.
Miller, R., Harstad, K., Bellan, J., 1998. Evaluation of equilibrium and non-equilibrium
evaporation models for many-droplet gas-liquid flow simulations. International Journal
of Multiphase Flow 24, 1025-1055.
Pope, S.B., 2000. Turbulent flows. Cambridge university press.
Wilcox, D.C., 1998. Turbulence modeling for CFD. DCW industries La Canada, CA.
Zhao, L., Boufadel, M.C., Socolofsky, S.A., Adams, E., King, T., Lee, K., 2014. Evolution of droplets
in subsea oil and gas blowouts: Development and validation of the numerical model
VDROP-J. Marine pollution bulletin 83, 58-69.
IA-E12PG00037 Final Report
Page 292
-------
EPA/600/R-16/152
Appendix H
IAA No. E12PG00037
Final Report
Dispersant Effectiveness, In-Situ Droplet Size Distribution and Numerical Modeling to Assess
Subsurface Dispersant Injection as a Deepwater Blowout Oil Spill Response Option
and
Evaluation of Oil Fluorescence Characteristics to Improve Forensic Response Tools
APPENDIX H
Weber Number Scaling Numerical Modeling Prediction of Droplet Size Distribution from
Subsurface Oil Releases with and without Chemical Dispersants
By
Haibo Niu and Linlu Weng
Water Studies Laboratory
Department of Engineering
Dalhousie University (Truro Campus)
Truro, NS, Canada
SgiJ DALHOUSIE
W UNIVERSITY
IA-E12PG00037 Final Report
Page 293
-------
EPA/600/R-16/152
Appendix H
1 Introduction
The increased in offshore oil and gas exploration in deep waters increases the risk of deepwater oil spills.
One recent example is the Deepwater Horizon (DWH) incident in the Gulf of Mexico. Oil released from
subsurface blowouts breaks up into droplets and the sizes of these droplets have strong impacts on the
subsequent fate of oil in the marine environment (Chen and Yapa, 2007; Bradvik et al., 2013; Johansen et
al., 2013). With a density smaller than that of the ambient seawater, larger oil droplets will rise to the sea
surface more rapidly than smaller droplets and will reach surface closer to the spill location than the smaller
droplets. Better knowledge on droplet size distributions resulting from subsurface oil releases will help us
predict whether the oil will surface and if so, when and where and what the oil slick thickness be (Chen and
Yapa, 2003).
Currently, both our knowledge on the droplet size distributions and our capability to predict the
distributions are limited. Before the DWH incident, only very few experimental work have been conducted
to measure droplet size distribution from subsurface releases and only few studied the effects of chemical
dispersant on droplet sizes. Topham (1975) was probably the earliest work studying droplets from
subsurface releases and he has reported droplet size ranging from 0.5 mm (detection limit) to 3 mm for
Norman Wells crude and a peak diameter of 15 ^m for Swan Hills crude. The field experimental data from
the Canadian Arctic gathered by Dome Petroleum gave a range from 50 |im to 2.1 mm (Buist et al., 1981).
Masutani and Adams (2001) conducted jet experiments on an oil-water system using four types of crude oil
and studied the different modes of jet breakup. Johansen et al. (2003) was the only full-scale deep water
experiment, they observed that droplet sizes resulting from the release of diesel at 844 m depth were from
1 to 10 mm.
While DWH is the first oil spill occurring at significant depth (~1500m), it is also the first time where
chemical dispersants were directly injected into the subsurface oil release to enhance the dispersion of oil
over a large water column (Louis et al., 2011). A total of 18,379 barrels of dispersant were used at the DWH
incident (The Federal Interagency Solutions Group, 2010). When a chemical dispersant is added at the depth
of the wellhead, the surfactant is expected to break the oil into small droplets. The only available data on
the effects of dispersant on droplet sizes is Brandvik et al. (2013). Brandvik et al (2013) have studied the
effects of dispersant by using seven different dispersant-oil-ratios (DORs) and the peak droplet sizes were
found strongly affected by DORs.
Very few publications are available on predicting the droplet sizes. Chen and Yapa (2007) developed a
method based on the maximum entropy formalism using the "deepspill" experimental data. Currently, this
method is mainly applied to subsurface releases without chemical dispersant. However, the feasibility of
this method is yet to be validated in the case of subsurface release with chemical dispersant. More recently,
Johansen et al. (2013) have incorporated new experimental data for the subsurface release cases with
chemical dispersant application developed a modified Weber number approach to predict the droplet sizes.
Zhao et al. (2014) used the same data set with a droplet breakup rate approach. However, all of these
available approaches were based on one single set of experimental data on subsurface oil-dispersant
interaction (Brandvik et al., 2013) by using one type of oil (Oseberg Blend). There is an urgent need to
validate these models with extensive experimental data on more oil types.
IA-E12PG00037 Final Report
Page 294
-------
EPA/600/R-16/152
Appendix H
Furthermore, although it appears likely that subsurface in-situ use of chemical dispersants may be very
effective for countering deepwater oil spills, many uncertainties still exist. For example, assumptions of the
optimum DOR are based on empirical data mostly obtained from bench-scale experimental protocols that
have been designed for testing at standard temperatures and pressures (STP), whereas conditions at a
wellhead on the ocean floor or anywhere along a riser beneath the ocean surface could be significantly
different. Dispersant effectiveness as a function of dispersant type, oil type, and DOR must be better
understood for application in deepwater environments. Furthermore, the interaction of dispersant and crude
oil at depth under different turbulence regimes may also have significant implication in optimizing
operational performance of subsurface dispersant injection. Improved understanding of these issues should
provide better support in decision-making for subsurface dispersant application.
To fill the existing knowledge gaps, extensive experimental studies have been conducted in a flow-
through wave tank located at the Bedford Institute of Oceanography (BIO) with an underwater high flow
rate oil release system. Accordingly, the objective of this project is to: 1) analyze these newly gained
experimental data from BIO; 2) develop a method that can effectively predict the droplet size distributions
of oil released from subsurface, with and without application of chemical dispersant; and 3) incorporate the
newly developed method with an oil spill model to study its effects on fate and transport of oil from
subsurface releases.
2 Methodology
2.1 Maximum Entropy Formalism (MEF) Approach
Probability density function (PDF) such as Rosin-Rammler or Nukiyama-Tanasawa distribution, are
established correlations for the droplet size distribution. However, more theoretical foundations were needed
for these correlation. Maximum entropy formalism (MEF) approach was used by Chen and Yapa (2007) to
develop model for estimating oil droplet size distribution.
To estimate a droplet spectrum, the probability density function (PDF) needs to be connected to a
characteristic size (e.g. Smax, S30, or S32) (Chen and Yapa, 2007). Smax is the maximum droplet size, S30 is the
mass mean volume equivalent diameter, and S32 is the Sauter mean (volume surface) diameter. Smax is
determined by diameter of the nozzle D and the Weber number (We) as follows:
kDWe
max
(i)
By knowing Smax, S30 and S32 can be estimated as follows:
(2)
(3)
where/is PDF defined as:
IA-E12PG00037 Final Report
Page 295
-------
EPA/600/R-16/152
Appendix H
/ = 38l exp - \ 3 - \813w1 - \ (S^u^ + S^B)
(4)
Where Si is nondimensional droplet diameter, ui is nondimensional droplet velocity. After solving the
Lagrangian multiplier Xt (/.a Xi, X2, X3), which are evaluated by several nonlinear constraint equations,
mentioned in the Chen and Yapa (2007) the droplet size number based distribution can be obtained in
Equation fN =
dN
d(S/S3 0)
= A exp
<5.
30
(5
-B
f 8 ^C
\S30J
( 5 (Chen and Yapa,
2007). It indicates that the droplet distributions are controlled by two tuning coefficients B and C:
/# -
dN
d(S/S30)
= A exp
30
<5
-B
f 8 ^C
(5)
Where /n is a number based probability density function, N is the droplet number, S30 is the volume mean
diameter, A is a term that accounts for normalization conditions. Their result seems to be less biased. Due
to the limited data, the effects of oil properties were neglected. The applicability of the formulation for
chemically dispersed oil will be tested in future study.
2.2 Droplet Breakup Approach
Maximum Entropy Formalism (MEF) Approach was widely used in flow atomization and spray;
there is less of consideration of oil property. Zhao et al. (2014) has developed a VDROP-J model which
considers the effects of both oil viscosity and oil-water interfacial tension (ITF). In a liquid-liquid dispersion
system, a population balance equation is proposed as follows:
d''(^',l) = Z ^Wi >')
01 j=i+1
V. J
V
droplet breakup
(6)
+ Z Z r(6/,' dk M6/,, , t) - niti,, t) J r(<, d, )i(d],/)
j=1 k=1 j=i+1
V. J
V
droplet coalescence
Where n is number concentration of droplets of diameter d, at a given time I. The term f->(dh dj is the
breakage probability density function (dimensionless)for the creation of droplet of diameter dj due to
breakage of droplets of (a larger) diameter dj, and g{dj) is the breakage frequency of droplets of diameter dj.
The first term represents the birth of droplets dt resulting from the breakup of droplets dj, while the second
term represents the death of droplets dt due to breakup into smaller droplets. For droplets coalescence, the
tQxmr(dk, dj) is the coalescence rate (m3/s). The first term of droplet coalescence represents the birth of
IA-E12PG00037 Final Report
Page 296
-------
EPA/600/R-16/152
Appendix H
droplets d, as a results of coalescence events occurring between droplets ch, and dj to form drops with the
size of di, while the second term represents deaths of droplets dj due to the coalescence of drops dj with all
other drops (including drops of size di themselves) to form larger drops.
The breakage rate g{di) is given by:
where Sed represents the collisional cross section of eddy and droplet (m2), ue is the turbulent velocity of an
eddy (m/s), uci is droplet velocity (m/s), ne is number concentration of eddies (number of eddies/m3), BE is
the breakup efficiency which is related with the IFT, dne is the number of eddies of size between uc and uj
are the velocities of eddies and droplet and Kb is a system-dependent parameter for droplet breakup, and
would need to be obtained by calibration to experimental data. Based on experimental data, the Kb was
found can be approximated by (Zhao et al., 2014):
where p is density (kg/m3), U is velocity (m/s), and D is droplet diameter (m) In Figure 1, an example is
given for the comparison ofVDROP-J with the experimental data (Brandvik et al., 2013). For a given release
condition (e.g., same oil type, discharge nozzle size, and exit velocity), same Kb (0.11 in this case) will be
obtained. Therefore, Equation 8 does not consider the effects of chemical dispersant on droplet sizes or
shape of the curves. To fit the droplet size distributions with model, other parameters such as ITF or known
dispersion efficiency must be used to adjust the shape of the curve. Both Zhao et al. (2014) and Johansen et
al. (2013) indicated IFTs (15.5, 0.05 and 0.09) from three experiments based on DOR of 0, 1:50, and 1:25,
respectively. The measured IFT (0.09) for DOR=l:25 is actually higher than the IFT (0.05) for
DOR= 1:50.This is against to the IFT fitting produced by Zhao et al. (2014) which indicated that the higher
IFT would lead to a closer curve to the untreated condition (DOR = 0). The author may use estimated
efficiencies of 10% and 80% for the case of DOR=1:50 and 1:25 during the fitting, respectively.
(7)
(8)
~ fctiKTirartHjJ duta - m • ifaiprreinls
4 hxpcnmcnul (fain DOR 130
X hiprfintL'iiul data - 1X)K J 25
rr*«ili» - rmtllffurnanli
— •MihI. Iiiiu - IX)R 1.50 vmh l(K4 ctlkicacy
— MinHmy iw»H> - IX>R 12? wilh cHllic*.?
0 S> UK) 150 2 250 300 350 4(HJ 454) 51H J
Droplet diameter (|Jinl
Figure 1: Comparison of oil droplet size distribution between VDROP-J and experimental data.
IA-E12PG00037 Final Report
Page 297
-------
EPA/600/R-16/152
Appendix H
(Source: Zhao et al. 2014)
2.3 Modified Weber Number Approach
There is no validation conducted for the MEF and droplet breakup approaches for droplet size prediction
with chemical dispersant application. However, chemical dispersion is one of the important technologies in
offshore oil spill response, and promising in responding to deepwater release. Thus, an approach in
predicting droplet size with chemical dispersant is desired.
Johansen et al. (2013) has proposed a modified Weber number approach for such purpose based on the
conventional Weber number approach by Wang and Calabreses (1986). In Johansen et al. (2013), Weber
number scaling law was used to fit their experimental data, expressed as:
'/D = Awe-35 (9)
where d' is characteristic droplet diameter (m) D is the nozzle diameter (m), A id a factor of proportionality
and We = pU2Dla is the Weber number; p is density of the liquid in the jet (oil) (kg/m3) U is the exit velocity
(m/s), and o is the interfacial tension between oil and water (N/m or kg/s2). However, this simple Weber
scaling law only fit well on DOR=0, for other DOR experiments, this scaling law do not fit it. Based on
these available data, a new prediction model (modified weber number) is used for oil droplet size distribution
with and without chemical dispersant.
The modified Weber number, We*, is defined as follows:
We
We = ? wl (10)
\ + BVi(d50/D)m
where We is the Weber number, Vi = We/Re is the viscosity number, d5o is the median droplet diameter
(m), D is the nozzle size (m), B is an empirical coefficient determined by experimental analysis. The relative
droplet size d50/D can be expressed as:
fe/Z)) = A(We*f'5 (11)
where A is an empirical constant. Based on the data from Brandvik (2013) and Johansen et al. (2013) the
value of A and B can be determined as A = 15.0 and B = 0.8.
Once clso is determined, the droplet size distribution can be estimated using either lognormal or Rosin-
Rammler distribution. Johansen et al. (2013) has concluded that Rosin-Rammler (Equation 12) distribution
gives better fit of experimental data overall.
V(d) = 1 -exp \-0.693(d/d50f\ (12)
where V(d) is the cumulative distribution, and a is the spreading-parameter.
IA-E12PG00037 Final Report
Page 298
-------
EPA/600/R-16/152
Appendix H
Although the mathematical formulations of three methods described above are of different level of
complexity, all three methods require two or three tuning coefficients determined from regression. It seems
that the efficiency and accuracy of droplet size prediction from these three methods are more or less the
same. Comparatively, the complexity of the modified Weber number approach is lower than the other two,
leading to advantage in real-world application. Therefore, the modified Weber number approach is selected
in this study to fit the new experimental data with performance validation.
