EPA/600/R-05/147
March 2006
Assessment of Vapor Intrusion in Homes
Near the Raymark Superfund Site Using
Basement and Sub-Slab Air Samples
Dominic C. DiGiulio and Cynthia J. Paul
U.S. Environmental Protection Agency
Office of Research and Development
National Risk Management Research Laboratory
Ground Water and Ecosystems Restoration Division
Ada, OK
Raphael Cody and Richard Willey
U.S. Environmental Protection Agency
Region I
Boston, MA
Scott Clifford and Peter Kahn
U.S. Environmental Protection Agency
Region I, New England Regional Laboratory
North Chelmsford, MA
Ronald Mosley
U.S. Environmental Protection Agency
Office of Research and Development
National Risk Management Research Laboratory
Research Triangle Park, NC
Annette Lee and Kaneen Christensen
Xpert Design and Diagnostics, LLC
Stratham, NH
Project Officer
Dominic C. DiGiulio
U.S. Environmental Protection Agency
Office of Research and Development
National Risk Management Research Laboratory
Ground Water and Ecosystems Restoration Division
Ada, OK
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
NATIONAL RISK MANAGEMENT RESEARCH LABORATORY
CINCINNATI, OH 45268
-------
Notice
The U.S. Environmental Protection Agency (EPA) through its Office of Research
and Development funded and managed the research described here through in-
house efforts and under Contract No. 68-C-02-092 to the Dynamac Corporation.
It has been subjected to the Agency's peer and administrative reviews and has
been approved for publication as an EPA document. Mention of trade names or
commercial products does not constitute endorsement or recommendation for
use.
All data generated in this report were subjected to an analytical Quality Assurance
Plan developed by EPA's New England Regional Laboratory. Also, a Quality
Assurance Project Plan was implemented at the Ground Water and Ecosystems
Restoration Division. Results of field-based studies and recommendations
provided in this document have been subjected to external and internal peer and
administrative reviews. This report provides technical recommendations, not
policy guidance. It is not issued as an EPA Directive, and the recommendations
of this report are not binding on enforcement actions carried out by the EPA or by
the individual states of the United States of America. Neither the United States
government nor the authors accept any liability or responsibility resulting from the
use of this document. Implementation of the recommendations of the document
and the interpretation of the results provided through that implementation are the
sole responsibility of the user.
-------
Foreword
The U.S. Environmental Protection Agency is charged by Congress with protecting the Nation's land, air,
and water resources. Under a mandate of national environmental laws, the Agency strives to formulate
and implement actions leading to a compatible balance between human activities and the ability of natural
systems to support and nurture life. To meet these mandates, 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 is the Agency's center for investigation of
technological and management approaches for reducing risks from threats to human health and the
environment. The focus of the Laboratory's research program is on methods for the prevention and
control of pollution to air, land, water, and subsurface resources; protection of water quality in public water
systems; remediation of contaminated sites and ground water; and prevention and control of indoor air
pollution. The goal of this research effort is to catalyze development and implementation of innovative,
cost-effective environmental technologies; develop scientific and engineering information needed by
EPA to support regulatory and policy decisions; and provide technical support and information transfer
to ensure effective implementation of environmental regulations and strategies.
This report describes the results of an investigation conducted to assist EPA's New England Regional
Office in evaluating vapor intrusion in homes and a commercial building near the Raymark Superfund
Site in Stratford, Connecticut. Methods were developed to sample sub-slab air and use basement and
sub-slab air measurements to evaluate vapor intrusion on a building-by-building basis. Using the methods
described in this report, volatile organic compounds detected in basement air due to vapor intrusion could
be separated from numerous other halogenated and non-halogenated (e.g., petroleum hydrocarbons)
compounds present in basement air.
Stephen G. Schmelling, Director/
Ground Water and EcosysigrnrRestoration Division
National Risk Management Research Laboratory
-------
Abstract
This report describes the results of an investigation conducted to assist EPA's New England Regional Office in
evaluating vapor intrusion at 15 homes and one commercial building near the Raymark Superfund Site in Stratford,
Connecticut. Methods were developed to sample sub-slab air and use basement and sub-slab air measurements to
evaluate vapor intrusion on a building-by-building basis. A volatile organic compound (VOC) detected in basement
air was considered due primarily to vapor intrusion if: (1) the VOC was detected in ground water or soil gas in the
vicinity (e.g., 30 meters) of a building, and (2) statistical testing indicated equivalency between basement/sub-slab
air concentration ratios of indicator VOCs and VOCs of interest. An indicator VOC was defined as a VOC detected
in sub-slab air and known to be only associated with sub-surface contamination. Using this method of evaluation,
VOCs detected in basement air due to vapor intrusion could easily be separated from numerous other halogenated
and non-halogenated (e.g., petroleum hydrocarbons) VOCs present in basement air. As a matter of necessity,
radon was used as an indicator compound at locations where an indicator VOC was not detected in basement air.
However, when basement/sub-slab air concentration ratios were compared for radon and indicator VOCs, statistical
non-equivalency occurred at three out of the four locations evaluated. Further research is needed to assess the
usefulness of radon in assessing vapor intrusion.
Holes for sub-slab probes were drilled in concrete slabs using a rotary hammer drill. Probes were designed to allow
for collection of air samples directly beneath a slab and in sub-slab media. Three to five probes were installed in each
basement. Placement of a probe in a central location did not ensure detection of the highest VOC concentrations.
Schematics illustrating the location of sub-slab probes and other slab penetrations (e.g., suction holes for sub-slab
permeability testing) were prepared for each building to document sample locations, interpret sample results, and
design corrective measures. Basement and sub-slab air samples were collected and analyzed for VOCs using
six-liter SilcoCan canisters and EPA-Method TO-15. Sub-slab air samples were also collected in one-liter Tedlar
bags using a peristaltic pump and analyzed on-site for target VOCs. Open-faced charcoal canisters were used
to sample radon gas in basement air. Scintillation cells and a peristaltic pump were used to sample radon gas in
sub-slab air.
Three methods were used to evaluate infiltration of basement air into sub-slab media during air extraction (purging
+ sampling). The first method consisted of sequentially collecting five one-liter Tedlar bag samples at a flow rate
of 1 standard liter per minute and comparing vapor concentration of four VOCs associated with vapor intrusion as
a function of extraction volume. This was performed at three locations with little effect on sample concentration.
This testing also indicated the absence of rate-limited mass exchange during air extraction. Replicate canister
iv
-------
samples representing extraction volumes of 5 to 9 and 10 to14 liters were compared at two locations with similar
results. A second method was then employed which utilized a mass balance equation and sub-slab and basement
air concentrations. When sensitivity of the method permitted, infiltration was shown to be less than 1% at sampled
locations. A third method involved simulating streamlines and travel time in sub-slab media during air extraction.
Air permeability testing in sub-slab media was conducted to obtain estimates of radial and vertical air permeability to
support air flow simulations. Simulations indicated that less than 10% of air extracted during purging and sampling
could have originated as basement air when extracting up to 12 liters of air. Overall, extraction volumes used in
this investigation (up to 14 liters) had little or no effect on sample results.
To assess the need for an equilibration period after probe installation, advective air flow modeling with particle
tracking was employed to establish radial path lengths for diffusion modeling. Simulations indicated that in sub-slab
material beneath homes at the Raymark site (sand and gravel), equilibration likely occurred in less than 2 hours.
Sub-slab probes in this investigation were allowed to equilibrate for 1 to 3 days prior to sampling. A mass-balance
equation was used to estimate the purging requirement prior to sampling. Simulations indicated that collection of
5 purge volumes would ensure that the exiting vapor concentration was 99% of the entering concentration even if
vapor concentration inside the sample system had been reduced to zero concentration prior to sampling.
-------
Table of Contents
Notice ii
Foreword iii
Abstract iv
List of Chemical Abbreviations viii
List of Figures ix
List of Tables xv
Acknowledgements xxi
Executive Summary xxii
1.0 Introduction 1
2.0 Site Description 3
3.0 Methods and Materials Used For Basement and Sub-Slab Air Sampling 6
3.1 Quality Control Measures for Sampling and Analysis Using EPA Method TO-15 6
3.2 Basement and Outdoor Air Sampling for VOCs 10
3.3 Quality Control Measures and Data Quality for Basement Air Sampling and Analysis
for Radon 11
3.4 Sub-Slab Probe Assembly and Installation 12
3.5 Sub-Slab Air Sample Collection for VOCs Using EPA Method TO-15 17
3.6 Quality Control Measures and Data Quality for Sub-Slab Air Sampling Using Tedlar
Bags and On-Site GC Analysis 17
3.7 Quality Control Measures and Data Quality for Sub-Slab Air Sampling for Radon
Using Scintillation Cells 19
4.0 Methods and Materials Used for Air Permeability Testing and Sub-Slab Flow Analysis 21
5.0 Discussion of Sampling Issues Associated with Sub-Slab Sampling 25
5.1 Assessment of Infiltration of Basement Air During Air Extraction 25
5.2 Assessment of Extraction Flow Rate 31
5.3 Evaluation of Equilibration Time 32
5.4 Selection of Purge Volume 34
5.5 Placement of Sub-Slab Vapor Probes 35
6.0 Use of Basement and Sub-Slab Air Measurement to Assess Vapor Intrusion 37
6.1 Method of Vapor Intrusion Evaluation 37
6.2 Summary of Results for Buildings Sampled in July and October 2002 39
6.3 Summary of Results for Buildings Sampled in March 2003 62
6.4 Results of Radon Testing to Assess Vapor Intrusion 96
6.5 Summary of Basement/Sub-Slab Concentration Ratios 98
7.0 Summary 100
References 105
VII
-------
List of Chemical Abbreviations
1,1,1-TCA
1,1-DCE
TCE
C-1.2-DCE
1,1-DCA
1,2-DCA
PCE
CH2CI2
CHCI3
CCI4
CCI3F
CCI2F2
CHBrCI2
CH3CH2CI
CCI3CF3
THF
MEK
MIBK
MTBE
1,2,4-TMB
1,3,5-TMB
CS_
1,1,1-trichloroethane
1,1-dichloroethylene
trichloroethylene
cis-1,2-dichloroethylene
1,1-dichloroethane
1,2-dichloroethane
perchloroethylene
methylene chloride
chloroform
carbon tetrachloride
trichlorofluoromethane (F-11)
dichlorodifluoromethane (F-12)
bromodichloromethane
chloroethane
trichlorotrifluoroethane (F-113)
tetrahydrofuran
methyl ethyl ketone
methyl isobutyl ketone
methyl tert-butyl ether
1,2,4-trimethylbenzene
1,3,5-trimethylbenzene
carbon disulfide
VIII
-------
List of Figures
Figure 1 Direction of ground-water flow (large arrows) and location of the residential area
of investigation near the Raymark Superfund Site
(modified from Tetra Tech NUS, Inc., 2000) 3
Figure 2 Location of geologic cross-sections and the residential area of investigation near the
Raymark Superfund Site (modified from Tetra Tech NUS, Inc., 2000) 4
Figure 3 Geologic cross-section G - G' (modified from Tetra Tech NUS, Inc., 2000) 5
Figure 4 Geologic cross-section H - H' (modified from Tetra Tech NUS, Inc., 2000) 5
Figure 5 Collection of a replicate basement air sample 7
Figure 6 Replicate precision as a function of mean basement concentration for the July 2002
and March 2003 sampling events 9
Figure 7 Coefficient of variation (COV) as a function of mean basement concentration
for July 2002 and March 2003 sampling events 9
Figure 8 Tripod and six-liter evacuated canister used to collect a 24-hour outdoor air
sample during the March 2003 sampling event 10
Figure 9 Coefficient of variation (COV) as a function of mean basement radon
concentration 12
Figure 10 General schematic of a sub-slab vapor probe 13
Figure 11 Brass materials used for sub-slab probe construction in homes near the Raymark
facility 13
Figure 12 Hex bushing used for probe construction when a concrete slab was less than
2.5cm thick 13
Figure 13 A comparison of VOC concentrations in outdoor air and outdoor air passing through
brass fittings used for probe construction during the July 2002 sampling event.
Dashed lines indicate detection limits 14
Figure 14 Stainless-steel materials now used for sub-slab probe assembly 14
ix
-------
List of Figures — continued
Figure 15 Drilling through a concrete slab using a rotary hammer drill 15
Figure 16 "Inner" and "outer" holes drilled in a concrete slab 15
Figure 17 Typical schematic illustrating location of sub-slab vapor probes 16
Figure 18 Sample train for sub-slab air collection using EPA Method TO-15 17
Figure 19 Sample train for sub-slab air collection using one-liter Tedlar bags 18
Figure 20 Comparison of EPA Method TO-15 and Tedlar bag sampling with on-site GC
analysis for 1,1,1-TCA, 1,1-DCE, TCE, and c-1,2-DCE, n = 91, r2 = 0.88 19
Figure 21 Sample train for sub-slab air collection for radon using scintillation cells 20
Figure 22 Regenerative blower used for air permeability testing 21
Figure 23 Variable-area flowmeter used for air permeability testing 22
Figure 24 Magnehelic gauges and suction hole used for vacuum measurement 22
Figure 25 Best-fit model results for permeability test conducted at House C with four vacuum
observation points and a flow rate of 255 SLPM
(kr/kz constrained between 1 - 2) 24
Figure 26a Sub-slab vapor concentration as a function of extraction volume at Probe A
in House L using Tedlar bag sampling and on-site GC analysis 26
Figure 26b Sub-slab vapor concentration as a function of extraction volume at Probe B
in House M using Tedlar bag sampling and on-site GC analysis 26
Figure 26c Sub-slab vapor concentration as a function of extraction volume at Probe A
in House N using Tedlar bag sampling and on-site GC analysis. Dashed lines
denote detection limit 27
Figure 27a Sub-slab vapor concentration as a function of extraction volume at Probe A
in House J using EPA Method TO-15 27
Figure 27b Sub-slab vapor concentration as a function of extraction volume at Probe A
in House M using EPA Method TO-15 28
Figure 28 Simulated streamline (solid lines) and travel time (s) (dashed lines) contours
in sub-slab media when k. = 7.4E-07 cm2, kr/kz = 1.5, i = 3.2E-09 cm,
flow rate = 1 SLPM, and depth to ground water = 500 cm 30
-------
List of Figures — continued
Figure 29 Simulated vacuum (Pa) (dashed lines) and pore-air velocity (solid lines) (cm/s)
in sub-slab media when kr = 7.4E-07 cm2, kr/kz = 1.5, 9a = 0.35, flow rate = 1 SLPM,
and depth to ground water = 500 cm 31
Figure 30 Time to reach C(t)/C5= 0.99 as a function of diffusion path length '5' and 9W for TCE
when C0 = 0, r\ = 0.4, pb = 1.68 g/cm3, Da = 7.4E-02 cm2/s, Dw = 9.3E-06 cm2/s,
and H = 0.38 (no sorption) 34
Figure 31 Purge volume as a function of C0/Cin and Cout/Cn 35
Figure 32 Total vapor concentration measured in one-liter Tedlar bags as a function of probe
location and house. Dark bars refer to centrally located probes. No VOCs
associated with subsurface contamination were detected at Location F. Locations
H, K, M, and P did not have a centrally located probe 36
Figure 33 Coefficient of variation (COV) as a function of mean sub-slab concentration (ppbv)
and method of analysis for VOCs associated with subsurface contamination 36
Figure 34 Basement/sub-slab concentration ratios using EPA Method TO-15 at House A
during the July 2002 sample event. Error bars represent one standard deviation.
Arrows indicate greater than or less than values due to non-detection in basement
or sub-slab air 42
Figure 35 Comparison of mean sub-slab air concentrations of VOCs collected in one-liter
Tedlar bags during the July and October 2002 sample events at House A. Error
bars represent one standard deviation 44
Figure 36 Basement/sub-slab air concentration ratios using EPA Method TO-15 at House B
during the July 2002 sample event - error bars represent one standard deviation.
Arrows indicate greater than or less than values due to non-detection in basement
or sub-slab air 47
Figure 37 Comparison of mean sub-slab air concentrations of VOCs collected in one-liter
Tedlar bags during the July and October 2002 sample events at House B. Error
bars represent one standard deviation 48
Figure 38 Basement/sub-slab concentration ratios using EPA Method TO-15 at House C
during the July 2002 sample event. Error bars represent one standard deviation.
Arrows indicate greater than or less than values due to non-detection in basement
or sub-slab air 51
Figure 39 Comparison of mean sub-slab air concentrations of VOCs collected in one-liter
Tedlar bags during the July and October 2002 sample events at House C. Error
bars represent one standard deviation 53
XI
-------
List of Figures — continued
Figure 40 Basement/sub-slab concentration ratios using EPA Method TO-15 at House D
during the July 2002 sample event. Error bars represent one standard deviation.
Arrows indicate greater than or less than values due to non-detection in basement
or sub-slab air 56
Figure 41 Comparison of mean sub-slab air concentrations of VOCs collected in one-liter
Tedlar bags during the July and October 2002 sample events at House D. Error
bars represent one standard deviation 57
Figure 42 Basement/sub-slab concentration ratios using EPA Method TO-15 at House E
during the July 2002 sample event. Error bars represent one standard deviation.
Arrows indicate greater than or less than values due to non-detection in basement
or sub-slab air 60
Figure 43 Comparison of mean sub-slab air concentrations of VOCs collected in one-liter
Tedlar bags during the July and October 2002 sample events at House E. Error
bars represent one standard deviation 61
Figure 44 Basement/sub-slab concentration ratios using EPA Method TO-15 at House G
during the March 2003 sample event. Error bars represent one standard deviation.
Arrows indicate greater than or less than values due to non-detection in basement
or sub-slab air 67
Figure 45 Basement/sub-slab concentration ratios using EPA Method TO-15 at House H
during the March 2003 sample event. Error bars represent one standard deviation.
Arrows indicate greater than or less than values due to non-detection in basement
or sub-slab air 71
Figure 46 Basement/sub-slab concentration ratios using one-liter Tedlar bags and on-site
GC analysis at House I during the March 2003 sample event. Error bars represent
one standard deviation 73
Figure 47 Basement/sub-slab concentration ratios using one-liter Tedlar bags and on-site
GC analysis at House J during the March 2003 sample event. Error bars represent
one standard deviation 76
Figure 48 Basement/sub-slab concentration ratios using one-liter Tedlar bags and on-site
GC analysis at House K during the March 2003 sample event. Error bars represent
one standard deviation 79
Figure 49 Basement/sub-slab concentration ratios using one-liter Tedlar bags and on-site
GC analysis at House L during the March 2003 sample event. Error bars represent
one standard deviation 82
XII
-------
List of Figures — continued
Figure 50 Basement/sub-slab concentration ratios using one-liter Tedlar bags and on-site
GC analysis at House M during the March 2003 sample event. Error bars represent
one standard deviation. Arrows indicate less than values due to non-detection
in basement air 85
Figure 51 Basement/sub-slab concentration ratios using EPA Method TO-15 at House N
during the March 2003 sample event. Error bars represent one standard deviation.
Arrows indicate greater than or less than values due to non-detection in basement
or sub-slab air 89
Figure 52 Basement/sub-slab concentration ratios using EPA Method TO-15 at House O
during the March 2003 sample event. Error bars represent one standard deviation.
Arrows indicate greater than or less than values due to non-detection in basement
or sub-slab air 92
Figure 53 Basement/sub-slab concentration ratios using EPA Method TO-15 at House P
during the March 2003 sample event. Error bars represent one standard deviation.
Arrows indicate greater than or less than values due to non-detection in basement
or sub-slab air 96
Figure 54 Comparison of basement/sub-slab air concentration ratios for radon and indicator
VOCs associated with vapor intrusion. Samples for VOCs collected in one-liter
Tedlar bags with on-site GC analysis 97
Figure 55 Coefficient of variation (COV) as a function of location and compound for VOCs
detected in basement air as a result of vapor intrusion 98
Figure 56 Summary of average basement/sub-slab concentration ratios for VOCs
present in basement air due to vapor intrusion using one-liter Tedlar bags and
on-site GC analysis. Arrows indicate less than values due to non-detection
in basement air 99
XIII
-------
List of Tables
Table 1 Computation of maximum percent infiltration of basement air into an evacuated
canister during sampling as a function of extraction volume, location, and probe.
