Use of Forensics in Petroleum Spill Cases

NYSBA Oil Spill Symposium

June 7, 2017

Jim Occhialini Alpha Analytical

Scott A. Stout, Ph.D.

NewFields, Rockland MA

Topics for Discussion

• Introduction • Chemical fingerprinting • Gasoline case study • Diesel case study • PAH case study • Age-dating

Introduction • About Alpha Analytical

– Conventional & Specialty Laboratory Services • Advanced hydrocarbon analysis

– Saturated hydrocarbons – Alkylated PAHs – Geochemical biomarkers – PIANO – PHI

• About NewFields Environmental Forensics – Consultants/Experts (technical and litigation) – Frequent collaboration with Alpha Analytical – Industrial and governmental clients

Introduction to Petroleum Analysis What’s your application?

• Regulatory compliance – Generally a quantitative determination i.e. “TPH” – Lots of methods

• Risk assessment – Quantitative, but can have qualitative aspects

• Risk based corrective action (RBCA) – ITRC TPH Workgroup

• Qualitative determinations – “What is it?”

• “Routine” (“& not so routine”) product identification – “Whose is it?”

• Forensic / source allocation

Qualitative Analysis Routine Environmental Lab

“Fingerprinting”

• Methods can identify - – Presence or absence of common products – ID based on pattern recognition

• “Forest vs. trees”

• Limitations – ~ 20 minute chromatographic run time – Trouble with mixtures, weathered samples

• “Unknown product” ID

Modified from Douglas et al. (2014) Introduction to Environmental Forensics, Elsevier, New York.

Conventional Lab

Fingerprint

Forensic Lab Fingerprint

WHAT and WHERE? • Unambiguous contaminant identification(s) • Well-defined spatial extent of contaminant(s)

WHEN and WHO? • Age-dating of contaminant(s) • Defensible Allocation of Responsibility Integrated Approach

– Chemical Fingerprinting – Site/Regulatory History – Process Forensics – Geology and Hydrology – Transport Modeling – Numerical Analysis – Allocation

Environmental Forensics

Limitations of Standard (SW-846) Methods

• Standard methods were developed to establish “nature and extent” of prescribed lists of COPCs – “Priority pollutant” chemicals

(n=~130) are only a subset of chemicals contained in complex mixtures of products released into the environment

– Many co-occurring, “nonpriority, pollutant” chemicals provide clues as to the source of the “priority pollutant” chemicals Peel the Onion!

Priority Pollutants

Diagnostic, Non-

priority Pollutants

Tiered Approach to Chemical Fingerprinting of Hydrocarbons

General Characterization

8015M GC/FID (C9+)

GC/FID/MS (C4-C44)

TIER 1 High-resolution “fingerprint”

TPH, DRO, ORO, PHI

Contamination vs. Natural OM Hydrocarbon Product ID

Overall Weathering

TIER 2 Detailed quantitative “fingerprint”

VOC: PIANO, ethers, ethanol, alkyl lead SVOC: PAHs, Alkyl-PAHs, Sulfur-PAHs,

PAH isomers, HPAHs, decalins, n-alkylcyclohexanes, biomarkers, biodiesels

Fuel, lube, waste specifics

Quantitative comparative analysis, weathering & mixture assessment

Detailed Characterization

8260M P&T GC/MS

(PIANO/Ethers)

8270M GC/MS

Alkyl Lead Ethanol

8270M GC/MS-SIM PAHs et al.

8270M GC/MS-SIM Biomarkers

Diamondoids FAMEs

Tier 1: What is it? TPH Interpretations

Diesel MGP Tar

Lube Oil

Tier 1: Not all TPH is contamination

Naturally-occurring organic matter

Coal ash

IS

IS

IS

29

31 27

25

Bacterial membrane debris

Plant waxes

Plant resins

Soil with 2000 mg/kg TPH

OTP

ANDROS

TANE

IS

IS

Soil with 1800 mg/kg TPH

Tier 2: Gasoline/PIANO Fingerprinting

• PIANO data (~90 analytes) can reveal meaningful differences – Weathering – General and

specific blending practices

C5 C13

alkylate

reformate

isomerate

straight-run

reformate

isopentane Methyl-naphthalenes

Former Site

Active Site

Gasoline Case Study

• Former site with historic impacts observes NAPL increase; suspected impact from upgradient active site

• NAPLs (10) and active site dispensed gasolines (3) analyzed via modified EPA Methods 8015, 8260 (PIANO), and 8270 (organic lead).