3 Prediction of Droplet Size Distribution
3.1 Experimental settings
A series of experiments of droplet size measurement for two types of oils (IFO-120 and ANS) have been
conducted by the COOGER in BIO. The current flow rate for the experiments is set to 1 cm/s and the oil
temperature is set to 80 °C. The detailed settings of the other parameters (i.e., oil amount, water temperature,
injection time, and flow in the tank) are listed in Tables 1 and 2. 24 experiments were conducted for each
types of oil by consideration of seasonal conditions (spring and summer). As shown in Tables 1 and 2, the
experiment No.l to 12 were set based on spring condition with slightly lower water temperature (mostly
lower than 10°C) . In contrast, the experiments of No. 13 to 24 were set based on summer condition with
warm water temperature (mostly higher than 10°C). The "R" marked in the experiment No. denoted a
repeated experiment with slightly adjusted conditions (e.g., different DOR). In addition, some of the
repeated experiments (i.e., 6R, 7R, 10R and 11R), which were scheduled in spring but not conducted due to
abnormal weather conditions with rising water temperature, were actually conducted late fall.
There were four dispersant-oil ratios (0, 1:250, 1:100, and 1:25) forthe spring condition. Comparatively,
the settings of dispersant-oil ratios are slightly different from which in the spring condition, which are 1:200
and 1:20.
Table 1: Experimental settings for droplet size analysis for IFO-120
No.
Factors
Measurements
Oil
DOR
Date
Oil Amount
(g)
Water
Temperature
(°C)
Injection
Time
(sec)
Flow in the
Tank
(gpm)
Injection
Pressure
(psi)
1
IFO-120
0
9-Jun-14
145.2
13.0
5
600
40
2
IFO-120
1:250
20-Jun-14
199.6
12.2
7
600
62
2R
IFO-120
1:200
04-Dec-14
208.2
6.7
7
600
60
IA-E12PG00037 Final Report
Page 299
-------
EPA/600/R-16/152
Appendix H
3
IFO-120
1:100
20-Jun-14
213.9
13.2
7
600
62
4
IFO-120
1:25
1 l-Jun-14
179.1
12.8
9
600
40
4R
IFO-120
1:20
05-Dec-14
219.6
5.6
10
600
30
5
IFO-120
0
17-Jun-14
275.1
12.0
7
600
62
6R
IFO-120
1:200
04-Dec-14
215.6
6.6
8
600
60
7R
IFO-120
1:100
10-Dec-14
239.3
7.5
8
600
60
8
IFO-120
1:25
1 l-Jun-14
255.8
13.2
9
600
40
8R
IFO-120
1:20
05-Dec-14
243.3
5.4
10
600
60
9
IFO-120
0
17-Jun-14
359.6
12.7
7
600
62
10R
IFO-120
1:200
04-Dec-14
221.7
6.6
8
600
60
11R
IFO-120
1:100
17-Dec-14
N/A
4.9
10
600
60
12
IFO-120
1:25
16-Jun-14
354.8
12.5
9
600
62
12R
IFO-120
1:20
10-Dec-14
204.8
6.8
9
600
60
13
IFO-120
0
12-Sep-14
256.8
14.9
7
600
60
14
IFO-120
1:200
15-Sep-14
279
13.5
8
600
60
15
IFO-120
1:100
16-Sep-14
336.2
14.0
8
600
60
16
IFO-120
1:20
17-Sep-14
315.9
14.7
7
600
60
17
IFO-120
0
12-Sep-14
293.3
14.7
8
600
60
18
IFO-120
1:200
15-Sep-14
331.8
13.8
8
600
60
19
IFO-120
1:100
16-Sep-14
353.8
14.7
7
600
60
20
IFO-120
1:20
17-Sep-14
345.6
15.2
7
600
60
21
IFO-120
0
12-Sep-14
303.6
15.2
8
600
60
22
IFO-120
1:200
15-Sep-14
363.3
14.0
8
600
60
23
IFO-120
1:100
16-Sep-14
352.6
14.7
7
600
60
IA-E12PG00037 Final Report
Page 300
-------
EPA/600/R-16/152
Appendix H
24
IFO-120
1:20
17-Sep-14
380
16.0
7
600
60
Note: R indicates repeated experiment.
Table 2: Experimental settings for droplet size analysis for ANS
No.
Factors
Measurements
Oil
DOR
Date
Oil Amount
(g)
Water T
(°C)
Injection
Time
(sec)
Flow in the
Tank
(gpm)
Injection
pressure
(psi)
1
ANS
0
22-May-14
208.0
11.4
4
600
40
2
ANS
1:250
23-May-14
280.0
10.6
5
600
40
2R*
ANS
1:200
02-Dec-14
290.5
6.4
5
600
40
3
ANS
1:100
23-May-14
284.5
11.2
5
600
40
4
ANS
1:25
26-May-14
283.0
8.4
5
600
40
4R
ANS
1:20
03-Dec-14
287.2
6.8
5
600
40
5
ANS
0
26-May-14
279.3
8.4
5
600
40
6
ANS
1:250
30-May-14
279.7
7.7
5
600
40
6R
ANS
1:200
02-Dec-14
335.0
6.1
5
600
40
7
ANS
1:100
30-May-14
276.3
8.5
5
600
40
8
ANS
1:25
02-Jun-14
277.4
9.4
5
600
40
8R
ANS
1:20
03-Dec-14
297.2
7.0
5
600
40
9
ANS
0
02-Jun-14
281.4
9.7
5
600
40
10
ANS
1:250
06-Jun-14
281.0
10.3
5
600
40
10R
ANS
1:200
17-Dec-14
344.5
5.4
5
600
40
11
ANS
1:100
06-Dec-14
276.8
10.7
5
600
40
IA-E12PG00037 Final Report
Page 301
-------
EPA/600/R-16/152
Appendix H
12
ANS
1:25
09-Jun-14
280.6
12.5
5
600
40
12R
ANS
1:20
10-Dec-14
295.7
7.3
5
600
40
13
ANS
0
05-Sep-14
303.7
17.7
5
600
40
14
ANS
1:200
08-Sep-14
295.2
16.0
5
600
40
15R
ANS
1:100
10-Sep-14
304.3
13.8
5
600
40
16
ANS
1:20
10-Sep-14
291.9
14.7
5
600
40
17
ANS
0
05-Sep-14
299.6
18.1
5
600
40
18
ANS
1:200
08-Sep-14
297.7
16.2
5
600
40
19
ANS
1:100
09-Sep-14
283.4
15.3
5
600
40
20
ANS
1:20
ll-Sep-14
289.6
14.1
5
600
40
21
ANS
0
08-Sep-14
297.1
15.1
5
600
40
22
ANS
1:200
09-Sep-14
281.8
14.2
5
600
40
23
ANS
1:100
10-Sep-14
284.4
13.4
5
600
40
24
ANS
1:20
ll-Sep-14
285.8
13.6
5
600
40
25
ANS
1:50
ll-Sep-14
316.2
17.6
6
600
40
Note: R indicates repeated experiment.
3.2 Measured Droplet Size Distributions
The droplet size distributions of IFO-120 based on different DOR and seasonal conditions are shown in
Figures 2 tolO. In addition, the droplet size distributions of ANS are listed in Figures 11 to 21. The ranges
of DOR for the ANS experiment (Figures 11 to 20) were the same as which for the IFO-120. A series of
experiments with DOR =1:50 are currently conducting by COOGER (one set of result is listed in Figure
21), further analysis will be conducted for this case.
As shown in Figure 2, the distribution and corresponding median of the droplet size distribution from
experiment No. 1, 5, and 9 based on same type of oil (IFO-120), DOR = 0, seasonal condition (spring,
similar water temperature) but different oil amount, injection time, and injection pressure. In addition, the
first two experiments (No. 1 and 5) have the same peak diameter (dp = 259 (jm), but slightly different dso
(258 (j,m in No. 1 and 176 in No. 5). The third experiment showed smaller dso (186 |im) and dp (100 |im).
This may be caused by relatively large plume or more smaller droplets caught by LISST.
IA-E12PG00037 Final Report
Page 302
-------
EPA/600/R-16/152
Appendix H
In summer condition, the experiment 13 and 21 has same dp (391 |im) and very similar dso (263 |im in
No. 13 and 264 in No. 21), while the experiment 17 has a slightly smaller dso (192 |im) and dp (293 (jm)
with similar settings from which in spring condition. However, the droplet size distributions form No. 13
and 21 are not completed due to the limited measuring window of the LISST. Thus, the data from these two
experiments will not be included in the further analysis. Comparing results from summer and spring, the dp
and dso from summer is relatively higher than which from spring. Since the only significant different setting
from summer to spring is the water temperature, which may be another factor that affecting the oil droplet
size.
The droplet size distributions with similar conditions but different DOR in spring are listed in Figures 4,
5, 7, and 9. By comparing the dp and dso in the experiments with different DOR, it indicates that the change
of droplet size is relatively insignificant with DOR from 0 to 1:100 (Figures 2, 5, and 7). However, a
significant decrease droplet size is observed with DOR increasing from 1:100 to 1:20. Therefore, there are
may be a threshold of DOR dosage that significantly changes the effects of dispersant on droplet size.
The droplet size distributions with similar conditions but different DOR in summer are listed in Figures
6, 8, and 10. Forthe warm cases (14, 18, and 22), experiment No. 18 showed strong effects oftruncation due
to the maximum diameter can be measured by LISST instrument was 500jjxn. Both experiments 14 and 22
have similar but slightly smaller dso compared with untreated cases (No.13 and 21), but the dp from warm
water are much smaller. This indicates that dispersant started to play a role in this case but the effects are
not very strong.
For the case of DOR= 1:100 with spring condition, the shape of the distribution and calculated dso and dp
in experiment No.3 are very similar to the untreated case and DOR=1:250 cases of experiment No.5 and
No.2, and the dispersant did not show a strong effects on the droplet distribution (Figure 7). Similar as
experiment No. 18, experiment No.7R also showed strong effects of truncation. For the summer condition
cases (Figure 8), although dso and dp for experiment No. 15 does not change significantly compared with
DOR=1:200 cases (e.g. No. 14), d50 from experiment No. 19 and 23 are much smaller and the overall oil
concentration are much higher. This indicates high dispersant effectiveness.
For the case of DOR=l:25 (or 20) with spring condition (Experiment No. 4, 8, and 12), while the first
experiment showed very low oil concentration compared with the other two experiments. The second and
the third experiments repeated very well with much higher oil concentration and smaller dP (128 |im for No.
8 and 104 |im for No. 12) and dso (99 |im for No. 8 and 93 (j,m for No. 12) (Figure 9). Similar trends can be
observed forthe summer condition cases (Figure 10).
Compared the droplet size distributions from spring to summer conditions with same DOR, the droplet
sizes from the results in summer experiment are significantly smaller than which in winter condition. The
only known parameter that is different from the spring and summer condition with same DOR is the water
temperature. Therefore, temperature may help facilitate the effect of dispersant on reduction the droplet size.
In general, the results from the cases with spring and summer conditions indicate very high effectiveness of
chemical dispersants.
Compared with the droplet sizes of IFO-120, the droplet sizes of ANS are significantly smaller. The
droplet size distributions from three experiments (No. 1, 5 and 9) with untreated ANS in spring conditions
are shown in Figure 11. The dp (75 - 88 |im) and dso (68 - 81 |im) are different but not significant in these
IA-E12PG00037 Final Report
Page 303
-------
EPA/600/R-16/152
Appendix H
three experiments. However, there is an abnormal peak observed in No. 5, which may due to unknown
effects (further experiments and analyses shall be needed). The droplet sizes from the experiments with
summer condition (No. 13, 17 and 21) (Figure 12) are similar (dp = 104 - 128 |im dsn = 89 - 101 (jm) and
higher than which from experiments with spring condition.
For the case of DOR= 1:250 with spring condition, three experiments (No. 2, 6 and 10) have been
conducted (Figure 13). In addition, three repeated experiments (No. 2R, 6R and 10R) with DOR=l :200 have
also been conducted (Figure 14). Experiment No. 2 shows two dp in one distribution which may due to
influences from environment, and thus is difficult to be analyzed. Nevertheless, the droplet size distributions
from No. 6 and 10 are highly similar with same dp (75 |im) and similar dso (63 |im for No. 6 and 66 |im for
No. 10). The repeated experiments with DOR= 1:200 show similar situation, the shape of the distribution
and calculated dso and dp are very similar between No. 2R and 6R; while the situation of No. 10R is similar
to which of No. 2. Compared with the untreated case (dp = 75 - 88 |im). the smaller dp (< 75 |im) in
DOR=1:200 (or 250) show the effect of dispersant on oil droplet distribution. The droplet size distributions
from the experiments (14, 18, and 22) based on summer condition are highly similar with identical dp(75
(mi) and very close dso (64 - 65 |im). Experiments with DOR=1:200 (or 250) have slightly smaller dso (64 -
65 |im) compared with untreated cases (13, 17 and 21) (dso = 68-81 (im), as well as the dp (75 |im for DOR
= 200 or 250 and 75 - 88 |im for DOR = 0). This indicates that the effect of dispersant on ANS is more
significantly than which on IFO-120 with very insignificant change of droplet size from DOR = 0 to 1:200.
Three experiments have been conducted for DOR=l: 100 with spring condition (Figure 16). The dso (55
- 58 (jm) in experiment No.3, 7 and 11 are smaller than the DOR=l :200 (or 250) cases (dso = 64 -65 |im)
while dp (75 |im) are same. For the summer condition cases (Figure 17), dso and dp for experiment No. 15R
does not change significantly compared with DOR=1:200 cases (e.g. No. 14), while the ones from
experiment 19 and 23 are relatively smaller.
For the case of DOR=l: 25 with spring condition, the dp (12 |im) and dso (3 - 10 |im) from corresponding
experiments (No. 8 and 12) are significantly lower than which from the experiments with DOR=l: 200 and
1: 100; while data from experiment No.4 appears abnormal distribution and could not be analyzed (Figure
18). The situations from the repeated experiments (No. 4R, 8R, and 12R) with DOR = 1:20 (Figure 19) are
very similar to the original one (DOR = 1:25). Furthermore, similar trends can be observed for the summer
condition cases (Figure 20). Figure 21 is atrial experiment of DOR=1:50 which is done in the summer of
2015, which indicate the droplet size of ANS is steadily decrease with increase DOR. It should also be noted
that the droplet size distributions are significantly different from the experiments with DOR = 1:20 or 25 to
the others. This may be cause by over dose of chemical dispersant. The other peaks in the distributions
(Figures 18, 19 and 20) may be caused by the over-dosed dispersant or the unknown background
components that were affected by the dispersant.
