1 Office of Research and DevelopmentNational Risk Management Research Laboratory, Ground Water and Ecosystems Restoration Division, Ada, Oklahoma

Flux-Based DNAPL SiteRemediation: Field Case StudiesSuresh Rao, School of Civil Engineering, Purdue University

Triad Conference – June 10, 2008

0 10 20 30 40 50 60 70 80 90

4660

4665

0 10 20 30 40 50 60 70 80 90

4660

4665

4670

0 10 20 30 40 50 60 70 80 90

4660

4665

0 10 20 30 40 50 60 70 80 90

4660

4665

4670

0 10 20 30 40 50 60 70 80 90

4660

4665

0 10 20 30 40 50 60 70 80 90

4660

4665

4670

Control Plane

z

y

x

nJ

q0

2

Flux-Based Characterizationat ~30 Field Sites

• Federal Sites– Hill AFB; Patrick AFB; Dover AFB; Vandenberg AFB;

Ft Lewis; Ft Devin; Indian Head; Port Hueneme; Paris Island; Cape Canaveral; Charleston; SRS; NASA

• Other Sites– Manufacturing Facilities in Indiana & Illinois – Dry Cleaner sites in Florida & Indiana

• International Sites– Borden CFB, Canada– Wales, UK– Australia (4 sites)

3

Four DNAPL Case Studies• Three large sites, one small site• Two source zones treated, one under

consideration• All sites had considerable archived data• At all sites new flux measurements made, PFM

deployments & Integral Pump Tests conducted• We will examine:

– Source & plume characterization– Source remediation performance assessment– Source & plume treatment options– Implications to long-term stewardship

4

DNAPL Site - 1

Patterson et al., EST 2006

MW46

-9-8-7-6-5-4-3-2-101

0 2 4 6 8qd (cm/day)

Elev

atio

n (m

AH

D)

0 2 4 6J(mg/m2/d)

q (cm/d) TriBE trans-DBE cis-DBE

MW47

-9-8-7-6-5-4-3-2-101

0 2 4 6 8qd (cm/day)

Elev

atio

n (m

AH

D)

0 25 50 75 100J(mg/m2/d)

q (cm/d) TriBE cis-DBE trans-DBE

MW22

-9-8-7-6-5-4-3-2-101

0 1 2 3 4 5qd (cm/day)

Elev

atio

n (m

AH

D)

0 2000 4000 6000J(mg/m2/d)

q (cm/d) TriBE cis-DBE trans-DBE

MW48

-9-8-7-6-5-4-3-2-101

0 2 4 6 8 10qd (cm/day)

Elev

atio

n (m

AH

D)

0 2000 4000 6000 8000

J(mg/m2/d)

q (cm/d) TriBE cis-DBE trans-DBE

MW24

-9-8-7-6-5-4-3-2-101

0 1 2 3 4 5qd (cm/day)

Elev

atio

n (m

AH

D)

0 100 200 300J(mg/m2/d)

q (cm/d) TriBE cis-DBE trans-DBE

Source Control Plane Fluxes

Unpublished Data: Do Not Cite or Distribute

6

Cross Section of Source Transect

MW46

MW47

MW22

MW48

MW24

q = 1-2 cm/day

q = 3-6 cm/day

2200

1500

2200

1700

660

4400

<1

48

<118

18

<1

<1

<1

Contaminant Fluxes are in mg/m2/day

<1

Mass Discharge:~ 38 kg/year

Unpublished D

ata: Do N

ot Cite or D

istribute

DNAPL Plume Mass• Plume mass (Mp) was estimated from integration of plume data (monitoring over 2002-2007 period)

• Most of the TriBE mass is found in the Intermediate aquifer

• Total mass of TriBE has stabilized at ~100-200 kg (additional mass in sorbed phase; R=??)

• Might be approaching “steady-state”conditions; stable plume??

• Need to consider mass of DBE, VB, Br.