Gasoline Case Study

• differences in degree of weathering, SRG abundance, and lead concentrations are evident

• specific alkylate type (%Iso) differences (independent of weathering) are also evident – alkylate type varies by

refining process

A

B

A

B

More weathered More SRG 1.6 glpg %Iso ~ 40 (H2SO4)

Less weathered Less SRG 0.15 glpg %Iso ~ 58 (HF)

distinct trimethylpentane isomer patterns

Gasoline Case Study

• Mixing model based on alkylate type used to estimate volume allocation

A

B

Distillate Fingerprinting

• Modified EPA Method 8015 (Tier 1) – whole oil and SHC

• Modified EPA Method 8270 (Tier 2) – Alkylated PAH – Sulfur-containing

aromatics – Low boiling

Biomarkers

Stoddard Solvent

Kerosene

Jet A

Diesel #2

sesquiterpanes

Diesel Case Study

• underground pipeline failure prompts investigation/clean-up in industrial area

Pipeline Product Pr/Ph ~ 2.0

Excavation Pr/Ph ~ 2.0

Nearby Storm Sewer

Pr/Ph ~ 2.0

Tier 1 GC/FID chromatograms (8015M)

Diesel Case Study

• nearby storm sewer diesel contains higher sulfur aromatics

Nearby Storm Sewer

DBT2/PA2: 0.40 DBT3/PA3: 0.57

Excavation DBT2/PA2: 0.29 DBT3/PA3: 0.43

Pipeline Product DBT2/PA2: 0.29 DBT3/PA3: 0.44

dibenzothiophenes phenanthrenes

Tier 2: 52 PAH-related analytes (8270M) 6-rings 2-rings

( )

Diesel Case Study

• nearby storm sewer contains distinct sesquiterpane biomarkers

Nearby Storm Sewer

Excavation

Pipeline Product

Tier 2: m/z 123 extracted ion profiles (8270M)

9 10

1

8

6

2 3

4

5 7

Abbrev. Compound Name/Group Abbrev. Compound Name/GroupD0 cis/trans-Decalin DBT0 DibenzothiopheneD1 C1-Decalins DBT1 C1-DibenzothiophenesD2 C2-Decalins DBT2 C2-DibenzothiophenesD3 C3-Decalins DBT3 C3-DibenzothiophenesD4 C4-Decalins DBT4 C4-DibenzothiophenesBT0 Benzothiophene BF Benzo(b)fluoreneBT1 C1-Benzo(b)thiophenes FL0 FluorantheneBT2 C2-Benzo(b)thiophenes PY0 PyreneBT3 C3-Benzo(b)thiophenes FP1 C1-Fluoranthenes/PyrenesBT4 C4-Benzo(b)thiophenes FP2 C2-Fluoranthenes/PyrenesN0 Naphthalene FP3 C3-Fluoranthenes/PyrenesN1 C1-Naphthalenes FP4 C4-Fluoranthenes/PyrenesN2 C2-Naphthalenes NBT0 NaphthobenzothiophenesN3 C3-Naphthalenes NBT1 C1-NaphthobenzothiophenesN4 C4-Naphthalenes NBT2 C2-NaphthobenzothiophenesB Biphenyl NBT3 C3-NaphthobenzothiophenesDF Dibenzofuran NBT4 C4-NaphthobenzothiophenesAY Acenaphthylene BA0 Benz[a]anthraceneAE Acenaphthene C0 Chrysene/TriphenyleneF0 Fluorene BC1 C1-ChrysenesF1 C1-Fluorenes BC2 C2-ChrysenesF2 C2-Fluorenes BC3 C3-ChrysenesF3 C3-Fluorenes BC4 C4-ChrysenesA0 Anthracene BBF Benzo[b]fluorantheneP0 Phenanthrene BJKF Benzo[j]fluoranthene/Benzo[k]fluoranthenePA1 C1-Phenanthrenes/Anthracenes BAF Benzo[a]fluoranthenePA2 C2-Phenanthrenes/Anthracenes BEP Benzo[e]pyrenePA3 C3-Phenanthrenes/Anthracenes BAP Benzo[a]pyrenePA4 C4-Phenanthrenes/Anthracenes PER PeryleneRET Retene IND Indeno[1,2,3-cd]pyrene