P[A], P[B], P[C], P[D], and P[E] denote probes evaluated at individual locations 29
Table 2 Outdoor air concentrations of VOCs during the July 2002 and March 2003 sample
events 39
Table 3a Basement and sub-slab concentrations of VOCs at House A using EPA Method
TO-15 during the July 2002 sample event 41
Table 3b Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House A using EPA Method TO-15 during the
July 2002 sample event 43
Table 3c Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House A using 1-liter Tedlar bags and on-site
GC analysis during the July 2002 sample event 43
Table 3d Sub-slab air concentrations of VOCs associated with sub-surface contamination
in House A using 1-liter Tedlar bags and on-site GC analysis during the
October 2002 sample event 43
Table 4a Basement and sub-slab concentrations of VOCs at House B using EPA Method
TO-15 during the July 2002 sample event 45
Table 4b Summary of basement/sub-slab air concentration ratios of VOCs associated
with sub-surface contamination in House B using EPA Method TO-15 during the
July 2002 sample event 46
Table 4c Summary of basement/sub-slab air concentration ratios of VOCs associated
with sub-surface contamination in House B using 1-liter Tedlar bags and on-site
GC analysis during the July 2002 sample event 46
Table 4d Sub-slab air concentrations of VOCs associated with sub-surface contamination
in House B using 1-liter Tedlar bags and on-site GC analysis during the
October 2002 sample event 48
XV
-------
List of Tables — continued
Table 5a Basement and sub-slab air concentrations of VOCs at House C using EPA Method
TO-15 during the July 2002 sample event 50
Table 5b Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House C using EPA Method TO-15 during the
July 2002 sample event 51
Table 5c Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House C using 1-liter Tedlar bags and on-site
GC analysis during the July 2002 sample event 51
Table 5d Sub-slab air concentrations of VOCs associated with sub-surface contamination
in House C using 1-liter Tedlar bags and on-site GC analysis during the October
2002 sample event 52
Table 6a Basement and sub-slab concentrations of VOCs at House D using EPA Method
TO-15 during the July 2002 sample event 54
Table 6b Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House D using EPA Method TO-15 during the
July 2002 sample event 55
Table 6c Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House D using 1-liter Tedlar bags and on-site
GC analysis during the July 2002 sample event 55
Table 6d Sub-slab air concentrations of VOCs associated with sub-surface contamination
in House D using 1-liter Tedlar bags and on-site GC analysis during the October
2002 sample event 56
Table 7a Basement/sub-slab air concentrations of VOCs at House E using EPA Method
TO-15 during the July 2002 sample event 58
Table 7b Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House E using EPA Method TO-15 during the
July 2002 sample event 59
Table 7c Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House E using 1 -liter Tedlar bags and on-site
GC analysis during the July 2002 sample event 59
XVI
-------
List of Tables — continued
Table 7d Sub-slab air concentrations of VOCs associated with sub-surface contamination
in House E using 1 -liter Tedlar bags and on-site GC analysis during the
October 2002 sample event 61
Table 8 Basement and sub-slab air concentrations for VOCs at House F using EPA Method
TO-15 during the March 2003 sample event 63
Table 9a Basement and sub-slab air concentration of VOCs at House G using EPA Method
TO-15 during the March 2003 sample event 65
Table 9b Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House G using EPA Method TO-15 during the March
2003 sample event 66
Table 9c Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House G using 1 -liter Tedlar bags and on-site
GC analysis during the March 2003 sample event 66
Table 9d Basement/sub-slab air concentration ratios for radon in House G using 48-hr
activated carbon canisters for basement air sampling (3/25-3/27/03) and scintillation
cells for sub-slab air sampling during the March 2003 sample event 67
Table 10a Basement and sub-slab air concentrations for VOCs at House H using EPA Method
TO-15 during the March 2003 sample event 69
Table 10b Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House H using EPA Method TO-15 during the March
2003 sample event 70
Table 10c Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House H using 1-liter Tedlar bags and on-site
GC analysis during the March 2003 sample event 70
Table 10d Basement/sub-slab air concentration ratios for radon in House H using 48-hr
activated carbon canisters for basement air sampling (3/21-3/24/03) and scintillation
cells for sub-slab air sampling during the March 2003 sample event 70
Table 11a Basement and sub-slab air concentrations for VOCs at House I using EPA Method
TO-15 during the March 2003 sample event 72
Table 11 b Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House H using 1 -liter Tedlar bags and on-site
GC analysis during the March 2003 sample event 73
XVII
-------
List of Tables — continued
Table 12a Basement and sub-slab air concentrations for VOCs at House J using EPA Method
TO-15 during the March 2003 sample event 74
Table 12b Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House J using 1-liter Tedlar bags and on-site GC
analysis during the March 2003 sample event 75
Table 12c Basement/sub-slab air concentration ratios for radon in House J using 48-hr
activated carbon canisters for basement air sampling (3/21-3/24/03) and scintillation
cells for sub-slab sampling during the March 2003 sample event 75
Table 13a Basement air concentrations for VOCs at House K using EPA Method TO-15
during the March 2003 sample event 77
Table 13b Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House K using 1-liter Tedlar bags and on-site
GC analysis during the March 2003 sample event 78
Table 13c Basement and sub-slab air concentration ratios for radon in House K using 48-hr
activated carbon canisters for basement air sampling (3/21 -3/24/03) and scintillation
cells for sub-slab air sampling during the March 2003 sample event 78
Table 14a Basement and sub-slab air concentrations for VOCs at House L using EPA Method
TO-15 during the March 2003 sample event 80
Table 14b Results of sequential sub-slab air sampling using 1 -liter Tedlar bags and on-site
GC analysis at House L during the March 2003 sample event 81
Table 14c Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House L using 1-liter Tedlar bags and on-site GC
analysis during the March 2003 sample event 81
Table 14d Summary of 48-hour indoor air measurements for radon using activated charcoal
at House L 81
Table 14e Basement/sub-slab air concentration ratios for radon in House L using 48-hr
activated carbon canisters for basement air sampling (3/26-3/28/03) and scintillation
cells for sub-slab air sampling during the March 2003 sample event 81
Table 15a Basement and sub-slab air concentrations for VOCs at House M using EPA Method
TO-15 during the March 2003 sample event 83
XVIII
-------
List of Tables — continued
Table 15b Results of sequential sub-slab air sampling using 1-liter Tedlar bags and on-site
GC analysis at House M during the March 2003 sample event 84
Table 15c Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House M using 1-liter Tedlar bags and on-site GC
analysis during the March 2003 sample event 84
Table 15d Basement/sub-slab air concentration ratios for radon in House M using 48-hr
activated carbon canisters for basement air sampling (3/22-3/24/03) and scintillation
cells for sub-slab air sampling during the March 2003 sample event 84
Table 16a Basement and sub-slab air concentrations for VOCs at House N using EPA Method
TO-15 during the March 2003 sample event 86
Table 16b Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House N using EPA Method TO-15 during the March
2003 sample event 87
Table 16c Results of sequential sub-slab air sampling and on-site GC analysis using 1-liter
Tedlar bags at House N during the March 2003 sample event 88
Table 16d Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House N using 1 -liter Tedlar bags and on-site
GC analysis during the March 2003 sample event 88
Table 16e Basement/sub-slab air concentration ratios for radon in House N using 48-hr
activated carbon canisters for basement air sampling (3/25-3/27/03) and scintillation
cells for sub-slab air sampling during the March 2003 sample event 88
Table 17a Basement and sub-slab concentrations for VOCs at House O using EPA Method
TO-15 during the March 2003 sample event 90
Table 17b Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House O using EPA Method TO-15 during the March
2003 sample event 91
Table 17c Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House O using 1 -liter Tedlar bags and on-site
GC analysis during the March 2003 sample event 91
Table 17d Basement/sub-slab air concentration ratios for radon in House O using 48-hr
activated carbon canisters for basement air sampling (3/25-3/27/03) and scintillation
cells for sub-slab air sampling during the March 2003 sample event 92
XIX
-------
List of Tables — continued
Table 18a Basement and sub-slab air concentrations for VOCs at House P using EPA Method
TO-15 during the March 2003 sample event 94
Table 18b Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House P using EPA Method TO-15 during the March
2003 sample event 95
Table 18c Summary of basement/sub-slab air concentration ratios of VOCs associated with
sub-surface contamination in House P using 1-liter Tedlar bags and on-site
GC analysis during the March 2003 sample event 95
Table 18d Basement/sub-slab air concentration ratios for radon in House P using 48-hr
activated carbon canisters for basement air sampling (3/26-3/28/03) and scintillation
cells for sub-slab air sampling during the March 2003 sample event 95
xx
-------
Acknowledgements
The authors would like to thank the following for their help and support in this project: Mike Jasinski,
Ron Jennings, Matt Hoagland, Mary Sanderson, and Don Berger of EPA Region I, David Burden of
NRMRL, Ada, OK, and William Bell of the Massachusetts Department of Public Health.
The authors would also like to acknowledge the following for their formal review of this manuscript:
Dr. John E. McCray
Colorado School of Mines
Environmental Science and Engineering Division
1500 Illinois Street
Golden, CO 80401
Dr. Blayne Hartman
HP Labs
432 N. Cedros Avenue
Solana Beach, CA 92075
Dr. Brian Schumacher
U.S. EPA
Office of Research and Development
National Exposure Research Laboratory
Environmental Sciences Division
RO. Box 93478
Las Vegas, NV 89193-3478
Dr. Helen Dawson
U.S. EPA, Region VIM
999 18th Street, Suite 300
Denver, CO 80401
XXI
-------
Executive Summary
This report describes the results of an investigation conducted to assist EPA's New England Regional Office
in evaluating vapor intrusion at 15 homes and one commercial building near the Raymark Superfund Site
in Stratford, Connecticut. Ground water beneath these homes is contaminated with 1,1,1-trichloroethane,
1,1-dichloroethylene, trichloroethylene, cis-1,1-dichloroethylene, and 1,1-dichloroethane. Methods were
developed to sample sub-slab air and use basement and sub-slab air measurements to evaluate vapor
intrusion on a building-by-building basis. A volatile organic compound (VOC) detected in basement air
was considered due primarily to vapor intrusion if: (1) the VOC was detected in ground water or soil gas
in the vicinity (e.g., 30 meters) of a building, and (2) the null hypothesis that the basement/sub-slab air
concentration ratio of the VOC was equal to the basement/sub-slab air concentration ratio of an indicator
VOC could not be rejected using a one-tailed Approximate t-Test at a level of significance less than
or equal to 0.05. An indicator VOC was defined as a VOC detected in sub-slab air and known to be
associated only with sub-surface contamination (i.e., no outdoor or indoor air sources). The VOCs 1,1-
dichloroethylene and 1,1-dichloroethane were considered indicator VOCs in this investigation because
they are degradation products of 1,1,1-trichloroethane and not commonly associated with commercial
products. The VOC cis-1,2-dichloroethylene was considered an indicator VOC because it is a degradation
product of trichloroethylene and also not commonly associated with commercial products. Using this
method of evaluation, VOCs detected in basement air due to vapor intrusion could easily be separated
from numerous other halogenated and non-halogenated (e.g., petroleum hydrocarbons) VOCs present
in basement air. The variance associated with each basement/sub-slab air concentration ratio was
calculated using the method of propagation of errors which incorporated the variance associated with
both basement and sub-slab air measurement. An average basement/sub-slab air concentration ratio
was computed using concentration ratios of all VOCs detected in basement air and associated with vapor
intrusion. The method of propagation of errors was then used to calculate the variance associated with
the average basement/sub-slab concentration ratio.
As a matter of necessity, radon was used as an indicator compound at locations where an indicator
VOC was not detected in basement air. However, when basement/sub-slab air concentration ratios
were compared for radon and indicator VOCs, statistical non-equivalency occurred at three out of the
four locations evaluated. At these three locations, the null hypothesis that the basement/sub-slab air
concentration ratio of radon was equal to the basement/sub-slab air concentration ratio of the indicator
VOC, 1,1-DCE, was rejected using a two-tailed Approximate t-Test at a significance level less than or
xxii
-------
equal to 0.1. There was a visual dissimilarity between the basement/sub-slab air concentration ratio of
radon and VOCs associated with vapor intrusion. This was in contrast to visual and statistical (levels of
significance always greater than 0.1) similarity of basement/sub-slab air concentration ratios of indicator
VOCs and other VOCs associated with vapor intrusion. These two observations indicate, at least in
this investigation, use of indicator VOCs was preferable to radon in assessing vapor intrusion. Further
research is needed at other sites containing indicator VOCs to determine the usefulness of radon in
assessing vapor intrusion.
Holes for sub-slab probes were drilled in concrete slabs using a rotary hammer drill. Probes were
designed to allow for collection of air samples directly beneath a slab and in sub-slab media. Three to
five probes were installed in each basement. Fifty-five probes were installed in 16 buildings which, on
average, resulted in placement of one probe every 220 ft2. Observation of high coefficients of variation in
sub-slab air concentrations (greater than 100% at several locations), and the need for statistical analysis
in assessing basement/sub-slab air concentration ratios, indicated that placement of multiple probes in
sub-slab media was necessary to evaluate vapor intrusion. Generally, one sub-slab vapor probe was
centrally located while two or more probes were placed within one or two meters of basement walls in
each building. In this investigation, placement of a probe in a central location did not ensure detection of
the highest VOC concentrations in sub-slab media. Schematics illustrating the location of sub-slab probes
and other slab penetrations (e.g., suction holes for sub-slab permeability testing) were prepared for each
building to document sample locations, interpret sample results, and design corrective measures.
Basement and sub-slab air samples were collected and analyzed for VOCs using six-liter SilcoCan
canisters and EPA-Method TO-15. Sub-slab air samples were also collected in one-liter Tedlar bags
using a peristaltic pump and analyzed on-site for target VOCs by EPA's New England Regional Laboratory
within 24 hours of sample collection. Open-faced charcoal canisters were used to sample radon gas in
basement air over a 48-hour period. Scintillation cells and a peristaltic pump were used to sample radon
gas in sub-slab air. Scintillation cells were analyzed within four hours using a portable radiation monitor
to count and amplify light pulses.
Three methods were used to evaluate infiltration of basement air into sub-slab media during air extraction
(purging + sampling). The first method consisted of sequentially collecting five one-liter Tedlar bag samples
at a flow rate of 1 standard liter per minute and comparing vapor concentration of four VOCs associated
with vapor intrusion as a function of extraction volume. This was performed at three locations with little
effect on sample concentration. This testing also indicated the absence of rate-limited mass exchange
during air extraction. Replicate canister samples representing extraction volumes of 5 to 9 and 10 to 14
liters were compared at two locations with similar results. A second method was then employed which
utilized a mass balance equation and sub-slab and basement air concentrations. When sensitivity of the
method permitted, infiltration was shown to be less than 1 % at sampled locations. A third method involved
XXIII
-------
simulating streamlines and travel time in sub-slab media during air extraction. Air permeability testing
in sub-slab media was conducted to obtain estimates of radial and vertical air permeability to support air
flow simulations. Simulations indicated that less than 10% of air extracted during purging and sampling
could have originated as basement air when extracting up to 12 liters of air. Overall, extraction volumes
used in this investigation (up to 14 liters) had little or no effect on sample results.
To assess the time required after probe installation for sampling (equilibration period), advective air flow
modeling with particle tracking was employed to establish radial path lengths for diffusion modeling.
Simulations indicated that in sub-slab material beneath homes at the Raymark site (sand and gravel),
equilibration likely occurred in less than 2 hours. Sub-slab probes in this investigation were allowed to
equilibrate for 1 to 3 days prior to sampling. A mass-balance equation was used to estimate the purging
requirement prior to sampling. Simulations indicated that collection of 5 purge volumes would ensure
that the exiting vapor concentration was 99% of the entering concentration even if vapor concentration
inside the sample system had been reduced to zero concentration prior to sampling. A purge volume
for the sample train used in homes near the Raymark site was typically less than 10 cm3.
In summary, this report constitutes an important first step in the development of a technical resource
document on sub-slab air sampling and use of indoor and sub-slab air samples to assess vapor
intrusion.
XXIV
-------
1.0 Introduction
In this report, vapor intrusion was defined as the entry
of volatile organic or inorganic compounds (elemental
mercury) and gases (e.g., methane) into a building
due to contaminated subsurface media (ground water,
soil, rock), non-aqueous phase liquids (NAPLs), or
waste material (e.g., landfills). Vapor intrusion was
not defined by the violation of health-based criteria
associated with these compounds or gases which
vary over time.
One commonly accepted conceptual model of vapor
intrusion involves diffusive transport of volatile organic
compounds (VOCs) or gases from contaminated
ground water and/or soil to sub-slab media. This
is followed by advective-diffusive transport in sub-
slab media through an overlying concrete slab and
basement walls until entry into a building is achieved.
Another conceptual model involves VOC and gas
transport into a building via preferential pathways (e.g.,
utility conduits). Entry of methane gas from landfills
and gasoline vapors from gas stations into buildings
via preferential pathways is well documented. Sub-
slab sampling taken alone is likely not an appropriate
method for assessing vapor intrusion for this latter
conceptual model.
Until recently, potential risk posed by vapor intrusion
was not consistently considered during sub-surface
investigations at sites regulated by the Comprehensive
Environmental Response and Liability Act (CERCLA)
and the Resource Conservation and Recovery Act
(RCRA). Also, risk posed by vapor intrusion has not
been consistently considered during investigations
involving leaking underground storage tank sites or
sites where residential or commercial construction
is proposed over known areas of soil and/or ground-
water contamination (i.e., Brownfields sites). Thus,
the number of buildings where vapor intrusion
has occurred, or is occurring, is unknown, and
the magnitude of the problem remains undefined.
Recognition of this exposure pathway necessitates its
consideration in regulatory decision making and may
require review of past regulatory decisions involving
VOC contamination in soil and/or ground water.
To assess this increasingly important regulatory issue,
the United States Environmental Protection Agency's
(EPA) Office of Solid Waste and Emergency Response
developed guidance (USEPA, 2002) to facilitate
assessment of vapor intrusion at sites regulated
by RCRA and CERCLA - sites where halogenated
organic compounds constitute the bulk of risk to
human health. EPA does not consider the guidance
applicable to underground storage tank sites where
petroleum compounds primarily determine risk, and
biodegradation in subsurface media may be a dominant
-------
fate process. In the guidance, EPA considers a site
as a regulated unit potentially consisting of numerous
buildings and subsurface monitoring points. The
guidance was not developed to conduct building-to-
building investigations. Recommendations provided
in this report directly support the guidance but also
support subsequent building-to-building investigations.
A number of state agencies have or are in the process
of developing state-specific guidance or advisories to
assess vapor intrusion.
In the guidance, EPA recommends concomitant use
of sub-slab air sampling with indoor air sampling to
differentiate outdoor and indoor sources of VOCs
(e.g., cosmetics, air fresheners, gasoline storage
or car parked in garage, cigarette smoke, solvents,
paints, furniture polish) from VOCs emanating from
contaminated soil or ground water. The agency,
however, does not provide detailed recommendations
on how to collect sub-slab air samples nor how to
use these samples to assess vapor intrusion. Also,
little is published in peer-reviewed literature on
sub-slab sample collection and interpretation. Sub-
slab sampling offers an opportunity to collect air
samples directly beneath the living space of a building
and thereby eliminates uncertainty associated with
interpolation or extrapolation of soil-gas and/or ground-
water concentrations from monitoring points away
from a building. Sub-slab sampling also provides
an opportunity to evaluate the validity of claims that
petroleum hydrocarbons of concern degrade prior to
vapor entry into sub-slab material.
The purpose of this report is to provide a method for
evaluating vapor intrusion using indoor and sub-slab air
samples. This report does not constitute guidance. It
is, however, an important first step in the development
of a technical resource document on sub-slab air
sampling and interpretation. Use of recommendations
provided in this report should increase the potential
of collecting samples representative of "true" sub-
slab air concentration even if the method of data
interpretation presented here is not utilized to assess
vapor intrusion.
-------
2.0 Site Description
The Raymark Superfund Site consists of 33.4 acres
of land previously occupied by the Raybestos-
Manhattan Company in Stratford, Connecticut,
where the company disposed of solid-waste from
settling lagoons during its operation. Between 1919
and 1989, the company produced asbestos and
asbestos compounds, metals, phenol-formaldehyde
resins, adhesives, gasket material, sheet packing,
clutch facings, transmission plates, and brake linings.
Between 1993 and 1996, EPA removed fill containing
asbestos, lead, and RGBs from a number of residential
properties and a middle school. EPA placed the fill
back on the facility property and isolated the waste
beneath a cap. In 1996 and 1997, EPA demolished
the facility buildings and placed a cap over the area
previously occupied by the buildings. The property
is now occupied by commercial buildings (e.g., Wal-
Mart, Home Depot).
As illustrated in Figure 1, ground water beneath
the residential area of this investigation generally
flows southeast from the former facility, underneath
a large residential community, and discharges into
the Housatonic River which eventually discharges
into Long Island Sound. Ground water in the vicinity
of the former facility is contaminated with a number
of VOCs including 1,1,1-trichloroethane (1,1,1-
TCA), trichloroethene (TCE), cis-1,2-dichloroethene
(c-1,2-DCE), 1,1 -dichloroethene (1,1 -DCE), and 1,1 •
dichloroethane (1,1-DCA).
Figure 1. Direction of ground-water flow (large arrows)
and location of the residential area of investigation near the
Raymark Superfund Site (modified from Tetra Tech NUS, Inc.,
2000).
-------
Figures 2, 3, and 4 illustrate glacio-fluvial deposits
and fractured granite bedrock valleys in the vicinity
of the residential area of investigation. The remedial
investigation (TetraTech NUS, 2000) and subsequent
studies financed by EPA indicate that ground-water flow
is heavily influenced by the location and orientation
of bedrock valleys. This results in a fairly complex
contaminant distribution profile making interpolation
and extrapolation of ground-water contaminant profiles
difficult. For instance, TCE was not detected during
ground-water sampling at well MW-215, illustrated
in Figure 4, but was detected in basement air at a
home less than 10 meters from this well. As will be
discussed, the basement/sub-slab air concentration
ratio of TCE observed at this location suggests that
the cause of TCE in basement air was vapor intrusion.
Non-detection of TCE in well MW-215 was likely
caused by a bedrock knoll close to this house where
ground-water flow may have been diverted. This
assertion is corroborated by significant drawdown
during sampling compared to other shallow wells.
Sub-slab and basement air samples were collected
in 15 homes and one business near the Raymark
Superfund Site in Stratford, Connecticut. The
investigation consisted of three separate sample
events. In July 2002, basement and sub-slab air
was sampled for VOCs at five homes using six-liter
SilcoCan canisters and EPA Method TO-15 (USEPA,
1999) analysis. Sub-slab air was also sampled using
one-liter Tedlar bags and analyzed on-site using gas
chromatography (GC). In October 2002, sub-slab air
at these five homes was re-sampled and re-analyzed
using one-liter Tedlar bags and on-site GC analysis to
assess temporal variability. In March 2003, basement
and sub-slab airwas sampled for VOCs at an additional
ten homes and one commercial building using six-liter
SilcoCan canisters and EPA Method TO-15 analysis.
Sub-slab air was also sampled using one-liter Tedlar
bags and on-site GC analysis. During this sample
event, basement and sub-slab air was sampled for
radon gas.
Figure 2. Location of geologic cross-sections and the
residential area of investigation near the Raymark Superfund
Site (modified from Tetra Tech NUS, Inc., 2000).
-------
(O
i
*>
o
CD
O.
O
IQ
o'
§
CO
CO
O
I
I
I
5;
0)'
Q.
u
o
to
o
O
2
ELEVATION (NGVD 1929)
SHORELINE HOUSATONIC RIVER
(Q
i
w
O
CD
O^
O
(Q
o'
O
en
I
o'
O
I
O
5;
CD'
Q.
CD
3
CD
O
C
C/)
D
O
ro
o
ELEVATION (NGVD 1929)
^ ro N>
SHORELINE HOUSATONIC RIVER
^ !3 ° °
ELEVATION (NGVD 1929)
ELEVATION (NGVD 1929)
-------
3.0 Methods and Materials Used for Basement and Sub-Slab Air Sampling
3.1 Quality Control Measures and Data
Quality for Sampling and Analysis
Using EPA Method TO-15
EPA's Compendium of Methods for Determination of
Toxic Organic Compounds in Ambient Air (TO Methods)
was developed for measurement of 97 VOCs listed in
Title III of the Clean Air Act Amendments of 1990. EPA's
TO Methods stipulate specific sampling and analytical
requirements for determination of VOCs in air. A
number of TO Methods are appropriate for indoor and
sub-slab air sampling. However, EPA Method TO-15
- Determination of Volatile Organic Compounds (VOCs)
in Air Collected in Specially-Prepared Canisters and
Analyzed by Gas Chromatography/Mass Spectrometry
(GC/MS) (USEPA, 1999) - was used to sample and
analyze basement and sub-slab air samples during
this investigation.
In EPA Method TO-15, two MS options are available.
In the MS-SCAN mode, a GC is coupled to a MS
programmed to scan all ions repeatedly over a specified
mass range. In the MS-SIM (selected ion monitoring)
mode, a GC is coupled to a MS programmed to scan
selected ions repeatedly. The MS-SCAN mode allows
wide identification of VOCs and detection in tenths of
a part per billion volume (ppbv) or hundreds of parts
per trillion volume while the MS-SIM mode allows
identification of a few select compounds and detection
in the tenths of parts pertrillion volume. The MS-SCAN
mode was used during this investigation because risk
levels for VOCs of concern were generally in the low
ppbv range, and VOCs associated with sub-surface
contamination were suspected to be present in sub-
slab material at concentrations ranging from tens to
hundreds ppbv.
EPA Method TO-15 requires that canisters be
meticulously cleaned prior to sampling. Six-liter
SilcoCan canisters were provided and analyzed by
EPA's New England Regional Laboratory. Canisters
were cleaned in accordance with EPA's New England
Regional Laboratory standard operating procedure
(USEPA, 1998). Canister cleaning involved three
evacuation/pressurization cycles. Each cycle
consisted of evacuation to 0.1 Pascal (Pa), heating to
150°C, and pressurization to 206.7 kPa with humidified
nitrogen. Canisters were then evacuated again to 0.1
Pa and vacuum tested with a Pirani sensor for a 24
hour period. Every canister used for basement and
sub-slab air sampling was then re-pressurized with
humidified, ultra-high purity nitrogen and analyzed
for VOCs using the same GC/MS utilized for sample
analysis. This process is known as certification. If all
canisters are subjected to this process prior to use,
as was the case in this investigation, there is 100%
-------
certification. Canisters were considered "clean" if
concentrations of target VOCs were less than 0.02
ppbv. Canisters were stored under pressure until the
day before sampling and then evacuated once more
to 0.1 Pa for sub-atmospheric pressure sampling
in accordance with EPA's New England Regional
Laboratory standard operating procedure (USEPA,
1996). A certification level of 100% is generally
desirable for indoor air sampling efforts because
risk-based concentrations are in the low ppbv range
for many compounds. However, a lower level of
certification may be suitable for sub-slab sampling
because vapor concentrations associated with sub-
surface contamination are typically one to three orders
of magnitude higher than indoor air.
EPA Method TO-15, EPA requires the use of duplicate,
replicate, and audit samples for quality control.
Performance is measured by relative percent difference
(RPD) defined by:
RPD = 100
_
X
where X, and X2 are values for sample 1 and 2,
respectively, and X is a sample mean. EPA Method
TO-15 requires duplicate and replicate precision less
than or equal to 25%. In Method TO-15, EPA defines
duplicate precision as a comparison between two
samples taken from the same canister. Duplicate
sampling is performed at an analytical laboratory
and is used to assess analytical precision. Duplicate
sampling was performed at EPA's New England
Regional Laboratory at a sampling frequency of 10%.
Relative percent differences did not exceed 30% for
any compound in any analysis. In Method TO-15, EPA
defines replicate precision as a comparison between
two canisters filled from the same air mass over the
same period of time. Replicate sampling is performed in
the field and can be used to assess precision associated
with the entire sample and analytical process. As
illustrated in Figure 5, replicate sampling consisted of
placing two canisters side-by-side. Replicate samples
were collected from basement air only. One replicate
sample was collected during the July 2002 sample
event which included 5 basement air samples and
one outdoor air sample. Two replicate samples were
collected during the March 2003 sampling event which
included 11 basement and two outdoor air samples.