In general, the chemical dispersant plays an importance role in reduce the droplet size of ANS no matter
in spring or summer conditions. The effectiveness of dispersant in reducing droplet size is higher on ANS
than which on IFO-120. There may be thresholds for the dose of chemical dispersant to some oils (e.g., IFO-
120) but will need further experiments to analyze. There may also be over dose of dispersant to some oils
(e.g., ANS) when the DOR is high, eventually affecting the droplet size distribution. Future experiment will
also need for this particular issue.
IA-E12PG00037 Final Report
Page 304
-------
EPA/600/R-16/152
Appendix H
0.1
I
0.09
0.08
o 0.07
0.06
0.05
u 0.
E
a) No. 1
d50=258.362um
dp=259um
¦ w ma m oi o
r ld ^ rv oo r
J ffl fO m lO N M '
Droplet size (um)
0.1
i
0.09
0.08
° 0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
b) No.5
d50=176.59um
dp=259um
j oi 1 ui m r
r fM r
0
c) No.9
d50=99.63um
dp=186um
r o cti m
COMflOimmiMr
m oi m ci vd
Droplet size (um)
Figure 2: Experimental droplet size distribution of IFO-120 based on experiment a) No. 1. b) No. 5, and c) No.9
with DOR = 0 in spring condition
IA-E12PG00037 Final Report
Page 305
-------
EPA/600/R-16/152
Appendix H
a) No.13
d50=263.27um
dp=391um
Droplet size (um)
0.1
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
b) No.17
d50=192.71um
dp=293um
droplet size (um]
0.1
"ao
5. 0.09
3
£ 0.08
3
| 0.07
ro
= 0.06
.1 0.05
c) No. 21
d50=263.95um
dp=391um
ro
| 0.04
__ i
0
rOfM<»oocn'3-iOCTii/imtvfor^cM,=j-oorviDCTiCT)fO(Nrir\i
Droplet size (um)
Figure 3: Experimental droplet size distribution of IFO-120 based on experiment a) No. 13, b) No. 17, and c) No. 21
with DOR = 0 in summer condition
IA-E12PG00037 Final Report
Page 306
-------
EPA/600/R-16/152
Appendix H
0.1
i
. 0.09
' 0.08
a) No.2
d50=195.14um
dp=259um
— 0.06
_o
o 0.05
£ 0.04
c
0
¦3- LT] l£> r»- 00 r
HmcoaimNMninNroMN-jcoMitaiaifniNHN
>i'!fii5CTinii<
-------
EPA/600/R-16/152
Appendix H
0.1
i
. 0.09
' 0.08
o 0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
a) No.2R
d50=293.51m
dp=293um
Droplet size (um)
o 0.025
E
b) N0.6R
d50=312.31um
dp=319um
COfMOOOOCTl^fvDCTir
r^^ro c
N rri ^ Lri
£ 0.03
0
a 0.02
E
1 0.01
>
0
c) No.lOR
d50=408.29um
dp=462um
roiNcococn't(£)
-------
EPA/600/R-16/152
Appendix H
0.1
j 0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
a) No.14
d50=230.22um
dp=293um
r iri r- 03 v
Droplet size(um)
0.1
>
0.09
' 0.08
0.07
0.06
0.05
b) No.18
d50=224.61um
dp=462um
3.02
3.01
m'tf-^DCTi;0ifvi*:,!rNlr>ir^^o<3-i£ff>(NT-iroooa}Lnr^r--LnLnr^mr--rNi'tcor--i£a>ror\ir-
^^ro^^^^^orvi^f^CT^roKrviodLj-iroro^odS^^^^jJCj^jrnn
't UllDMOr
H
-------
EPA/600/R-16/152
Appendix H
0.1
a) No.3
d50=220.51um
dp=259um
.09
.08
.07
.06
.05
.04
.03
.02
.01
0
H fNl
CD ID
Droplet size (um)
0.1
•§ 0.09
b) No.7R
~Z 0.08
d50=325.85um
| 0.07
dp=462um
= 0.06
o
| 0.05
S 0.04
c
£ 0.03
o
V 0.02
E
1 001
0
fO(NooooCTi^-iBCTifMrHfOoocimr^r>.ir)mr^for^.(N'^oor-i£aic»fO(NrH(N
N ffl ^lOifll^ODHHHHHNNWlffl^iOVDNOO
Droplet size (um)
Figure 7: Experimental droplet size distribution of IFO-120 based on experiment a) No. 3, and b) No. 7, with DOR
= 1:100 in spring condition
IA-E12PG00037 Final Report
Page 310
-------
EPA/600/R-16/152
Appendix H
0.09
Ij-
3- 0.08
c
O 0.07
£
= 0.06
,o
| 0.05
£ 0.04
0
j ^ to CT) ro 1^ . , ~
(HdHlSNrflm-tullfll-iu
droplet size distribution (um)
0.1
00
0.09
c) No.23
3
0.08
d50=163.84um
C
dp=219um
o
0.07
E
ro
0.06
o
c
0.05
o
2
0.04
C
O)
0.03
£
i—i
o
0.02
i_rl n—i
01
! 1
£
0.01
| 1
o
0
>
fOrMOOOOCT1*tlDCTl(NrHfOOO-fOI^
-------
EPA/600/R-16/152
Appendix H
0.1
"53
< 0.09
~ 0.08
c
| 0.07
ro
- 0.06
o
| 0.05
a) No.4
d50=122.23um
dp=186um
ro
t 0.04
c
a»
£ 0.03
o
a> 0.02
|
-f 0.01
>
0
ro
rorsiooooai'st-ioairsirHroooaiLnt^r^.LnLni^ror^rvi
r>:rSro*t(>ir>lr,l
0
miNMCoai'fiDaiNHniMaiLnsNinuiNniNiN
r~;r^oST,!rSr^r'lu^O(N'tiDCTiror--'(NCdLrirorri'tod
fslro "^Ln^Or>-0O*HrHrHrHrHfMfMrOfO"S-U1lDr>-0O
Droplet size (um)
104
128
157
186
219
259
293
332
391
462 [
Figure 9: Experimental droplet size distribution of IFO-120 based on experiment a) No. 4, b) No. 8, and c) No. 12
with DOR = 1:25 in spring condition
IA-E12PG00037 Final Report
Page 312
-------
!. 0.09
: 0.08
— 0.06
o 0.05
i "d- i/i to r-^ oo
HmcoCTiLr)r^r~-uiLnr~~ror-(NTtcor^y5aiqirn
j^uiairoSr^cOLfifrifri^rcd^I^l1"
^HHH!N(Nfnm'fl^l^D^COrtH'
U1 fll (fl fll ID
Droplet size(um)
0.1
5
\ 0.09
r 0.08
c) No.24
d50=52.38um
dp=63.3um
— 0.06
o
| 0.05
£ 0.04
c
ai
£ 0.03
ii 0.02
Droplet size (um)
Figure 10: Experimental droplet size distribution of IFO-120 based on experiment a) No. 16, b) No. 20, and c) No.24 with DOR =
1:20 in summer condition
IA-E12PG00037 Final Report
Page 313
-------
EPA/600/R-16/152
Appendix H
. 0.09
' 0.08
0.07
0.06
0.05
a) No.l
d50=81.9um
dp=88.2um
0.03
0.02
Droplet size (um)
0.1
00
0.09
3
TT 0.08
c
| 0.07
ro
— 0.06
o
g 0.05
S 0.04
c
0)
£ 0.03
o
V 0.02
E
-? 0.01
b) No.5
d50=70.51um
dp=74.7 um
) s id ffi ffi m r
D 0.05
o
S 0.04
S 0.03
o 0.02
o
<1J
£ 0.01
o 0
c) No.9
d50=68.13um
dp=88.2um
CO (N CO 00 CI *
*rintorNOOrH»H»-i*-i«-if«JiOO
¦coMnmaimrjHn
iNi/iWHinmmoiio
1 CM (N CM fO CO *r
Droplet size (um)
Figure 11: Experimental droplet size distribution of ANS based on experiment a) No. 1. b) No. 5, and c) No. 9 with
DOR = 0 in spring condition
IA-E12PG00037 Final Report
Page 314
-------
EPA/600/R-16/152
Appendix H
0.1
)
. 0.09
' 0.08
a) No.13
d50=88.67um
dp=104um
0.06
0.05
0.04
0.03
0.02
0.01
0
) 00 "=}¦ tD cnmr-.r-.Lr)Lnr--mr-.aicr)m.u-iu")r>.mr-c\i'tf,oor--i£>aiaim.(N*toor^iDaia^rO(NHrsi
'tiOiCMOi
<(Ni0rv-co'
Droplet size (um)
Figure 12: Experimental droplet size distribution of ANS based on experiment a) No. 13, b) No. 17, and c) No. 21
with DOR = 0 in summer condition
IA-E12PG00037 Final Report
Page 315
-------
EPA/600/R-16/152
Appendix H
a) No.2
d50= 398.7757um
dp=462um
¦UlliNCO
m n w m
Droplet size (um)
0.1
0.09
~ 0.08
c
o 0.07
E
ro
- 0.06
| 0.05
ro
£ 0.04
c
0
b) No.6
d50=62.96um
dp=74.7um
j co co en m cti t
1 -3- LT) ID r-- 00 T
(rnco(TiLnr~-r~-LnLnr~-rOf^-rsi'^cor~-uDa^a>m(NTHrM
!^>£iairor<«NCOirifOro^co3^22^lCSSSS
(^^rHrsifNmmTtLni£.r-.oo^HrH^rHrvJ,N,Nr0rOTt
Droplet size (um)
0.1
5
[ 0.09
' 0.08
c) No.10
d50=66.325um
dp=74.7um
£ 0.04
c
a>
£ 0.03
(mcooiiflNMniONrnMN^WMUCTicimiNt
^ lo tn r-- co t
J ^ ID Ol (fl N f
co i/i m rn co
<
-------
EPA/600/R-16/152
Appendix H
a) No.2R
d50=65.75um
dp=74.7um
-= 0.06
I 0.05
TO
h 0.04
c
V
g 0.03
0
£ 0.02
1
-i 0.01
ft 8 2S
JCOCOffifUSWNrtmKOliflNMflUl
J. pri ^ ^ ^ o
-------
EPA/600/R-16/152
Appendix H
0.09
— 0.08
C
o 0.07
= 0.06
o
| 0.05
T
S 0.04
c
a»
c 0.03
o
£ 0.02
o 0.01
0.1
*oB
^ 0.09
¦§r 0.08
C
o 0.07
= 0.06
0
1 0.05
£ 0.04
g 0.02
"I 0.01
>
0
a) No.14
d50=64.66um
dp=74.7um
mr-icoooHrsi
(NiOMHuicirnoim
HrlrlNNNfflfO't
0.1
< 0.09 c) No.22
S nnp d50=63.75um
dp=74.7um
o 0.07
£
droplet size (um)
Figure 15: Experimental droplet size distribution of ANS based on experiment a) No. 14, b) No. 18, and c) No. 22 with DOR =
1:200 in summer condition
IA-E12PG00037 Final Report
Page 318
-------
EPA/600/R-16/152
Appendix H
0.1
5
. 0.09
' 0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
a) No.3
d50=56.875um
dp=74.7um
icoijiLnr^r~-LnLnr~-mr--(N'
fuSCTJfoKrJoduSfOfii^oo0
Droplet size (um)
) Ol ffl Ol ic
!. 0.09
' 0.08
b) No.7
d50=55.49um
dp=74.7um
- 0.06
O 0.05
-r>-Lj-)Lf)rvmr~(Nttcor>-i£)i^:(I5CTifrir>.'(NOdirifOfO'd:Od2!^l22rdrjSf
^L/itfir^ooHHHHHNNfOrn^uivDr^oo
Droplet size (um)
c) No.ll
d50=57.589um
dp=74.7 um
LD ID 00 '
Droplet size (um)
Figure 16: Experimental droplet size distribution of ANS based on experiment a) No. 3, b) No. 7, and c) No. 11
with DOR = 1:100 in spring condition
IA-E12PG00037 Final Report
Page 319
-------
EPA/600/R-16/152
Appendix H
0.09
' 0.08
= 0.06
o 0.05
0.03
0.02
d) N0.15R
d50=63.61um
dp=74.7um
miNcocaoi'JiDci
-------
EPA/600/R-16/152
Appendix H
a) No.4
d50=9.534um
dp=12.1um
^y5airri[
-------
EPA/600/R-16/152
Appendix H
0.09
— 0.08
§ 0.05
| 0.02
o 0.01
a) No.4R
d50=2.34um
dp=12.1um
t co r*» to oi oi m r
¦ u"i r-- co '
Droplet size (um)
b) N0.8R
d50=2.739um
dp=12.1um
j ui co h m ai r
¦ m to r»- oo r
Droplet size (um)
c) No.lZR
d50=6.57um
dp=12.1um
2 0.07
= 0.06
h to cn
j ro to
r uo ui r~-' co
imcooiiflNNinmMflMN
j^^OCTiroSrJodirifriforfoo
IHrlrlNNfOfO^ifl^NCO
Droplet size(um)
fl'COSDOlCTirflNHN
Or-jLOCOrHUIOlfOCTlin
HHHHNNN(OfO*f
Figure 19: Experimental droplet size distribution of ANS based on experiment a) No. 4R, b) No. 8R, and c) No.