DA Dibenz[ah]anthracene/Dibenz[ac]anthraceneGHI Benzo[g,h,i]perylene

Comparison of Methods PAH-based (EPA 8270)

Parent PAH

Alkylated PAHs

C1-

C2-

C3-

C4-

Extended PAHs

Examples of PAH Fingerprinting

Pyrogenic II • Alkyl < Parent • High 4 to 6 Ring

Pyrogenic I • Alkyl < Parent • High 2 and 3 Ring

Petrogenic • Alkyl > Parent • Little 4 to 6 Ring

MGP Tar

Urban Runoff

2 3 4 5-6 O Rings

N0

N1

N2

N3

N4 B

DF

AY

AE

F0

F1

F2

F3

A0

P0

PA1

PA2

PA3

PA4

DBT

0 D

BT1

DBT

2 D

BT3

DBT

4 FL

0 PY

0 FP

1 FP

2 FP

3 FP

4 N

BT0

NBT

1 N

BT2

NBT

3 N

BT4

BA0

C0

BC1

BC2

BC3

BC4

BB

BJK BB

BE

P BA

P IN

D

DA

GH

I PE

R

RET

BF

2 3 4 5-6 O Rings

“bell-shaped” patterns

“skewed” patterns

Petroleum

PAH Case Study

• Former fuel storage facility located in industrial area along canal

• cPAH source(s) in surface soils elevated and attributed to spilled fuel by regulator

• Chemical fingerprinting study conducted to evaluate source(s) of PAH in surface soils – Tier 1: 8015M – Tier 2: 8270M

PAH Case Study

• PAHs sources can be more confidently determined when TPH is understood

• Tier 1 TPH fingerprinting via 8015M revealed four distinct hydrocarbon sources

GRO RRO DRO

PAH Case Study 1090 mg/kg TPH

40 mg/kg TPH 0.2 mg/kg TPAH51 0.06 mg/kg TPAH16

0.03 mg/kg cPAH

1600 mg/kg TPH 33 mg/kg TPAH51 2.0 mg/kg TPAH16 0.33 mg/kg cPAH

900 mg/kg TPH 3.4 mg/kg TPAH51 0.43 mg/kg TPAH16

0.19 mg/kg cPAH

35 mg/kg TPAH51 19 mg/kg TPAH16

2.4 mg/kg cPAH

Fluoranthenes/ Pyrenes cPAH

cPAH

Phenanthrenes/ Anthracenes

Naphthalenes

PAH Case Study

MGP Tar

Combustion Residue/Ash

Diesel Fuel

Asphalt

PAH Case Study

• Spilled petroleum was a limited source of cPAH

• Historic fill (MGP tar and ash) from canal dredging is dominant source of cPAH

Increasingly Pyrogenic

Surface Soils

Increasingly Pyrogenic

Age-Dating of Gasoline Contamination

• Never simple • Mixing always

confounding • Chemistry is not ‘magic’ • Constrain the age thru

combination of: – chemistry – site or regulatory history – F&T modeling

Gasoline additives (concentration vs. presence/absence)

Blending Practices Sulfur content Lead Isotopes 206Pb/207Pb Degree of Weathering

(simple ratios, volatiles)