Thus, replicate sampling frequency was 3 out of 19
samples, or about 16%.
Figure 5. Collection of replicate basement air sample.
-------
As illustrated in Figure 6, RPDs for replicate
sampling were generally less than, or near to, 25%.
The highest RPDs were for methyl ethyl ketone and
trichlorotrif luoroethane analyzed during the July 2002
sample event. An alternative method of assessing
replicate precision is to express mean concentration
as a function of a coefficient of variability (COV) which
is simply the standard deviation divided by the mean
times 100. As illustrated in Figure 7, the mean COV
for the three homes used for replicate sampling during
the July 2002 and March 2003 sampling events was
only 5.5%. As will be discussed, the COV for replicate
basement air sampling was useful for statistical
analysis in assessing vapor intrusion.
In EPA Method TO-15, EPA defines audit accuracy
as the difference between analyses provided in an
audit canister and the nominal value as determined
by an audit authority. Audit canisters containing target
VOCs were analyzed to assess analytical accuracy.
Relative percent differences did not exceed 30%
for any compound. As an additional quality control
check, ultra-high purity humidified nitrogen was
introduced into the analytical instrument inlet line prior
to analyzing canisters to serve as laboratory blanks
and to demonstrate lack of background contamination
in analytical instrumentation. A laboratory blank was
analyzed every six canister samples. During the
July 2002 sampling event, acetone was detected in
17 of 21 laboratory blanks at concentrations slightly
below reporting limits. Acetone concentrations
were not high enough though to exceed acceptance
criteria (observed concentration in samples less
than 5 times the concentration in laboratory blanks).
During the March 2003 sampling event, 12 laboratory
blanks were analyzed. Acetone was detected in one
laboratory blank at 0.11 ppbv. Prior to analyzing each
canister, surrogate compounds 1,2-dichloroethaned4,
p-bromofluorobenzene, and toluene d8 were
introduced into the analytical instrument inlet line
to assess the accuracy of the analytical system.
Acceptable recovery (88%-116%) was attained for
all surrogate compounds in all samples.
Sub-slab gas sample concentrations were typically
muchhigherthanbasementairsamples. Fiftymilliliters
(ml) of air were withdrawn from each canister for
preliminary analysis to determine if concentrations
were within the calibration range. If so, 500 ml were
withdrawn from canisters for a second round of
analysis. Otherwise, a smaller volume of sample was
withdrawn and diluted to ensure analysis within the
calibration range. Analytical results were reported in
units of pg/m3 and ppbv. A conversion from ppbv to
pg/m3 can be obtained by use of the Ideal Gas Law:
MVP
m3 1000'KT
C (ppbv)
where Mv = molecular weight of VOC (g/mole),
P = pressure (atm), T = temperature (°K), and 9? =
the ideal gas constant (8.204E-05 atm m3/°K mole).
For instance, 10 ppbv benzene (Mv = 78 g/mole) is
equivalent to 32 ^g/m3 at a pressure of 1 atmosphere
and a temperature of 298°K (25°C).
-------
70
60-
50-
•S5 40-
'o
£
D.
! 30H
o
20-
10-
0-
0.01
» Mar-03 mean = 7.9 %
•- Jul-02 mean = 7.2 %
TO-15 Standard
• * • »
T— (H • 1 #-
0.1 1 10
Mean Basement Concentration (ppbv)
100
Figure 6. Replicate precision as a function of mean basement concentration for the July 2002 and March 2003 sampling events.
50-i
30-
^~
5^
> 20-
0
O
ID "
o-
—
'
i
[
1
-
_,_
• ;
i
1
I 1
; :
'
O ; ; ,
T ; ;
T ,
*oy
°6y
v* *
i ;
;
-i J -
•
<
0;
«t*
' ' 11
,
o
- .
T:
o
^ T ; ^
' ^ y
!
: ' T
>
*
-^r
: •';•,'
1 • i ! j 1 ! i
: • i ' •
i i ; i i
• House B (avg cov = 5.2%, n=28, 7/02)
0 House F (avg cov = 6 2%, n=18, 3/03)
T House H (avg cov = 5.2%, n=28, 3/03)
global avg = 5.5%, n= 74
!
I
T
I
i
I
,
*%,
«
(
._
T
1
»
i —
*
j
1
'
i
— _ _.._j
I
'
T' ;
[ ,
-j-
10'2
101
10° 101
Concentration (ppbv)
102
103
Figure 7. Coefficient of variation (COV) as a function of mean basement concentration for July 2002 and March 2003 sampling
events.
-------
3.2 Basement and Outdoor Air Sampling
for VOCs
Building surveys were conducted prior to basement
and sub-slab sampling. Surveys provide an opportunity
to explain the purpose and rudimentary mechanics of
indoor air and sub-slab air sampling to homeowners
and to check buildings for household solvents which
could potentially hinder a vapor intrusion investigation.
A survey guide similar to that included in EPA's vapor
intrusion guidance (USEPA, 2002) was used for each
building.
Basement and outdoor air sampling was conducted
prior to sub-slab sampling for VOCs. Figure 8
illustrates a tripod used to collect a 24-hour outdoor
air sample during the March 2003 sampling event.
Figure 8. Tripod and six-liter evacuated canister used to
collect a 24-hour outdoor air sample during the March 2003
sampling event.
Each evacuated canister used for basement and
outdoor airsampling was equipped withaflow controller
and particulate filter. A particulate filter was attached
to the high pressure inlet port of a flow controller.
The low pressure outlet port of the flow controller
was connected to a canister inlet port. Vacuum was
measured in each canister prior to sampling. Air in
each basement was sampled by placing a six-liter
SilcoCan canister approximately 1 to 1.5 meters (3 to
5 feet) above the floor in a centrally located position.
With the exception of closing basement doors and
windows, no other precautions were taken to reduce
air exchange in homes before or during sampling.
During the July 2002 sample event, 5 basements
were sampled at a flow rate of 73 to 80 ml/min over a
1 -hour period. During the March 2003 sample event,
11 basements were sampled at a flow rate of 3.1 to
3.3 ml/min over a 24-hour period.
As illustrated in Figure 5, an Aalborg Electronic Mass
Flow Meter (Model GFM-1700) was used during
sampling of basement airto ensure that flow controllers
maintained constant sample flow rates (±1 ml/min).
While not performed in this investigation, flow rate can
also be estimated by periodically measuring vacuum
in a canister through use of the Ideal Gas Law by:
AV,
1
can(2)
' can(l) ' STP "can
At
At Tcan(:
can(l)
where: AVSTP = change in standard volume of air in
canister (L),At= change in time (min), TSTpand PSTP
= standard temperature (273°K), and pressure
(1 atm), Vcan = volume of canister (L), Pcan(10r2)
= pressure (atm) in canister at time 1 or 2, and
Tcan(1or2) = temperature (°K) in canister at time 1 or
2. For instance, if absolute pressure in a canister is
0.1 atm and 293°K at time (1) and 0.2 atm and 298°K
10
-------
at time (2) over a period of 60 minutes, then sample
flow rate during this period is estimated at 0.008
standard liters per minute (SLPM) or 8 standard cubic
centimeters per minute. At the cessation of sampling,
pressure was measured to document the presence
of residual vacuum. Mechanical flow controllers have
difficulty maintaining constant flow near atmospheric
pressure.
3.3 Quality Control Measures and Data
Quality for Basement Sampling and
Analysis for Radon
Radon was sampled in basement and sub-slab air to
evaluate the feasibility of using radon as an indicator
compound to assess vapor intrusion. In the absence
of significant off-gassing of radon in a water supply or
radon precursors in building materials, measurement
of radon in sub-slab and indoor air may provide an
indoor air/sub-slab air concentration ratio unaffected by
source terms outside or inside a building. Radon gas
has the potential to be widely utilized as an indicator
compound during vapor intrusion investigations
because it is present in virtually all sub-surface
media, albeit at low concentrations in some areas of
the United States.
Open-faced, activated charcoal canisters were used to
measure basement radon gas concentrations during
the March 2003 sample event in accordance with
sampling procedures outlined in EPA 402-R-93-003
(USEPA, 1993). Cylindrical 1.5 cm diameter, 5 cm
deep canisters containing Calgon-type 1193-coconut
shell charcoal were supplied by AccuStar Labs in
Medway, MA. Canisters were placed with the open
side up 1.2 to 1.5 meters above a floor in a central
location with unimpeded air flow and left undisturbed
for a period of 48 hours. EPA 402-R-93-003 requires
placement of canisters a minimum distance of 75
cm from a floor, one meter from an exterior wall, and
deployment for a minimum of 48 continuous hours.
Property owners were advised to close windows and
doors during sampling to minimize ventilation. At the
cessation of sampling, canisters were closed with
protective covers, resealed, and submitted to AccuStar
Labs for analysis.
Compliance with EPA 402-R-93-003 requires
attainment of a lower limit of detection (sensitivity) of
0.5 picoCuries per liter (pCi/L) for an exposure period
of two to seven days and attainment of five primary
quality assurance factors: (1) routine instrument
checks, (2) calibration of canisters and equipment
utilized to analyze canisters at least once every 12
months, (3) agreement within ± 25% between known
and measured concentrations at or above 4 pCi/L with
a testing frequency of 3 per 100 samples submitted,
(4) measurement of background concentration in
field blanks with a testing frequency of at least 5% of
canisters deployed, and (5) a coefficient of variation
(COV) less than 10% for concentrations at or above
4 pCi/L for replicate (comparison of two canisters)
analysis with a sampling frequency of at least 10% of
samples collected. A detection limit of 0.4 pCi/L was
attained in this study. As illustrated in Figure 9, the
average COV was 10%, but a COV greater than 10%
was observed for several samples at concentrations
greater than 4 pCi/L.
11
-------
I\J-
60-
50-
40-
3?
> 30-
O
o
20-
10-
o-
! i
j
j
i
i — i — i —
I I
i i
! !
I I
I i
i !
i I
', i
1 t
- ! !
| 1
j
-! j-
s
*
1
1
I
1
\
t
i
!
|
1 i
>
i i I i
i i !.
i r ' --
i i i
! i 1
! t !
I {
1 i
! !
i 1
I i
S 1
i !
! !
i j
1 T
! 1
I i
i '
i i
t 0
1 1
I i
'• k*
i r
i ..
• • j •
i !
! !
•
"
i
j j
| " t
' I
|
I
i \
! 1
i
i
i
|
!
.
'
j
j
: : '• ^ 1 — i — : — r~
t
•
• *
Vverage CpV = ^OiO%
1 ' i i '
i
i
,
i
i
i
,
I
L
: i
s ;
i (
t
! j
1
i i
! i
I !
s
i
-l
i ;
' * !
i • . •
10'1 10° 101
Figure 9. Coefficient of variation (COV) as a function of mean basement radon concentration.
102
3.4 Sub-Slab Probe Assembly and
Installation
To minimize the potential for drilling through utility
lines, local utility companies were contacted to mark
off entry points of water, gas, electrical, and sewer lines
outside each building. Utility companies, however,
will not trace utility lines inside a building. During this
investigation, it was often possible to see points of
entry of gas, water, and sewer lines through basement
walls and floors. At most homes, a sewer line entered
a basement in a central location and could be traced
outside a building by using utility company markings.
Tracing utility lines inside a building with a slab on
grade would have been considerably more difficult.
In this case, a local plumber and electrician would
have been contracted to provide recommendations
on safe locations to drill.
Sub-slab vapor probes were installed several days prior
to sampling. Figure 10 illustrates general construction
details of sub-slab vapor probes installed in concrete
slabs. Sub-slab vapor probes were designed to lie
flush on the upper surface of a slab to not interfere
with daily building use and to "float" in a slab to enable
gas sample collection from sub-slab material in direct
contact with a slab orfrom an air pocket directly beneath
a slab created by sub-slab material subsidence. Use
of a screen was unnecessary because probes were
not inserted directly into sub-slab material. Probes
were assembled prior to drilling to minimize exposure
time of sub-slab soils to an open hole.
As illustrated in Figure 11, sub-slab vapor probes
were assembled from 5.08 cm (2") long, 1/8" brass
pipe nipples having inner and outer diameters of
0.64 cm (1/4") and 0.95 cm (3/8"), respectively, and
12
-------
Recessed
-Threaded Cap
ement Grout/ //
// // // // // ,
Brass or Stainless
Steel Threaded',
Fitting or
Compression
Fitting
Brass or Stainless
Steel Tubing
*<4&&2*& ^;^%f %£$i
'^^ •*-*.-% '&:{ $?'£* ?^- &-
^^Sub-slabson^ A^?-P^:-Sr:^^i
> '-'..^j^i *§:>?•£ rt-*:x^ ^^p ^
- ;.<3f .f^ .•°n^-;..e^;°\>-.-J°.g :°
-------
Figure 13. A comparison of VOC concentrations in outdoor air and outdoor air passing through brass fittings used for probe
construction during the July 2002 sampling event. Dashed lines indicate detection limits.
Initial probe design emphasized the use of materials
readily availableatatypical hardware store. However,
cutting oils are often used to machine brass nipples
and couplings and thus require analytical testing as
described here to ensure cleanliness. Sub-slab vapor
probes at other sites are now assembled from 2.5 cm
(1") long, 0.64 cm (1/4") OD x 0.46 cm (0.18") ID gas
chromatography grade 316 stainless-steel tubing and
2.5 cm (1") long, 0.64 cm (1/4") OD x 0.32 cm (1/8")
NPT Swagelok stainless-steel compression fittings.
Use of gas chromatography grade stainless-steel
materials minimizes potential VOC contamination due
to probe assembly. The components of this probe
design are illustrated in Figure 14.
Figure 14. Stainless-steel materials now used
for sub-slab probe assembly.
14
-------
As illustrated in Figure 15, a Model 11224E Bosch
1.25-inch rotary hammer drill was used to create a 2.5
cm (1") "outer" diameter hole approximately 2.5 cm (1")
into a slab. Initial depth of penetration was equivalent
to the length of the brass couplings to ensure that the
probes were flush with the upper surface of the slab.
The inside of the outer hole was cleaned with a damp
towel priorto creating aO.95 cm (3/8") "inner" diameter
hole through the remainder of the concrete. The drill
bit was then allowed to penetrate an additional 5 cm
(2") into sub-slab material (e.g., sand or sand and
gravel) to create an open cavity to prevent potential
obstruction of probes during sampling. The outer
diameter hole was then cleaned once more with a
damp towel to increase the potential of a good seal
during cement application. Inner and outer holes are
illustrated in Figure 16. Probe tubing was then inserted
into the inner diameter hole allowing couplings or hex
fittings to rest at the base of the outer diameter hole.
A quick-drying, lime-based cement which expanded
upon drying (to ensure a tight seal) was mixed with
tap water to form a slurry and placed into the annular
space between the probe and inside of the 2.5 cm
(1") diameter hole using a small metal rod. Tap water
was not analyzed for VOCs during our investigations.
Tap water at these homes was chlorinated and likely
contained trihalomethanes but not VOCs of concern
in sub-surface media. Nevertheless, it would appear
prudent in future investigations to use VOC-free water
for cement application. The cement was allowed to
cure for at least 24 hours prior to sampling. Using
this procedure, 3 probes could be installed in less
than 2 hours.
Figure 15. Drilling through a concrete slab using a rotary
hammer drill.
Figure 16. "Inner" and "outer" holes drilled in a concrete slab.
15
-------
Schematics illustrating the location of sub-slab probes
and other slab penetrations (e.g., suction holes for
sub-slab permeability testing), evacuated canisters for
basement air sampling, and other investigative devices
(e.g., permeation tubes for air exchange testing) were
prepared for each building. Schematics were used
to document sample locations, interpret sub-slab
sample results, and design corrective measures.
Figure 17 illustrates a typical schematic developed
for each building.
I
Prob« C
Lolly Column
Chimney
Lolly Column
i W
on Lol
on
Heater
Mice Ooer^f
Suction B
Stain to
1st Floor
illy Column
0Lolly Column @U>»y Column
its
Suction A
Probe B
o f«n s
Electric Box and Water Meter/-)
i—i i V
LEGEND
Q T«*t Monitoring Point and Sub-Slab VOC Sampling Location
• Multi-Point Airflow T«st Extraction Point
(g) Proposed Location for Mitigation Fan
O Proposed SSDS Extraction Point Location
gj^ Sub-Slab Radon T«st Location and Sample Stz* R«f«r«nc« Number
4& Indoor Air Radon Test Location and Sample Size Reference Number
Figure 17. Typical schematic illustrating location of sub-slab vapor probes.
16
-------
3.5 Sub-Slab Air Sample Collection for
VOCs Using EPA Method TO-15
As illustrated in Figure 18, a brass NPTto Swagelok
union fitting was used to connect vapor probes to a
"T" fitting made of a stainless-steel flexible line and an
in-line valve. A portable vacuum pump was used to
purge vapor probes and sampling lines. Samples were
collected by closing the in-line valve on the pump end
of the "T" fitting and opening a valve for entry into
a six-liter SilcoCan canister. A particulate filter was
attached to the inlet port. Samples were collected by
opening the canister valve and waiting until canister
pressure approached atmospheric pressure (grab
sampling). This took approximately two minutes.
Time-integrated sub-slab sampling will be evaluated
in future investigations.
Figure 18. Sample train for sub-slab air collection using EPA
Method TO-15.
3.6 Quality Control Measures and Data
Quality for Sub-Slab Air Sampling
Using Tedlar Bags and On-Site GC
Analysis
In addition to sample collection in evacuated canisters,
sub-slab air samples were collected in 1-liter Tedlar
bags. Tedlar bags were filled in about one minute
resulting in an average flow rate of 1 SLPM. As
illustrated in Figure 19, sub-slab vapor samples were
collected from the vapor probes using a threaded (1 /8")
brass or plastic nipple, a peristaltic pump, Tygon, and
Masterflex tubing. All tubing was disposed of after
sampling at each probe to avoid cross contamination.
High purity FEP-lined polyethylene tubing could be
used in lieu of Tygon tubing because it offers very low
vapor and gas permeability, is non-photo reactive,
and is a low cost alternative to fluoropolymer tubing.
Tedlar bags were stored in a cooler without ice to
avoid condensation and analyzed for target VOCs
(1,1,1-TCE, 1,1-DCE, TCE, c-1,2-DCE) by EPA's
New England Regional Laboratory within 24 hours
of sample collection.
Unlike TO Methods, EPA does not have explicit quality
assurance guidelines for on-siteGC analysis. Thus, a
site-specific quality assurance (QA) plan or standard
operating procedure (SOP) for on-site GC analysis
is critical to collecting defensible data. On-site GC
analysis was conducted by EPA's New England
Regional Laboratory using their SOP (USEPA, 2002b).
Hartman (2004) discusses the use of QA requirements
in EPA Methods 8021 and 8260 (water analysis) for
on-site vapor GC analysis.
17
-------
Figure 19. Sample train for sub-slab air collection using one-liter Tedlar bags.
A brief review of EPA's New England Regional
Laboratory SOP for on-site GC analysis is provided
to document work here and to aid development of
QA plans developed for other sites. Air samples from
each Tedlar bag were injected into two portable GCs
with results compared for consistency. The first GC
was a Shimadzu 14A equipped with a 30 m, 0.53 mm
megabore capillary column, a Photoionization Detector
(PID), and an Electron Capture Detector (ECD). The
second GC was a Photovac 10A10 equipped with a
1.2 m (4 ft), 0.32 cm (1/8") SE-30 column and a PID.
A Hamilton 250 /j\ steel barrel syringe with a 2 inch,
25-gauge needle was used to directly inject 200 y\
of sample into both GCs. Standards were prepared
from readily available commercial methanol stock
solutions and diluted in VOC-free water in Class A
volumetric glassware to a concentration of 10 jug/I.
Standards were then immediately transferred from
the volumetric glassware into 40 ml VOA vials and
stored on ice. Prior to air sample analysis, 10 ml of
standard was withdrawn from the 40 ml VOA vial to
create a headspace above the liquid standard. After
a period of equilibrium in an ice bath (0 - 1°C), field
GCs were calibrated for target compounds using the
headspace above the 10/L/g/l standard. This proved
to be a simple, quick, dependable, and inexpensive
method for calibration.
Figure 20 presents a comparison of SilcoCan canister
and TO-15 analysis with Tedlar bag sampling with on-
site GC analysis for 1,1,1-TCA, 1,1-DCE, TCE, and
c-1,2-DCE analysis (n = 91, r2 = 0.88). There is no
systematic bias in the data set (i.e., Tedlar bag analysis
consistently lower or higher than TO-15 analysis).
Four out of five of the outliers were associated with
sampling one probe during one sample event. The
regression coefficient increased to 0.95 when these
four points were omitted. In general, there was good
agreement between Tedlar bag sampling with on-site
GC analysis and EPA Method TO-15.
Tedlar bag sampling and on-site analysis provided
near real-time data compared to EPA Method TO-15.
However, this method provided analysis for a limited
number of compounds with higher detection limits.
Detection limits for on-site analysis were 2-5 ppbv
compared to 0.1 - 0.5 ppbv for EPA Method TO-15.
18
-------
10'
101 102
EPA Method TO-1 5 (ppbv)
Figure 20. Comparison of EPA Method TO-15 Tedlar bag sampling with on-site GC analysis for 1 ,1 ,1-TCA, 1 ,1-DCE, TCE, and
c-1 ,2-DCE analysis, n = 91 , r 2 = 0.88.
3.7 Quality Control Measures and Data
Quality for Sub-Slab Air Sampling for
Radon Using Scintillation Cells
Scintillation cells are air-tight metal cylinders lined
with activated silver zinc sulfide ZnS(Ag). They have
a transparent window at one end for scintillation
counting and connectors at the other end for sample
intake or flow through. When an alpha particle from the
radioactive decay of radon isotopes Rn-222, Rn-220,
and Rn-219 strikes the lining of a scintillation cell, the
alpha particle becomes a helium atom, and the sulfide
de-excites by emitting photons or light pulses. The
transparent window and a radiation monitor equipped
with a photomultiplier tube are used to amplify and
count light pulses.
Scintillation cells used for this investigation were
originally purchased from the Pylon Electronic
Development Company. Pylon Models 110Aand 300A
scintillation cells have internal volumes of 151 ±3 and
270±3 ml, respectively. The larger internal volume
of Model 300A allows greater sensitivity compared
to Model 110A. The scintillation cells were equipped
with two gas-tight Swagelok connectors allowing
continuous monitoring (flowthrough)orgrabsampling
(vacuum collection or disconnection after a period of
flow through or purging).
The sub-slab sampling train is illustrated in Figure 21.
A threaded (1/8") barbed nipple was used to attach
Tygon tubing to a sub-slab probe. Plastic barbed fitti ngs
were used to connect Tygon to Masterf lex tubing used
19
-------
by a peristaltic pump to create vacuum
in a probe for sample collection. A
particulate filter was placed on the inlet
sideofascintillationcell. Quick-connect
assemblies were used for connection
of Tygon tubing to Pylon cells. A flow
meter was placed on the outlet side of a
scintillation cell to ensure a flow rate of
approximately 1 SLPM. The outlet end
of the flow meter was vented outside.
Purging ceased when 10 cell volumes
were exchanged in each cell.
Figure 21. Sample train for sub-slab air collection for radon using scintillation cells.
Samples were analyzed by the MA Department of
Public Health (Mr. William Bell) within four hours
as recommended in EPA 402-R-93-003. A Pylon
AB-5 portable radiation monitor was used to amplify
and count light pulses. Counts per minute were
determined by recording total counts over six 10 minute
measurement intervals and dividing by total counting
time. The relationship between the number of light
pulses counted per time and the concentration of radon
gas in a cell is outlined in EPA 402-R-93-003
cpm(s)-cpm(bkg)C
Rn= (3)(2.22)(E)(V)(A)
where: CRn = concentration of radon gas (pCi/l),
cpm(s)= counts per minute forthe scintillation cell after
sample collection, cpm(bkg) = counts per minute in the
scintillation cell priorto sample collection (background
count), C = a correction factor for radon decay during
the counting interval provided in EPA (1993), 3 = total
number of alpha particle emitters, 2.22 = a conversion
factor relating disintegrations per minute (dpm) to pCi,
E = counting efficiency (cpm/dpm), V = volume of the
scintillation cell (liters), and A = a correction factor
for radon decay between sample collection and start
of measurement provided in EPA (1993). Certificates
of calibration for counting efficiencies for scintillation
cells used in this study were included with analytical
results. Knowledge of counting efficiency is necessary
because not all alpha particle impacts on the activated
zinc sulfide lining result in detection of light pulses.