12R with DOR = 1:20 in spring condition
IA-E12PG00037 Final Report
Page 322
-------
EPA/600/R-16/152
Appendix H
Volume concentration/oil amount (uL/Lg)
ooooooooo
bbbbbbbboP
a) No.16
d50=7.987um
dp=12.1um
L
(V
r>
r>
(NCOCOCTlctiDailNr
ro Ll-I in CO rHr-
IHHHNNlflffl^uliBNCO
Droplet size (um)
Volume concentration/oil amount (uL/Lg)
ooooooooo
bbbbbbbboP
OMMWJSitnCT^VlOOkOI—»
b) No.20
d50=6.998um
1 dp=12.1um
Lt^
NMCOCl'JUJOlNp
NfntNNmiDor
m ^ LO l£l CO rt r
-imooaiu")r-.r--ir)ior--mr-.rM*tcQf--<£>CTia>ro
-------
EPA/600/R-16/152
Appendix H
0.1
w> 0.09
3- 0.08
E
o 0.07
E
= 0.06
No. 25
d50=47.77um
dp=53.7um
~ 0.04
c 0.03
o 0.01
i m oi is
g (N rn d
l Is- oo
HmooaiLnr^i^mmr-^rnr^oi^tcor
j-^tDaifoSojcoLrifororfcdS^^
i(N(NfOrO'3-Lninr~-co""T
Ifl ffl ffl ffl (M r
Droplet size (um)
Figure 21: Experimental droplet size distribution of ANS based on experiment with DOR = 1:50 in summer
condition
3.3 Data Fitting with Modified Weber Number Approach
Based on experimental settings (Table 1 and 2) and measured droplet size distributions (Figures 2 to 21),
as well as the additional measurements on oil viscosity and IFT, the Weber number (We), Viscosity number
(Vi) and Reynold number (Re) were calculated. The values of calculated and additional measured parameters
for IFO-120 and ANS are listed in Tables 3 and 4. By normalize the dsn with the preset nozzle size in the
experiments (D = 2.387 mm), the relationship between relative volume median droplet sizes (dWD) and
modified Weber number (We* in Equation 11) for corresponding oils can be determined as in Figures 22
and 23. In comparison purpose, the corresponding data for Oseberg Blend based on the SINTEF tower tank
experiments are also included in these figures.
As shown in Figure 22, for the treated IFO-120 crude oil with DOR< 1:100, the modified Weber number
approach fits the measured data IFO-120 well. The empirical constant A has been determined based on
Equation 11 with regression approach. The empirical constant A for IFO-120 with DOR < 1:100 is A = 5
which is significantly lower than the one for Oseberg Blend (A = 15, Johnsen et al., 2013). In the case of
DOR > 1:100, the value of regressed constant is A = 2.54 for IFO-120 and A = 8.7 for Oseberg Blend. It
indicates an about 45% of A values for both oils from DOR < 1:100 to DOR > 1:100.
The regressions of constant^ for ANS with different DOR conditions are listed in Figure 23. A reduction
of 45% of A values is observed for ANS from DOR < 1:100 to DOR > 1:100. It can be seen that the fitting
situation for the regression of IFO-120 is better than which of ANS. Nevertheless, the trends of A with the
change of DOR are consistent for IFO-120, ANS, and Oseberg Blend. Furthermore, the change of A values
may be caused by the significant reduction of IFT. For the Oseberg Blend, when the DOR changed from 0
to 1:100 to 1:25, the corresponding IFTs were reduced from 15.5 to 0.5 to 0.09 mN/m (Johansen et al.,
2013). However, the change of IFTs measured in the COOGER's experiments are from 46.78 (mN/m) to
56.97 (DOR=1:100), and 49.09 (1:20), which are much less significant than which from Johansen et al.,
2013 If similar magnitude of reduction as Johansen et al. (2013) is applied to IFO-120, the two fitted line
could get much closer. Therefore, besides the oil properties, measured IFT played a significant role in
determining the values of empirical constant A and it must be examined further.
IA-E12PG00037 Final Report
Page 324
-------
EPA/600/R-16/152
Appendix H
Table 3: Data analyses for droplet size distribution of IFO-120
No.
Factors
Parameters
Oil
DOR
Q
(L/min)
Viscosity
(mPa-s)
dso
(nm)
dp
(nm)
U
(m/s)
IFT
(mN/m)
We
Vi
Re
1
IFO-120
0
1.8063
44
230
259
5.6
46.78
1.55xl03
5.27
293.5
1R
IFO-120
0
2
IFO-120
1:250
1.7729
45
197.3
259
6.6
57.84
1.74xl03
5.14
338.0
2R
IFO-120
1:200
1.849
45
293.510
293
6.887
57.84
1.89x10s
5.36
352.556
3
IFO-120
1:100
1.8999
42
223.1
259
7.1
56.97
2.02x10s
5.22
388.1
A*
4
IFO-120
1:25
1.2373
40
122.2
186
4.6
49.09
9.96xl02
3.75
265.4
4R
IFO-120
1:20
1.365
40
195.310
462
5.058
49.09
1.21x10s
4.14
292.84
5
IFO-120
0
2.4435
44
176.6
259
9.1
46.78
4.08x10s
8.56
476.4
6R
IFO-120
1:200
1.676
45
312.310
319
6.241
57.84
1.55x10s
4.86
319.451
7R
IFO-120
1:100
1.86
42
341.750
462
6.927
56.97
1.94x10s
5.11
379.893
8
IFO-120
1:25
1.7672
40
98.9
128
6.6
49.09
2.02x10s
5.36
379
8R
IFO-120
1:20
1.513
40
177.920
293
5.634
49.09
1.49x10s
4.59
324.444
9
IFO-120
0
3.1941
44
100.4
186
11.9
46.78
6.97x10s
11.19
622.7
IA-E12PG00037 Final Report
Page 325
-------
EPA/600/R-16/152
Appendix H
10R
IFO-120
1:200
1.723
45
408.290
462
6.417
57.84
1.64xl03
4.99
328.489
11R
IFO-120
1:100
N/A
42
370.340
462
N/A
56.97
N/A
N/A
N/A
12
IFO-120
1:25
2.4511
40
93.8
128
9.1
49.09
3.91x10s
7.44
525.7
12R
IFO-120
1:20
1.415
40
211.340
293
5.27
49.09
1.3x10s
4.29
303.449
13*
IFO-120
0
2.281
44
263.3
391
8.5
46.78
3.13x10s
7.89
444.7
14
IFO-120
1:200
2.1684
45
230.2
259
8.1
57.84
2.29x10s
6.28
413.4
15
IFO-120
1:100
2.613
42
215.2
259
9.7
56.97
3.37x10s
7.17
533.7
16
IFO-120
1:20
2.8059
40
82.8
88.2
10.5
49.09
4.51x10s
8.52
601.8
17
IFO-120
0
2.2795
44
192.7
293
8.5
46.78
3.13x10s
7.99
444.5
CO
IFO-120
1:200
2.5788
45
224
462
9.6
57.84
3.24x10s
7.47
491.6
19
IFO-120
1:100
3.1426
42
179.8
259
11.7
56.97
4.88x10s
8.63
641.9
20
IFO-120
1:20
3.0697
40
69.38
74.7
11.4
49.09
5.40x10s
9.32
658.4
21*
IFO-120
0
2.3596
44
254.6
391
8.8
46.78
3.35x10s
8.27
460.1
22*
IFO-120
1:200
2.8236
45
245.9
293
10.5
57.84
3.88x10s
8.18
538.3
23
IFO-120
1:100
3.1319
42
167.8
219
11.7
56.97
4.85x10s
8.60
639.7
24
IFO-120
1:20
3.3753
40
52.6
63.3
12.6
49.09
6.53x10s
10.24
723.9
Table 4: Data analyses for droplet size distribution of ANS
IA-E12PG00037 Final Report
Page 326
-------
EPA/600/R-16/152
Appendix H
Factors
Parameters
No.
Q
Viscosity
dso
dp
U
IFT
Oil
DOR
(L/min)
(mPa-s)
(nm)
(nm)
(m/s)
(mN/m)
We
Vi
Re
1
ANS
0
3.617
7.2
81.9
88.2
13.471
63.97
5.84xl03
1.52
3852.38
2*
ANS
1:250
3.895
8.2
398.780
462
14.507
60.52
7.16xl03
1.97
3642.77
2R
ANS
1:200
4.041
8.2
65.750
74.7
15.051
60.52
7.71x10s
2.04
3779.38
3
ANS
1:100
3.958
8.3
56.875
74.7
14.740
55.94
8.00x10s
2.19
3656.72
A*
4
ANS
1:25
3.937
7.6
9.534
10.2
14.663
42.07
1.08x10s
2.64
3972.47
4R
ANS
1:20
3.995
7.6
2.340
12.1
14.880
42.07
1.08xl04
2.69
4031.43
5
ANS
0
3.885
7.2
70.512
74.7
14.471
63.97
6.74x10s
1.63
4138.34
6*
ANS
1:250
3.891
8.2
62.961
74.7
14.492
60.52
7.14x10s
1.96
3638.87
6R
ANS
1:200
4.66
8.2
64.140
74.7
17.357
60.52
1.02xl04
2.35
4358.32
7
ANS
1:100
3.844
8.3
55.487
74.7
14.316
55.94
7.54x10s
2.12
3551.38
8*
ANS
1:25
3.859
7.6
3.095
12.1
14.373
42.07
l.OlxlO4
2.6
3893.87
8R
ANS
1:20
4.134
7.6
2.739
12.1
15.398
42.07
1.16xl04
2.78
4171.8
9
ANS
0
3.915
7.2
68.131
88.2
14.580
63.97
6.84x10s
1.64
4169.46
10*
ANS
1:250
3.909
8.2
66.325
74.7
14.559
60.52
7.21x10s
1.97
3655.78
10R*
ANS
1:200
4.792
8.2
212.55
462
17.849
60.52
1.08xl04
2.42
4481.91
IA-E12PG00037 Final Report
Page 327
-------
EPA/600/R-16/152
Appendix H
11
ANS
1:100
3.851
8.3
57.589
74.7
14.341
55.94
7.57xl03
2.13
355.76
12*
ANS
1:25
3.904
7.6
6.301
12.1
14.538
42.07
1.03xl04
2.63
3938.78
12R
ANS
1:20
4.144
7.6
6.570
12.1
15.321
42.07
1.15xl04
2.77
4150.74
13
ANS
0
4.225
7.2
88.870
104
15.735
63.97
7.97xl03
1.77
4499.9
14
ANS
1:200
4.107
8.2
64.661
74.7
15.295
60.52
7.96x10s
2.07
3840.5
15R
ANS
1:100
4.233
8.3
63.604
74.7
15.766
55.94
9.15x10s
2.34
3911.2
16
ANS
1:20
4.061
7.6
7.987
12.1
15.124
42.07
1.12xl04
2.73
4097.4
17
ANS
0
4.168
7.2
97.212
128
15.523
63.97
7,76x10s
1.75
4439.1
18
ANS
1:200
4.141
8.2
65.183
74.7
15.424
60.52
8.09x10s
2.09
3873
19
ANS
1:100
3.942
8.3
59.305
63.3
14.683
55.94
7.94x10s
2.18
3642.6
20
ANS
1:20
4.029
7.6
6.999
12.1
15.005
42.07
l.lxlO4
2.71
4065.1
21
ANS
0
4.133
7.2
101.396
128
15.393
63.97
7.63x10s
1.73
4402.1
22
ANS
1:200
3.920
8.2
63.747
74.7
14.600
60.52
7.25x10s
1.98
3666.2
23
ANS
1:100
3.956
8.3
57.583
63.3
14.735
55.94
7.99x10s
2.19
3655.4
24
ANS
1:20
3.976
7.6
8.391
12.1
14.808
42.07
1.07xl04
2.68
4011.8
Note: * mark means these data were not considered in the prediction of droplet size distribution due to incomplete measured
distribution.
IA-E12PG00037 Final Report
Page 328
-------
iooooo.o:
Modified weber number. We*
~ Oseberg, Untreated (DOR=0)
O Oseberg, Treated (0
-------
3.4 IFT and Reynolds Number Scaling
EPA/600/R-16/152
Appendix H
Due fact that the effects of oil/dispersant in water concentration affects IFT and in situ sampling may be
impractical, as suggested by Johansen et al. (2013), some method for prediction of IFT related to a given
DOR will be useful. IFT measurements with a variety of oils premixed with different dosages of dispersants
might help to establish such relationships in more general terms. As demonstrated by MacKay and Hossain
(1982), with same amount of oil and dispersant, the water volume affects the IFT significantly. In the direct
sampling methods, the amount of oil/dispersant in 1L of sample from different experiments could vary
significantly and therefore affects the IFT measurements. For example for Murban oil with DOR=l: 1333,
the IFT was 3.7 and 7.9 (mN/m) for 100 and 800 mL of water, respectively. Brandvik et al. (2013) provide
a more advantage method for more consistent IFT measurement compared with the direct sampling
methods. In this method, oil/water samples were collected at 1.5 m height above the nozzle in 1 L long
necked measuring flask. Oil appeared as droplets in the water with size distribution depending on the DOR
and method of dispersant application. The surface oil layer in the narrow neck of the bottle and was
collected for IFT measurements after 24 h. using spinning drop method as described by Khelifa and So
(2009), the Dataphysics Spinning Drop Tensiometer SVT-20N with control and calculation software SVTS
20 IFT was used. The IFT in this study were measured using a different method by premix 10 mg oil-
dispersant in 100 mL seawater.
Before such a relationship is establish, we believe that the use of IFT should be avoided and the use of
Modified Weber number approach should be re-considered. Wang and Calabrese (1986) have found that
droplet breakup was governed by the Weber number scaling for small viscosity numbers (Vi —~ 0), but that
a Reynolds number scaling would apply for large viscosity number (Vi » 1):
where p is the density of oil, U is the exit velocity, D is the nozzle diameter, and u is the dynamic viscosity.
Using of Reynolds scaling instead of modified Weber number scaling have the apparent advantage of
avoiding the inconsistency IFT measurements and can make comparison of data from different sources
easier.
The application of this concept for existing experimental data has been shown in Figure 24. The
calculated and observed dso/D correlates very well. In addition, the volume median diameters for IFO-120
are plotted against Reynolds number in Figure 25 together with data for Oseberg Blend by Brandvik et al.
(2013). It can be seen from the plot that Reynolds scaling fits the data well. Values of empirical constants
A were obtained for all IFO-120 combined (exclude DOR=l:25 (or 20)) and Oseberg Blend through
regression analysis. A was 6.1 for combined data while the A for Oseberg Blend is 16.8. The data has shown
that with the d50/D is slightly bigger (higher A) for summer condition cases than winter condition cases
(13)
where C = A5 4/i3 4. and the Re is the Reynolds number given by
(14)
IA-E12PG00037 Draft Final Report
Page 330
-------
EPA/600/R-16/152
Appendix H
with same Reynolds number (Figure 25). The cases for ANS show quite difference compared with IFO-
120 cases. The DOR=0 and <1:100 experimental data points are more closed to Oseberg Blend data, the A
for combined data of ANS (excluded DOR= 1:25 (or 20)) is 10.5 (Figure 26). It is unclear if this is associated
with uncertainties due to limited experimental data points or it is actually due to the effects of different
water temperature. With more experimental data available, this observation will be revisited. Without
considering the effects of temperature, the difference in A between IFO-120 and Oseberg Blend are
considered to be the effects of oil type.
