biovapor model: use, application,...

1

Use this area for cover image(Maximum height 6.5cm & width 8cm)

Biovapor Model: Use, Application, Comparisons

in:

Petroleum Vapor Intrusion Workshop

Sunday, September 13, 2015 1:00 to 5:00 pm at:

25th National Tanks Conference and Expo

Phoenix

September 14 – 16, 2015

George DeVaull

george.devaull@shell.com

2

Workshop Agenda

Welcome, Introductions, Safety Issues

Update on ITRC PVI Guide

Update on EPA OUST

Screening Criteria

BioVapor Model; Use, Comparison. Applications

Lead Scavengers: BioVapor

Lead Scavengers: Case Studies

Sampling and Analysis

Case Studies/ Lessons

Summary30 minutes

3

BioVapor Use, Comparison. Application

To Be Covered:

Model Introduction

Application Examples

4

Model Use in Evaluating the PVI Pathway

Model Use in regulatory program varies

In states where VI modeling may be applied:

� The sole basis for eliminating consideration of the VI pathway (11 states)

�A line of evidence in the investigation (7 states)

� If applied, may require confirmatory sampling (8 states)

And continues to evolve …

From:

MADEP. 2010. Vapor Intrusion/Indoor Air Guidance Survey. Report prepared for Massachusetts

Department of Environmental Protection by Parsons, Boston, MA. Final, July.

5

Johnson and Ettinger (1991): Heuristic model for predicting the intrusion rate of

contaminant vapors into buildings, Environ. Sci. Tech., 25:1445-1452.

� Applied: ASTM E2081-00; E1739-95; USEPA, 2003; others

USEPA OSWER - Subsurface Vapor Intrusion Guidance (2002):

� “The draft guidance recommends certain conservative assumptions that may not be

appropriate at a majority of the current 145,000 petroleum releases from USTs. As

such, the draft guidance is unlikely to provide an appropriate mechanism for

screening the vapor pathway at UST sites.”

Tillman, F.D. and J.W. Weaver, 2005, Review of recent research on vapor intrusion,

EPA/600/R-05/106

� “While caution would require the evaluation of the soil-to-indoor air pathway for

all subsurface contamination, there are, in fact, not many cases of proven vapor

intrusion documented in the scientific literature. This is particularly true for organic

vapors subject to aerobic biodegradation, such as gasoline compounds (petroleum

hydrocarbons).

J&E Model: Subsurface Vapors to Indoor Air Vapor Intrusion

6

Questions (API): Roger Claff, claff@api.org, 202-682-8399; Bruce Bauman, Bauman@api.org, 202-686-8345

Acknowledgements: Tom McHugh, Paul Newberry, GSI Environmental, Houston.

American Petroleum Institute BioVapor Model

Download at: www.api.org/pviOR Navigate www.api.org to Environment, Health & Safety > Soil & Groundwater Research > Vapor IntrusionFree, asks for registration information (update notification)

7

Model Use Comparison

Complex: numerical, multi-dimensions, time-dependent

� intensive computation, potentially few users

� Explore building / foundation interaction details

� Lateral building / foundation to source separation

� Can be ‘stiff’ (numerically unstable)

Simple: analytical, semi-analytical, one-dimension

� Very fast calculations

� Multiple chemicals, oxygen sinks, no problem

� Sensitivity estimates are realistically possible

� Insight into trends, sensitivity, key parameters

� Easily coded and run

Yao and Suuberg, 2013: A Review of Vapor Intrusion Models, ES&T

Many models are available … tradeoffs

8

API BioVapor: Use

Structure

� Menu-driven

� Microsoft Excel™ spreadsheet

� Open, unlocked, reference guidance

Input:

� Same or similar parameters as Johnson & Ettinger model

� Similar conceptual model & caveats on model applicability and use.

� Includes ‘oxygen-limited aerobic biodegradation’ (DeVaull, ES&T 2007)

� Additional Parameters and Information

� Either can be readily estimated, or

� Included in database (example: chemical-specific aerobic degradation rates)

Key:• Quantify the contribution of aerobic biodegradation• Available and relatively easy to use

9

BioVapor: Menus & output

10

Petroleum Biodegradation Conceptual Model

Key Idea: oxygen consumption andhydrocarbon attenuation are directly correlated

ambient

air

petroleum

vapor

source

oxygen flux

(down)

petroleum flux

(up)

transition point

outdoor air

below foundation

indoor air

source

Building Resistance (walls, roof)

Foundation Resistance

Soil Resistance (aerobic)

Soil Resistance (anaerobic)

11

Oxygen below Buildings: Basis

Aerobic Biodegradation

� Hydrocarbon to Oxygen use ratio: 1 : 3 (kg/kg)

� Atmospheric air (21% Oxygen; 275 g/m3 oxygen) provides the capacity to

degrade 92 g/m3 hydrocarbon vapors (92,000,000 ug/m3)

Oxygen below a Foundation: can it get there?