EPA 402-R-93-003 requires: (1) a lower limit of
detection of 1 .OpCi/Lorless, (2) collection of replicates
at 10% of sample load with attainment of a coefficient
of variation of 10% or less at concentrations 4 pCi/L
or greater, (3) use of field blanks kept sealed in a low
radon (lessthan 0.2 pCi/L) environment, and analyzed
in the same manner as other samples at 5% of sample
load, and (4) calibration of cells at least once every
12 months.
20
-------
4.0 Methods and Materials Used for Air Permeability Testing
and Sub-Slab Air Flow Analysis
Air permeability testing was conducted in sub-slab
media to support design of corrective action (sub-
slab depressurization) and airflow simulations used
to evaluate infiltration of basement air into sub-slab
mediaduring sampling. A Bosch 1.25-inch capacity
hammer drill (Model 11224E) was used to drill a
5.08 cm (2") diameter hole 2.5 cm (1") through a
slab for installation of suction points. A 2.5 cm (1")
diameter drill bit was then advanced through the
remainder of a slab and 7 to 8 cm (3") into sub-slab
material to create an open cavity for air extraction.
Sub-slab material consisted of coarse sand and
gravel similar to native deposits. Suction points
were assembled from 5.1 cm (2") long threaded
brass pipe having inner and outer diameters of 1.9
cm (3/4") and 2.5 cm (1") respectively and 3.2 cm
(1 Vi") outer diameter brass couplings. The length
of the threaded pipe used was determined by the
thickness of the concrete slab. Couplings were
installed without threaded pipe when the slab was
less than 7 to 8 cm (3") thick. All suction points
were completed flush with the top of the concrete
slab with recessed brass plugs so as not interfere
with day-to-day use of the basements.
As illustrated in Figure 22, a small regenerative blower was
used to extract air from sub-slab material. A variable-area
flowmeter illustrated in Figure 23 was used to measure
flow rate. As illustrated in Figure 24, air pressure was
measured with magnehelic gauges. A digital manometer
(± 1.0% of range accuracy) was also used for pressure
measurement.
Figure 22. Regenerative blower used for air permeability testing.
21
-------
Figure 23. Variable-area flowmeter used for air permeability
testing.
Radial and vertical air permeability of sub-slab media
was estimated using Baehr and Joss's (1995) analytical
solution for two-dimensional, axisymmetric, steady-
state gas flow in a semi-confined domain.
4>(r,z)=
where
2abnQn.u.g
-------
Baehr and Joss (1995) represent I as the vertical
permeability of a semi-confining boundary layerdivided
by the thickness of the boundary layer. Given the
potential of sub-slab material subsidence, i can be
taken in this application to represent a term used in
describing lateral air flow beneath a slab but above
sub-slab material and vertical airflowthrough cracks in
aslab. Airflowthrough cracks is not explicitly simulated
in this model. Also, the vertical permeability of slab
material is not estimated in this application.
Parameter estimation involves finding values of kr, kz,
and i that minimize the difference between observed
versus simulated data for pressure or the root mean
squared error (RMSE) defined as:
RMSE =
where P is observed pressure and p is modeled
pressure. When there are few pressure measurements
(some of which are of minimum usefulness because
of near atmospheric pressure response), non-
uniqueness in parameter estimation and convergence
to a local but not global minimum RMSE can occur. That
is, a similar pressure distribution in a domain can be
simulated using various parameter inputs. To address
this issue, 5000 random initial guesses were applied
for kr, kz, and I to determine a lowest RMSE. This
process was repeated with decreasing intervals of kr,
kz, and i until the same or similar RMSE values were
obtained. A FORTRAN program, MFRLKINV (USEPA,
2001), was used to facilitate computations.
Figure 25 illustrates the best five fits of kr, kr/kz,
and I for an air permeability test at House C
consisting of four observation points conducted at
a flow rate of 255 SLPM. Estimation of kr/kz was
constrained between 1 and 2 because sand and
gravel typically exhibit kr/kz values within this range
(USEPA, 2001 -Sections). Unconstrained estimation
of kr/kz resulted in greater variation in I to provide
comparable fits to the observation points. When
pressure measurement of the closest observation point
was eliminated from consideration during parameter
estimation, estimates of kr varied from 1010 cm2 to
107 cm2. This demonstrates the need to locate one or
more pressure monitoring points or sub-slab probes
fairly close (e.g., one meter) to the source of vacuum
during sub-slab air permeability testing.
Estimates of kr, kz, and I were then used for air flow
simulations. Radial volumetric specific discharge (qr)
and pore-air velocity (vr) (cm/s) are defined as:
1 kr
vr =
where Q, = volumetric air content and
f\ I IX I 00
— = - —IZqnAnCosfq^b-zVbjK,
SbL
ab
Vertical volumetric specific discharge (qz) and pore-air
velocity vz (cm/s) are defined as:
^=—^ -£*
where
|^|ZqAsin[qn(b-z)/b]K0f^l .
23
-------
50
45-
40-
35-
I
TO 30-
*=
§
i
b
£ 25-1
5 20 H
in
0)
D.
15-
10-
5-
0
kr = 7.90E-07 cm2, kr/kz = 1.65, leakance == 3.21 E-09 cm
kr = 7.20E-07 cm2, kr/k2 = 1.42, leakance = 3.27E-09 cm
kr = 8.16E-07 cm2, kr/kz = 1.65, leakance = 3.25E-09 cm
kr = 6.69E-07 cm2, kr/kz = 1.39, leakance = 3.30E-09 cm
kr = 7.12E-07 cm2, kr/k2 = 1.22, leakance = 3.13E-09 cm
observed pressure differential (Pa)
0
200
400 600
Radial Distance (cm)
800
1000
Figure 25. Best fit model results for permeability test conducted at House C with four observation points and a flow rate of 255
SLPM (kr/kz constrained between 1-2).
For analysis of air flow during sampling, the norm
of the radial and vertical pore-air velocity vector (v)
(cm/s) is calculated by:
dt
ea
dt
and the algorithm:
The stream function, \i/(r,z), for axisymmetric flow with
anisotropy may be written as:
dz
dr
d\\i
dr
vr(r*lfz*+1)]At
a dz
Solving for \|/(r,z) yields:
The path of an air particle in a flow field at location
(r0,z0) at time zero was solved by the following set of
ordinary differential equations:
where i = 0,1 ..... N. The particle tracking terminates
when the particle reaches the probe. Total travel time
= N At . The equations were solved numerically using
a second-order Runge-Kutta method. A FORTRAN
program, SAIRFLOW, (USIEPA, 2001) was used to
facilitate computation of P(r,z), vr, vz, v, v|/(r,z), and
travel time to a vapor probe.
24
-------
5.0 Discussion of Sampling Issues Associated with Sub-Slab Air Sampling
5.1 Assessment of Infiltration of Basement
Air During Air Extraction
Generally, sub-slab air samples were collected by first
purging two liters of air from probes at a flow rate of
1 SLPM, then collecting a sample into a one liter Tedlar
bag at a flow of 1 SLPM, followed by purging one liter
again at a flow rate of 1 SLPM, then finally collecting
a five-liter sample into a six-liter evacuated canister
over a period of approximately one minute.
If during sub-slab sampling, basement or indoor air
enters openings in a slab (e.g., cracks, utility entry
locations) and is collected into a sampling vessel, then
measured sub-slab concentrations should decrease
for VOCs having sub-slab concentrations higher than
basement air. The opposite effect should occur for
VOCs having sub-slab concentrations lower than
basement air. One way to directly evaluate infiltration
of basement or indoor air into a sampling vessel
during air extraction is to collect a series of sequential
samples and measure vapor concentration as a
function of extraction volume. Constant concentration
in sequential samples would indicate the absence
of significant infiltration of basement air during the
extraction period. A reduction in concentration during
air extraction could indicate significant infiltration
during extraction or reduced sub-slab air concentration
away from the probe (spatial variability). An increase
in concentration during air extraction could indicate
increased concentration away from the probe (spatial
variability).
Sub-slab samples were collected sequentially in a
probe at three homes (L, M, and N) infive 1 -liter Tedlar
bags at a flow rate of 1 SLPM and analyzed on site by
GC analysis. Results of sampling at each location are
summarized numerically in Tables 14b, 15b, and 16c
(see pages 81,84 and 88) and graphically in Figures
26a, 26b, and 26c. At each location, extraction of
5 liters at a flow rate of 1 SLPM had little effect on
sample concentration indicating a lack of significant
infiltration. Concrete slabs at these three buildings
consisted of approximately 2-4" of relatively intact
concrete (few cracks). A comparison of replicate
canister samples was also used to assess the effect
of extraction volume. After Tedlar bag sampling, a
canister sample represented an extraction volume of
5 to 9 liters. A replicate canister sample represented
an extraction volume of 10 to 14 liters. Results of
sampling are summarized numerically in Tables 12a
and 15a (see pages 74 and 83) and graphically in
Figures 27a and 27b. Sampling at Probe A in House
J indicated little difference in sample results. Sampling
at Probe A in House M revealed a slight decrease in
vapor concentration as afunction of extraction volume
for most VOCs. Thus, replicate sampling in canisters
indicated little or no effect in sample concentration
due to air extraction.
25
-------
300
Extracted Volume (L)
Figure 26a. Sub-slab vapor concentration as a function of extraction volume at Probe A in House L using Tedlar bag sampling and
on-site GC analysis.
600
500
400
O
"co
-b 300
I
o
O
O 200
Sample Volume (L)
Figure 26b. Sub-slab vapor concentration as a function of extraction volume at Probe B in House M using Tedlar bag sampling and
on-site GC analysis.
26
-------
Extracted Volume (L)
Figure 26c. Sub-slab vapor concentration as a function of extraction volume at Probe A in House N using Tedlar bag sampling and
on-site GC analysis. Dashed lines denote detection limit.
8.
101
10"-
Extraction Volume = 9 L
Extraction Volume = 14 L
o o o o
o
111
o
CL
o
o
o
0)
o
HI
*
m
UJ
m
n 5
0 O
-------
101-
Q.
Q.
C
O
10°-
o
o
ID'1
extracted volume = 9 liters
extracted volume = 14 liters
v-1—-r—-i"—~f
r — ™
"
o J=
O csi
f
E
S Z t
(D O -5
4 II
Figure 27b. Sub-slab vapor concentration as a function of extraction volume at Probe A in House M using EPA Method TO-15.
A second direct method to evaluate infiltration of
indoor air into sub-slab media during extraction is to
compare basement and sub-slab concentrations of
VOCs known not to be associated with subsurface
contamination. Statistical testing to distinguish VOCs
associated with vapor intrusion from other VOCs
detected in indoor air is discussed in section 6.0. A
simple mass-balance equation:
_ ^ss^ss + ^ indoor^ leak
is used to express a measured sub-slab vapor
concentration (Cmeas) as a function of "true" sub-slab
concentration (Css) and indoor air concentration (Cndoor)
where Qleak = flow rate of air through cracks or other
openings in the slab and Qss = sub-slab air flow to a
vapor probe. If infiltration of indoor air into sub-slab
media can be expressed as:
Q,™
then
and
GL+ Q,.
1-1 =
Q,.
GL+
r - r '
^indoor '-'ss
The assumption that Cmeas is due entirely to infiltration
of indoor air into sub-slab media (Css = 0) leads
to computation of a maximum value of I for each
VOC detected in indoor air and not associated with
vapor intrusion. The lowest value of I can then be
28
-------
selected to represent maximum infiltration during air
extraction. For instance, 1,4-dichlorobenzene (not
a VOC associated with sub-surface contamination)
was detected in basement air at House H at 36 ppbv
but was not detected in a sub-slab probe (detection
limit 0.086 ppbv). Thus, less than 0.24% of sampled
air originated from above the slab. This method
provides reasonable results (I < 100%) only if the ratio
C /C is greater than 1. Table 1 summarizes
indoor' meas a
maximum infiltration of indoor air during air extraction
at each probe. When the sensitivity of the test was
Table 1. Computation of Maximum Percent Infiltration of
Basement Air into an Evacuated Canister During Sampling
as a Function of Extraction Volume, Location, and Probe.
P[A], P[B], P[C], P[D], and P[E] Denote Probes Evaluated at
Individual Locations
tracer over a specified period of time. Infiltration then
could be estimated by:
Location
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Extraction
Volume (L)
9
9
9
9
9
9
9
9
9
14
NA
9
14
9
9
9
P[A]
(%)
<42
<2.1
IND
<27
< 12
<8.6
<0.60
ND
<7.5
<3.0
NA
NA
<78
NA
<0.42
NA
P[B]
(%)
<9.0
<6.7
IND
<27
< 19
NA
NA
<0.24
NA
NA
NA
<2.7
NA
< 44
NA
< 1.3
P[C]
(%)
IND
<0.35
IND
NA
<39
<22
<0.60
<0.33
NA
NA
NA
NA
NA
<24
<0.63
< 1.4
P[D]
(%)
IND
NA
NA
NA
NA
NA
P[E]
(%)
NA
NA = not analyzed IND=mdetermmate
satisfactory (e.g., sensitivity < 1%), infiltration during
sampling was evidently very low. Use of this method
however, requires detection of elevated levels of
VOCs not associated with sub-surface contamination
in indoor air and low levels or low detection limits for
these compounds in sub-slab air. Sensitivity could
be increased by enclosing an area around a probe
with a chamber during air extraction and injecting a
r
^tracer
sample
where C, = tracer concentration within the chamber,
trscGr
Umpie 's total sampling time, and ttracer is time of tracer
application. Tracer concentration would have to be held
constant during the tracer application period. Also,
flow analysis would have to be conducted to estimate
the potential area of infiltration during testing. If a
pure phase solvent is exposed to air within a chamber
surrounding a sub-slab probe and equilibration within
the chamber is assumed, CX r could be estimated
iracer
in units of jug/m3 using the solvent's saturated vapor
concentration (Cf)
C,, = 106
or in units of ppbv by
-lO'-S-
p
where Pv = vapor pressure (atm), and Mv, T, R
and W are as previously defined. For instance, the
saturated vapor concentration of isopropanol, which
has a vapor pressure of 0.058 atm at 25°C and a
molecular weight of 60.1 g/mole, is 1.43E+08 jug/m3
or 5.80E+07 ppbv at a pressure of 1 atmosphere. If
a pure phase solvent is used as a tracer for infiltration
testing, it would be prudent to minimize ttracer/tsample
because detection of a solvent at a high concentration
could result in high detection limits for other VOCs of
interest. For instance, if I = 0.01 or 1 %, and ttracer/tsample
= 1, then isopropanol would be present in a sampling
canister at a concentration of 1.43E+06jug/m3 which
would likely preclude detection of other VOCs of
29
-------
interest. The ratio t,racer /tsample would have
to be reduced to 0.001 (e.g., sample time
of 10,000s, tracer time of 10s) to observe
a concentration of 1.43E+03 fjg/m3 in the
canister.
A third, but indirect, method of evaluating
infiltration of indoor air during air extraction
is to simulate streamlines and particle
transport during flow. Mean estimated
parameters at House C (kr = 7.4E-07 cm2,
kr/kz =1.5, i = 3.2E-09 cm) were used to
generate streamlines and travel time contours
of air particles in sub-slab material with Oa =
0.35 at a flow rate of 1 SLPM. Depth to a
no flow lower boundary was set at 500 cm.
This simulation is illustrated in Figure 28.
Dashed contour lines for 60 and 300 seconds
reflect collection of 1 and 5 liters of sample,
respectively, at a flow rate of 1 SLPM. Areas
between streamlines reflect fractional flow.
As previously stated, air flow through the
slab is not simulated because the leakance
term represents a combination of lateral
flow beneath a slab and sub-slab material
and vertical flow through a slab. The solid
contour lines for 0.90 and 0.95 streamlines are
highlighted. The 0.95 streamline at the top of
the figure indicates that approximately 5% of
air that had exited a probe at 100 seconds (1.7
liters) originated from above sub-slab material
of which some fraction of this 5% could have
been from infiltration from basement air. The
0.90 streamline indicates that approximately
10% of air that exited a probe at 700 seconds
(11.7 liters) originated from above sub-slab
material of which some fraction of this 10%
could have come from infiltration from basement air. Thus,
when extracting 12 liters of air, less than 10% of air entering
an evacuated canister could have been from infiltration of
basement air. The results of sequential sampling and mass
balance analysis revealed that less than 1% of air entering
evacuated canisters during this investigation were from
basement air.
In this investigation, extraction volumes up to 14 liters had
little effect on sample results. However, data generated in
this investigation cannot be extrapolated to justify the use of
large extraction volumes (e.g., 20, 50, or 100 liters) during
sampling. Further research is needed in this area.
probe
fertical Distance Beneath Slab
C? O O1 O
Slab
i
1
i- i ; ;---;- ;-.-• -.^trnrr
17 / ; i § r -«- -i --oe,
' . / , / r ' 4-- '—-075*
: ' / / / / '- ' - s
" '' ' ' ' ' ' / ' /
/ II 1 ' 1 Otis
N° , / / ill 1
'' / > / ^ /,./"•«
^x ^ / / -/ x /ftss '
^.'' / / *' ~ '' ' °-s '
on** x x ^r / r t
^00''"^ >r^ X>''' /^/* /
'^-' 'x''-V x- '- /''
"200' ' : - - ' ' \ -'' ' ^ '+*'' /'" ^ *'-
„ .300- " " - „ -" " ~-, x ^ x x ^ ' v x X X?
-2QJ
-25^
_ 400-
^ -600-
- .800-
'"- x "^ ~ d
20 25
5 10 15
Radial Distance From Probe (cm)
Figure 28. Simulated streamlines (solid lines) and travel time (s)
(dashed lines) contours in sub-slab media when kr = 7.4E-07 cm2,
kr/kz = 1.5, i = 3.2E-09 cm, flow rate = 1 SLPM, and depth to ground
water = 500 cm.
30
-------
5.2 Assessment of Extraction Flow
Rate
Laboratory- and field-scale research conducted
on soil venting indicates that rate-limited air-
water and/or solid-water mass exchange can
occur in sub-surface media during air flow.
Rate-limited mass exchange could decrease
vapor concentration in a sample container
below what would be expected from equilibrium
partitioning. EPA (Section 9, 2001) provides
a comprehensive summary of mass transfer
coefficients for air-water and solids-water
exchange determined for several soil types
in laboratory column studies. These studies
indicate that for sandy, non-oven dried soils
typically found directly beneath a slab, solids-
water partitioning and hence rate-limited solids-
water rate-limited mass exchange should be
insignificant. Rate-limited air-water exchange,
however, can be significant at high pore-air
velocities. In one case, rate-limited air-water
exchange was observed at a pore air velocity as
lowasO.01 cm/s. In another case, no rate-limited
behavior was observed at a pore-air velocity as
high as 0.16 cm/s. Rate-limited mass transfer
is a function of media-to-media mass transfer
coefficients and a characteristic length over
which mass transport occurs. The transport
length in laboratory column studies is typically
on the order of 30 cm. The characteristic length
in sub-slab media could be considerably longer
(resulting in increased potential of attainment of
local equilibrium) depending on the thickness
and permeability of sub-slab media as illustrated
by streamlines in Figure 28.
Figure 29 illustrates pore-air velocity and travel time at a
sampling at a rate of 1 SLPM for sub-slab conditions present
at HouseC. Because of convergent flow to a relatively small
probe cross-sectional area (radius = 0.32 cm, length = 5
cm), pore-air velocity likely exceeded 0.01 cm/s throughout
most of the domain during sampling (radius «17 cm for 5
liters of air extracted from Figure 28). However, constant
concentration in sequential samples indicated an absence
of rate-limited mass transport during air extraction. This
could be due to a long characteristic transport length and
low extraction volume relative to the sampled domain. The
California Environmental Protection Agency (Cal EPA)
in conjunction with the California Department of Toxic
Substances and Los Angeles Regional Water Quality Control
Board recently published an advisory on soil-gas sampling
probe
Slab
CN *- •«— T- (
-10-
-------
(Cal EPA, 2003) specifying maximum flow rate during
sampling. This advisory is being used for sub-slab
sampling. Cal EPA recommends a maximum sampling
rate of 0.1 to0.2SLPM. Given simulations presented
here, this recommendation appears reasonable.
5.3 Evaluation of Equilibration Time
The process of drilling through a concrete slab would
be expected to reduce sub-slab vapor concentration
in the immediate vicinity of a probe. For strictly
diffusive transport, concentration reduction would be
a function of chemical properties of a VOC (Henry's
constant, organic carbon - water partition coefficient,
aqueous diffusion coefficient, air diffusion coefficient),
material properties of sub-slab media (water content,
porosity, bulk density, and organic carbon content),
and temperature. For advective-diffusive transport,
additional factors such as air permeability and the
pressure differential between basement and sub-slab
air are relevant. Concentration reduction would be
expected to be greatest in relatively dry permeable
material. However, these conditions would also
expedite equilibration of the vapor concentration
around a sub-slab probe. If sub-slab material consists
of silt or clay, equilibration time may not necessarily
be significantly longer because the initial vapor
concentration perturbation may be reduced by a lower
media permeability and lumped diffusion coefficient.
Robust estimation of equilibration time would require
knowledge of the extent and magnitude of vapor
concentration reduction in sub-slab media and three-
dimensional advective-diffusive modeling. The problem
can be simplified by using advective air flow modeling
with particle tracking to estimate a maximum radius
of perturbation for various sub-slab conditions when
clean air flows into an open hole. This radius could
then be utilized as a path length in diffusion modeling
to calculate a maximum equilibration time when a
hole is sealed. For instance, if a maximum pressure
differential of 15 Pa (highest pressure differential or
most conservative value used in EPA's vapor intrusion
guidance) is present between sub-slab media and
basement air during probe installation and mean
estimated parameters at House C (kr = 7.4E-07 cm2,
kr/kz = 1 .5, I = 3.2E-09 cm) with 6a= 0.35 are used for
flow analysis, then this is equivalent to 0.22 SLPM of
air flow into a probe. The radius of perturbation then
is a function of the time in which the probe is open.
In homes near the Raymark facility, holes drilled
for probes were open for a maximum period of one
hour. Using particle tracking, this results in a radius
of perturbation of 27 cm.
Now consider concentration C(r,t) of a VOC at radius
'r' and time 't' in a sphere around a sub-slab probe.
If VOC concentration inside the sphere is initially at
a uniform initial concentration C(r,0) or C0 and the
surface concentration of the sphere at radius '5 ' is
maintained at a constant concentration C(5 ,t) or Cs,
then a normalized concentration at the center of the
sphere or at the probe C(0,t) or C(t) can be estimated
by (Crank, p.91, 1975):
T is a dimensionless time defined by
32
-------
D is a lumped diffusion coefficient (cm2/s) defined
by
D =
TaDa+iwDw
and ta = soil-air phase tortuosity, Da = free air diffusion
coefficient (cm2/s), TW = soil-water phase tortuosity, Dw
= free water diffusion coefficient, H = dimensionless
Henry's constant, 9a =volumetric air phase content,
6W= volumetric water phase content, Kd = soil - water
partition coefficient (cm3/g), and pb= bulk density of
soil (g/cm3). Tortuosity factors can be estimated by
Millington and Quirk (1961):
t, =
where r)= porosity.