Furthermore, A has been reduced from 16.8 to 8.7 (49% reduction) for Oseberg, from 6.1 to 3.21 (47%
reduction) for IFO-120 and 10.5 to 1.75 (83% reduction) for ANS (Figures 24 and 25). This is reduction
can be used to model the effects of chemical dispersant on droplet size. Based on the experimental data on
the three oils, it is proposed that a constant value A could be selected for Reynolds number scaling
depending on oil types for cases of DOR < 1:100. For DOR of 1:25, a 50% reduction of A maybe used and
a linear interpolation may be used to estimate A values for other DOR greater than 1:100 but less than 1:25
for Oseberg Blend and IFO-120. However, the change A values for ANS does not follow the linear relation.
Data points of DOR = 1:50 for ANS is close to the one of DOR = 1:100 but relatively far from which of
DOR=l:25 (or 20). This may be caused by the effects of oil type and further interpolations for the relation
of DOR and A value for ANS will be needed in future study.
0.20
0.15
¦H).10
re
o
Q
©
to
T3
0.05
0.00
0.00 0.05 0.10 0.15 0.20
d50/D obs
~
~
¦ IFO-120, Cold
water
¦ IFO-120,
Warm water
~ Oseberg
Blend
1
Figure 24: Measured (obs) and computed (calc) relative droplet sizes d50/D from experiments with IFO-120 and
Oseberg Blend
IA-E12PG00037 Draft Final Report
Page 331
-------
EPA/600/R-16/152
Appendix H
1000
Reynolds Number (Re)
Oseberg, Untreated (DOR=0)
Oseberg, Treated (0
-------
EPA/600/R-16/152
Appendix H
two most commonly used distribution are lognormal and Rosin-Rammler distributions. Johansen et al.
(2013) has also concluded that there is currently no theoretical basis for choosing the right distribution
function and the choice of function must be based on empirical data.
Johansen et al. (2013) have found that Rosin-Rammler could provide better overall fit of the experiment
data and they have derived a spreading coefficient a = 1.8 for the corresponding distribution. In this study,
Rosin-Rammler distribution was also selected and corresponding regression analysis has been conducted
to calculate the best spreading coefficients (Tables 5 and 6).
The initial data analysis has indicated that the distributions of the data with d/dso <= 1 and d/dso > 1 are
significantly varied. Thus, it would be difficult and/or inaccurate to predict the measured IFO-120 and ANS
data by only a single distribution.
In order to address this challenge, a two-step Rosin-Rammler approach was introduced by advancing
from the Rosin-Rammler approach proved by Johansen et al. (2013). -The proposed approach uses two
separate spreading coefficients: ai for d/dso <= 1 and 012 for d/dso > 1, providing better fit of the data in all
cases. The data distribution and the corresponding regression results are shown in Figures 27 to 42.
Regressed based on the single Rosin-Rammler distribution, the overall spreading coefficient (a) for IFO-
120 is 2.33 which is larger than that for Oseberg Blend (1.8). For ANS, a = 1.77, is smaller than which for
Oseberg Blend (a = 1.8). According to the two-step Rosin-Rammler approach,the average ai for IFO-120
is 2.01 and 012 is 2.74. In addition, the average ai for ANS is 1.78 and a.2 is 1.63. Furthermore, the regression
coefficients (R2) for the regressions based on single and two-step Rosin-Rammler distributions were also
calculated for both IFO-120 and ANS under different DOR and seasonal conditions (Figures 27 to 42). The
R2 for two-step Rosin-Rammler are higher than which for the single one in most of the case, indicating the
advantage of the proposed two-step Rosin-Rammler approach.
Table 5: Spreading coefficient for Rosin-Rammler distribution of IFO-120
All Data
Average
Single
2-step
Single
2-step
a
ai
a.2
a
ai
a.2
Summer
Untreated
No. 13*
/
/
/
No. 17
1.86
1.53
2.20
No.21*
/
/
/
1.86
1.53
2.20
1:20
No. 16
1.75
2.13
1.44
No.20
1.55
1.95
1.18
No.24
1.85
2..05
1.50
1.72
2.04
1.37
1:100
No. 15
1.96
1.50
2.54
No. 19
1.57
1.30
2.00
IA-E12PG00037 Draft Final Report
Page 333
-------
EPA/600/R-16/152
Appendix H
No.23
1.59
1.31
1.975
1.71
1.37
2.17
1:200
No. 14
2.39
1.85
3.10
No. 18*
/
/
/
No.22*
/
/
/
2.39
1.85
3.10
Spring
Untreated
No.l*
/
/
/
No.5
1.66
1.32
2.14
No.9
1.54
1.49
1.632
1.60
1.41
1.89
1:25
No.4
2.13
2.24
2.10
No.8
2.05
2.31
1.83
No. 12
1.77
1.99
1.62
1.98
2.18
1.85
1:100
No.3
2.61
1.98
3.31
No.7R*
/
/
/
No.llR*
/
/
/
2.61
1.98
3.31
1:250
No.2
2.39
1.72
3.20
N0.6R*
/
/
/
No.l OR*
/
/
/
2.39
1.72
3.20
Average
2.33
2.01
2.74
Note: "/" indicates that the data is unavailable due to incomplete droplet size distribution from
measurement
Table 6: Spreading coefficient for Rosin-Rammler distribution of ANS
All Data
Average
Single
a
2-Step:
Single
2-step
a
ai
(X2
a
ai
(X2
summer
Untreated
No. 13
1.93
2.32
1.57
1.90
2.30
1.55
No. 17
1.87
2.29
1.51
No. 21
1.9
2.3
1.58
1:20
No. 16
1.12
0.62
1.39
1.14
0.64
1.41
No. 20
1.12
0.61
1.34
No. 24
1.17
0.68
1.49
1:100
No. 15R
1.99
2.24
1.65
No. 19
2.05
2.24
1.67
IA-E12PG00037 Draft Final Report
Page 334
-------
EPA/600/R-16/152
Appendix H
No. 23
2.03
2.20
1.66
2.02
2.23
1.66
1:200
No. 14
1.96
2.26
1.49
2.02
2.27
1.57
No. 18
2.03
2.29
1.61
No. 22
2.06
2.25
1.622
Spring
Untreated
No. 1
2.08
2.00
1.90
2.04
2.01
1.93
No. 5*
/
/
/
No. 9
1.99
2.02
1.95
1:20
No. 4R*
/
/
/
0.98
0.49
1.10
No. 8R*
/
/
/
No. 12R
0.98
0.49
1.10
1:100
No. 3
2.14
2.17
2.11
2.10
2.11
2.09
No. 7
2.15
2.13
2.18
No. 11
2.00
2.03
1.97
1:200
(and 250)
No. 2R
1.87
2.11
1.50
2.01
2.19
1.73
No. 6R
1.92
2.16
1.54
No. 10
2.23
2.30
2.15
Average
1.77
1.78
1.63
Note: "/" indicates that the data is unavailable due to incomplete droplet size distribution from
measurement
IA-E12PG00037 Draft Final Report
Page 335
-------
EPA/600/R-16/152
Appendix H
(single Rosin-Rammler)
(single Rosin-Rammler]
0.9961x
= 0.9939
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
0 Single Rosin-Ramler
• 2-step Rosin Rammler
Linear (Single Rosin-
Ramler)
Linear (2-step Rosin
Rammler)
0 Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
- Linear (2-step
Rosin-Rammler)
Experiment No. 5
Experiment No. 1
100%
Experiment No. 5
o
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
100%
Experiment No. 1
o.i 1
D/d50
0.2 0.4 0.6 0.8
Cumulative Volume Fraction, Measurement
IA-E12PG0Q037 Draft Final Report
Page 336
-------
EPA/600/R-16/152
Appendix H
Experiment No. 9
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
Experiment No. 9
(single Rosin-RaiVimler)
0 02 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
Figure 27: Cumulative distribution of d/dso and regression results for IFO-120 with DOR = 0 in spring conditions
(Note: the left figures are distributions and the right ones are regression results)
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
80%
| 7W6
u
ra
t 6096
E
0 50%
>
v
1 40%
Jt
"5
| 3W6
u
20%
Experiment No.2
Experiment No.2
o
0 02 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
IA-E12PG0Q037 Draft Final Report
Page 337
-------
EPA/600/R-16/152
Appendix H
Figure 28: Cumulative distribution of d/dso and regression results for IFO-120 with DOR = 1:250 in spring
conditions
(Note: the left figures are distributions and the right ones are regression results)
[single Rosin-Rammler)
[7 step Rosin R,immler|
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
o
0 m 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
Experiment No. 3
Experiment No, 3
100%
0.1 1
D/d50
Figure 29: Cumulative distribution of d/dso and regression results for IFO-120 with DOR = 1:100 in spring
conditions
(Note: the left figures are distributions and the right ones are regression results)
IA-E12PG0Q037 Draft Final Report
Page 338
-------
EPA/600/R-16/152
Appendix H
Experiment No. 4
100%
90%
70%
30%
10%
0%
0.01
0.1
D/d50
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
Experiment No. 4
£
3
S
I 0.4
§
RJ
1
0
0
038
99E
"(sing
e
Ros
n
R
1 1
amrnler)
1
,
<
~
2
%
9
;(2-St
Rosin-R
ammler
i
»
/l
m
w
i
•
»
i
w
O Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
0 0.2 0.4 0.6 0.8 l
Cumulative Volume Fraction, Measurement
IA-E12PG0Q037 Draft Final Report
Page 339
-------
EPA/600/R-16/152
Appendix H
Experiment No. 8
100%
90%
80%
¦? 70%
« 60%
50%
40%
20%
10%
0%
0.01
0.1
D/d50
• measurement
Single Rosin-
Rammler
^—2-step Rosin-
Rammler
Experiment No. 8
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
O Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
IA-E12PG0Q037 Draft Final Report
Page 340
-------
EPA/600/R-16/152
Appendix H
single Rosin-Rammler)
0 Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
Experiment No. 12
100%
g 70%
M
*= 60%
0)
E
O 50%
>
£ 40%
3
E
3 30%
Experiment No. 12
o.i 1
D/d50
0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
Figure 30: Cumulative distribution of d/dso and regression results for IFO-120 with DOR = 1:25 in spring
conditions
{Note: the left figures are distributions and the right ones are regression results)
'(single Rosin Fammler)
:ep Rosin- idrnmler'j ¦>
O Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
Experiment No. 17
o
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
100%
Experiment No. 17
Figure 31: Cumulative distribution of d/d» and regression results for IFO-120 with DOR = 0 in summer conditions
{Note: the left figures are distributions and the right ones are regression results)
IA-E12PG0Q037 Draft Final Report
Page 341
-------
EPA/600/R-16/152
Appendix H
Experiment No. 14
O ?i>:
measurement
¦ Single ROsin-
Rammler
-2-step Rosin-
Rammler
D/d50
Experiment No. 14
i
Y =
R*
0.93
= 0.9
66x
936
single
Rosin-
Ram
mle
i
•
¦
•v>
y =
i.UU
uyx
(2-ste
Rosin-Ran
imle
r\
I'
R2
= 0
9
9S
1
r)
71 •
6
9
•
%
•
¦JS
1
d
0
J
F
<> Single Rosin-Rammler
• 2-step Rosin-Rammler
¦ Linear (Single Rosin-
Rammler)
¦ Linear (2-step Rosin-
Rammler)
0 0.2 0.4 0.6 0.8 i
Cumulative Volume Fraction, Measurement
Figure 32: Cumulative distribution of d/dso and regression results for IFO-120 with DOR = 1.200 in summer
conditions
{Note: the left figures are distributions and the right ones are regression results)
IA-E12PG0Q037 Draft Final Report
Page 342
-------
EPA/600/R-16/152
Appendix H
Rosin-Rammler)
(2-stf p Rosin-Ramm^)-'
V = 0.9293X
Rosin-Ra
RH 0.9993
|2-step Rosin-Rammlei'
W
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
0 Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
<> Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
Experiment No. 15
Experiment No. 19
o
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
o
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
0 50%
100%
% 40%
Experiment No. 15
Experiment No. 19
IA-E12PG0Q037 Draft Final Report
Page 343
-------
EPA/600/R-16/152
Appendix H
Rosin-Rartimler)
-step Rosin-Rammler)
measurement
• Single Rosin-
Rammler
-2-step Rosin-
Rammler
O Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
100%
c 70%
0
re
J: 60%
01
|
0 50%
>
01
»
% 40%
3
E
U 30%
Experiment No. 23
0.1 1 10
D/d50
Experiment No. 23
o
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
Figure 33: Cumulative distribution of d/dso and regression results for IFO-120 with DOR =1: 100 in summer
conditions
{Note: the left figures are distributions and the right ones are regression results)
• measurement
Single Rosin-Rammler
2-step Rosin-Rammler
O Single Rosin-Rammler
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
Experiment No.16
100%
Experiment No. 16
02 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
IA-E12PG0Q037 Draft Final Report
Page 344
-------
(single Rqsin-Rammier)
• measurement
Single Rosin-Rammler
2-step Rosin-Rammler
• measurement
Single Rosin-Rammler
2-step Rosin-Rammler
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
0 Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
EPA/600/R-16/152
Appendix H
Experiment No. 20
Experiment No. 20
80%
c
0
5 70%
u
n
1 60%
|
0 50%
>
01
I 40%
"5
£ 30%
u
20%
80%
c
0
+= 70%
ro
1 60%
£
0 50%
>
ai
¦| 40%
"5
£ 30%
3
U
20%
0
0 02 0.4 0.6 OS 1
Cumulative Volume Fraction, Measurement
Experiment No. 24
Experiment No. 24
o
0 02 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
Figure 34: Cumulative distribution of d/'ck, and regression results for IFO-120 with DOR = 1:20 in summer
conditions
{Note: the left figures are distributions and the right ones are regression results)
IA-E12PG0Q037 Draft Final Report
Page 345
-------
EPA/600/R-16/152
Appendix H
1.0049X
= 0.9996
RosinR.ammler)
1.0029X
Rosin-Rairimter)
Rosin-Rammler)
measurement
• Single Rosin-
Rammler
-2-step Rosin-
Rammler
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
Experiment No. 9
100%
Experiment No. 9
o
0 02 0.4 0.6 0£ 1
Cumulative Volume Fraction, Measurement
Experiment No. 1
O.l 1 10
D/d50
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Experiment No. 1
o
0 02 0,4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
O Single Rosin-Rammler
Figure 35: Cumulative distribution of d/dso and regression results for ANS with DOR = 0 in spring conditions
{Note: the left figures are distributions and the right ones are regression results)
IA-E12PG0Q037 Draft Final Report
Page 346
-------
EPA/600/R-16/152
Appendix H
Experiment No. 2R
measurement
• Single Rosin-
Rammler
-2-step Rosin-
Rammler
D/d50
Experiment No. 2R
E 0.6
41
£
3
0
>
1 04
2
3
E
3
u
02
cW
y = 1
R2 =
¦bX
D.9984
single Rosin-Rammler)
0,f
Y =
H j (
.DDflO
2-step Rosin-R
mmler]
M
1
•
9
i
i
o
•
J
J
V
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
Cumulative Volume Fraction, Measurement
Experiment No. 6R
Experiment No. 6R
«= 60%
S 10%
measurement
¦ Single Rosin-
Rammler
-2-step Rosin-
Rammler
«i
f
r-
R*
Sx
85
•
= 0
9S
(single
Ros
n
R
ammle
r)
I
l.UU
fx
?