Common Approaches

Regulatory Limits on Lead in Gasoline

Leaded Unleaded1926 3.17 Surgeon General1959 4.23 Surgeon General

Jul-74 0.05b Federal Register 38(6), Part II, Jan. 10, 1973

Oct-82 1.1a

Jul-85 0.5a

Jan-86 0.1Jan-92 banned in CAJan-96 banned nationwide Federal Register, 1990

Leaded UnleadedJan-76 3.0 0.05Jan-87 1.1Dec-90 banned nationwideDec-90 0.1c

aaverage quarterly leaded gasoline productionbincidental lead in unleaded gasolineconly permitted in off-highway and marine use

Date

RegulationDate

Clean Air Act, Section 22, Canada Gazette, Part II, 108(15), Aug. 14, 1974

Federal Register, June 8, 1977

Canada

United States Regulation

Stout et al. (2006) Automotive Gasoline. Environmental Forensics, Academic Press, p. 465-531.

Lead Concentration (Avg) LEAD IN LEADED GASOLINES - 1946-1987 (data from Dickson et al., 1987; Shelton et al. 1982)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Winter

1946

Winter

1947

Winter

1948

Winter

1949

Winter

1950

Winter

1951

Winter

1952

Winter

1953

Winter

1954

Winter

1955

Winter

1956

Winter

1957

Winter

1958

Winter

1959

Winter

1960

Winter

1961

Winter

1962

Winter

1963

Winter

1964

Winter

1965

Winter

1966

Winter

1967

Winter

1968

Winter

1969

Winter

1970

Winter

1971

Winter

1972

Winter

1973

Winter

1974

Winter

1975

Winter

1976

Winter

1977

Winter

1978

Winter

1979

Winter

1980

Winter

1981

Winter

1982

Winter

1983

Winter

1984

Winter

1985

Winter

1986

glpg

(avg

.)

REGULARPREMIUM

Regional Datasets show considerable ‘scatter’ Lead in Motor Gasoline Survey 1960-1987 (District 2: Mid-Atlantic Coast Region)

0

0.5

1

1.5

2

2.5

3

3.5S

umm

er 1

960

Win

ter 1

960-

Sum

mer

196

1W

inte

r 196

1-S

umm

er 1

962

Win

ter 1

962-

Sum

mer

196

3W

inte

r 196

3-S

umm

er 1

964

Win

ter 1

964-

Sum

mer

196

5W

inte

r 196

5-S

umm

er 1

966

Win

ter 1

966-

Sum

mer

196

7W

inte

r 196

7-S

umm

er 1

968

Win

ter 1

968-

Sum

mer

196

9W

inte

r 196

9-S

umm

er 1

970

Win

ter 1

970-

Sum

mer

197

1W

inte

r 197

1-S

umm

er 1

972

Win

ter 1

972-

Sum

mer

197

3W

inte

r 197

3-S

umm

er 1

974

Win

ter 1

974-

Sum

mer

197

5W

inte

r 197

5-S

umm

er 1

976

Win

ter 1

976-

Sum

mer

197

7W

inte

r 197

7-S

umm

er 1

978

Win

ter 1

978-

Sum

mer

197

9W

inte

r 197

9-S

umm

er 1

980

Win

ter 1

980-

Sum

mer

198

1W

inte

r 198

1-S

umm

er 1

982

Win

ter 1

982-

Sum

mer

198

3W

inte

r 198

3-S

umm

er 1

984

Win

ter 1

984-

Sum

mer

198

5W

inte

r 198

5-S

umm

er 1

986

Win

ter 1

986-

g/ga

l (av

g.)

Regular

Premium

11,382 Regular gasolines9,448 Premium gasolines

Avg. ± 1σ

Blending and Lead Content Changes Coincident in 1970s

MTBE

CA ban

NY ban

O2 min. lifted

Age-Dating of Diesel Contamination

• Never simple • Mixing always

confounding • Chemistry is not ‘magic’ • Constrain the age thru

combination of: – chemistry – site or regulatory history – F&T modeling

Blending Practices Sulfur content Hydrotreating Biodiesel Degree of Weathering

(simple ratios)

Common Approaches

Sulfur Concentration (Avg) On-Road Diesels

High Sulfur Era

Low Sulfur

Era Ultra-Low Sulfur Era

Distillate Hydrotreatment

No Desulfurization

Hydrodesulfurized

( )