If attainment of C(t)/C8 = 0.99 is desired and
C0=0 (most conservative condition), then T approaches
0.537. The initial concentration would be greaterthan
zero if air exited a probe during installation. The
assumption of air entry adds an additional degree
of conservativeness. Figure 30 illustrates time to
C(t)/C5 =0.99fortrichloroethylene(TCE)asafunction
of diffusion path length and ew when T=0.537 and
r|= 0.4, pb = 1.68 g/cm3, Da=7.4E-02 cm2/s,
Dw = 9.3E-06 cm2/s, H=0.38, and no sorption. This
relationship is simply t = 0.537 52/D. At homes near
the Raymark site, sub-slab and underlying soils
underlying each building consisted of relatively dry
sand and gravel. Little or no sorption would be expected
in this material and Owwould be relatively low (e.g.,
ew= 0.05). Figure 30 indicates that for a diffusion path
lengthof27cm,timetoC(t)/C5 =0.99 would occur in
less than 2 hours. During this investigation, sub-slab
probes were allowed to equilibrate for 1 to 3 days. If
probes were immediately sampled after installation
without regard to grout setting time, approximately
14.4 liters (hemispheric domain with a radius of
27 cm and 6a = 0.35) of air would have to be extracted
to remove sub-slab air potentially affected by probe
installation. For sub-slab material consisting of silt or
clay with kr = kz = 1 .OE-09 cm2 and 6a= 0.10, the radius
of perturbation would be approximately 6 cm over an
exposure period of 1 hour requiring an equilibration of
time of approximately 10 hours. However, most sub-
slab material consists of sand and gravel or sand even
for homes built directly on clay. Thus, in most cases,
an equilibration time of 2 hours should be sufficient
for sampling.
33
-------
10000
O5
O5
CD
II
to
O
O
o
-4—*
CD
E
1000 - =-=---=: --=7 •.£.-=:: r~-i=--: ;F --> ->-^n~-1-;
100
10 -
0.01
0.001
10
100
Diffusion Path Length '8' (cm)
Figure 30. Time to reach C(t)/Ce = 0.99 as a function of diffusion path length ' 8' and 0W for TCE when C0 = 0, ^1 = 0.4,
pb = 1.68 g/cm3, Da = 7.4E-02 cm2/s, Dw = 9.3E-06 cm2/s, and H = 0.38 (no sorption).
5.4 Selection of Purge Volume
Sub-slab vapor probes and associated tubing must be
purged prior to sampling because air in a probe and
tubing will initially have VOC concentrations reflective
of indoor air upon removal of recessed socket plugs.
A purge volume consists of the total internal volume
of: (1) sample tubing and associated fittings between
a probe and sample container, (2) tubing and fittings
associated with sub-slab vapor probes, (3) the open
hole in slab below a probe, and (4) the cavity created
in sub-slab material during drilling. Minimum purge
volume prior to sampling can be estimated using a
mass balance equation:
where C is a well-mixed vapor concentration within and
exiting the sample system, Cin = concentration entering
the system, Cout = concentration exiting system,
Q = flow rate entering and exiting sample system,
and V = internal volume of sampled system. When
subject to an initial condition C(0)=C0, (concentration
at time zero), purge volume (tQ/V) can be expressed
as a function of C and C by:
tQ .
— = In
V
dC =Q
dt V
c,n-c
34
-------
Figure 31 illustrates a simulation of purge volume as
a function of C0/Cin and Cout/Cin. Collection of 5 purge
volumes ensures that the exiting vapor concentration
is 99% of the entering concentration even when vapor
concentration inside the sample system has been
reduced to zero prior to sampling (C0 = 0). A purge
volume for the sample train used in homes near the
former Raymark site was typically less than 10 cm3.
4 -
Q) o
E 3
_3
o
CD
I5 2
Q_
1 -
0
coul/c,n
cou,/c,n
cout/c,n
cout/cm
cn,,,/c,n
= 0.99
= 0.95
= 0.90
= 0.85
= 0.80
0.0
0.2
0.4 0.6
C0/C,n
Figure 31. Purge volume as a function of C0/Cin and Cout/Cin.
0.8
1.0
5.5 Placement of Sub-Slab Vapor Probes
Generally, during this investigation, one sub-slab vapor
probe was centrally located, while two or more probes
were placed within one or two meters of basement walls
in each building. This was done to ensure detection
of vacuum throughout the entire sub-slab during air
permeability testing for corrective action. Figure 32
illustrates total vapor concentration of VOCs detected
in Tedlar bags as a function of probe location in each
building tested. There appears to be little correlation
of probe placement with VOC concentration. That
is, placement of a probe in a central location did not
ensure detection of the highest VOC concentrations.
Figure 33 illustrates COVs for VOCs detected
in sub-slab air and associated with sub-surface
contamination. In many instances, COVs exceeded
100%, indicating substantial spatial variability in sub-
slab air concentration and the need for placement of
multiple probes during asub-slab investigation. In this
investigation, 55 probes were installed in 16 buildings
which on average resulted in the placement of one
probe every 20 m2 (220 ft2).
35
-------
104
103 -
Q.
a.
c
g
E
-i—'
§ 102
c
o
O
O
O
5
"o m1
10°
H
I
Location
M
N
O
Figure 32. Total vapor concentration measured in one-liter Tedlar bags as a function of probe location and house. Dark bars refer
to centrally located probes. No VOCs associated with subsurface contamination were detected at location F. Locations H, K, M,
and P did not have a centrally located probe.
^
B
o
160 -i
140 -
120 -
100 -
80 -
60 -
40 -
20 -
0 -
• EPA Method TO-1 5
o Tedlar Bag and On-Site GC
o
' , o o
o
O o
c» o
• •o
• o ••
, : ., i • *o •
, ' ° O
\ T 9 w Q
_ O , O
• m o • ° o o
••.,,•* o ' , 0
. , - - , * •' • '• * O
• °° ° 0 * ° °
9 * * ' o ° o • •
. ; : °: . . o 0 o o
10-'
10° 101 102
Mean Sub-Slab Concentration (ppbv)
103
Figure 33. Coefficient of variation (COV) as a function of mean sub-slab concentration (ppbv) and method of analysis for VOCs
associated with sub-surface contamination.
36
-------
6.0 Use of Basement and Sub-Slab Air Measurements to Assess Vapor Intrusion
6.1 Method of Vapor Intrusion Evaluation
A dimensionless basement/sub-slab concentration
ratio for a VOC or radon was defined by:
/.\ Xbsmt(i)
CC(|) = — .
Xss(i)
where Xbsmt(i) = mean concentration of a VOC in
basement air and Xss(i) - mean concentration of a VOC
in sub-slab air. When a VOC was detected in sub-
slab air but not basement air, a (/) was reported to be
less than the ratio of the detection limit in basement
air to sub-slab air. For instance, if X .. for TCE was
SB(\)
48 ppbv but was not detected in basement air at a
detection limit of 0.24 ppbv, then
-------
0.06 X
bsmt(i) .
An average basement/sub-slab air concentration
ratio (a) was calculated using basement/sub-slab
air concentration ratios for all VOCs associated with
vapor intrusion by:
The variance associated with a (e£) was calculated
by:
Basement/sub-slab air concentration ratios, a (i), and
associated standard deviations, ^/o^, were computed
for every VOC detected in basement or sub-slab air
and plotted. A VOC detected in basement air was
considered due to vapor intrusion if: (1) the VOC
was detected in ground water and/or soil gas in the
"vicinity" of the house, and (2) the null hypothesis that
basement/sub-slab air concentration ratio of the VOC
was equal to the basement/sub-slab air concentration
ratio of an indicator VOC could not be rejected using a
one-tailed Approximate t-Test at a level of significance
or Type I error less than or equal to 0.05. A Type I
error is committed when the null hypothesis is rejected
when it is true. The alternative hypothesis was that
the basement/sub-slab air concentration ratio of the
VOC was greater than the basement/sub-slab air
concentration ratio of an indicator VOC. An indicator
VOC is defined as a VOC detected in sub-slab air
and known to be associated only with subsurface
contamination. The VOCs, 1,1-dichloroethyleneand
1,1 -dichloroethane, were considered indicator VOCs
in this investigation because they are degradation
products of 1,1,1-trichloroethane and not commonly
associated with commercial products. The VOC, cis-
1,2-dichloroethylene, was considered an indicator
VOC because it is a degradation product of
trichloroethylene and not commonly associated with
commercial products. Vicinity is a subjective term
but generally refers to a number of ground-water or
soil-gas measurements within 30 meters of a building.
The primary purpose of these two questions was to
ascertain whether or not VOCs detected in basement
air were due to vapor intrusion. Sources (e.g., indoor,
outdoor) of VOCs not associated with vapor intrusion
were not investigated. Evaluation of outdoor and
indoor air source terms would require additional
building-related information or characterization needs
(e.g., air exchange rates). The Approximate t-Test
for Independent Sample of Unequal Variance was
employed for statistical testing where the test statistic
t' is defined as:
t' =
n i,
n*0)
where a *, Js the variance of the basement/sub-slab
concentration ratio for the indicator VOC, and n(i)
and n*(i) are the number of sub-slab measurements
used in determining a (i) and a * (i), respectively. The
degrees of freedom (df) are defined as:
df =
nfi -1 n*i -1
where
n*(0-iy+(l-c)2(n(0-l
c =
n*(i)
38
-------
6.2 Summary of Results for Buildings
Sampled in July and October 2002
Basement and sub-slab air samples were collected
for VOC analysis in houses A, B, C, D, and E during
the July 2002 sample event. Basement (1 -hr) and
sub-slab (grab) samples were collected in six-liter
evacuated canisters using EPA Method TO-15. Sub-
slab samples were also collected in one-liter Tedlar
bags with on-siteGC analyses. An outdoor air sample
(1 -hr) was collected outside of House B. The results
of this sample, as well as two outdoor air samples
collected during the March 2003 sample event, are
presented in Table 2. One of the VOCs associated with
sub-surface contamination, 1,1,1 -TCA, was detected
at 0.58 ppbv during the July 2002 sample event.
Only sub-slab air samples using one-liter Tedlar bags
with subsequent on-site GC analysis were collected
during the October 2002 sample event. Mean sub-
slab air concentrations for 1,1,1 -TCA, 1,1 -DCE, TCE,
and c-1,2-DCE collected in one-liter Tedlar bags were
compared for the July 2002 and October 2002 sample
events using a two-tailed Approximate t-Test. The null
hypothesis was that the mean concentration of a VOC
during the July 2002 sample event was equal to the
mean concentration of the VOC during the October
2002 sample event. The alternate hypothesis was
that the means were not equal. The rejection criteria
was a Type I error or level of significance less than
or equal to 0.1 (twice the level of significance for
one-tailed tests used to assess vapor intrusion). A
Type I error is committed when the null hypothesis is
rejected when it is true.
Table 2. Outdoor Air Concentrations of VOCs During July 2002
and March 2003 Sample Events
VOC
1,1,1-TCA
1,1 -DCE
TCE
c-1 ,2-DCE
1,1 -DCA
1 ,2-DCA
PCE
CHA
CHCI3
CCI4
CCI3F(F-11)
CCI2F2(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1,2,4-TMB
1 ,3,5-TMB
1 ,3-butadiene
1 ,3-DCB
1,4-DCB
4-ethyltoluene
isopropyl alcohol
ethyl/vinyl acetate
CS2
Outdoor-
1hr
07/16/02
House B
(ppbv)
0.58
ND(0.25)
ND(0.25)
ND(0.25)
ND(0.25)
ND(0.25)
ND(0.25)
0.44
0.10
0.09
0.27
0.66
ND(2.2)
ND(0.25)
ND(0.25)
0.10
4.50
ND(0.25)
ND(2.3)
ND(0.46)
ND(0.21)
0.23
ND(0.24)
1.0
ND(0 50)
0.15
0.85
0.20
0.65
0.25
ND(0.23)
ND(0.24)
ND(0.25)
ND(0.50)
ND(0.24)
ND(0.24)
ND(0.25)
ND(0.25)
ND(0.48)
ND(0 23)
Outdoor-
24hr
03/24/03
House K
(ppbv)
ND(0.09)
ND(0.092)
ND(0.092)
ND(0.09)
ND(0.092)
ND(0.092)
0.12
0.19
ND(0.09)
0.080
0.25
0.5
ND(0.084)
ND(0.09)
ND(0.95)
ND(0.09)
2.0
ND(0.086)
ND(0.088)
0.45
0.081
0.46
0.22
0.69
ND(0.09)
0.33
0.63
0.09
0.23
0.1
ND(0.084)
ND(0.088)
ND(0.09)
ND(0.18)
ND(0.09)
ND(0.088)
0.09
0.31
ND(0.16)
ND(0.086)
Outdoor-
24hr
03/27/03
House G
(ppbv)
ND(0.11)
ND(0.11)
ND(0.11)
ND(0.11)
ND(0.11)
ND(0.11)
ND(0.10)
0.70
ND(0.11)
ND(0.11)
0.23
0.47
ND(0.10)
ND(0.11)
0.95
ND(0.11)
2.0
ND(0.10)
ND(0.10)
0.51
ND(0.096)
0.61
0.3
0.62
ND(0.11)
0.38
3.5
0.14
0.45
0.17
ND(0.10)
0.16
ND(0.11)
ND(0.21)
ND(0.11)
ND(0.10)
0.16
0.93
ND(0.19)
ND(0.10)
ND ( ) = Not detected above (reporting limits)
39
-------
The slabs atthese houses were located approximately
two meters below grade. All buildings tested were
in locations of known ground-water and soil-gas
contamination from the Raymark Superfund Site.
During the October 2002 sample event, basement
(48-hr activated charcoal) and sub-slab (scintillation
cells) air samples were collected for radon analysis.
However, since the results of radon testing in the
October 2002 sample event were not used to assess
vapor intrusion during the July 2002 sample event, the
results of sub-slab air radon testing during the October
sample event were not included in this report.
House A
Atthe time of probe installation, nosignificant cracks or
holes were observed in the concrete slab or in painted
cinderblock walls. Concentrations of VOCs detected in
basement and/or sub-slab air using EPA Method TO-
15 are summarized in Table 3a. The only constituent
associated with sub-surface contamination detected
in basement air was 1,1,1-TCA at a concentration
of 0.20 ppbv. The detection limit for other VOCs
associated with sub-surface contamination was 0.24
ppbv. Other chlorinated VOCs detected in basement
air were methylene chloride, chloroform, and carbon
tetrachloride at concentrations of 0.78,0.14, and 0.13
ppbv, respectively. Freons, F-11, F-12, and F-113were
detected at 0.39, 0.71, and 0.12 ppbv, respectively.
Hydrocarbons, hexane, cyclohexane, benzene,
toluene, ethylbenzene, m/p-xylenes, o-xylene, styrene,
1,2,4-trimethylbenzene, 1,3,5-trimethylbenzene,
and 4-ethyltoluene were detected in basement air
at concentrations up to 1.8 ppbv. Acetone, methyl
isobutyl ketone, and methyl tertiary-butyl ether were
detected in basement air at concentrations of 8.3,
0.11, and 0.49 ppbv, respectively.
Three probes were installed for sub-slab sampling. All
three probes were sampled using EPA Method TO-15
and with one-liter Tedlar bags. As indicated in Table
3a, when sampling using EEPA Method TO-15,1,1,1-
TCA, 1,1-DCE, TCE, c-1,2-DCE, and 1,1-DCAwere
detected at maximum concentrations of 100, 62, 60,
21, and 13 ppbv, respectively. The only other VOCs
detected in sub-slab air using EPA Method TO-15 were
acetone and chloroform at maximum concentrations
of 3.8 and 1.3 ppbv, respectively. Detection limits of
other VOCs varied from 1.8 to 18 ppbv. As indicated
in Table 3c, when sampling with one-liter Tedlar
bags, 1,1,1 -TCA, 1,1 -DCE, TCE, and c-1,2-DCE were
detected at maximum concentrations in Probe A at
164, 75, 78, and 29 ppbv, respectively.
Basement/sub-slab air concentration ratios for
VOCs using EPA Method TO-15 are illustrated in
Figure 34. The basement/sub-slab air concentration
ratios for all five VOCs associated with sub-surface
contamination were lower than basement/sub-slab
air concentration ratios for other VOCs detected in
basement air. However, since indicator VOCs, 1,1-
DCE , c-1,2-DCE, and 1,1 -DCA, were not detected in
basement air at the time of sampling, their basement/
sub-slab air concentration ratios and associated
standard deviations could not be computed. All that
can be inferred from available data is that the actual
basement/sub-slab air concentration ratios of these
indicator VOCs were less than the values indicated.
The null hypothesis that the basement/sub-slab air
concentration ratio of 1,1,1-TCA was equivalent to
the basement/sub-slab air concentration ratios of
indicator VOCs, 1,1-DCE , c-1,2-DCE, and 1,1-DCA
could not be evaluated. Thus, there was insufficient
data to determine whether or not the presence of
1,1,1 -TCA in basement air was due to vapor intrusion
40
-------
Table 3a. Basement and Sub-Slab Air Concentrations for VOCs at House A Using EPA Method TO-15 During the July 2002
Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
c-1,2-DCE
1,1 -DCA
1 ,2-DCA
PCE
CHA
CHCI3
CCI4
CCI3F(F-11)
CCI2F2(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1,2,4-TMB
1,3,5-TMB
1 ,3-butadiene
1,3-DCB
1,4-DCB
4-ethyltoluene
isopropyl alcohol
ethyl/vinyl acetate
CS2
bsmt
1-hr
07/16/02
(ppbv)
0.20
ND(0.24)
ND(0.24)
ND(0.24)
ND(0.24)
ND(0.24)
ND(0.25)
0.78
0.14
0.13
0.39
0.71
ND(2.2)
ND(0.25)
ND(2 3)
0.12
8.30
ND(2 3)
ND(2.3)
ND(0.26)
0.11
0.49
ND(0.24)
1.2
0.19
0.25
1.8
0.26
0.89
0.27
0.12
0.29
0.11
ND(0.50)
ND(0.24)
ND(0.24)
0.19
ND(0 25)
ND(0.48)
ND(0.48)
scaled
stdev
cov=6%
(ppbv)
0.01
IND
IND
IND
IND
IND
IND
0.05
0.01
0.01
0.02
0.04
IND
IND
IND
0.01
0.50
IND
IND
IND
0.01
0.03
IND
0.07
0.01
0.02
0.11
0.02
0.05
0.02
001
0.02
0.01
IND
IND
IND
0.01
IND
IND
IND
P[A]
grab
07/16/02
(PPbv)
99
62
60
21
13
ND(2.1)
ND(2.1)
ND(2.1)
1.0
ND(2.1)
ND(2.1)
ND(2.1)
ND(2.1)
ND(2.1)
ND(2.1)
ND(2.1)
3.5
ND(18)
ND(18)
ND(1.8)
ND(1.8)
ND(2.0)
ND(18)
ND(2.2)
ND(4.3)
ND(2.1)
ND(2.1)
ND(2.1)
ND(4.3)
ND(2.1)
ND(2.0)
ND(2.1)
ND(2.1)
ND(2.1)
ND(2.1)
ND(2.1)
ND(2.1)
ND(2.2)
ND(2.1)
ND(2.1)
ND( )=Not detected (reporting limit)
P[B]
grab
07/16/02
(ppbv)
100
56
55
18
13
ND(2.1)
ND(2.1)
ND(1.9)
1.3
ND(1.9)
ND(1.9)
ND(1 9)
ND(2.1)
ND(2.1)
ND(2.1)
ND(1.9)
3.8
ND(18)
ND(18)
ND(1.7)
ND(1.7)
ND(1.8)
ND(18)
ND(2.0)
ND(3.9)
ND(1 .9)
ND(1.9)
ND(1.9)
ND(3.9)
ND(1.9)
ND(1 8)
ND(1.9)
ND(1.9)
ND(1.9)
ND(1.9)
ND(1.9)
ND(1.9)
ND(2.0)
ND(1.9)
ND(1.9)
P[C]
grab
07/16/02
(PPbv)
43
20
28
6.0
4.0
ND(2.1)
ND(2.1)
ND(1.8)
0.73
ND(1.8)
ND(1.8)
ND(1 8)
ND(2.1)
ND(2.1)
ND(2.1)
ND(1.8)
ND(18)
ND(18)
ND(18)
ND(1.6)
ND(1.6)
ND(1.7)
ND(18)
ND(1.9)
ND(3.7)
ND(1.8)
ND(1.8)
ND(1.8)
ND(3.7)
ND(1.8)
ND(1.7)
ND(1.8)
ND(1.8)
ND(1.8)
ND(1.8)
ND(1.8)
ND(1.8)
ND(1 .9)
ND(1 8)
ND(1.8)
sub-slab
mean
07/16/02
(PPbv)
81
46
48
15
10
ND(<2.1)
ND(<2.1)
ND(<1 9)
1.0
ND(<1.9)
ND(<1.9)
ND(<1.9)
ND(<2.1)
ND(<2.1)
ND(<2.1)
ND(<1.9)
3.7
ND(<18)
ND(<18)
ND(<1.7)
ND(<1.7)
ND(<2.0)
ND(<18)
ND(<2.0)
ND(<4.0)
ND(<1.9)
ND(<1.9)
ND(<1.9)
ND(<4.0)
ND(<1.9)
ND(<1.7)
ND(<1.9)
ND(<1.9)
ND(<1.9)
ND(<1.9)
ND(<1.9)
ND(<1.9)
ND(<2.0)
ND(<1.9)
ND(<1 9)
sub-slab
stdev
07/16/02
(PPbv)
29
19
14
6.2
4.6
IND
IND
IND
0.29
IND
IND
IND
IND
IND
IND
IND
0.21
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
sub-slab
cov
07/16/02
(%)
36
40
29
42
46
IND
IND
IND
28
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
bsmt/
sub-slab
ratio
(-)
2.5E-03
< 4.3E-03
< 4.2E-03
< 1 .3E-02
< 2.0E-02
IND
IND
>4.1E-01
1 .4E-01
> 6.8E-02
>2.1E-01
> 3.7E-01
IND
IND
IND
> 6.3E-02
2.3E+00
IND
IND
IND
> 6.5E-02
> 2.5E-01
IND
> 6.0E-01
> 4.8E-02
> 1 .3E-01
> 9.5E-01
> 1.4E-01
> 2.2E-01
> 1.4E-01
>7.1E-02
> 1 .5E-01
> 5.8E-02
IND
IND
IND
IND
IND
IND
IND
bsmt/
sub-slab
stdev
(-)
9.0E-04
IND
IND
IND
IND
IND
IND
IND
4.0E-02
IND
IND
IND
IND
IND
IND
IND
1 .9E-01
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND = indeterminate
41
-------
1,3,5-TMB -
1,2,4-TMB -
styrene -
o-xylene -
m/p-xylenes -
ethylbenzene -
toluene
benzene -
cyclohexane -
hexane -
MTBE -
MIBK
acetone -
CCI3CF3(F-113) -
CCI2F2(F-12) -
CCI3F(F-11) -
CCI4 -
CHCI3 -
CH2CI2 -
1,1-DCA -
c-1,2-DCE -
TCE -
1,1-DCE -
1,1,1-TCA -
10'3
10'2 10-1 10°
Basement/Sub-Slab Concentration Ratios
Figure 34. Basement/sub-slab concentration ratios using EPA Method TO-15 at House A during the July 2002 sample event. Error
bars represent one standard deviation. Arrows indicate greater than or less than values due to non-detection in basement or sub-
slab air.
at the time of sampling. However, it was evident that
significant levels of VOCs associated with sub-surface
contamination were present in sub-slab air at the time
of sampling.
A statistical analysis of VOCs associated with sub-
surface contamination and sampled using EPA Method
TO-15issummarizedinTable3b. Since the basemen V
sub-slab air concentration ratios of indicator VOCs, 1,1-
DCE, c-1,2-DCE, and 1,1 -DCA were less than 4.3E-03,
1.3E-02, and 2.0E-02, respectively, the basement/sub-
slab air concentration ratio of VOCs associated with
sub-surface contamination was less than 4.3E-03. If
the presence of 1,1,1-TCA in basement air was not
due to vapor intrusion at the time of sampling, then
the basement/sub-slab air concentration ratio of VOCs
associated with sub-surface contamination was less
than the basement/sub-slab air concentration ratio
of 1,1,1-TCA or 2.5E-03. Coefficients of variation in
sub-slab air concentration ranged from 29 to 46%.