� Through the foundation

� Equate to same transport parameters as other VI chemicals

� Around the foundation edges (bonus)

� Additional oxygen

Key: Oxygen below a foundation• Can oxygen get there?• Is there enough oxygen to support significant aerobic biodegradation?

12

Oxygen in the BioVapor Model

Three Options:

1. Specify Aerobic depth

� Measure vapor profile

2. Specify Oxygen concentration under a foundation

� Measure oxygen

3. Let the model balance hydrocarbon & oxygen consumption

� Specify vapor source composition (gasoline vapor, etc.)

� Estimate or measure hydrocarbon source

Key:• Pick one method; the others are related (and predicted)• Relatively unique to this model (particularly #3)

13

Aerobic Petroleum Biodegradation Rates in SoilAerobic Petroleum Biodegradation Rates in SoilAerobic Petroleum Biodegradation Rates in SoilAerobic Petroleum Biodegradation Rates in Soil

kw = 0.48 /hr (0.08 to 3.0)

Aromatic Hydrocarbons

kw = 40 /hr (7.8 to 205)

Aliphatic Hydrocarbons

• Chemical-Specific Rates

DeVaull, 2011: Biodegradation rates for petroleum hydrocarbons in aerobic soils: A summary of measured data, International Symposium on Bioremediation and Sustainable Environ. Technol., June 2011, Reno.

ww

ieff

Rk

HDL

⋅

⋅=

θ‘reaction length’

Rates for ‘air-connected’ soils

14

Diffusion-Limited Homogeneous Reaction Rates in Soils

Hydrocarbon and oxygen may deplete before permeating through

soil agglomerate. The effect is a reduced observed degradation rate

when diffusion coefficients are lower.

�Quantify: Thiele (1939); Best (1955); Bosma (1997)

Bulk Soil

Oxygen at

Surface

Hydrocarbon

At Depth

Pore Scale

plane of

symmetry

pore

Lp

Concentration

In Soil

soil

particleO2

CxHyO2

CxHy

Concentration in Soil Pores

or microbes, flocs,

biofilm, or soil

agglomerate scale

15

Data: Reduced observed degradation rate

Slower observed water-phase rates at lower air-filled soil porosity

�Predictable by reduced diffusion through soil matrix

BTEX data

median

diffusive column

advective column

soil gas profiles

unsaturated microcosm

kw = kw,0 / ( 1 +Deff / Deff,0 )

0.001

0.01

0.1

1

10

100

1000

0 0.1 0.2 0.3 0.4 0.5

Ae

rob

ic D

eg

rad

ati

on

Ra

te k

w(1

/ho

ur)

Soil Pore Air (cm3/cm

3)

quartiles (range)

Asymptote is high-end

attenuation rate

(aerobic) in groundwater

omit ~

time

16

Reaction Length

For diffusion-dominated transport in (aerobic) vadose zone soils

Distance over which degradation occurs approaches a

lower limit (does not approach zero; not instantaneous)

omit ~

time

�Distance: Diffusion-Reaction Length

� Implication: Modeling could be extended into capillary fringe

17

BioVapor Use, Comparison. Application

To Be Covered:

Model Introduction

Application Examples

Model / Data & Sensitivity Examples

18

Model Application 1: Compare 1-D to 3-D Estimates

3D: Abreu 2009: GWM&R

�& API Publ. 4555

Basement Scenario

Matched Parameters

� Except “Depth”

10-15

10-14

10-13

10-12

10-11

10-10

10-9

10-8

10-7

10-6

10-5

10-4

10-3

10-2

0.1 1 10 100 1000

Ind

oo

r to

So

urc

e V

ap

or

Co

nce

ntr

ati

on

Ra

tio

Source Vapor Concentration (mg/L) (g/m3)