( )

( )

( )

4

1 2

3

4

3

4,6

2,4

2,6

3,6

2,8 2,7 3,7

1,4 1,6 1,8

1,3 3,4 1,7 1,9

2,3 1,2

3E

4,6

3,6

1,4 1,6 1,8

3E

2,4,6

2,4,7 2,4,8

1,4,7

1,3,7

3,4,7

2,4,6

2,4,7 2,4,8

1,4,7

1,3,7

160 mg/kg

178 mg/kg

125 mg/kg

22.9 mg/kg (85% decrease)

45.7 mg/kg (74% decrease)

46.4 mg/kg (62% decrease)

OTP

OTP

A

B

C

Christensen & Larsen Model • Weathering-based “age-dating” method

Age from Historical Records (Years)

2219181817141211998

0.5

Table 1Location, Type of Installation,

and Age of Known Diesel Oil Spills

Location Name Type of Installation

Provestenen, DK Oil TerminalHengelo, depot, NL Oil TerminalFredericia, DK Oil TerminalIshoj, DK Service StationHaarlem, NL Service StationVanlose, DK Service StationHorsholm, DK Service StationNieuwesluis, NL Oil TerminalBrunnik, NL Service StationHengelo, loading rack, NL Oil Terminal

*The site was included because the location was in all respects similar to the other locations. Heating and diesel oils are basically the same, except for additives.

Thisted, DK Service StationEjby, DK Heating Oil Tank*

Christensen & Larsen (1993). Method for determining the age of diesel oil spills in the soil. Ground Water Monitoring & Remediation. 13(4); 142-149.

Premise to Christensen & Larsen

FRESH DIESEL abundant n-alkanes

n-C17/Pristane ~ 1.4

C14

C10 C20

IS IS

IS IS

IS

IS

IS

IS

BIODEGRADED DIESEL no n-alkanes

n-C17/Pristane ~ 0 nPr

Ph

Pr

nPr Ph

Pr

C17

n-Alkanes are more susceptible to biodegradation than acyclic isoprenoids TRUE

Pr

C17

Pr

C17

T (yr) = -8.4(n -C17/Pr) + 19.8Kaplan et al. (1995)

0

4

8

12

16

20

24

0 0.4 0.8 1.2 1.6 2 2.4

n -C17/pristane ratio

Age

of t

he D

iese

l Spi

ll

Data points visually picked from Christensen and Larsen (1993) Figure 4.

Christensen & Larsen Model

Kaplan et al. (1995). Pattern of chemical changes in fugitive hydrocarbon fuels in the Environment. SPE Paper No. 29754.

11.4 ± 2.1 yrs.

nC17/Pr = 1

Principal Critique of C&L Model • Too many site-specific variables control rate(s) of

biodegradation to expect a single, universal rate – O2, nutrient availability, etc. – NAPL mass/concentration

• Insufficient data presented by C&L to evaluate correlation/statistics

• Starting ratio of spilled fuels vary • Almost never know if a single, multiple, or long-

term release has occurred

Stout et al. (2002). Invited commentary on the Christensen and Larsen Technique. Environ. Forensics 3:9-11.

T (yr) = -8.4(n -C17/Pr) + 19.8Kaplan et al. (1995)

0

4

8

12

16

20

24

0 0.4 0.8 1.2 1.6 2 2.4

n -C17/pristane ratio

Age

of t

he D

iese

l Spi

ll

Data points visually picked from Christensen and Larsen (1993) Figure 4.

Elegantly Simple or Overly Simple

? ?

?

?

?

?

?

?

? ?

1.98 ± 0.83

? r2 = 0.89

r2 = 0.66

Conclusions • Environmental forensics (what,

who, when?) requires appropriate data and interpretation

• Tiered analytical approach whose design depends on questions/ objectives

• Integration of good data with knowledgeable interpretation yields greater defensibility in conclusions

Questions?

Jim Occhialini Alpha Analytical

jocchialini@alphalab.com (508) 380-8618

Scott Stout, Ph.D. NewFields

sstout@newfields.com (781) 681-5040 X105