•
= 0.99
.V
1
a
.yl
>
I
*
(
>'
9
•
4
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
D/d50
Cumulative Volume Fraction, Measurement
IA-E12PG0Q037 Draft Final Report
Page 347
-------
EPA/600/R-16/152
Appendix H
Experiment No. 10
Experiment No. 10
measurement
• Single Rosin-
Rammler
-2-step Rosin-
Rammler
r
y
= ]
.005
Bx
nmler
•
2 -
0.99
)7
mgic r\u3in-r\d
=
1.002
fiv
.4
(2
-ste
tosin-Ra
mmler
0.9-
i
<
\
•
•
*
•
•
1
I
i
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
D/d50
0 02 0.4 0.6 0.8
Cumulative Volume Fraction, Measurement
Figure 36: Cumulative distribution of d/ck, and regression results for ANS with DOR = 1:200 or (250) in spring
conditions
(Note: the left figures are distributions and the right ones are regression results, Experiment No. 10 is DOR=1:250)
1.0043X
Rosin-Rammler)
0.9998
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
Experiment No. 3
100%
Experiment No. 3
o
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
IA-E12PG0Q037 Draft Final Report
Page 348
-------
EPA/600/R-16/152
Appendix H
(singje Rosin-Rammler]
0.9996
y=1.0Ol4x
(single Rosin-Rammler)
ammler)
>:-p Rosin-R
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
0 Single Rosin-Rammler
t 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
Experiment No. 7
Experiment No. 11
Experiment No. 7
Cumu ative Volume Fraction. Measurement
0.1 1
D/d50
100%
0.2 0.4 0.6 0.8
Cumulative Volume Fraction, Measurement
Experiment No. 11
0.1 1
D/d50
Figure 37: Cumulative distribution of d/dso and regression results for ANS with DOR = 1:100 in spring conditions
(Note: the left figures are distributions and the right ones are regression results)
IA-E12PG0Q037 Draft Final Report
Page 349
-------
EPA/600/R-16/152
Appendix H
Experiment No. 12R
• measurement
Single Rosin-Rammler
2-step Rosin-Rammler
D/d50
Experiment No. 12R
2 0.6
i
¦
~t
«
Y =
R*
).
9!
).<
42x
no
(
ingle Rosin-Rar
lmler;
•
0
Y =
3.99
99x
(
-steo Rosin-Ra
imler)
0
R2 =
0
9<
94
5
J>
s
•
Y
1
9
r
•
I
•
0
0
O
0 02 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
• Linear (Single Rosin-
Rammler)
¦ Linear (2-step Rosin-
Rammler)
Figure 38: Cumulative distribution of d/d50 and regression results for ANS with DOR = 1:20 in spring conditions
(Note: the left figures are distributions and the right ones are regression results)
IA-E12PG0Q037 Draft Final Report
Page 350
-------
EPA/600/R-16/152
Appendix H
Experiment No. 13
100%
80%
70%
50%
40%
20%
10%
0%
0.01
0.1
10
D/d50
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
Experiment No. 13
J
R2
= i.oix
= 0.997
single Rosin-Rammler)
0
•
= 0.
3999
(2-step Rosin-Rammler^
4
•
4
•1
•
%
Cj.
•
0
•
>•'
|
9
0 02 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
0 Single Rosin-Rammler
• 2-step Rosin-Rammier
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
IA-E12PG0Q037 Draft Final Report
Page 351
-------
EPA/600/R-16/152
Appendix H
Experiment No. 17
100%
90%
80%
c 70%
o
'¦H
U
ra
£ 60%
v
|
O 50%
>
O)
>
« 40%
3
|
3 30%
20%
10%
Rosin-Rammler)
Rosin-Rammler)
Experiment No. 17
o
0 02 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
Experiment No. 21
100%
90%
80%
| 70%
a
^ 60%
a
|
o 50%
>
| 40%
3
E
3 30%
20%
10%
(single Rosin-Rammler]
-step Rosin-Rammler)
0.1 1
D/d50
Experiment No. 21
o
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
measurement
¦ Single Rosin-
Rammler
-2-step Rosin-
Rammler
0 Single Rosin-Rammler
• 2-step Rosin-Rammler
Linear (Single Rosin-
Rammler)
Linear (2-step Rosin-
Rammler)
Figure 39: Cumulative distribution of d/dso and regression results for ANS with DOR = 0 in summer conditions
(Note: the left figures are distributions and the right ones are regression results)
IA-E12PG00037 Draft Final Report Page 352
-------
EPA/600/R-16/152
Appendix H
Experiment No. 14
• measurement
• Single Rosin-
Rammler
-2-step Rosin-
Rammler
D/d50
Experiment No. 14
1
£
a
s
A
•
1
O Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
0 0.2 0.4 0.6 0.8 l
Cumulative Volume Fraction, Measurement
Experiment No. 18
Experiment No. 18
.= M
¦P 40%
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
*
1 =
R2
1.0
= o.<
>36
sin
jle Rosi
n-
3a
mmle
r)
5
•
Y =
Lx
(?-s
r)
R2
= 0
998
8
,0
•
»
J--'"
41
i
%
0
L
1
s
k
f
O Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
D/d50
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
IA-E12PG0Q037 Draft Final Report
Page 353
-------
EPA/600/R-16/152
Appendix H
(2-step Rosin-RSmmter)
0 Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
100%
Experiment No. 22
Experiment No. 22
Cumulative Volume Fraction, Measurement
Figure 40: Cumulative distribution of d/dso and regression results for ANS with DOR = 1:200 in summer conditions
{Note: the left figures are distributions and the right ones are regression results)
Rammler)
Rosin-
-Rammlerp"
measurement
¦ Single Rosin-
Rammler
-2-step Rosin-
Rammler
O Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
c 70%
0
(0
£ 60%
01
I
0 50%
01
>
'% 40%
3
|
<3 30%
Experiment No. 15R
100%
Experiment No. 15R
0.1 1
D/d50
IA-E12PG0Q037 Draft Final Report
Page 354
-------
EPA/600/R-16/152
Appendix H
Experiment No. 19
100%
90%
80%
- 60%
01
E
measurement
• Single Rosin-
Rammler
-2-step Rosin-
Rammler
Experiment No. 19
i*r
3J999 (sin6lle Rosin-Rammler)
2-step Rosin- tammler}. £
y = I.OjOIx
^=^01)987
Single Rosin-
Rammler
2-step Rosin-
Rammler
• Linear (Single
Rosin-Rammler)
• Linear (2-step
Rosin-Rammler)
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
Experiment No. 23
£ 60%
E
o 50%
measurement
• Single Rosin-
Rammler
-2-step Rosin-
Rammler
D/d50
IA-E12PG0Q037 Draft Final Report
Page 355
-------
EPA/600/R-16/152
Appendix H
1
Experiment No. 21
f
>
<*
Y =
1.
31
85x
Ra
m
R2
= 0
.9
991
(s
n
gi
•
O Single Rosin-
2 0.6
LL
Rammler
01
E
3
•
• 2-step Rosin-
Rammler
5
i
Jj 0.4
Rosin-Rammler)
"5
§
£
3
U
,
.
r
Rosin-Rammler)
0.2
•
%
J
f
A
9
0 (
0.2 0.4 0.6 0.8
Cumulative Volume Fraction, Measurement
1
Figure 41: Cumulative distribution of d/dso and regression results for ANS with DOR =1:100 in summer conditions
(Note: the left figures are distributions and the right ones are regression results)
IA-E12PG00037 Draft Final Report
Page 356
-------
EPA/600/R-16/152
Appendix H
Experiment No. 16
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
Experiment No. 16
y
=
0.9809>
2
D.
97
7
sing
le Kosm-K
d-
nmier)
•
*0
Y
=
1.0C
I35>
2-st
ep Rosin-f
ammler)
1
<
R
yyy
1
<
>
1
•
0
~
»
•
•
i
•
P<
>
0
0
o
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
O Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
IA-E12PG0Q037 Draft Final Report
Page 357
-------
EPA/600/R-16/152
Appendix H
Experiment No. 20
100%
70%
60%
50%
30%
20%
10%
0%
0.01
0.1
10
1
D/d50
• measurement
Single Rosin-
Rammler
2-step Rosin-
Rammler
Experiment No. 20
2 0.6
v
E
3
0
>
| 0.4
n
3
E
3
u
y
R
/
o
0.
9
,
528>
>784
sing
e
Rosin-R
immle
)
«
y
1.
DC
>15*
»
0
Cf
osin-F
ammler)
R
0
.£
964
•
0
1
5
•
>
O Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
0 0.2 0.4 0.6 0.8 1
Cumulative Volume Fraction, Measurement
IA-E12PG0Q037 Draft Final Report
Page 358
-------
EPA/600/R-16/152
Appendix H
Experiment No. 24
Experiment No. 24
100%
I.9784X
0/781 (single Rosin-Rpmmler)
L0lll8x (2-st;pRosin-
0.H976 U L
90%
80%
= 70%
0 Single Rosin-
Rammler
• 2-step Rosin-
Rammler
Linear (Single
Rosin-Rammler)
Linear (2-step
Rosin-Rammler)
60%
S 0.6
• measurement
¦S 50%
Single Rosin-
Rammler
n 40%
> 0.4
2-step Rosin-
Rammler
u 30%
20%
0.2
0.2
Cumulative Volume Fraction, Measurement
0.4
0.6
0.01
0.1
D/d50
Figure 42: Cumulative distribution of d/d50 and regression results for ANS with DOR = 1:20 in summer conditions
(Note: the left figures are distributions and the right ones are regression results)
4 Summary
In this study, research has been conducted for the droplet size distributions of two types of oils (IFO-
120 and ANS) release from subsurface injection with/without application of chemical dispersant in different
seasonal conditions (i.e., spring and summer). Firstly, a series of experiments have been conducted via
wave tank experiment by COOGER in BIO to measure the droplet sizes. These data were analyzed and
utilized to determine the relative volume median diameter (dsn) and the peak diameter (c/,,). Accordingly to
the droplet size distribution and the modified Weber number approach, the Weber number (We), as well as
the additional measurements on oil viscosity and IFT, the Viscosity number (Vi) and Reynold number (Re)
were calculated. In addition, the relation between the droplet size distributions and dispersant-oil-ratios
(DORs) has also been analyzed. Finally, the corresponding empirical coefficients have been determined for
the droplet size prediction.
Furthermore, the data analysis has also indicated that the distributions of the data with d/dso <= 1 and
d/dso > 1 are significantly varied. Thus, it would be difficult and/or inaccurate to predict the measured IFO-
120 and ANS data by only a single distribution. Therefore, a two-step Rosin-Rammler approach was
introduced by advancing from the Rosin-Rammler approach proved by Johansen et al. (2013). The proposed
approach uses two separate spreading coefficients: ai for d/dso <= 1 and a.: for d/dso > 1, providing better fit
of the data in all cases. The regression coefficients for the two-step Rosin-Rammler are higher than which
IA-E12PG00037 Draft Final Report Page 359
-------
EPA/600/R-16/152
Appendix H
for the original single one in most of the case, indicating the advantage of the proposed two-step Rosin-
Rammler approach.
In general, the chemical dispersant plays an importance role in reduce the droplet size of ANS no matter
in spring or summer conditions. The effectiveness of dispersant in reducing droplet size is higher on ANS
than which on IFO-120. There may be thresholds for the dose of chemical dispersant to some oils (e.g.,
IFO-120) but will need further experiments to analyze. There may also be over dose of dispersant to some
oils (e.g., ANS) when the DOR is high, eventually affecting the droplet size distribution. Future experiment
will also need for this particular issue.
The measured IFT for the IFO-120 and ANS with different DORs appeared significant difference
compared with the ones measured from SINTEF for the modified Weber number approach. This may due
to the characteristics of different oil. Further experiments will be needed to address this issue.
Acknowledgement
This study was supported by the Centre for Offshore Oil, gas, and Energy Research, Fisheries and
Oceans Canada, Bedford Institute of Oceanography and US Environmental Protection Agency (US EPA).
References
Buist, I.A., W. M. Pistruzak, and D.F. Dickins, "DOME Petroleum's Oil and Gas Undersea Ice Study", in
Proceedings of the Fourth Arctic Marine Oilspill Program Technical Seminar, Ottawa, ON, 1:647-686,
1981,
Brandvik, P. J., et al. (2013). "Droplet breakup in subsurface oil releases - Part 1: Experimental study of
droplet breakup and effectiveness of dispersant injection." Marine Pollution Bulletin 73(1): 319-326.
Chen, F. H. and P. D. Yapa (2007). "Estimating the oil droplet size distributions in deepwater oil spills."
Journal of Hydraulic Engineering-Asce 133(2): 197-207.
Chen, F. H. and P. D. Yapa (2003). "A model for simulating deepwater oil and gas blowouts - Part II:
Comparison of numerical simulations with "deepspill" field experiments." Journal of Hydraulic Research
41(4): 353-365.
The Federal Interagency Solutions Group, Oil Budget Calculator Science and Engineering Team (2010).
Oil Budget Calculator: Deepwater Horizon: 217.
Johansen, O., et al. (2003). "DeepSpill - Field study of a simulated oil and gas blowout in deep water." Spill
Science & Technology Bulletin 8(5-6): 433-443.