Unpublished Data: Do Not Cite or Distribute

Patterson et al., EST 2006

Mass in Plume (kg)

8

Lessons Learned: DNAPL Site 1•Source mass: ~1.5-3 mT? (500-1,000L?)•Contaminant mass discharge from a small area of control plane

•Archived data integrated with flux data for an improved Conceptual Site Model

•Partitioning Inter-well Tracer Tests & Integral Pump Tests recently completed

•Aggressive source remediation & plume management planned

•Post-remediation monitoring to establish effectiveness & design long-term stewardship

HistoryTCE plume created due to activities associated with the production of detonators from the Second World to the 1970s

Hydrogeology• Multiple interconnected aquifers

• 4 quaternary aquifers and three tertiary aquifers

• v = 20 – 40 m/yr

• Water table ~ 10 to 15 m bgs

Plume Characterization• Plume monitored since 1996

•1,500 m long, 300 m wide and 16 m deep Remedial activities to dateShallow excavation in source area

Proposed Site RemediationChemical oxidation of source??

Permeable Reactive Barrier??

Remediation driven by:• Industrial redevelopment• Flexible regulatory environment• Cost constraints (passive vs aggressive)

DNAPL Site-2

MW 205

MW 227

MW 222

MW 18

MW 1

MW 225

MW 224

MW 223

MW 219

Plume Control Plane Wells

MW219

MW6

MW5B

MW5A

MW241

MW21810

15

20

25

30

35

0 20 40 60

length along transect (m)

dept

h bg

s (m

)Source Control Plane Wells

MW4A

MW4B

MW3AMW1A

MW1B

MW1C

MW2A

MW2B

MW3B

10

15

20

25

30

35

0 5 10 15 20 25

length along transect (m)

dept

h bg

s (m

)

10 μg/L

MW 221

MW 220

200 0 200 400 meters

TCE Plume with Monitoring Wells& Two Flux Control Planes

11

Source Transect Well

Plume Transect Well

13

16

19

22

25

28

31

0 2 4

qd (cm/day)

dept

h bg

s (m

)

0 50 100J(mg/m2/d)

• Source Strength = 3 g/day (other sites: 100 – 400 g/day)

• Negative Correlation

• Plume Strength = 6 g/day• Positive Correlation• No degradation

Basu et al., JCH 2008 (in review)

Well 241&218

13

16

19

22

25

28

31

0 2 4

qd (cm/day)

dept

h bg

s (m

)

0 50 100J(mg/m2/d)

q (cm/d)

J (mg/m2/d)

Clayey sand

Sandy Clay

Clay

Clay

TCE & Groundwater Flux Profiles:PFM Deployments

• Several small areas with conc >1500 ug/L • Plume disconnecting from a depleted source zone• Some shifts in plume shape• Decrease in source mass discharge

TCE Plume Dynamics

Source Mass Estimation

Method A:

( )( ) ( )2006 1996

1996

exp 10

exp 10d t

M M k

V AC k k=

= −

= −

Method B:

M2006 = present source mass

< 10 kg

,0,0

00

exp 2250t

DP D

MM M t dt kg

M⎛ ⎞

= − =⎜ ⎟⎝ ⎠

∫

,0, ,0

0

exp 1 /DD t D

MM M t kg year

M⎛ ⎞

= − =⎜ ⎟⎝ ⎠

M0 = 2,260 kg

MD,0 = 170–365 kg/yr

M2006 ~ 10 kg

14

Proposed Remediation at Site: Chemical Oxidation of SourceObservations

1. Source strength small compared to other sites (3 g/day)

2. Flux data indicate negatively correlated source distribution – high concentrations in low flow regions

3. Mass discharge at source and plume control planes of similar magnitude: plume degradation rates ~ 0

4. Source concentrations are decaying and will attain irrigation standards in <10 years

5. Source mass <10 kg, Plume mass ~ 3800 kg

Interpretations1. Source removal is not important

at this site??