A statistical analysis of VOCs associated with sub-
surface contamination and sampled using one-liter
Tedlar bags is summarized in Table 3c. Since the
basement/sub-slab air concentration ratio of the
indicator VOC, 1,1 -DCE, was less than 4.0E-03, the
basement/sub-slab air concentration ratio of VOCs
associated with sub-surface contamination was
less than 4.0E-03. If the presence of 1,1,1-TCA in
basement air was not due to vapor intrusion at the
time of sampling, then the basement/sub-slab air
concentration ratio of VOCs associated with sub-
surface contamination was less than the basement/
sub-slab air concentration ratio of 1,1,1-TCA or
42
-------
1.6E-03forthe July 2002 sampling event. Coefficients
of variation in sub-slab air concentration ranged from
24 to 31%.
The results of sub-slab sampling using one-liter Tedlar
bags during the October 2002 sample are summarized
in Table 3d. A comparison of mean sub-slab air
concentrations of 1,1,1-TCA, 1,1-DCE,TCE,andc-1,2-
DCE during the July 2002 and October 2002 sample
events is illustrated in Figure 35. The null hypothesis
that the mean sub-slab air concentrations of 1,1,1-
TCA were equivalent during the July and October
2002 sample events was rejected using a two-tailed
Approximate t-Test at a level of significance less than
orequaltoO.1 (p = 0.08). The null hypotheses that the
mean sub-slab concentrations of 1,1 -DCE, TOE, and
c-1,2-DCEwereequivalentduringthe July and October
2002 sample events were not rejected at significance
levels of 0.48, 0.63, and 0.22, respectively.
Table 3b. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House A Using EPA Method TO-15 During the July 2002 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
c-1,2-DCE
1,1 -DCA
bsmt
1-hr
07/16/02
(ppbv)
0.20
ND(0.24)
ND(0.24)
ND(0.24)
ND(0.24)
scaled
stdev
cov=6%
(ppbv)
0.01
IND
IND
IND
IND
P[A]
grab
07/16/02
(ppbv)
99
62
60
21
13
P[B]
grab
07/16/02
(ppbv)
100
56
55
18
13
P[C]
grab
07/16/02
(ppbv)
43
20
28
6.0
40
sub-slab
mean
07/16/02
(ppbv)
81
46
48
15
10
sub-slab
stdev
07/16/02
(ppbv)
29
19
14
62
4.6
sub-slab
cov
07/16/02
(%)
36
40
29
42
46
mean and standard deviation of basement/sub-slab ratio
bsmt/
sub-slab
ratio
(-)
2.5E-03
<4.3E-03
<4.2E-03
< 1 .3E-02
<2.0E-02
<4.3E-03
bsmt/
sub-slab
stdev
(-)
9.0E-04
IND
IND
IND
IND
IND
Table 3c. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House A Using 1-Liter Tedlar Bags and On-Site GC Analysis During the July 2002 Sample Event
VOC
1,1,1-TCA
1,1-DCE
TCE
C-1.2-DCE
bsmt-1 hr
#1582
07/16/02
(ppbv)
0.20
ND(0.24)
ND(0.24)
ND(0.24)
scaled
stdev
cov=6%
(ppbv)
0.01
IND
IND
IND
P[A]
grab
07/16/02
(ppbv)
164
75
78
29
P[B]
grab
07/16/02
(ppbv)
144
65
58
20
P[C]
grab
07/16/02
(ppbv)
69
38
33
ND(25)
sub-slab
mean
07/16/02
(ppbv)
126
59
56
25
sub-slab
stdev
07/16/02
(ppbv)
39
14
sub-slab
cov
07/16/02
(%)
31
24
14 25
6 26
mean and standard deviation of basement/sub-slab ratio
bsmt/
sub-slab
ratio
(-)
1 .6E-03
<4.0E-03
<4.3E-03
IND
<4.0E-03
bsmt/
sub-slab
stdev
(-)
5.0E-04
IND
IND
IND
IND
Table 3d. Sub-Slab Air Concentrations of VOCs Associated with Sub-Surface Contamination in House A Using 1-Liter Tedlar Bags
and On-Site GC Analysis During the October 2002 Sample Event
VOC
1,1,1-TCA
1,1-DCE
TCE
C-1.2-DCE
P[A]
grab
10/30/02
(ppbv)
402
170
137
48
P[B]
grab
10/30/02
(ppbv)
256
102
75
28
P[C]
grab
10/30/02
(ppbv)
197
72
47
15
sub-slab
mean
n=3
10/30/02
(ppbv)
285
115
86
30
sub-slab
stdev
n=3
10/30/02
(ppbv)
106
50
46
17
sub-slab
cov
n=3
10/30/02
(%)
37
44
53
55
43
-------
500
TCE
C-1.2-DCE
1,1,1-TCA 1,1-DCE
Figure 35. Comparison of mean sub-slab air concentrations of VOCs collected in one-liter Tedlar bags during the July and October
2002 sample events at House A. Error bars represent one standard deviation.
House B
At the time of probe installation, no significant cracks
or holes were observed in the visible portion of the
concrete slab or in cinderblock walls. Most of the
basement was finished with carpeting and paneled
walls. Concentrations of all VOCs detected in
basement and/or sub-slab air using EPA Method TO-
15 are summarized in Table 4a. A replicate sample
was collected for basement air analysis. VOCs
associated with sub-surface contamination, 1,1,1-
TCA, 1,1 -DCE, and TCE, were detected in basement
air at concentrations of 0.41, 0.11, and 0.41 ppbv,
respectively. The detection limit for c-1,2-DCE and
1,1 -DCA was 0.25 and 0.26 pppv, respectively. Other
chlorinated VOCs detected in basement air were
perchloroethylene, methylene chloride, chloroform,
and carbon tetrachloride at concentrations of 0.53,
6.2, 0.91, and 0.14 ppbv, respectively. Freons, F-
11, F-12, and F-113, were detected in basement
air at concentrations of 0.71, 1.3, and 0.19 ppbv,
respectively. Hydrocarbons, heptane, hexane,
cyclohexane, benzene, toluene, ethylbenzene, m/
p-xylenes, o-xylene, 1,2,4-trimethylbenzene, 1,3,5-
trimethylbenzene, 4-ethyltoluene were detected
in basement air at concentrations up to 58 ppbv.
Acetone, methyl ethyl ketorie, methyl isobutyl ketone,
and methyl tertiary-butyl ether were detected in
basement air at concentrations of 40, 2.4, 0.22, and
9.8 ppbv, respectively. During basement sampling, the
homeowner stated that a latex paint had been used on
the second floor room two days prior to sampling.
44
-------
Table 4a. Basement and Sub-Slab Air Concentrations for VOCs at House B Using EPA Method TO-15 During the July 2002
Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
C-1.2-DCE
1,1 -DCA
1,2-DCA
PCE
CH2CI2
CHCI3
CCI4
CCI3F
(F-11)
CCI2F2
(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3
(F-11 3)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1,2,4-TMB
1,3,5-TMB
1,3-
butadiene
1,3-DCB
1 ,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
1-hr
07/16/02
(ppbv)
040
0.10
0.41
ND(0 25)
ND(0 26)
ND(0 25)
0.53
62
0.91
014
071
1 3
ND(2 2)
ND(0.25)
086
0.10
38
ND(0.25)
ND(2.3)
24
0.22
98
3.4
3.8
1.5
1 1
17
17
58
20
ND(0 25)
1 5
0.56
ND(0.25)
ND(0 25)
ND(0 25)
2.2
ND(0.25)
12
014
bsmt
1-hr
07/16/02
(ppbv)
041
011
040
ND(0.25)
ND(0 26)
ND(0 25)
0.52
6.2
0.83
0 14
071
1 3
ND(2 2)
ND(0.25)
086
019
40
ND(0 25)
ND(2.3)
2.2
0 15
95
3.4
4.1
1 5
1 0
17
16
56
19
ND(0 25)
1.5
0.54
ND(0 25)
ND(0.25)
ND(0 25)
2.0
ND(0 25)
11
013
bsmt
mean
n=2
07/16/02
(ppbv)
041
011
0.41
ND(<0.25)
ND(<026)
ND(<025)
0.53
62
0.87
014
071
1 3
ND(<22)
ND(<0.25)
086
0 14
39
ND(<025)
ND(<2.3)
2.3
019
97
3.4
40
1 5
1 1
17
17
57
20
ND(<025)
1 5
055
ND(<0.25)
ND(<0.25)
ND(<025)
21
ND(<0.25)
12
0.14
ND( )=Not detected (reporting limit)
bsmt
stdev
n=2
37/16/02
(ppbv)
001
001
0.01
IND
IND
IND
0.01
000
0.06
000
000
000
IND
IND
000
006
1 4
IND
IND
0 14
005
0.21
0.00
021
000
0.07
000
071
1 4
071
IND
000
001
IND
IND
IND
0.14
IND
0.71
0.01
bsmt
cov
n=2
J7/16/02
(ppbv)
1 7
67
1.7
IND
IND
IND
1.3
00
6.5
00
00
00
IND
IND
00
445
36
IND
IND
61
26.8
2.2
0.0
54
00
6.7
0.0
43
25
36
IND
00
26
IND
IND
IND
6.7
IND
6.1
5.2
P[A]
grab
07/16/02
(ppbv)
79
37
20
43
1.6
ND(0.23)
0.37
36
43
0 12
0.75
34
0.13
ND(0 23)
ND(0.24)
0.10
1.1
ND(0 23)
ND(18)
ND(0.43)
ND(0 20)
1 8
ND(0.22)
044
0.53
026
1 6
0.51
1 2
055
ND(0.22)
ND(0.22)
ND(0.23)
ND(0 23)
ND(0.23)
ND(0 23)
ND(0.23)
ND(0 23)
ND(45)
ND(0.21)
IND = indeterminate
P[B]
grab
07/16/02
(ppbv)
120
44
80
16
11
ND(1 9)
0.79
ND(1 9)
7.2
ND(1 9)
ND(1 9)
11
ND(17)
ND(1 9)
ND(2.0)
ND(1 9)
ND(18)
ND(1 9)
ND(18)
ND(3.5)
ND(1 6)
ND(1.8)
ND(1 9)
ND(1 9)
ND(3 8)
ND(1 9)
ND(1 9)
ND(1 9)
ND(3 8)
ND(1 9)
ND(1 9)
ND(1.9)
ND(1 9)
ND(1 9)
ND(1 9)
ND(1.9)
ND(1 9)
ND(1 9)
ND(3 7)
ND(1.8)
P[C]
07/16/02
(ppbv)
81
094
30
1 3
1.2
ND(0 42)
0.30
0.15
32
ND(0.42)
0.50
570
0.20
ND(0 42)
ND(0 44)
ND(0.42)
1.2
ND(0.42)
ND(18)
ND(0.79)
ND(0 36)
0 16
ND(0 42)
ND(0.43)
ND(0.86)
ND(0 42)
ND(0.42)
ND(0.42)
0.20 L
ND(0.42)
ND(0.42)
ND(0.42)
ND(0 42)
ND(0 42)
ND(0 42)
ND(0 42)
ND(0.43)
ND(0.42)
ND(0 83)
ND(0.39)
sub-slab
mean
n=3
07/16/02
(ppbv)
45
16
43
72
4.6
ND(<0 85)
0.49
1.9
14.5
<081
0.63
205
0.17
ND(<0.85)
ND(<0.9)
<081
1.2
ND(<0.85)
ND(<18)
ND(<1 6)
ND(<0 72)
1 0
ND(<0 85)
<0.92
<1 7
<086
<1.3
<094
<1 7
<096
ND(<0 85)
ND(<0.85)
ND(<0 85)
ND(<0.85)
ND(<0 85)
ND(<0 85)
ND(<0.84)
ND(<0.85)
ND(<17)
ND(<0.84)
sub-slab
stdev
n=3
07/16/02
(ppbv)
65
24
32
7.8
5.5
IND
0.27
2.4
15
IND
0.18
316
0.05
IND
IND
IND
007
sub-slab
cov
n=3
07/16/02
(%)
143
149
74
108
121
IND
54
130
105
IND
28
154
30
IND
IND
IND
61
IND | IND
IND
IND
IND
1.2
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
118
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
bsmt/
sub-slab
ratio
(-)
8.9E-03
6.5E-03
9 3E-03
<3.5E-02
<5.7E-02
IND
1.1E+00
3.3E+00
6.0E-02
> 1 7E-01
1.1E+00
6.3E-03
< 1 .3E+01
IND
>96E-01
>2.3E-01
34E+01
IND
IND
> 1 .4E+00
>31E-01
9.8E+00
>4.0E+00
>45E+00
>88E-01
> 1 3E+00
> 1 3E+01
> 1 8E+01
>34E+01
>2 1E+01
IND
> 1 8E-01
>66E-01
IND
IND
IND
>2.6E+00
IND
>7.1E-01
> 1 7E-01
bsmt/
sub-
slab
stdev
(-)
1 .3E-02
9.6E-03
6 9E-03
IND
IND
IND
5.9E-01
4.3E+00
6.3E-02
IND
3.2E-01
9.8E-03
IND
IND
IND
IND
2 4E+00
IND
IND
IND
IND
1.2E+01
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
mean and stdev calculated from 2 or more measurements
45
-------
Three probes were installed for sub-slab sampling. All
three probes were sampled using EPA Method TO-15
and one-liter Tedlar bags. As indicated in Table 4b,
when sampling using EPA Method TO-15,1,1,1 -TCA,
1,1 -DCE, TCE, c-1,2-DCE, and 1,1 -DCA were detected
at maximum concentrations in Probe B at 120,44,80,
16, and 11 ppbv, respectively. Other chlorinated VOCs
detected in sub-slab air using EPA Method TO-15 were
perchloroethylene, methylene chloride, chloroform,
and carbon tetrachloride at maximum concentrations
of 0.79,3.6,32, and 0.12 ppbv, respectively. Freons,
F-11, F-12, and F-113, were detected in sub-slab air
at maximum concentrations of 0.75, 570, and 0.10
ppbv, respectively. The high concentration of F-12
at Probe [C] may have been associated with a leak
from the central air conditioning system located in
the basement. Hydrocarbons, hexane, cyclohexane,
benzene, toluene, ethylbenzene, m/p-xylenes, and
o-xylene were detected in sub-slab air at maximum
concentrations up to 1.6 ppbv. Acetone and methyl
tertiary-butyl ether were detected in sub-slab air
at maximum concentrations of 1.2 and 1.8 ppbv,
respectively. As indicated in Table 4c, when sampling
with one-liter Tedlar bags, 1,1,1-TCA, 1,1-DCE,
TCE, and c-1,2-DCE were detected at maximum
concentrations in Probe EJ at 137, 48, 75, and 25
ppbv, respectively.
Basement/sub-slab air concentration ratios for VOCs
using EPA Method TO-15 are illustrated in Figure
36. With the exception of F-12 and chloroform,
basement/sub-slab air concentration ratios for all five
VOCs associated with sub-surface contamination
were lower than other VOCs detected in basement
air. The standard deviations of basement/sub-slab air
concentration ratios of F-12 and chloroform exceeded
Table 4b. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House B Using EPA Method TO-15 During the July 2002 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
c-1 ,2-DCE
1,1 -DCA
bsmt
1-hr
07/16/02
(ppbv)
0.40
0.10
041
ND(0.25)
ND(0 26)
bsmt
1-hr
07/16/02
(ppbv)
041
0.11
040
ND(0 25)
ND(0.26)
bsmt
mean
n=2
07/16/02
(ppbv)
041
0 11
041
ND(<0 25)
ND(<0.26)
bsmt
stdev
n=2
07/16/02
(ppbv)
001
001
0.01
IND
IND
bsmt
cov
n=2
07/16/02
(ppbv)
1 7
67
1.7
IND
IND
P[A]
grab
07/16/02
(ppbv)
79
37
20
43
1 6
P[B]
grab
07/16/02
(ppbv)
120
44
80
16
11
P[C]
grab
07/16/02
(ppbv)
81
0.94
30
1 3
1 2
sub-slab
mean
n=3
07/16/02
(ppbv)
45
16
43
7.2
4.6
sub-slab
stdev
n=3
07/16/02
(ppbv)
65
24
32
7.8
5.5
sub-slab
cov
n=3
07/16/02
(%)
143
149
74
108
121
mean and standard deviation of basement/sub-slab ratio
bsmt/
sub-slab
ratio
(-)
8.9E-03
6.5E-03
9 3E-03
<3.5E-02
<5.7E-02
8.3E-03
bsmt/
sub-slab
stdev
(-)
1 3E-02
9 6E-03
6 9E-03
IND
IND
5.8E-03
Table 4c. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House B Using 1-Liter Tedlar Bags and On-Site GC Analysis During the July 2002 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
o-1,2-DCE
bsmt
1-hr
07/16/02
(ppbv)
0.40
0.10
041
ND(0.25)
bsmt
1-hr
07/16/02
(ppbv)
041
0.11
040
ND(0.25)
bsmt
mean
n=2
07/16/02
(ppbv)
0.41
0.11
041
ND(<0.25)
bsmt
stdev
n=2
07/16/02
(ppbv)
0.01
0.01
001
IND
bsmt
cov
n=2
07/16/02
(ppbv)
1.7
67
1 7
IND
P[A]
grab
07/16/02
(ppbv)
10
ND(10)
20
ND(25)
P[B]
grab
07/16/02
(ppbv)
137
48
75
25
P[C]
grab
07/16/02
(ppbv)
14
ND(10)
27
ND(25)
sub-slab
mean
n=3
07/16/02
(ppbv)
54
<23
41
<25
sub-slab
stdev
n=3
07/16/02
(ppbv)
72
IND
30
IND
sub-slab
cov
n=3
07/16/02
(%)
135
IND
74
IND
mean and standard deviation of basement/S'jb-slab ratio
bsmt/
sub-slab
ratio
(-)
7 5E-03
>4.8E-03
1 OE-02
IND
8 8E-03
bsmt/
sub-slab
stdev
(-)
1 OE-02
IND
7 3E-03
IND
6 3E-03
46
-------
ethylvinylacetate -
4-ethyltoluene -
1,3,5-TMB -
1,2,4-TMB -
o-xylene -
m/p-xylenes -
ethylbenzene -
toluene -
benzene -
cyclohexane -
hexane -
heptane -
MTBE -
MIBK -
MEK -
acetone -
CCI3CF3(F-113) -
CH3CH2CI -
CHBrCI2 -
CCI2F2(F-12) -
CCI3F(F-11) -
CCI4 -
CHCI3 -
CH2CI2 -
PCE -
1,1-DCA -
c-1,2-DCE -
TCE -
1,1-DCE -
1,1,1-TCA -
" ' : ; i i i ! i
.-: i-i, ;uii - '• .
f 1 fT i Mi " ;
i U Li ! Li* ; -
i • i ; - 1 1 it 1
".I ! litlii i
"' r; I'iiiJ " i
. : M ' 1 1 1
Li).::i!j -
: r ' • i : 1 1
T i r i . > rl '
!. 1 . -t I 1|
- , ., , pi,! - ,
1 ', - •' ' , ' i
• ! H^HH — f '-
} * r • ' * ; i ~ *
-! ^ , 'f4;; '
j
L ' ? ' ! L J ( '
i '.- ; . -<-' *s-
.j ..<£..
i i • A i
I • I
!
i
"'
1
!
1
j
i
i — j_
iil!
-! 1 H
• ' ' i
I [ , ,
i r, .
,
i i • '
1 | !
: j i i
i i !;
, '. :
i 'i!
rt ;
,41,
!:]r
Tri^^~
•«hM
L j '
\ ' ' '
1 j
I ' ^i i
- 1 j- j f-
•f ' *~~L I
! i' 1 ; t
r ; ] ' i
i t i i.
i i * •
1 1 1
t i -1 - *-
. . / i-i
' f \ I
iA. I t-fc
+ | 4 j
• J 'M-
-f i i -i '
n T
' '• t i ..
I-L.
i i
; " t ,! .r ";
1 1 f
4!!
I M
In
"M
u
!!
« S
^±i
-T'
14
il
if
! I
"4
ii
!]
i i j
'ti
! I
.^3
•^
" -->
-
A
^
-
M
. ,'
• I
-"
u 1-
- ^j__
^
^s*.^
__:.iu3.
^__^
- !
^__>
-^
r
+-
-J
1 t ' M
1 i *'!
t~hrn
' ! i «
- , i ,+
rui
1 i ' ! !
i , l:j
A! i t ( )
•i.ni
P , f ; r
' * a J
j ! n '
1 1 , •
-f ' J !
i !M
i 1 1 1
M ' !
' 1 l
j . ;
j_ L...
j
i '
t i
f *_
•' i -
A ;
t * t
1 '
!— -,
[^
1
;
^ j
i
^
i r-
4 i
j
(
0.1), it was inferred that detection
of 1,1,1 -TCA, 1,1 -DCE, and TCE in basement air was
due to vapor intrusion at the time of sampling. 1,1,1-
TCA was detected in outside air at 0.58 ppbv during
the July 2002 sampling event. Unlike other VOCs
such as benzene, toluene, ethylbenzene, and xylenes,
1,1,1 -TCA was not detected in outside air during the
March 2003 sampling event nor in previous outdoor
sampling activities conducted by EPA's New England
Laboratory.
Results and statistical analysis of VOCs associated
with sub-surface contamination and sampled using
EPA Method TO-15 are summarized in Table 4b. Use
of basement/sub-slab air concentration ratios of 1,1,1 -
TCA, 1,1-DCE, and TCE resulted in computation of
an average basement/sub-slab air ratio of 8.3E-03 for
VOCs associated with vapor intrusion. Coefficients
of variation in sub-slab air samples ranged from 74
to 149%. Results and statistical analysis of VOCs
associated with sub-surface contamination sampled
using one-liter Tedlar bags are summarized in Table
4c. Use of basement/sub-slab air concentration ratios
of 1,1,1 -TCA and TCE resulted in computation of an
average basement/sub-slab ratio of 8.8E-03 for VOCs
associated with vapor intrusion. The coefficients of
variation for sub-slab air sampling ranged from 74 to
135%.
47
-------
The results of sub-slab sampling using one-liter
Tedlar bags during the October 2002 sampling
event are presented in Table 4d. A comparison of
mean sub-slab air concentrations for 1,1,1 -TCA and
TCE for the July 2002 and October 2002 sampling
events using one-liter Tedlar bags is illustrated in
Figure 37. The null hypotheses that the mean sub-
slab air concentrations of 1,1,1-TCA and TCE were
equivalent during the July and October 2002 sample
events were not rejected using atwo-tailed Approximate
t-Test at a level of significance of 0.1 or less.
Table 4d. Sub-Slab Air Concentrations of VOCs Associated with Sub-Surface Contamination in House B Using 1 -Liter Tedlar Bags
and On-Site GC Analysis During the October 2002 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
C-1.2-DCE
P[A]
grab
10/30/02
(ppbv)
101
39
50
4.6
P[B]
grab
10/30/02
(ppbv)
169
82
80
12
P[C]
grab
10/30/02
(ppbv)
42
18
29
ND(4.1)
sub-slab
mean
n=3
10/30/02
(ppbv)
104
46
53
8
sub-slab
stdev
n=3
10/30/02
(ppbv)
64
33
26
5.2
sub-slab
cov
n=3
10/30/02
(%)
61
70
48
63
180
160 -
> 140 H
a.
a.
^ 120 H
c
o
'73 100 H
c
CD
o
c
o
O
80
60
O
a.
to 40
20 -
0
mean 7/02
mean 10/02
1,1,1-TCA TCE
Figure 37. Comparison of mean sub-slab air concentrations of VOCs collected in one-liter Tedlar bags during the July and October
2002 sample events at House B. Error bars represent one standard deviation.
48
-------
House C
During the home survey, a strong petroleum odor was
noticed inside the basement. It was determined that
a lawnmower was leaking gasoline from inside a shed
attached to the house. The lawnmower was removed
from the shed. The shed door was kept open for
three days prior to installation of sub-slab probes and
basement airsampling. Atthe time of probe installation,
no significant cracks or holes were observed in the
concrete slab or in unpainted cinderblock walls.