Johansen, O., et al. (2013). "Droplet breakup in subsea oil releases - Part 2: Predictions of droplet size
distributions with and without injection of chemical dispersants." Marine Pollution Bulletin 73(1): 327-335.
IA-E12PG00037 Draft Final Report
Page 360
-------
EPA/600/R-16/152
Appendix H
Louis J. Thibodeaux, et al. (2011). "Marine Oil Fate: Knowledge Gaps, Basic Research, and Development
Needs; A Perspective Based on the Deepwater Horizon Spill." Environmental Engineering Science 28(2):
87-93.
Lefebvre, A. H. (1989). Atomization and Sprays, Taylor & Francis, P. 421
Stephen M. Masutani and E. E. Adams (2001). Experimental Study of Multi-Phase Plumes with Application
to Deep Ocean Oil Spills. Deep Spill JIP, University of Hawaii (UH) and Massachusetts Institute of
Technology (MIT). 147.
Topham, D.R., Hydrodynamics of an Oilwell Blowout, Beaufort Sea Technical Report #33, Department of
the Environment, Victoria, BC, 52p., 1975
Wang, C. Y. and R. V. Calabrese (1986). "Drop Breakup in Turbulent Stirred-Tank Contactors .2. Relative
Influence of Viscosity and Interfacial Tension." Aiche Journal 32(4): 667-676.
Zhao, L., et al. (2014). "Evolution of droplets in subsea oil and gas blowouts: Development and validation of the
numerical model VDROP-J." Marine Pollution Bulletin 83(1): 58-69.
IA-E12PG00037 Draft Final Report
Page 361
-------
Supplemental Material A. Fluorescence Analyses of 25 Oil Types at 4 Dispersant to Oil Ratios (listed in alphabetical order within oil type: Light, Intermediate, Heavy)
FWHM data in gray are estimates only as intensity did not return to half maximum before rising toward next adjacent peak.
Sample
Name
API
Gravity
Sulfur
Content
(weight %)
Point of
Origin
DOR
overall maximum peak
Fex
(nm)
C 1
r Em
(nm)
(RU)
c 1
' FWHM
(±4.5 nm)
Inner
Filter
Effect (IFE)
2nd intensity peak, lower ex/em*
Fex
(nm)
r Em
(nm)
(RU)
c z
' FWHM
(±4.5 nm)
minor peak, similar ex/higher em'
Fex
(nm)
P 5
r Em
(nm)
(RU)
c 3
' FWHM
(±4.5 nm)
minor peak, higher ex/similar em*
Fex
(nm)
P 4
r Em
(nm)
(RU)
c 4
' FWHM
(±4.5 nm)
F 281/340
(RU)
F 281/446
(RU)
FIR
Fchelsea-R
239/358
(RU)
Fchelsea-C
239/441
(RU)
rTriOS
254/358
(RU)
¦~Cyclops-R
254/349
(RU)
¦~Cyclops-C
320/511
(RU)
r ECO
371/460
(RU)
Arabian Light
Light
(32.2")
1.85
Saudi Arabia
0
1:200
1:100
1:20
224 335 400.4168 36.4960
224 335 357.6249 36.4960
224 335 426.8238 36.4960
224 335 701.7454 41.0730
1.1798
1.0714
1.2633
4.1859
218 290 151.5028 31.7020
218 285 148.2625 31.7020
218 285 147.4888 31.7020
218 290 224.3619
275
284
254
236
469 4.
478 2.
473 15.
441 236.
7933
6419
6626
8910 255.2810
278 335 36.8206 41.0730
275 335 33.5329 41.0730
278 335 39.8975 41.0730
278 335 82.6645
34.1437
30.9667
38.1081
78.7067
4.6837
2.6676
14.1205
203.6622
7.2898
11.6086
2.6988
0.3865
28.8119
26.0505
36.0842
113.5339
4.9367
3.8100
13.5018
231.9759
16.8258
15.6252
22.3495
93.1478
20.8790
19.1433
24.4236
73.9108
3.5586
2.2830
12.2223
124.4251
Brent
Light
(38.20
0.40
East Shetland
Basin, North Sea,
UK (water depth
140 m)
0
1:200
1:100
1:20
224 335 646.1843 36.4960
224 335 660.3712 36.4960
221 335 708.1563 36.4960
224 335 1098.4158 41.0730
1.1518
1.1618
1.3653
3.0468
218 303 75.9188
221 290 63.9468
221 299 78.6673
221 303 144.6971
254
248
251
233
441 10.
432 11.
441 37.
432 261.
1605
6829
7209
3387
275 326 63.5367 45.6200
275 326 65.5055 45.6200
275 326 71.9189 50.2000
278 335 133.8242
56.1012
58.2578
64.9403
127.8678
7.3951
8.7661
32.8876
189.0144
7.5863
6.6458
1.9746
0.6765
40.2460
42.9915
58.0009
173.4207
8.9941
9.3088
34.0570
231.1604
31.0871
32.0393
44.3147
146.2237
37.3653
38.7053
46.2510
116.6535
5.6402
5.6408
22.0010
85.0858
Federated
Light
(39.4")
0.32
NW Alberta,
Canada
0
1:200
1:100
1:20
224 335 574.3489 41.0730
224 335 607.9723 41.0730
224 335 645.2807 41.0730
224 335 1223.1724 50.2370
1.1437
1.2130
1.3698
4.3653
218 285 91.7448
218 290 112.8743
218 290 91.7746
218 303 213.5284
251
248
254
233
441 15.
441 30.
441 72.
436 661.
6750
9314
6915
2244 226.9050
272 326 54.4863
272 326 57.8010
278 335 65.0589
275 340 165.3826
50.2000
59.3710
51.3190
55.3673
62.4421
163.8397
13.8540
27.5293
66.5376
460.6894
3.7043
2.0112
0.9384
0.3556
68.2091
79.0319
99.7517
371.8531
15.0958
29.2400
70.1628
614.1385
46.4210
55.2828
71.2351
289.5725
48.6704
54.6837
65.5325
207.3393
9.2681
18.4455
39.0617
168.2464
Gullfaks
Light
(32.7")
0.31
North Sea,
Norway (water
depth 230 m)
0
1:200
1:100
1:20
221 335 937.0018 36.4960
221 335 934.4237 36.4960
221 335 933.0803 36.4960
221 335 1524.2070 41.0730
1.2434
1.2715
1.3451
3.7319
218 290 97.9804
218 290 96.3758
218 294 96.5534
215 281 179.2527
9.0400
257 441 15.8195
251 436 18.0573
257 436 28.6241
233 441 352.3552
275 326 90.0457 41.0430
275 326 91.9918 41.0430
275 326 93.8404 45.6200
275 335 193.9621
77.5575
80.0347
80.6821
179.7879
13.4001
14.3665
24.8691
253.8018
5.7878
5.5709
3.2443
0.7084
44.3429
43.6717
46.8422
212.7369
12.4754
14.9723
25.7671
324.5520
37.0081
37.1213
39.3444
187.2627
45.7345
46.3369
47.9996
146.4968
10.1643
10.3233
18.4141
122.7369
Hibernia
Light
(35.6")
0.41
Newfoundland,
Canada(water
depth 80 m)
0
1:200
1:100
1:20
221 335 938.0775 36.4960
221 335 951.4881 36.4960
221 335 978.6157 36.4960
221 335 1812.4055 41.0730
1.2390
1.3366
1.4740
4.0175
218 285 173.3257
218 290 168.1463
215 281 170.8232
218 290 305.6178
22.6080
22.6450
290
260
254
225
492 11.
483 27.
478 55.
422 567.
2138
3287
8991
3160
275 326 87.9497 41.0430
275 326 93.5200 45.6200
275 326 96.1672 45.6200
275 335 201.2828
74.9422
81.0861
85.2278
182.7473
10.8507
25.1700
50.5145
374.9851
6.9067
3.2215
1.6872
0.4873
42.8229
48.8423
55.1197
236.6556
10.5048
23.7015
50.2643
480.2082
33.8845
41.7274
47.7291
199.5141
44.5846
51.5979
56.0714
152.8903
8.2628
22.5015
40.6007
192.4787
MC252
(Discoverer
Enterprise)
Light
(37.2")
<0.1
Louisiana, US
(water depth up
to 1500 m)
0
1:200
1:100
1:20
221 335 998.4975 36.4960
221 331 1009.1816 36.4960
221 335 1085.5377 36.4960
221 331 1998.5995 36.4960
1.2686
1.3728
1.4799
4.7580
218 290 181.4689
215 290 196.1398
218 285 189.5899
221 308 437.3233
257 474 17.8789
260 473 31.0582
254 473 81.0433
230 487 744.6895 200.4510
272 326 94.8271 41.0430
275 326 99.4748 50.1630
272 326 106.1992 45.6200
275 335 231.8561
76.9368
82.3071
88.2869
203.9413
15.3431
26.9270
72.3715
527.5106
5.0144
3.0567
1.2199
0.3866
46.3171
53.4529
64.3268
229.0701
15.2539
28.4331
72.8640
670.7186
40.7658
45.5031
55.0190
200.3698
50.3429
53.5737
63.5313
175.3657
15.0873
25.1354
59.5572
268.1784
MC252
(generic)
Light
(35.2")
0.21
Louisiana, US
(water depth up
to 1500 m)
0
1:200
1:100
1:20
221 335 857.3517 36.4960
221 335 877.7767 41.0730
221 335 964.0212 41.0730
221 331 1795.1270 41.0730
1.2351
1.2462
1.5711
4.7317
215 285 159.8695
215 303 108.8697
215 290 92.4174
218 303 376.3732
22.6800
251
254
260
233
441 16.
450 26.
446 92.
441 704.
8587
8996
3028
3134 204.5770
269 326 80.8431 45.5860
272 326 83.4356 45.6200
275 331 93.3841 54.7840
272 335 212.5289
65.0029
68.4320
79.1424
193.3704
14.7138
19.2422
83.1330
483.6686
4.4178
3.5563
0.9520
0.3998
36.3515
42.3193
62.4186
260.6423
15.9277
21.1766
89.9889
622.6480
33.1775
34.9381
56.6115
217.6143
41.2382
46.2312
58.4282
178.0065
10.0270
16.2865
52.7589
199.9642
Scotian Shelf
Condensate
Light
(51.4")
0.02
Nova Scotia,
Canada(water
depth 12-20 m)
0
1:200
1:100
1:20
224 335 946.5222 41.0730
221 335 1408.5872 36.4960
221 335 1487.1614 41.0730
221 335 1337.9812 41.0730
1.1527
1.4413
1.5326
1.5025
218 303 88.8169
218 285 595.5652 36.2160
218 285 588.8325 36.2160
218 285 565.7260 36.2160
275 331 95.6155 45.6200
275 331 141.1720 45.6200
275 331 151.9375 45.6200
275 331 126.3912 45.6200
81.6878
118.0016
125.7985
104.9922
2.0109
2.1982
2.1669
2.2033
40.6224
53.6817
58.0552
47.6513
46.0174
59.7045
70.0388
52.5865
3.3515
3.9548
5.2573
5.2818
33.2896
48.4607
51.3959
42.6106
45.6231
66.0981
74.0406
59.2394
0.1522
0.0828
0.0638
0.1270
Terra Nova
Light
(33.8")
0.43
Newfoundland,
Canada(water
depth 90-100 m)
0
1:200
1:100
1:20
221 335 665.4964 36.4960
221 335 719.7159 36.4960
221 335 821.2414 36.4960
221 335 1380.3396 41.0730
1.1604
1.2207
1.3653
3.7647
218 290 124.9477
218 290 118.4676
215 290 150.9355
215 303 248.2621
272 511 9.1321
305 506 19.5982
269 511 52.2634 233.2450
230 492 486.6393
275 326 62.5324 41.0430
278 335 68.9711 45.6200
275 335 76.9978 45.6200
291 335 160.6443
54.4188
60.1640
67.8490
149.7883
7.8571
15.8660
46.1358
371.4435
6.9261
3.7920
1.4706
0.4033
30.6780
37.7650
48.1195
186.0975
6.8939
15.5717
45.2714
438.2137
24.2375
29.9080
39.0681
161.0352
30.8974
36.8026
43.5464
125.2096
7.5572
14.9163
41.2224
216.3706
Alaska North
Slope
Intermediate
(29.8")
1.09
Prudhoe Bay,
Alaska, US
0
1:200
1:100
1:20
221 335 697.0711 36.4960
221 335 715.0080 36.4960
221 335 839.5968 36.4960
221 335 1171.6320 36.4960
1.1643
1.1515
1.3153
3.3276
218 285 131.7351
218 285 134.0672
218 294 148.9155
218 285 196.5901
36.2160
36.2160
293
266
236
473
2.6573
497 11.7574
446 173.1327
275 326 66.6243 41.0430
275 326 67.4011 41.0430
272 326 79.6627 41.0430
275 331 126.9662
56.0902
56.3347
68.4291
114.5851
2.5976
1.5911
10.3806
130.0100
21.5932
35.4051
6.5920
0.8814
33.3178
32.4303
46.8092
116.4722
2.3807
2.0669
11.2163
158.2507
26.9739
26.2117
37.8401
96.4654
34.4954
33.9454
44.1656
90.0891
1.7365
1.3257
8.2902
73.0147
Alaska North
Slope (10%
weathered)
Intermediate
(26.8")
1.20
Prudhoe Bay,
Alaska, US
0
1:200
1:100
1:20
221 335 812.9654 36.4960
221 335 831.6980 36.4960
221 335 828.0589 36.4960
224 335 1109.5088 36.4960
1.1927
1.2071
1.2843
3.0146
215 290 147.4303
215 285 151.4220
215 285 136.5127
218 285 168.7047
284
263
233
501
3.1155
492 9.0535
441 159.5431
275 326 76.0907 41.0430
275 326 77.5943 41.0430
275 326 82.1870 41.0430
278 335 117.9003
64.7766
64.7794
67.8410
106.5389
2.9855
2.9855
7.4701
117.7190
21.6973
21.6982
9.0817
0.9050
40.3102
40.2480