2. Source treatment maybe inefficient due to accessibility of isolated ‘hotspots’

3. Plume remediation or containment is vital

4. Source removal is not important?

5. Plume remediation more important than source

Recommendations:

1. No source removal is necessary

2. Plume treatment or containment maybe required

Integration of Historic Data with Mass Flux Measurements

0 100 200 300 400

Feet

Area 1Area 2

Area 3

• TCE source zone delineated (three major zones: NA1, NA2, NA3)• Electrical Resistive Heating (TRS) of source zones• Measured mass flux using Passive Flux Meters (PFMs) and Integral Pump Tests (IPTs) -- before and after source treatment at NA1 & NA3

Ft Lewis, WA

16

Flux Transect Wells

NA1Source

Area

17

Hill AFB OU2HistoryCreated due to disposal of chlorinated solvents from degreasing operations during 1967 to 1975

Hydrogeology• Shallow unconfined aquifer

• K ~ 2 m/day

Plume Characterization• DNAPL pooled in the paleo-channel forming a source

• 900 m long, 150 m wide, and 10 m deep plume

Source Remediation Activities• 40,600 gallons DNAPL recovered

• In 1996 containment wall constructed around the source area

• In 1997, additional DNAPL discovered in a depression in the clay surface, just north of the containment wall in Panel 5.

• Study initiated to investigate effects of mass removal on mass discharge

• SEAR resulted in TCE mass depletion 1,300-2,200 kg; >70% reduction??

7 August 2003

Panel 5 Clay Aquitard SurfaceFigure 1

Operable Unit 2

Hill AFB, Utah

0

4

8

12

16

20

U2-154 U2-152 U2-150 U2-148 U2-116 U2-149 U2-151 U2-153

Well

Pre

TCE

Mas

s D

isch

arge

(g/d

ay)

0.0

0.4

0.8

1.2

1.6

2.0

Psot

TC

E M

ass

Dis

char

ge

(g/d

ay)

PrePost

0

0.5

1

1.5

2

2.5

3

1 2 3 4 5 6 7 8 9 10

Well ID

Pre

Mol

ar D

isch

arge

(m

oles

/day

)

0

0.05

0.1

0.15

0.2

0.25

0.3

Post

Mol

ar D

isch

arge

(m

oles

/day

)

PrePost

Field studies indicate significant decrease in mass discharge from source zones after DNAPL mass removal.

Effect of Source Remediation

Hill AFB Surfactant Remediation

Flux reduction >98%

Ft Lewis Thermal Remediation

Flux reduction >90%

Brooks et al. JCH In review

20

DNAPL Mass Depletion:Ft Lewis EGDY Thermal

NA1

NA2

NA3USACE, 2007 Draft Report

NA2

21

N

LC-213

LC-212

LC-211

LC-207

LC-206

LC-205

LC-204

LC-203

LC-202

LC-201

-0.005

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

0.045

0.05

0.055

0.06

0.065

0.07

TCE Flux

(mg/cm2/hr)2313 332313 33

(Feet bgs)

Fort Lewis EGDY NA1

Pre-Remediation(Baseline GW Data)

USACE, 2007 Draft Report

Fort Lewis EGDY NA1

LC-213

LC-212

LC-211

LC-207

LC-206

LC-205

LC-204

LC-203

LC-202

LC-201

-0.005

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

0.045

0.05

0.055

0.06

0.065

0.07

TCE Flux

(mg/cm2/hr)2313 332313 33

(Feet bgs)Thermal Treatment Cost ≈ $6M; Flux cost ≈ 2%

Peak-Thermal(Performance Monitoring)

USACE, 2007 Draft Report

USACE

April 2004

N

23

Lessons Learned: Ft Lewis EGDY• Detailed site characterizations required to locate and

delineate sources (cost & level of effort?)• Initial NAPL mass estimates; later revised with additional

analysis & interpretation.• Thermal treatment of NA1, NA2, NA3• ~2,600 kg TCE recovered during thermal treatment

(~70% reduction)• GW fluxes are large (q ~ 25 cm/day)• Thermal treatment reduced (>95%) TCE & DCE mass

discharge • Flux distributions across NA1 source control plane show

small area contributed most of mass discharge• How would the source remediation decisions have

changed if the flux distribution data were used?

24

Lesson Learned: Hill AFB OU2• TCE Mass discharge estimates using three

methods are in agreement.• TCE mass discharge is across a fraction of the

source control plane.• Source treatment (SEAR) resulted in TCE mass

depletion (1,300-2,200 kg removed; ~70% reduction) and an associated decrease (>90%) in TCE mass discharge.