Basement Scenario

BioVapor

0.65 m

1.3 m

1.8 m

2.5 m

3.1 m

4.3 m

6.1 m

Abreu (2009) 3D

1 m

2 m

3 m

4 m

5 m

7 m

10 m

0

2

4

6

8

Ae

rob

ic D

ep

th (

m)

NAPL Dissolved Phase

19

Application 1A: Scenario Type Classification

� Lower Concentration Source� Dissolved Groundwater Source

� Clean Soil Model

� Lower VOC flux

� Lower Oxygen Demand

� Higher Concentration Source� LNAPL Source

� Dirty Soil Model

� Higher VOC Flux

� Higher Oxygen Demand

O2

HC

reaction

zone

O2

HC

reaction

zone

ex

clu

sio

n

dis

tan

ce

ex

clu

sio

n

dis

tan

ce

20

Application 1A: Exclusion Distances

DeVaull, G. E., Environ. Sci. Technol. 2007, 41, 3241-3248.

Increase separation distance by a factor of 2, attenuation factor decreases by a factor of 8E-06

Distance is a much more robust screening factor than an

attenuation ratio.

21

Application 1A: Exclusion distance

No detects at all in this quadrant

Lahvis, M.A., et al., Vapor Intrusion Screening at Petroleum UST Sites, Groundwater Monitoring and

Remediation [Article first published online: 21 Feb 2013].

Low % detect & conc. in this quadrant

Scatter plot – soil gas vs. distance from water table

22

Application 2 – Measured Data to BioVapor Comparison

Beaufort, South Carolina

� Favorable comparison of petroleum & oxygen concentrations

Data: Lahvis et al., Water Resources Research, 1999, 35, 3, 753-765.

23

Application 2A – Measured Data to BioVapor Comparison

Ratio of indoor to source vapor concentration: BTEX

24

Model Application 3: Extreme ConditionsPotential “worst case” indoor air concentrations

Key Ideas: “Worst Case” Conditions• Same for or Building, Soils and Vapor Source• Opposite Extreme for Foundation Type

Building Foundation Types:• Non-degrading chemicals:

• High Vapor Flow Through Foundation

• Aerobically degrading petroleum:

• Low Oxygen (Air) Flow through Foundation

25

Sensitivity Analysis 1:

“Some required or optional model inputs parameters such as oxygen

concentration below the building foundation and baseline soil oxygen

respiration rate are not commonly measured during site investigation.

…the user should conduct a sensitivity analysis in order to evaluate

the effect of input parameter value uncertainty on the model results”

“Users of this model should not rely exclusively on the information

contained in this document. Sound business, scientific, engineering,

and safety judgment should be used in employing the information

contained herein.”

Neither API nor any….

Weaver, J. (2012). BioVapor Model Evaluation, For 23rd National Tanks

Conference Workshop St. Louis, Missouri, March 18, 2012

BioVapor User’s Guide:

26

Sensitivity Analysis 2:

Parameter importance ranking

� Primary

� Depth, source concentration

� Oxygen content, biodegradation rate, foundation air flow, soil moisture

content

� Secondary

� Air exchange rate, other factors in J&E

� Results will be more strongly dependent on source depth and strength

than analogous J&E, and unless the source is right below foundation, less

dependent on building parameters.

Weaver, J. (2012). BioVapor Model Evaluation, For 23rd National Tanks Conference Workshop St. Louis, Missouri, March 18, 2012.

Picone, S. et al., 2012: Environmental Toxicology and Chemistry, Vol. 31, No. 5, pp. 1042–1052, 2012.

BioVapor versus Johnson and Ettinger:

27

BioVapor Use, Comparison. Application

To Be Covered:

Model Introduction

Application Examples

The Lead Scavengers EDB and 1,2-DCA

28

Background Information: EDB and 1,2-DCA

Lead scavengers ethylene dibromide (EDB) and 1, 2-dichlorethane (1,2-

DCA) were first added to leaded gasoline in the 1930s to prevent alkyl lead

compounds from forming solid lead oxide deposits that could foul engines.

Usage […] was virtually eliminated from gasoline in the United States in the

1980s, although use in certain aviation fuels continues (as of 2015).