Concentrations of VOCs detected in basement and/or
sub-slab air using EPA Method TO-15 are summarized
in Table 5a. All five VOCs associated with sub-surface
contamination were detected in basement air. 1,1,1-
TCA, 1,1 -DCE, TCE, c-1,2-DCE, and 1,1 -DCA were
detected at concentrations of 3.8, 2.3,1.5, 0.57, and
0.52 ppbv, respectively. Other chlorinated VOCs
detected in basement air were perchloroethylene,
methylene chloride, chloroform, carbon tetrachloride,
1,3-dichlorobenzene, and 1,4-dichlorobenzene at
concentrations of 0.17,3.9,0.34,0.23,0.10, and 0.15
ppbv, respectively. Freons, F-11, F-12, and F-113,
were detected at concentrations of 0.82, 1.1, 0.22
ppbv, respectively. Hydrocarbons, heptane, hexane,
cyclohexane, benzene, toluene, ethylbenzene, m/p-
xylenes, o-xylene, styrene, 1,2,4-trimethylbenzene,
1,3,5-trimethylbenzene, 4-ethyltoluene were detected
in basement air at concentrations up to 5.3 ppbv.
Acetone, methyl ethyl ketone, and methyl tertiary-butyl
ether were detected in basement air at concentrations
of 5.3,1.0, and 8.5 ppbv, respectively. The compound,
1,3-butadiene, was detected at a concentration of 0.35
ppbv.
Four probes were installed for sub-slab sampling.
All four probes were sampled using EPA Method
TO-15 and in one-liter Tedlar bags. As indicated in
Table 5b, when sampling using EPA Method TO-15,
1,1,1 -TCA, 1,1 -DCE, TCE, c-1,2-DCE, and 1,1 -DCA
were detected at maximum concentrations in Probe
A at 590, 410, 280, 120, and 94 ppbv, respectively.
Total VOCs in probes exceeded 1000 ppbv. The only
other VOCs detected in sub-slab air were acetone,
toluene, and m/p-xylenes at 28, 4.2, and 7.0 ppbv,
respectively. Detection limits for other VOCs ranged
from 13 to 200 ppbv. As indicated in Table 5c, when
sampling with one-liter Tedlar bags, 1,1,1-TCA, 1,1-
DCE, TCE, and c-1,2-DCE were detected at maximum
concentrations in Probe [A] at 833,486,260, and 120
ppbv, respectively.
Basement/sub-slab air concentration ratios of VOCs
detected in basement air and sampled using EPA
Method TO-15 are illustrated in Figure 38. With the
exception of the bromodichloromethane, which had
a basement/sub-slab air concentration ratio at some
value greater than 9.2E-04, basement/sub-slab air
concentration ratios for all five VOCs associated
with sub-surface contamination were significantly
lower than other VOCs detected in basement air.
Bromodichloromethane, a trihalomethane, was not
present in ground water or soil gas in the vicinity of
the house and thus was removed from consideration
of vapor intrusion. Since the null hypotheses that
basement/sub-slab air concentration ratios of 1,1,1-
TCAandTCEwereequivalenttoindicatorVOCs, 1,1-
DCE, c-1,2-DCE, and 1,1 -DCA, could not be rejected
using a one-tailed Approximate t-Test at a level of
significance less than or equal to 0.05 (p > 0.1), it was
inferred that detection of 1,1,1-TCA, 1,1-DCE, TCE,
c-1,2-DCE, and 1,1-DCA in basement air was due to
vapor intrusion at the time of sampling.
49
-------
Table 5a. Basement and Sub-Slab Air Concentrations for VOCs at House C Using EPA Method TO-15 During the July 2002
Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
C-1.2-DCE
1,1 -DCA
1 ,2-DCA
PCE
CH2CI2
CHCI3
CCI4
Ppl p/p H H\
CCI2F2
(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3
(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1 ,2,4-TMB
1 ,3,5-TMB
1,3-
Dutadiene
1,3-DCB
1,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
1-hr
07/16/02
(ppbv)
38
23
1.5
057
052
ND(0.25)
0.17
39
0.34
023
0.82
1 1
0 14
ND(0 25)
ND(0.25)
022
5.3
ND(0 25)
ND(2 3)
1 0
ND(0 22)
85
0.97
2.4
0.57
0.80
5.3
1 0
3.7
1 5
0.29
24
0.77
0.35
010
0.15
1.70
ND(0 25)
1.5
018
scaled
stdev
cov=6%
(ppbv)
0.23
0 14
0.09
003
003
IND
001
023
002
001
0.05
007
0.01
IND
IND
0.01
032
IND
IND
0.06
IND
051
0.06
0.14
003
005
0.32
006
022
009
0.02
0 14
005
002
0.01
0.01
0.10
IND
0.09
001
P[A]
grab
07/16/02
(ppbv)
590
410
280
120
94
ND(21)
ND(21)
ND(21)
ND(21)
ND(21)
ND(21)
ND(21)
ND(190)
ND(21)
ND(21)
ND(21)
ND(200)
ND(21)
ND(18)
ND(39)
ND(18)
ND(20)
ND(21)
ND(22)
ND(43)
ND(21)
42
ND(21)
ND(42)
ND(21)
ND(20)
ND(21)
ND(21)
ND(43)
ND(21)
ND(20)
ND(22)
ND(22)
ND(42)
ND(20)
ND( ) = Not detected above reporting limits
P[B]
grab
07/16/02
(PPbv)
420
290
200
84
66
ND(19)
ND(20)
ND(20)
ND(19)
ND(20)
ND(20)
ND(19)
ND(180)
ND(19)
ND(19)
ND(20)
28
ND(19)
ND(18)
ND(36)
ND(17)
ND(18)
ND(19)
ND(20)
ND(40)
ND(20)
ND(20)
ND(20)
70
ND(20)
ND(19)
ND(19)
ND(20)
ND(40)
ND(19)
ND(19)
ND(20)
ND(20)
ND(39)
ND(18)
P[C]
grab
07/16/02
(PPbv)
510
290
200
55
57
ND(13)
ND(13)
ND(13)
ND(13)
ND(13)
ND(13)
ND(13)
ND(110)
ND(13)
ND(13)
ND(13)
14
ND(13)
ND(18)
ND(24)
ND(11)
ND(12)
ND(13)
ND(13)
ND(26)
ND(13)
ND(13)
ND(13)
ND(26)
ND(13)
ND(12)
ND(13)
ND(13)
ND(26)
ND(13)
ND(12)
ND(13)
ND(13)
ND(25)
ND(12)
P[D]
grab
07/16/02
(PPbv)
460
300
180
64
61
ND(15)
ND(15)
ND(15)
ND(15)
ND(15)
ND(15)
ND(15)
ND(130)
ND(15)
ND(15)
ND(15)
20
ND(15)
ND(18)
ND(28)
ND(13)
ND(14)
ND(15)
ND(15)
ND(90)
ND(15)
ND(15)
ND(15)
ND(30)
ND(15)
ND(14)
ND(15)
ND(15)
ND(30)
ND(15)
ND(14)
ND(15)
ND(15)
ND(29)
ND(14)
IND = indeterminate
sub-slab
mean
n=4
07/16/02
(PPbv)
495
323
215
81
70
ND(<17)
ND(<17)
ND(<17)
ND(<17)
ND(<17)
ND(<17)
ND(<17)
ND(<152)
ND(<17)
ND(<17)
ND(<17)
21
ND(<17)
ND(18)
ND(<32)
ND(<15)
ND(<16)
ND(<17)
ND(<18)
ND(<50)
ND(<17)
<13
ND(<17)
<21
ND(<17)
ND(<16)
ND(<17)
ND(<17)
ND(<35)
ND(<17)
ND(<16)
ND(<18)
ND(<18)
ND(< 34)
ND(<16)
sub-slab
stdev
n=4
07/16/02
(PPbv)
73.3
58.5
443
28.8
17
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
70
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
ND
ND
sub-slab
cov
11=4
07/16/02
l'%)
148
181
206
36
?A
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
34
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
ND
ND
bsmt/
sub-slab
ratio
(-)
7 7E-03
7.1E-03
7.0E-03
7 1 E-03
7.5E-03
IND
>1.0E-02
>2.0E-02
> 1 4E-02
>4.8E-02
>65E-02
IND
IND
> 1 3E-02
2 5E-01
IND
IND
>3.1E-02
IND
>53E-01
>4.6E-02
>1.3E-01
>1.1E-02
>4.7E-02
>4.1E-01
>5.9E-02
> 1 .8E-02
>88E-02
> 1 .8E-02
> 1 4E-01
>4.5E-02
> 1 .OE-02
>59E-03
>9.4E-03
>9.4E-02
IND
>4.4E-02
>1.1E-02
bsmt/
sub-slab
stdev
(-)
1 2E-03
1 4E-03
1 .5E-03
2.6E-03
1.9E-03
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
8 9E-02
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
ND
ND
ND
ND
ND
mean and stdev calculated from 2 or more measurements
50
-------
Table 5b. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House C Using EPA Method TO-15 During the July 2002 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
c-1,2-DCE
1,1 -DCA
bsmt
1-hr
07/16/02
(ppbv)
3.8
2.3
1.5
057
0.52
scaled
stdev
cov=6%
(ppbv)
0.23
0.14
0.09
0.03
0.03
P[A]
grab
07/16/02
(ppbv)
590
410
280
120
94
P[B]
grab
07/16/02
(ppbv)
420
290
200
84
66
P[C]
grab
07/16/02
(ppbv)
510
290
200
55
57
P[D]
grab
07/16/02
(ppbv)
460
300
180
64
61
sub-slab
mean
n=4
07/16/02
(ppbv)
495
323
215
81
70
sub-slab
stdev
n=4
07/16/02
(ppbv)
73.3
58.5
44.3
28.8
17
sub-slab
cov
n=4
07/16/02
(%)
14.8
18.1
20.6
36
24
mean and standard deviation of basement/sub-slab ratio
bsmt/
sub-slab
ratio
(-)
7.7E-03
7.1E-03
7.0E-03
7.1E-03
7.5E-03
7.3E-03
bsmt/
sub-slab
stdev
(-)
1 .2E-03
1 .4E-03
1 .5E-03
2.6E-03
1 .9E-03
7.9E-04
Table 5c. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House C Using 1-Liter Tedlar Bags and On-Site GC Analysis During the July 2002 Sample Event
VOC
1,1,1-TCA
1,1-DCE
TCE
0-1,2-DCE
bsmt
1-hr
07/16/02
(ppbv)
3.8
2.3
1.5
057
scaled
stdev
cov=6%
(ppbv)
0.23
014
0.09
0.03
P[A]
grab
07/16/02
(ppbv)
833
486
260
120
P[B]
grab
07/16/02
(ppbv)
650
374
201
98
P[C]
grab
07/16/02
(ppbv)
757
423
249
61
P[D]
grab
07/16/02
(ppbv)
722
416
195
74
sub-slab
mean
n=4
07/16/02
(ppbv)
741
425
226
88
sub-slab
stdev
n=4
07/16/02
(ppbv)
761
46.2
33.0
26
sub-slab
cov
n=4
07/16/02
(%)
10.3
10.9
14.6
30
mean and standard deviation of basement/sub-slab ratio
bsmt/
sub-slab
ratio
(-)
5.1E-03
5.4E-03
6.6E-03
6.5E-03
5.9E-03
bsmV
sub-slab
stdev
(-)
6.1E-04
6.7E-04
1 .OE-03
2.0E-03
6.0E-04
CS2 -
ethylvinylacetate -
4-ethyltoluene -
1,4-DCB
1,3-DCB
1,3-butadiene
1,3,5-TMB
1,2,4-TMB
styrene
o-xylene
m/p-xylenes -
ethylbenzene -
toluene -
benzene -
cyclohexane -
hexane
heptane
MTBE
MEK
acetone
CCI3CF3(F-113)
CHBrCI2
CCI2F2(F-12) -
CC,3F(Fc11) 1
CHCI3
CH2CI2
PCE
1,1-DCA
0-1,2-DCE
TCE
1,1-DCE
1,1,1-TCA -
10-'
10°
10-3 -10-2 -10-1
Basement/Sub-Slab Concentration Ratio
Figure 38. Basement/sub-slab concentration ratios using EPA Method TO-15 at House C during the July 2002 sample event.
Error bars represent one standard deviation. Arrows indicate greater than or less than values due to non-detection in basement or
sub-slab air.
51
-------
Results and statistical analysis of VOCs associated
with sub-surface contamination and sampled using
EPA Method TO-15 are summarized in Table 5b.
Use of basement/sub-slab air concentration ratios of
1,1,1 -TCA, 1,1 -DCE, TCE, c-1,2-DCE, and 1,1 -DCA
resulted in computation of an average basement/
sub-slab ratio of 7.3E-03. Coefficients of variation of
sub-slab air concentrations varied from 15 to 36%.
Results and statistical analysis of VOCs associated
with sub-surface contamination and sampled using
one-liter Tedlar bags are summarized in Table 5c.
Use of basement/sub-slab air concentration ratios of
1,1,1-TCA, 1,1-DCE, TCE, and c-1,2-DCE resulted
in computation of an average basement/sub-slab
air concentration ratio of 5.9E-03. Coefficients of
variation of sub-slab air concentrations varied from
10 to 30%.
The results of sub-slab sampling with one-liter Tedlar
bags during the October 2002 sample event are
summarized in Table 5d. A comparison of mean
sub-slab air concentrations for 1,1,1-TCA , 1,1-DCE,
TCE, and c-1,2-DCE for the July 2002 and October
2002 sampling events using one-liter Tedlar bags is
illustrated in Figure 39. The null hypotheses that the
mean sub-slab air concentrations of 1,1,1 -TCA , 1,1-
DCE, TCE, and c-1,2-DCE were equivalent during the
July and October 2002 sample events was not rejected
using a two-tailed Approximate t-Test at significance
less than or equal to 0.1.
Table 5d. Sub-Slab Air Concentrations of VOCs Associated with Sub-Surface Contamination in House C Using 1 -Liter Tedlar Bags
and On-Site GC Analysis During the October 2002 Sample Event
voc
1,1,1 -TCA
1,1-DCE
TCE
C-1.2-DCE
P[A]
grab
10/01/02
(ppbv)
941
618
469
167
P[B]
grab
10/01/02
(ppbv)
630
387
276
91
P[C]
grab
10/01/02
(ppbv)
623
323
200
34
P[D]
10/01/02
(ppbv)
505
288
188
57
sub-slab
mean
n=4
10/01/02
(ppbv)
675
404
283
87
sub-slab
stdev
n=4
10/01/02
(ppbv)
187
148
130
58
sub-slab
cov
n=4
10/01/02
(%)
27.6
36.7
45.8
67
52
-------
1000
mean 7/02
mean 10/02
1,1,1-TCA
1,1-DCE
TCE
C-1.2-DCE
Figure 39. Comparison of mean sub-slab air concentrations of VOCs collected in one-liter Tedlar bags during the July and October
2002 sample events at House C. Error bars represent one standard deviation.
House D
At the time of probe installation, significant cracks
and holes were observed in the concrete slab. At
some locations, the slab resembled a veneer of
plaster-like material. Concentrations of VOCs
detected in basement and/or sub-slab air using EPA
Method TO-15 are summarized in Table 6a. 1,1,1-
TCA and 1,1-DCE were detected in basement air at
concentrations of 0.48 and 0.13 ppbv, respectively.
The detection limit for TCE, c-1,2-DCE, and 1,1 -DCA
was 0.28 ppbv. Other chlorinated VOCs detected in
basement air were methylene chloride, chloroform,
and carbon tetrachloride at concentrations of 7.1,
0.11, and 0.11 ppbv, respectively. Freons, F-11.F-12,
and F-113 were detected at concentrations of 0.30,
0.61,0.10 ppbv, respectively. Hydrocarbons, hexane,
cyclohexane, benzene, toluene, ethylbenzene,m/p-
xylenes, o-xylene, styrene, 1,2,4-trimethylbenzene,
4-ethyltoluene were detected in basement air at
concentrations uptol .4 ppbv. Acetone, tetahydrofuran,
methyl ethyl ketone, and methyl tertiary-butyl ether
were detected at concentrations of 6.9, 3.7, 6.2, and
0.57 ppbv, respectively.
Three probes were installed for sub-slab sampling. All
three probes were sampled with Tedlar bags. Only
two probes were sampled using EPA Method TO-15
because one probe, Probe P[B], became loose during
sampling with a Tedlar bag. As indicated in Table
6b, when sampling using EPA Method TO-15,1,1,1-
TCA, 1,1-DCE, TCE, c-1,2-DCE, and 1,1-DCA were
detected at maximum concentrations at 110, 110,
28, 6.1, 16 ppbv, respectively. The only other VOCs
detected in sub-slab air were acetone and chloroform at
4.5 and 1.4 ppbv, respectively. Detection limits for other
53
-------
Table 6a. Basement and Sub-Slab Air Concentrations for VOCs at House D Using EPA Method TO-15 During the July 2002
Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
C-1.2-DCE
1,1 -DCA
1,2-DCA
PCE
CH,CI2
CHCI3
CCI4
CCI3F
(F-11)
CCI2F2
(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3
(F-11 3)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1 ,2,4-TMB
1 ,3,5-TMB
1,3-
butadiene
1,3-DCB
1 ,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
1-hr
07/16/02
(ppbv)
048
013
ND(0.28)
ND(0.27)
ND(0 28)
ND(0 28)
ND(0 25)
7.1
011
0.11
0.30
061
ND(2.2)
ND(0.25)
ND(0 25)
010
69
ND(0 25)
37
6.2
ND(0.21)
057
ND(0.24)
1.3
0.21
0.28
1 4
039
1 3
035
011
0.71
ND(0.25)
ND(0.50)
ND(0.24)
ND(0 24)
0.41
ND(0 25)
ND(0.48)
ND(0.23)
scaled
stdev
cov=6%
(ppbv)
0.03
0.01
IND
IND
IND
IND
IND
043
0.01
001
002
0.04
IND
IND
IND
001
0.41
IND
0.22
0.37
IND
0.03
IND
0.08
0.01
002
0.08
002
0.08
002
001
004
IND
IND
IND
IND
002
IND
IND
IND
P[A]
grab
07/16/02
(ppbv)
52
22
16
46
12
ND(1 9)
ND(1.9)
ND(1 9)
1.4
ND(1.9)
ND(1 9)
ND(1 9)
ND(1 9)
ND(1.9)
ND(1 9)
ND(1 9)
2.7
ND(1.9)
ND(18)
ND(3.5)
ND(1.9)
ND(1 8)
ND(1.8)
ND(1 9)
ND(3.8)
ND(1.9)
ND(1 9)
ND(1.9)
ND(3 8)
ND(1 9)
ND(1 8)
ND(1.9)
ND(1 9)
ND(1 9)
ND(1.9)
ND(1.9)
ND(1.9)
ND(1 9)
ND(1.9)
ND(1 9)
ND( )=Not detected (reporting limit)
P[B]
grab
07/16/02
(ppbv)
110
110
28
61
16
ND(1.9)
ND(1.9)
ND(1.9)
098
ND(1.9)
ND(1 9)
ND(1.9)
ND(1 .9)
ND(1 9)
ND(1 .9)
ND(1.9)
45
ND(1 9)
ND(18)
ND(3 6)
ND(1 9)
ND(1 8)
ND(1 8)
ND(2 0)
ND(3 9)
ND(1 9)
ND(1 9)
ND(1 9)
ND(3 9)
ND(1.9)
ND(1 8)
ND(1 9)
ND(1 9)
ND(1.9)
ND(1 9)
ND(1 9)
ND(2.0)
ND(1 9)
ND(1 9)
sub-slab
mean
n=2
07/16/02
(ppbv)
81
66
22
5.4
14
ND(<1 9)
ND(<1 9)
ND(<1.9)
1.2
ND(<1 9)
ND(<1.9)
NO(<1 9)
ND(<1 9)
ND(<1.9)
ND(<1 9)
ND(<1 9)
36
ND(<1 9)
ND(<18)
ND(<3 6)
ND(<1 9)
ND(<1.8)
ND(<1 8)
ND(<2.0)
ND(<3 9)
ND(<1.9)
ND(<1 9)
ND(<1 9)
ND(<3 9)
ND(<1.9)
ND(<1 8)
ND(<1.9)
ND(<1 9)
ND(<1.9)
ND(<1 9)
ND(<1 9)
ND(<20)
ND(<1.9)
ND(<1.9)
ND(1.9) ND(<1 9)
sub-slab
stdev
n=2
07/16/02
(ppbv)
41
62
85
1 1
28
IND
IND
IND
030
IND
IND
IND
IND
IND
IND
IND
1 3
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
sub-slab
cov
n=2
07/16/02
(%)
51
94
39
20
20
IND
IND
IND
25
IND
IND
IND
IND
IND
IND
IND
35
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
bsmt/
sub-slab
ratio
(-)
5.9E-03
2 OE-03
< 1 3E-02
<50E-02
<20E-02
IND
IND
>37E+00
9 2E-02
>58E-02
> 1 6E-01
>32E-01
IND
IND
IND
>53E-02
1 9E+00
IND
>21E-01
>1 7E+00
IND
>32E-01
IND
>6.5E-01
>5.4E-02
>1 5E-01
>7.4E-01
>21E-01
>3.3E-01
>1.8E-01
>61E-02
>3.7E-01
IND
IND
IND
IND
IND
> 2 1 E-01
IND
IND
bsmt/
sub-slab
stdev
(-)
3 OE-03
1 9E-03
IND
IND
IND
IND
IND
IND
2 4E-02
IND
IND
IND
IND
IND
IND
IND
6 9E-01
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND = indeterminate
54
-------
compounds ranged from 1.8 to 18 ppbv. As indicated
in Table 6c, when sampling with one-liter Tedlar bags,
maximum concentrations of 1,1,1 -TCA, 1,1 -DCE, and
TCE were 168, 111, and 30 ppbv, respectively. The
detection limit for c-1,2-DCE was 25 ppbv.
Basement/sub-slab air concentration ratios of VOCs
detected in basement air and sampled using EPA
Method TO-15 are illustrated in Figure 40. Since
the null hypothesis that the basement/sub-slab air
concentration ratio of 1,1,1 -TCA was equivalent to the
indicator VOC, 1,1 -DCE, could not be rejected using
a one-tailed Approximate t-Test at a significance level
less than or equal to 0.05 (p > 0.1), it was inferred that
detection of 1,1,1 -TCA and 1,1 -DCE in basement air
was due to vapor intrusion at the time of sampling.
Results and statistical analysis of VOCs associated
with sub-surface contamination and sampled using
EPA Method TO-15 are summarized in Table 6b.
Use of basement/sub-slab air concentration ratios
of 1,1,1-TCA and 1,1-DCE resulted in computation
of an average basement/sub-slab air concentration
ratio of 3.9E-03. Coefficients of variation of sub-slab
air concentrations ranged from 20 to 94%. Results
and statistical analysis of VOCs associated with
sub-surface contamination and sampled using one-
liter Tedlar bags are summarized in Table 6c. Use
of basement/sub-slab air concentration values of
1,1,1-TCA and 1,1-DCE resulted in computation of
an average basement/sub-slab air concentration ratio
of 3.7E-03. Coefficients of variation of sub-slab air
concentrations ranged from 35 to 95%.