44.5102
97.2009
2.9991
4.7037
9.4948
141.3493
31.4807
31.4551
37.6410
86.6070
39.4282
40.9625
45.0597
81.7803
2.2554
2.3404
6.3725
70.4364
Heidrun
Intermediate
(28.6")
0.46
Norwegian Sea
(water depth 350
m)
0
1:200
1:100
1:20
221 335 902.6871 36.4960
221 335 909.4710 36.4960
221 335 964.3148 36.4960
224 335 1098.8957 36.4960
1.2818
1.2575
1.3330
2.1944
218 299 117.2393
218 303 120.8474
218 299 111.2935
218 308 164.0666
257 441 21.1729
254 432 16.4071
251 441 27.0994
236 441 174.9308
275 326 91.3695 41.0430
275 326 93.9774 41.0430
275 326 99.1424 41.0430
275 326 116.9711
75.1032
76.8515
81.5684
104.4772
16.6835
12.3624
21.8320
135.8026
4.5016
6.2166
3.7362
0.7693
46.6748
47.9185
57.6495
103.6075
17.2943
13.4413
22.3738
162.7506
37.5950
36.5101
40.8470
86.9668
47.5629
46.3281
51.2695
80.7585
12.6075
10.0746
17.2084
74.9609
Lago
Intermediate
(27.3")
0.30
Maracaibo Basin,
Venezuela
0
1:200
1:100
1:20
221 335 352.2231 36.4960
221 335 398.3964 36.4960
224 335 367.7451 41.0730
221 335 453.0976 41.0730
1.0723
1.1277
1.1202
1.7484
218 285 48.9594
218 290 51.1749
215 290 45.9607
218 299 55.7396
236
446 63.5925 283.0880
278 335 32.0908 41.0730
275 335 35.8690 41.0730
275 335 33.0880 41.0730
275 335 44.7620
29.1256
32.5430
29.6885
41.5534
2.4010
2.7752
3.3449
45.3767
12.1308
11.7265
8.8756
0.9157
23.5723
29.2280
26.8896
53.7942
2.4668
3.4148
3.7673
57.8881
17.6000
20.7464
19.4312
41.9449
21.1205
24.3474
22.3621
37.3306
1.7177
1.8901
2.3043
27.6019
MESA
Intermediate
(30.3")
0.87
Orinoco Basin,
Venezuela
0
1:200
1:100
1:20
221 335 757.8380 36.4960
221 335 806.7615 36.4960
221 335 745.1679 36.4960
221 335 1107.0899 41.0730
1.1875
1.2196
1.2636
2.7673
218 290 128.7504
218 290 143.3521
218 290 127.0369
218 299 185.4842
31 CTTfl
233
418 142.3828
275 326 72.1471 50.2000
275 335 74.2263 45.6200
275 326 70.6516 45.6200
275 335 116.5099
62.9403
64.0487
61.6738
106.3210
3.4698
5.3486
9.5702
102.5629
18.1396
11.9748
6.4443
1.0366
47.3344
51.1556
51.2559
123.4786
4.6808
7.1721
11.9492
126.8977
34.9579
37.4703
37.8598
95.9839
41.9985
42.8810
44.1276
88.7799
2.3492
3.4450
6.6796
48.8107
Intermediate
Newfoundland,
0
221
335 1145.2869 36.4960
1.2784
218
303 152.5516
275
326 108.7189 41.0430
88.6457
9.0486
9.7967
43.5595
11.9753
40.9107
50.2525
6.7127
-------
(White Rose)
(29.8")
Canada(water
1:200
221
335
1223.9796
36.4960
1.3272
218
290
202.7941
--
--
--
--
275
326
115.3793
41.0430
93.9090
12.3688
7.5924
50.1617
14.9794
42.3903
56.7183
9.6724
5.1417
depth 100 m)
1:100
221
335
1236.6347
36.4960
1.5486
215
294
159.5234
260
441
50.7172
275
326
116.7994
45.6200
98.9043
45.3497
2.1809
63.0618
47.2371
54.5822
63.8962
31.4917
15.6276
1:20
221
335
1973.5528
36.4960
3.5550
218
303
280.6617
9.0750
233
441
359.0106
171.7950
278
335
211.3166
186.5334
261.0755
0.7145
182.6322
322.1702
163.5870
147.7333
124.8090
49.8540
Vasconia
Intermediate
0.56
Colombia
0
224
335
844.9348
36.4960
1.1536
215
303
100.6205
--
--
--
--
275
326
79.9347
41.0430
68.0535
1.9690
34.5629
45.3441
4.3971
36.3864
45.3699
1.0751
0.6805
(26.3°)
1:200
224
335
828.3719
36.4960
1.1583
218
303
97.9280
--
--
--
--
272
326
80.8601
41.0430
67.4266
1.5176
44.4299
44.2001
4.0596
35.4004
46.8488
0.5749
0.4581
1:100
224
335
835.6222
36.4960
1.1724
218
290
98.3668
--
--
--
--
275
326
80.5851
41.0430
67.9124
1.9921
34.0901
43.5892
4.6579
37.7414
46.1524
1.1150
0.6503
1:20
221
335
935.7946
36.4960
1.6723
215
308
139.7536
--
--
--
--
272
326
92.7718
45.6200
80.0346
23.4436
3.4139
60.3238
33.0451
52.4788
59.7833
12.6010
5.4375
Access
Heavy
3.91
Athabaska region,
0
221
335
39.5805
41.0730
1.0197
218
285
33.2874
36.2160
215
558
2.4487
56.3880
272
326
4.9297
4.1833
0.3668
11.4056
7.2683
0.8325
4.6137
4.4813
0.1885
0.1454
Western
(21-3°)
Alberta, Canada
1:200
221
335
46.5153
41.0730
1.0204
215
285
38.9976
36.2160
218
567
2.9473
51.7750
272
326
5.3889
4.6291
0.3983
11.6208
7.7770
1.0395
4.5552
4.5711
0.0901
0.0734
Blend Dilbit
1:100
221
335
49.8423
45.6200
1.0266
218
285
45.8912
36.2160
215
571
2.9107
51.7510
272
326
5.9759
5.1332
0.6010
8.5418
8.7321
1.3962
5.2585
5.1513
0.1711
0.1227
1:20
224
335
60.1881
45.6530
1.0777
218
285
47.3804
40.7590
233
441
13.3766
148.9860
272
326
7.8131
7.1753
6.3520
1.1296
15.7712
11.7630
9.8247
8.5862
2.5057
0.9710
Belridge
Heavy
1.03
San Joaquin
0
230
344
118.6885
50.2740
1.0688
221
294
25.0773
--
--
--
--
290
353
20.8257
68.6540
17.6309
3.5970
4.9016
67.1562
7.8788
37.7229
30.6641
1.1081
0.8613
Heavy
(13.6°)
Valley, California,
1:200
230
344
161.7504
50.2740
1.0812
218
294
34.0575
--
--
--
--
290
353
27.5169
68.6540
23.7155
4.5438
5.2193
88.9294
10.7576
48.5595
39.1353
1.2513
1.1570
US
1:100
227
344
140.9613
50.2740
1.0744
221
308
33.8168
--
--
--
--
290
353
25.1983
73.2570
21.4500
4.6412
4.6216
82.0319
9.3816
45.6629
36.5600
1.2524
0.9988
1:20
227
340
147.0897
50.2740
1.1216
221
290
31.8670
--
--
--
--
290
353
27.1402
73.2570
22.3664
6.7721
3.3027
86.7568
13.2676
50.2330
39.1433
2.4427
1.6384
Cold Lake
Heavy
3.77
NE Alberta,
0
224
335
120.6093
41.0730
1.0503
218
285
55.0742
--
--
--
--
275
326
13.9909
63.8180
12.5727
0.6645
18.9218
17.8324
2.0608
11.2944
11.8179
0.2705
0.2268
Dilbit
(21.5°)
Canada
1:200
224
331
120.6515
41.0730
1.0454
218
285
50.9834
40.7590
--
--
--
--
275
326
13.4098
63.8180
12.3240
0.7111
17.3317
17.8461
1.7589
10.9085
11.6320
0.2531
0.2152
1:100
224
335
125.8546
41.0730
1.0587
218
285
44.4385
31.7020
--
--
--
--
275
326
14.3305
63.8180
13.2762
1.1542
11.5027
21.0296
2.6073
12.3514
12.9223
0.4954
0.2521
1:20
224
340
133.1499
41.0730
1.1657
218
285
55.7758
233
441
15.7453
275
326
15.3840
72.9890
14.0151
7.2938
1.9215
23.0364
13.6412
15.1511
14.9083
3.6527
1.4960
Hondo
Heavy
4.41
Santa Barbara
0
221
331
283.0362
41.0730
1.0889
215
290
55.1274
--
--
--
--
278
335
27.6105
54.7840
27.0154
1.4698
18.3809
43.2085
5.8964
26.0841
26.9209
0.4113
0.2916
(19.5°)
Channel,
1:200
221
335
312.2745
41.0730
1.0758
218
290
62.6687
--
--
--
--
275
335
30.6188
54.7840
29.9090
1.9443
15.3831
50.0240
6.6702
30.1853
31.3212
0.3143
0.3773
California, US
1:100
221
335
274.7955
41.0730
1.0745
215
285
49.9639
--
--
--
--
278
335
26.6222
54.7840
26.0164
1.4917
17.4407
39.2872
5.2774
25.4749
25.4756
0.3729
0.3969
(water depth 260
1:20
221
335
288.0102
41.0730
1.0747
215
285
51.1261
31.6770
--
--
--
--
278
335
27.9249
54.7840
27.6591
1.6937
16.3307
44.4916
6.3517
27.1693
27.9998
0.3853
0.3321
IFO-40
Heavy
2.51
unknown
0
224
335
1173.9084
41.0730
1.2005
221
303
120.4664
--
--
--
--
275
335
114.0423
50.2000
109.7049
3.9261
27.9426
155.5248
7.4990
103.6229
111.3150
1.6747
0.8991
(21.9°)
1:200
221
335
1246.6349
41.0730
1.2208
218
303
130.7611
--
--
--
--
275
335
119.5841
50.2000
115.7084
2.4241
47.7317
160.5839
6.8963
105.1693
115.5723
0.9567
0.4653
1:100
224
335
1338.5599
41.0730
1.2469
218
303
144.2245
--
--
--
--
278
335
128.8407
50.2000
126.7588
3.5465
35.7420
181.6527
7.1909
118.8126
128.3301
2.0773
0.8103
1:20
224
335
1458.7885
41.0730
2.2908
218
303
168.9769
248
520
54.9662
278
335
151.8776
54.7840
148.7639
32.5496
4.5704
222.4382
41.1861
153.5833
156.2830
36.6263
13.9537
IFO-120
Heavy
2.89
unknown
0
221
335
3030.6917
36.4960
1.6769
218
299
356.9362
--
--
--
--
275
326
288.5176
45.6200
253.8040
2.3142
109.6730
238.3292
10.3519
156.8753
198.5147
0.6383
0.7026
(18.4°)
1:200
221
335
2903.2101
36.4960
1.6130
221
299
284.1346
--
--
--
--
275
335
278.0117
45.6200
245.6221
2.4122
101.8258
216.1829
14.6200
151.8306
189.2919
1.0311
0.7444
1:100
221
335
3090.2251
36.4960
1.6501
218
303
372.9288
--
--
--
--
275
335
289.7281
45.6200
260.1135
3.0331
85.7579
229.2055
6.2012
156.3515
190.8314
1.2379
0.4351
1:20
221
335
2527.7304
41.0730
1.6704
215
299
362.2292
--
--
--
--
278
326
236.1044
45.6200
215.0700
6.2661
34.3231
189.9474
15.3716
139.1934
165.5095
4.2343
1.3665
IFO-180
Heavy
1.54
unknown
0
224
335
1263.0471
41.0730
1.2176
221
294
93.4542
--
--
--
--
278
335
118.9519
45.6530
115.4168
3.4277
33.6718
155.2826
9.0125
105.8880
114.2930
1.5280
0.7810
(14.1°)
1:200
224
335
1394.4230
41.0730
1.2311
221
303
126.8108
--
--
--
--
278
335
131.1659
45.6530
126.7468
3.5129
36.0808
172.7134
10.4629
122.5754
123.2982
1.5977
0.8673
1:100
224
335
1703.5484
45.6530
2.3720
218
303
158.7036
--
--
--
--
278
335
164.8318
50.2370
158.9337
4.3777
36.3057
229.6268
43.9836
166.9271
164.3761
20.2543
7.7073
1:20
224
335
1532.9881
41.0730
1.6277
218
303
148.5740
--
--
--
--
275
335
149.2341
54.7840
144.6928
11.8678
12.1921
213.5766
20.6104
148.9958
150.1388
8.5394
3.8566
IFO-300
Heavy
1.72
unknown
0
224
335
720.5500
41.0730
1.1149
221
299
58.8131
--
--
--
--
278
335
66.4979
45.6200
61.3247
1.0011
61.2559
58.9409
2.4890
44.3442
46.5557
0.3989
0.2483
(11.9°)
1:200
224
335
443.5128
41.0730
1.1449
215
303
41.1199
--
--
--
--
275
335
42.2335
45.6200
38.6145
1.4936
25.8539
41.4059
1.6533
31.7921
32.6736
0.3091
0.2193
1:100
224
335
465.9054
41.0730
1.0654
215
290
28.6954
--
--
--
--
275
335
43.3745
45.6200
40.2103
0.8799
45.6987
42.1867
2.8316
32.0261
33.8555
0.3175
0.1768
1:20
224
335
661.4977
41.0730
1.0959
221
290
44.2314
--
--
--
--
275
335
63.3623
45.6200
58.2410
1.4631
39.8066
62.7861
3.4138
45.4621
49.9825
0.5846
0.3138
Santa Clara
Heavy
2.85
Ventura County,
0
224
335
157.2981
41.0730
1.0544
218
290
39.5859
--
--
--
--
278
335
15.1363
45.6530
14.2036
0.5478
25.9289
16.7196
2.0912
9.4978
10.3678
0.1712
0.1965
(22.1°)
California, US
1:200
224
335
147.5528
41.0730
1.0530
218
285
37.7030
--
--
--
--
278
335
13.8842
45.6530
13.0702
0.5749
22.7335
15.6790
2.0057
8.4954
9.2278
0.1158
0.1535
1:100
224
335
154.9767
41.0730
1.0753
218
285
39.5709
--
--
--
--
278
335
15.3927
45.6530
14.4400
0.7907
18.2618
16.0674
2.3119
10.0251
10.3353
0.2387
0.2503
1:20
224
335
169.3872
41.0730
1.1187
218
289
40.8290
--
--
--
--
275
335
16.4747
45.6530
15.6909
2.2520
6.9675
20.2081
4.4623
11.7720
12.5627
1.0095
0.5811
------- |