• Increase in DCE mass discharge suggests increased biodegradation after SEAR.

• How does the source treatment influence plume evolution?

DNAPL Site Attributes

P&TNone?None?P&TPlume RemediationRecommendation

Flushing?None?ChemOx?ERH (99%+)Source Remediation

Site Management~10~1~0??Avg. Deg Rate Constant, k (yr-1)

~500~3,800~1,000~10,000Mass, Mp (kg)

250x50x151500x40x11500x120x155,000x500x50Dimensions(m x m x m)

Current Plume

30 x 10 = 30025x15 = 37510x5 = 5070x12 = 840Control Plane, Acp (m2)

~0.10.40?0.2Depletion Rate Constant, α (yr-1), α=[qAcpC0/M0]

~100~3~300~800Mass Discharge, MD, (g/day)

~150~570~10Max. Conc., Cmax (mg/L)

~2,500<100??~5,000Mass, m0 (kg)Current Source

5 (3 & 20)2.5 (1 & 13)10 (6 & 40)30Groundwater Flux, q (cm/day)

Site Hydrogeology

Oz-2Oz-1MidwestUS

NorthwestUSSITE PARAMETERS

REMChlor Simulations of Management Options

x ( m )

y(m

)

0 1 0 0 2 0 0 3 0 0 4 0 0 5 0 0 6 0 0-1 0 0

0

1 0 0t = 3 0 y e a rs

C o m p lia n c e B o u n d a ry

x (m)

y(m

)

0 1000 2000 3000-200

0

200 t = 60 yearsx(m)

y(m

)

0 500 1000 1500 2000 2500 30-200

0

200

50025055

TotalChlorinatedVOCs(ug/L)t=now

Plume shrinks at both ends Plume marches on at leading edgeBoth cases 90% mass depletion

SITE 1SITE 2

Control Plane

z

y

x

MD(t)

Total Chlorinated Solvents t=now

27

What Guides Remediation Choices?

1. Remedial ObjectivesCompliance boundary (“everywhere” or at specified POC?)Performance Metrics:

concentration or flux?Mass depletion & residual mass

Source and/or plume?2. Remediation timeframe

Short-term responsesLong-term site stewardship

3. Site CharacteristicsSource strength and longevityDegradation rates in the plume

Source Remediation Options?

MD (t) decreases over time (exponential?); so, if λ (t) is constant (?), eventually all plumes start shrinking–** requires long term stewardship for at least a century

1. No remediation

4.Integrated "Treatment Train" Approach: (2) + (3) → Implement "aggressive" short-term action to deplete most of the source mass (say, ~80%?), and then use the "passive" source treatment (nZVI or eZVI) to sustain essentially zero source flux, or a low-grade “chaser” of say chem ox

Plume response similar to 2; but, since the source mass hasn't been reduced, the source treatment has to be maintained for a very long time

3. Reduced Source Flux as in the case of n-ZVI or enhanced bio

Plume shrinks inwards from both ends

Plume “pinched off” at the head but tail continues to move forward –split plumes?

Plume “pinched off” at the head; tail starts shrinking when reduced MD reaches it

2. Reduce source mass through some aggressive source depletion action (e.g., flushing, chem ox, thermal, etc) MD(t) ↓

ShrinkingMD(t) < λ (t)

AdvancingMD(t) > λ (t)

StableMD(t) = λ(t)

options

29

Summary Comments• Groundwater & contaminant flux characterization

at DNAPL source control planes provides critical information needed in source remedy selection and performance assessment.

• “Source Strength” can be used to compare sites, and the required source strength reduction through mass depletion can be determined based on likely plume response.

• Regulatory framework and policy guidance lacking for adoption.

30

Discussion Items• How much longer will we debate the

benefits/limitations of partial mass removal for DNAPL source zones?

• If “MCL everywhere” can’t really be achieved “at reasonable cost and in reasonable time,” what are the alternatives for site remediation?

• If some DNAPL mass is left behind in the source zone, what are the implications to site stewardship (costs, risks, liabilities, etc)?

• How do short-term discounting procedures influence site remediation decisions at government & corporate sites?

31

Questions?