8 minutesFrom: Falta, R.W., GWMR, 2004

Remaining:

Piston-Driven Aircraft

� FAA (2013) - underway

http://www.faa.gov/about/initiatives/avgas/

29

Properties, Use, Information

Ethylene Dibromide, EDB

� CASRN 106-93-4, C2H4Br2, MW = 187.86, Tbp = 131°C,Tmp = 10°C, log10Kow=1.96, S = 4310 mg/L, Pv = 11.2 mmHg, H = 6.5E-4 atm-m3/mol. rL = 2.169 g/mL liquid

� Uses: agricultural fumigant (soil, grain, citrus, tobacco), chemical manufacture, solvent, catalyst, exhaust system scavenger in gasolines containing lead antiknock compounds. Naturally produced by marine algae.

� Aerobic Degradation: aerobic – ubiquitous (HSDB): ‘in one laboratory screening study using

100 soils, half-lives ranging from 1.5 to 18 weeks were determined’

1,2-Dichloroethane, 1,2-DCA. ethylene dichloride (EDC)� CASRN 107-06-2, C2H4Cl2, MW = 98.95, Tnbp = 84°C, Tmp = -35°C, log10Kow = 1.45, S = 5100

mg/L, Pv = 80 mmHg, H = 9.79E-4 atm-m3/mol.

� Uses: production of vinyl chloride monomer, a precursor to PVC. Reported in indoor air from plastic [from China] {Doucette, WJ, et al., 2010: GWM&R 30 (1): 67–73.} Solvent, exhaust system scavenger in gasolines containing lead antiknock compounds. Agricultural fumigant (grain).

� Aerobic Degradation: Davis GB, Patterson BM, Johnston CD. Aerobic bioremediation of 1,2

dichloroethane and vinyl chloride at field scale, J Contam Hydrol. 2009 107(1-2):91-100.

30

Prior Evaluation and Information

Groundwater Evaluation: Risk, Natural Attenuation

• Anaerobic Degradation

• Dehalogenation in the presence of fuel hydrocarbons

31

Risk Evaluation - Narrative

USEPA, 2013: Evaluation Of

Empirical Data To Support Soil

Vapor Intrusion Screening Criteria

For Petroleum Hydrocarbon

Compounds, US Environmental

Protection Agency, Office of

Underground Storage Tanks,

Washington, DC. EPA 510-R-13-

001

�Appendix F. Analysis of Lead

Scavengers: Ethylene Dibromide

and 1,2-Dichloroethane

Presuming No Degradation

32

Vapor Intrusion: Exclusion Distance Evaluation

Supportive, but Sparse Detection Data; mostly non-detect

Example: Peargin and Kolhatkar. 2013

33

Different Approach: Estimate Aerobic Degradation Rates

Example: CA LUFT Site #90099 [data from GeoTracker]� Rejzek, T: Vapor Intrusion of the Lead Scavenger 1,2-Dichloroethane (EDC) at LUFT Sites, presentation at

March 2015 AEHS Conference

� One measured soil gas profile [1,2-DCA] with detections; analysis: same profile, same date, no residual

34

1,2-DCA Aerobic Biodegradation

Data Compilation. Derived water-phase aerobic degradation rates.

kw = 0.087 /hr

1

1

35

EDB Aerobic Biodegradation

Data Compilation. Derived water-phase aerobic degradation rates.

kw = 0.0093 /hr

1

1

36

Data References

Pignatello, JJ, CR Frink, PA Marin, EX Droste, Field-Observed Ethylene Dibromide in an Aquifer After Two Decades, Journal of Contaminant Hydrogeology, 5, 1990, 195-214.

Freitas dos Santos, L. M., A. G. Livingston, Mineralisation of 1,2-dibromoethane and other brominated aliphatics under aerobic conditions, Water Science and Technology, 1997, 36(10), 17-25.

Olaniran, A.O.; A. Balgobind, B. Pillay, Quantitative assessment of the toxic effects of heavy metals on 1,2-dichloroethane biodegradation in co-contaminated soil under aerobic condition, Chemosphere 85 (2011) 839–847.

Wang, S.Y., Y.C. Kuo, Y.Z. Huang, C.W. Huang, C.M. Kao, Bioremediation of 1,2-dichloroethane contaminated groundwater: Microcosm and microbial diversity studies, Environmental Pollution 203 (2015) 97-106.

Herbst, B., U. Wiesmann, Kinetics and Reaction Engineering Aspects of the Biodegradation of Dichloromethane and Dichloroethane, Water Res., 30, 5, 1069-1076, 1996.

Rejzek, T., Vapor Intrusion of the Lead Scavenger 1,2-Dichloroethane (EDC) at LUFT Sites, presentation at the 25nd Annual International Conference on Soil, Water, Energy, and Air, Mission Valley, San Diego, California. March 23 - 26, 2014.

Soil Gas Assessment Report, 101 East Victoria Street, Santa Barbara, California (LUFT Site #90099), GeoTracker [accessed June 2015]. http://geotracker.waterboards.ca.gov/

37

BioVapor - Amend the Chemical Database: EDB, 1,2-DCA

edit the database

addchemical

data

* EDB, 1,2-DCA not included in downloadable BioVapor at API.org

38

Model Comparison: 1-D to 3-D

Example nomagram

�used previously

Soil Gas Screening

3D: Abreu 2009: GWM&R

� & API Publ. 4555

Basement Scenario

Matched Parameters

� Except “Depth”

all aerobic all anaerobic

Application: • Repeat this plot & scenario for EDB & 1,2-DCA•Use BioVapor•This is a sensitivity analysis for depth and source concentration

(att

en

ua

tio

n f

ac

tor)

39

BioVapor Modeling

Benzene

Okay for:

� Total Source Vapor > 10.2E+6 ug/m3; Benzene < 3080 ug/L-water; Separation Distance > 1.22 m (4 ft)

� Target indoor air: 0.31 ug/m3 ; Source Concentration: 7.0E5 ug/m3 in soil gas ; need attn. < 4.4E-7

� Dry Sand (porosity 0.375 cm3/cm3, moisture 0.054 cm3/cm3), cair = H · cgroundwater, cair (measured)

� Less effect for lower soil porosity & higher soil moisture

(att

en

ua

tio

n f

ac

tor)

40

BioVapor Modeling

EDB

Okay for:

� Total Source Vapor < 6.0E+6 ug/m3; EDB < 8200 ug/L-water; Separation Distance > 1.79 m (6 ft)

� Target indoor air: 0.0041 ug/m3 ; Source Concentration: 2.1E4 ug/m3 in soil gas ; need attn. < 1.9E-7

� Dry Sand (porosity 0.375 cm3/cm3, moisture 0.054 cm3/cm3) , cair = H · cgw / 10 , cgw (measured)

� Less effect for lower soil porosity & higher soil moisture

(att

en

ua

tio

n f

ac

tor)

41

BioVapor Modeling

1,2-DCA

Okay for:

� Total Source Vapor < 1.8E+7 ug/m3; 1,2-DCA < 1310 ug/L-water; Separation Distance > 1.22 m (4 ft)

� Target indoor air: 0.094 ug/m3 ; Source Concentration: 5.2E3 ug/m3 in soil gas ; need attn. < 1.8E-5

� Dry Sand (porosity 0.375 cm3/cm3, moisture 0.054 cm3/cm3) , cair = H · cgw / 10 , cgw (measured)

� Less effect for lower soil porosity & higher soil moisture

(att

en

ua

tio

n f

ac

tor)

42

Lead Scavengers EDB and 1,2-DCA

Aerobic degradation is observed for EDB & 1,2-DCA in soils

�Degradation rates estimated; manually added to BioVapor database

BioVapor modeling & relative risk of EDB & 1,2-DCA in leaded

gasoline, with respect to vapor intrusion:

� EDB, or 1,2 DCA, when present, has a potential risk envelope similar

to or less than benzene (close source to foundation separation

distance, high source concentrations)

Results shown tend to conservative; a general sensitivity analysis may

also appropriate

�Presumed upper bound soil gas estimates of EDB & 1,2-DCA may be

conservative (20,000 & 5200 ug/m3 respectively)

Field data for EDB & 1,2-DCA in soil gas is very sparse

�Mostly below and near detection limits

43

Summary

BioVapor

� Introduction & Description

�Sensitivity

� Separation Distance, Source Strength, Oxygen Availability

Case Examples

�Model / Data

�Sensitivity

� EDB, 1,2-DCA

Covered:

44

Questions (API): Roger Claff, claff@api.org, 202-682-8399; Bruce Bauman, Bauman@api.org, 202-686-8345

Acknowledgements: Tom McHugh, Paul Newberry, GSI Environmental, Houston.

American Petroleum Institute BioVapor Model

Download at: www.api.org/pviOR Navigate www.api.org to Environment, Health & Safety > Soil & Groundwater Research > Vapor IntrusionFree, asks for registration information (update notification)