Table 6b. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House D Using EPA Method TO-15 During the July 2002 Sample Event
VOC
1,1,1 -TCA
1,1-DCE
TCE
C-1.2-DCE
1,1 -DCA
bsmt
1-hr
07/16/02
(ppbv)
0.48
0.13
ND(0.28)
ND(0.27)
ND(0.28)
scaled
stdev
cov=6%
(ppbv)
0.03
0.01
IND
IND
IND
P[A]
grab
07/16/02
(ppbv)
52
22
16
4.6
12
P[C]
grab
07/16/02
(ppbv)
110
110
28
6.1
16
sub-slab
mean
n=2
07/16/02
(ppbv)
81
66
22
5.4
14
sub-slab
stdev
n=2
07/16/02
(ppbv)
41
62
8.5
1 1
2.8
sub-slab
cov
n=2
07/16/02
(%)
51
94
39
20
20
mean and standard deviation of basement/sub-slab ratio
bsmV
sub-slab
ratio
(-)
5.9E-03
2.0E-03
<1.3E-02
<5.0E-02
<2.0E-02
3.9E-03
bsmt/
sub-slab
stdev
(-)
3.0E-03
1.9E-03
IND
IND
IND
1 .8E-03
Table 6c. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House D Using 1-Liter Tedlar Bags and On-Site GC Analysis During the July 2002 Sample Event
VOC
1,1,1 -TCA
1,1-DCE
TCE
c-1 ,2-DCE
bsmt
1-hr
07/16/02
(ppbv)
0.48
0.13
ND(0.28)
ND(0.27)
scaled
stdev
cov=6%
(ppbv)
0.03
0.01
IND
IND
P[A]
grab
07/16/02
(ppbv)
77
22
18
ND(25)
P[B]
grab
07/16/02
(ppbv)
16
ND(10)
ND(6)
ND(25)
P[C]
grab
07/16/02
(ppbv)
168
111
30
ND(25)
sub-slab
mean
n=3
07/16/02
(ppbv)
87
67
24
<25
sub-slab
stdev
n=3
07/16/02
(ppbv)
76
63
8
IND
sub-slab
cov
n=3
07/16/02
(%)
87.9
95
35.4
IND
mean and standard deviation of basement/sub-slab ratio
bsmt/
sub-slab
Tedlar
(-)
5.5E-03
2.0E-03
<1.1E-02
IND
3.7E-03
bsmt/
sub-slab
stdev
(-)
4.9E-03
1 .9E-03
IND
IND
2.6E-03
55
-------
The results of sub-slab sampling with one-liter Tedlar
bags during the October 2002 sample event are
summarized in Table 6d. A comparison of mean sub-
slab sample concentrations for 1,1,1-TCA, 1,1 -DCE,
and TCE during the July and October sampling events
is illustrated in Figure 41. The null hypotheses that
the mean sub-slab air concentrations of 1,1,1 -TCA ,
1,1-DCE, and TCE were equivalent during the July
and October 2002 sample events were not rejected
using a two-tailed Approximate t-Test at a level of
significance less than or equal to 0.1.
isopropylalcohol -
1,2,4-TMB -
styrene -
o-xylene -
m/p-xylenes -
ethylbenzene -
toluene -
benzene -
cyclohexane -
hexane -
MTBE -
MEK -
THF -
acetone -
CCI3CF3(F-113) -
CCI2F2(F-12) -
CCI3F(F-11) -
CCI4 -
CHCI3 -
CH2CI2 -
1,1-DCA -
c-1,2-DCE -
TCE -
1,1-DCE -
1,1,1-TCA -
10-5
., 0-4
io-3 io-2
Basement/Sub-Slab Concentration Ratios
Figure 40. Basement/sub-slab concentration ratios using EPA Method TO-15 at House D during the July 2002 sample event.
Error bars represent one standard deviation. Arrows indicate greater than or less than values due to non-detection in basement or
sub-slab air.
Table 6d. Sub-Slab Air Concentrations of VOCs Associated with Sub-Surface Contamination in House D Using 1-Liter Tedlar Bags
and On-Site GC Analysis During the October 2002 Sample Event
voc
1,1,1-TCA
1,1-DCE
TCE
c-1 ,2-DCE
P[A]
grab
11/01/02
(ppbv)
39
8.2
7.9
ND(4)
P[C]
grab
11/01/02
(ppbv)
61
52
23
11
sub-slab
mean
n=2
11/01/02
(ppbv)
50
30
15
11
ND( ) = Not detected (reporting limit)
sub-slab
stdev
n=2
11/01/02
(ppbv)
16
31
11
IND
sub-slab
cov
n=2
11/01/02
(%)
31
103
69
IND
IND = indeterminate
56
-------
200
0 J
1,1,1-TCA
1,1-DCE
TCE
Figure 41. Comparison of mean sub-slab air concentrations of VOCs collected in one-liter Tedlar bags during the July and October
2002 sample events at House D. Error bars represent one standard deviation.
House E
At the time of probe installation, there was a two
centimeterseparation between theslab and cinderblock
walls where underlying sandy soil was exposed. A
portion (approximately 30%) of the basement was
finished with ceramic tile. Poured concrete walls
were painted. An oil furnace was centrally located in
the basement. Concentrations of VOCs detected in
basement and/or sub-slab air using EPA Method TO-15
are summarized in Table 7a. 1,1,1 -TCAand 1,1 -DCE
were detected in basement air at concentrations of
0.57 and 0.12 ppbv, respectively. The detection limits
forTCE, c-1,2-DCE, and 1,1 -DCA were between 0.26
and 1.1 ppbv. Other chlorinated VOCs detected in
basement air were methylene chloride, chloroform,
and carbon tetrachloride at concentrations of 9.5,
0.81, and 0.10 ppbv, respectively. Freons, F-11, F-12,
and F-113, were detected at concentrations of 0.44,
0.59, and 0.09 ppbv, respectively. Hydrocarbons,
hexane, benzene, toluene, ethylbenzene, m/p-xylenes,
o-xylene, 1,2,4-trimethylbenzene were detected in
basement airat concentrations upto 1.2 ppbv. Acetone,
methyl ethyl ketone, and methyl tertiary-butyl ether
were detected at concentrations of 9.6,1.0, and 0.27
ppbv, respectively.
Three probes were installed for sub-slab sampling. All
three probes were sampled using EPA Method TO-
15 and one-liter Tedlar bags. As indicated in Table
7b, when sampling using EPA Method TO-15,1,1,1-
TCA, 1,1 -DCE, TCE, c-1,2-DCE, and 1,1 -DCA were
57
-------
Table 7a. Basement and Sub-Slab Air Concentrations for VOCs at House E Using EPA Method TO-15 During the July 2002
Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
c-1 ,2-DCE
1,1 -DCA
1 ,2-DCA
PCE
CHA
CHCI3
CCI4
CCI3F(F-11)
CCI/2
(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3
(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1,2,4-TMB
1 ,3,5-TMB
1,3-
butadiene
1 ,3-DCB
1 ,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
1-hr
07/16/02
(ppbv)
057
012
ND(0 26)
ND(0 26)
ND(1.1)
ND(0 25)
ND(0 25)
9.5
081
0.10
044
0.59
0.15
ND(0 25)
ND(0 25)
0.09
scaled
stdev
cov=6%
(ppbv)
0.03
0.01
IND
IND
IND
IND
IND
057
005
001
0.03
004
0.01
IND
IND
0.01
P[A]
grab
07/16/02
(ppbv)
48
13
20
2.5
70
ND(1 1)
ND(1 1)
ND(1 1)
0.73
ND(1 1)
047
0.84
ND(9 8)
ND(1 1)
ND(1.1)
ND(1 1)
P[B]
grab
07/16/02
(ppbv)
78
59
32
12
15
ND(1 8)
ND(1 8)
ND(1 8)
1 5
ND(1 8)
ND(1.8)
ND(1.8)
ND(16)
ND(1.8)
ND(1 8)
ND(1 8)
96 0.58 2.6 5.0
ND(0 25)
ND(2.3)
1 0
ND(0 25)
027
ND(0 25)
1.1
ND(0 50)
0.22
1.2
0.14
037
0.17
ND(0.23)
019
ND(0 26)
ND(0 50)
ND(0 24)
ND(0 24)
ND(0.25)
ND(0 25)
1.0
ND(0 23)
IND
IND
0.06
IND
0.02
IND
0.07
IND
0.01
007
001
002
001
IND
0.01
IND
IND
IND
IND
IND
IND
006
IND
ND(1 1)
ND(18)
ND(2.0)
ND(1 1)
ND(1.0)
ND(1 1)
ND(1.1)
ND(1.1)
ND(1 1)
ND(1 1)
ND(1 1)
ND(2 2)
ND(1 1)
ND(1 1)
ND(1 1)
ND(1.1)
ND(1.1)
ND(1.1)
ND(1 1)
ND(1 1)
ND(1.1)
ND(2.2)
ND(1.1)
ND( ) = Not detected (reporting limit)
ND(1.8)
ND(18)
ND(3 3)
ND(1 8)
ND(1 7)
ND(1.8)
ND(1 8)
ND(1.8)
ND(1.8)
ND(1 8)
ND(1 8)
ND(3.6)
ND(1 8)
ND(1.8)
ND(1 8)
ND(1 8)
ND(1.8)
ND(1 8)
ND(1 8)
ND(1.8)
ND(1 8)
ND(3.5)
ND(1 8)
P[C]
grab
07/16/02
(ppbv)
170
170
66
26
32
ND(3 7)
ND(3 7)
ND(3 7)
1 4
ND(3 7)
ND(3.7)
ND(3.7)
ND(33)
ND(3.7)
ND(3 7)
ND(3 7)
57
ND(3 7)
ND(18)
ND(7 0)
ND(3.7)
ND(3 5)
ND(3.7)
ND(3 8)
ND(3 7)
ND(3 7)
ND(3 7)
ND(3.7)
ND(7 5)
ND(3 7)
ND(3.7)
ND(3.7)
ND(3.7)
ND(3 7)
ND(3.7)
ND(3.7)
ND(3 7)
ND(3 7)
ND(7 3)
ND(3.7)
sub-slab
mean
n=3
07/16/02
(ppbv)
99
81
39
14
18
ND(<22)
ND(<22)
ND(<2 2)
1 2
ND(<2 2)
<20
<21
ND(<20)
ND(<2.2)
ND(<22)
ND(<2 2)
44
ND(<2 2)
ND(18)
ND(<4.1)
ND(<2.2)
ND(<2.1)
ND(<2.2)
ND(<2 2)
ND(<2 2)
ND(<2.2)
ND(<2 2)
ND(<2.2)
ND(<44)
ND(<2.2)
ND(<22)
ND(<2 2)
ND(<2.2)
ND(<2 2)
ND(<2.2)
ND(<2.2)
ND(<2.2)
ND(<2.2)
ND(<4.3)
ND(<2 2)
sub-slab I sub-slab
stdev cov
n=3 n=3
07/16/02
(PPbv)
64
81
24
12
13
IND
IND
IND
042
IND
IND
IND
IND
IND
IND
IND
1 6
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
07/16/02
bsmt/
sub-slab
ratio
(%> ! (->
64
100
61
88
71
IND
IND
IND
35
IND
IND
IND
IND
IND
IND
IND
37
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
5 8E-03
1.5E-03
< 6 7E-03
< 1 9E-02
<612E-02
IND
IND
>43E+00
367
>4.5E-02
>22E-01
>2.8E-01
>73E-03
IND
IND
..4.12E-02
2.2E+00
IND
IND
:»24E-01
IND
:> 1 3E-01
IND
>50E-01
IND
> 1 OE-01
>5.5E-01
>64E-02
>8.4E-02
> 7.7E-02
IND
> 8.6E-02
HMD
IND
IND
IND
IND
IND
>23E-01
IND
bsmt/
sub-slab
stdev
(-)
3 7E-03
1.5E-03
IND
IND
IND
IND
IND
IND
2 4E-01
IND
IND
IND
IND
IND
IND
IND
8.2E-01
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND = indeterminate
58
-------
detected at maximum concentrations in Probe C at
170,170,66,26, and 32 ppbv, respectively. The only
other VOCs detected in sub-slab air were acetone,
chloroform, F-11, and F-12 at maximum concentrations
of 5.7,1.5,0.47, and 0.84 ppbv, respectively. Detection
limits for other compounds ranged from 1.1 to 18 ppbv.
As indicated in Table 7c, when sampling with one-liter
Tedlar bags, 1,1,1 -TCA, 1,1 -DCE, TCE, and c-1,2-DCE
were detected at maximum concentrations at Probe
[C] at 234, 130, 65, and 26 pppv, respectively.
Basement/sub-slab air concentration ratios of
VOCs detected in basement air and sampled using
EPA Method TO-15 are illustrated in Figure 42.
Basement/sub-slab air concentration ratios for all
five VOCs associated with sub-surface contamination
were significantly lower than other VOCs detected in
basement air. The basement/sub-slab air concentration
ratio of bromodichloromethane was some value
greater than 7.3E-03. This VOC was eliminated from
consideration of vapor intrusion because it was not
detected in soil gas or ground water in the vicinity
of the building. Since the null hypothesis that the
basement/sub-slab air concentration ratio of 1,1,1 -TCA
was equal to the basement/sub-slab air concentration
ratio of the indicator VOC, 1,1-DCE, could not be
rejected using a one-tailed Approximate t-Test at a
level of significance less than or equal to 0.05 (p =
0.1), it was inferred that the presence of both 1,1,1-
TCA and 1,1 -DCE in basement air was due to vapor
intrusion at the time of sampling.
Table 7b. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House E Using EPA Method TO-15 During the July 2002 Sample Event
VOC
1,1,1 -TCA
1,1-DCE
TCE
c-1 ,2-DCE
1,1 -DCA
bsmt
1-hr
07/16/02
(ppbv)
057
0.12
ND(0.26)
ND(0.26)
ND(1.1)
scaled
stdev
cov=6%
(ppbv)
0.03
0.01
IND
IND
IND
P[A]
grab
07/16/02
(ppbv)
48
13
20
2.5
70
ND( ) = Not detected (reporting limit)
P[B]
grab
07/16/02
(ppbv)
78
59
32
12
15
P[C]
grab
07/16/02
(ppbv)
170
170
66
'26
32
IND = indeterminate
sub-slab
mean
n=3
07/16/02
(ppbv)
99
81
39
14
18
sub-slab
stdev
n=3
07/16/02
(ppbv)
64
81
24
12
13
sub-slab
cov
n=3
07/16/02
(%)
64
100
61
88
71
mean and standard deviation
bsmt/
sub-slab
ratio
(-)
5.8E-03
1.5E-03
<6.7E-03
< 1 .9E-02
<6.12E-02
3.6E-03
bsmV
sub-slab
stdev
(-)
3.7E-03
1 .5E-03
IND
IND
IND
2.0E-03
Table 7c. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House E Using 1-Liter Tedlar Bags and On-Site GC Analysis During the July 2002 Sample Event
VOC
1,1,1 -TCA
1,1-DCE
TCE
c-1,2-DCE
bsmt
1-hr
07/16/02
(ppbv)
0.57
0.12
ND(0.26)
ND(0 26)
scaled
stdev
cov=6%
(ppbv)
0.03
0.01
IND
IND
P[A]
grab
07/16/02
(ppbv)
59
10
20
ND(25)
ND( ) = Not detected (reporting limit)
P[B]
grab
07/16/02
(ppbv)
107
44
36
ND(25)
P[C]
grab
07/16/02
(Ppbv)
234
130
65
26.00
IND = indeterminate
sub-slab
mean
n=3
07/16/02
(ppbv)
133
61
40
<25
sub-slab
stdev
n=3
07/16/02
(ppbv)
90
62
23
IND
sub-slab
cov
n=3
07/16/02
(%)
68
101
57
IND
mean and standard deviation
bsmt/
sub-slab
ratio
(-)
4.3E-03
2.0E-03
<6.4E-03
IND
3.1E-03
bsmt/
sub-slab
stdev
(-)
2.9E-03
2.0E-03
IND
IND
1.8E-03
59
-------
ethylvinylacetate -
1,2,4-TMB -
o-xylene -
m/p-xylenes -
ethylbenzene -
toluene -
benzene -
hexane -
MTBE -
MEK -
acetone -
CCI3CF3(F-113) -
CHBrCI2 -
CCI2F2(F-12) -
CCI3F(F-11) -
CCI4 -
CHCI3 -
CH2CI2 -
1,1-DCA -
0-1,2-DCE -
TCE -
1,1-DCE -
1,1,1-TCA -
10-'
-10-3 10-2 1Q-1 10°
Basement/Sub-Slab Concentration Ratios
Figure 42. Basement/sub-slab concentration ratios using EPA Method TO-15 at House E during the July 2002 sample event. Error
bars represent one standard deviation. Arrows indicate greater than or less than values due to non-detection in basement or sub-
slab air.
Results and statistical analysis of VOCs associated
with sub-surface contamination and sampled using
EPA Method TO-15 are summarized in Table 7b.
Use of basement/sub-slab air concentration ratios
of 1,1,1-TCA and 1,1-DCE resulted in computation
of an average basement/sub-slab ratio of 3.6E-03.
Coefficients of variation in sub-slab airsamples ranged
from 61 to 100%. Results and statistical analysis of
VOCs associated with sub-surface contamination and
sampled using one-liter Tedlar bags are summarized in
Table 7c. Use of basement/sub-slab air concentration
values of 1,1,1-TCA and 1,1-DCE resulted in
computation of an average basement/sub-slab ratio
of 3.1 E-03. Coefficients of variation in sub-slab air
samples ranged from 57 to 101%.
The results of sub-slab sampling using one-liter Tedlar
bags during the October 2002 sample event are
summarized in Table 7d. A comparison of mean sub-
slab sample concentrations for 1,1,1 -TCA, 1,1 -DCE,
and TCE during the July and October 2002 sampling
events is illustrated in Figure 43. The null hypotheses
thatthe mean sub-slab air concentrations of 1,1,1 -TCA
, 1,1 -DCE, and TCE were equivalent during the July
and October 2002 sample events were not rejected
using a two-tailed Approximate t-Test at a level of
significance less than or equal to 0.1.
Hence, acomparison of July and October 2002 sample
events indicated that levels of VOCs associated with
sub-surface contamination found in sub-slab air were
statistically different (p < 0.10) in only 1 out of 16
comparisons. This indicated little temporal variability
in sub-slab air concentrations between the July 2002
and October 2002 sampling events.
60
-------
Table 7d. Sub-Slab Air Concentrations of VOCs Associated with Sub-Surface Contamination in House E Using 1 -Liter Tedlar Bags
and On-Site GC Analysis During the October 2002 Sample Event
voc
1,1,1-TCA
1,1-DCE
TCE
C-1.2-DCE
P[A]
grab
11/01/02
(ppbv)
30
10
13
6
P[B]
grab
11/01/02
(ppbv)
78
37
29
18
P[C]
grab
11/01/02
(ppbv)
370
198
152
84
sub-slab
mean
n=3
11/01/02
(ppbv)
159
82
65
36
sub-slab
stdev
n=3
11/01/02
(ppbv)
184
102
76
42
sub-slab
cov
n=3
11/01/02
(%)
115
124
118
117
400
300 -
.0
Q.
CL
O
£ 200 -
§
o
O
o
Q.
TO
mean (7/02)
mean (10/02)
100 -
1,1,1-TCA
1,1-DCE
TCE
Figure 43. Comparison of mean sub-slab air concentrations of VOCs collected in one-liter Tedlar bags during the July and October
2002 sample events at House E. Error bars represent one standard deviation.
61
-------
6.3 Summary of Results for Buildings
Sampled in March 2003
Basement and sub-slab air samples were collected for
VOC analysis in Houses F through P during the March
2003 sample event. Basement (24-hr) and sub-slab
(grab) samples were collected in six-liter evacuated
canisters. Sub-slab samples were also collected in
one-liter Tedlar bags with on-site GC analyses. Two
outdoor air samples (24-hr) were collected outside of
Houses G and K with results summarized in Table 2.
(See page 39.)
During the March 2003 sample event, basement
(48-hr activated charcoal) and sub-slab (scintillation
cells) air samples were collected for radon analysis.
When one or more indicator VOCs were present in
basement air and more than one probe was sampled
for radon, the basement/sub-slab air concentration
ratio of radon was compared with the basement/sub-
slab air concentration ratio of indicator VOCs using
a two-tailed Approximate t-Test. The null hypothesis
was that the basement/sub-slab air concentration
ratio of radon was equal to the basement/sub-slab
air concentration ratio of an indicator VOC. The
alternate hypothesis was that the basement/sub-slab
concentration ratios were not equal. The rejection
criteria was a Type I error or level of significance less
than or equal to 0.1 (twice the level of significance
for one-tailed tests used to assess vapor intrusion).
As a matter of necessity, radon was used as an
indicator compound to assess vapor intrusion when
indicator VOCs, 1,1-DCE, c-1,2-DCE, and 1,1-DCA,
were not detected in basement air. Basement slabs
were approximately 1.6 meters below ground surface.
House F
Concentrations of all VOCs detected in basement
and/or sub-slab air using EPA Method TO-15 are
summarized in Table 8. A replicate basement air
sample was collected at House F. VOCs associated
with sub-surface contamination were not detected
in basement air at detection limits ranging from
0.078 to 0.086 ppbv. Thus, assessment of vapor
intrusion was not necessary at this location. Other
chlorinated VOCs detected in basement air were
methylene chloride, carbon tetrachloride, and 1,1-
dichlorobenzene at concentrations of 0.87, 0.09,
and 0.09 ppbv, respectively. Freons, F-11, F-12, and
F-113, were detected at concentrations of 0.40, 1.4,
and 0.07 ppbv, respectively. Hydrocarbons, hexane,
benzene, toluene, ethylbenzene, m/p-xylenes, o-
xylene, styrene, and 1,2,4-trimethylbenzene were
detected in basement air at concentrations up to
1.25 ppbv. Acetone, methyl ethyl ketone, and methyl
tertiary-butyl ether were detected at concentrations of
2.8,0.55, and 0.32 ppbv, respectively. The compound,
1,3-butadiene, was detected at a concentration of
0.33 ppbv.
Four probes were installed for sub-slab sampling.
Two probes were sampled using EPA Method TO-
15. All four probes were sampled using one-liter
Tedlar bags. VOCs associated with sub-surface
contamination were not detected using EPA Method
TO-15. Detection limits ranged from 0.083 to 0.085
ppbv. Other chlorinated VOCs detected in sub-slab
air were perchloroethylene, methylene chloride,
chloroform, and carbon tetrachloride at maximum
concentrations of 0.13, 0.70, 0.09, and 0.09 ppbv,
respectively. Freons, F-11 and F-12, were detected
at maximum concentrations of 0.42 and 1.7 ppbv,
62
-------
Table 8. Basement and Sub-Slab Air Concentrations for VOCs at House F Using EPA Method TO-15 During the March 2003
Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
C-1.2-DCE
1,1 -DCA
1,2-DCA
PCE
CHA
CHCI3
CCI4
CCI3F(F-11)
CCI2F2
(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3
(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1 ,2,4-TMB
1 ,3,5-TMB
1,3-
butadiene
1,3-DCB
1,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
24-hr
03/26/03
(ppbv)
ND(0.084)
ND(0.086)
ND(0.086)
ND(0 084)
ND(0.086)
ND(0.084)
ND(0.084)
077
ND(0 084)
ND(0.086)
0.38
1 4
ND(0.079)
ND(0 086)
05
0.07
28
ND(0.081)
ND(0.082)
053
ND(0 076)
030
ND(0 082)
033
ND(0.084)
0.41
1 2
016
0.48
0.16
010
0080
ND(0 084)
0.31
ND(0.084)
009
0.11
29
032
ND(0.081)
bsmt
24-hr
03/26/03
(ppbv)
ND(0.082)
ND(0.084)
ND(0.084)
ND(0 082)
ND(0.084)
ND(0.082)
ND(0.082)
0.87
ND(0.082)
0.09
04
1.4
ND(0.077)
ND(0.085)
ND(0.87)
ND(0 082)
2.5
ND(0.079)
ND(0.08)
0.55
ND(0.074)
0.32
ND(0.08)
0.36
ND(0.082)
045
1 3
0.17
0.52
016
010
0.090
ND(0 082)
033
ND(0.082)
ND(0.08)
0 10
40
0.35
ND(0 079)
bsmt
mean
03/26/03
(ppbv)
ND(<0.083)
ND(<0.085)
ND(<0.085)
ND(<0 083)
ND(<0.085)
ND(<0.083)
ND(<0.083)
0.82
ND(<0 083)
<0088
0.39
1 4
ND(<0.078)
ND(<0 086)
<0.66
<0076
2.7
ND(<0 080)
ND(<0081)
0.54
ND(<0 075)
0.31
ND(<0.081)
0.35
ND(<0.083)
043
1 25
0.17
050
0 16
010
0.085
ND(<0.083)
032
ND(<0 083)
<0.09
011
3.5
034
ND(<0 080)
ND( ) = Not detected (reporting limit)
bsmt
stdev
n=2
03/26/03
(Ppbv)
IND
IND
IND
IND
IND
IND
IND
0.07
IND
IND
0.01
0.00
IND
IND
IND
IND
0.21
IND
IND
0.01
IND
001
IND
002
IND
0.03
0.07
0.01
003
0.00
000
001
IND
001
IND
IND
0.01
078
0.02
IND
P[A]
grab
03/28/03
(ppbv)
ND(0 078)
ND(0 08)
ND(0 08)
ND(0 078)
ND(0.08)
ND(0.078)
ND(0 078)
0.49
ND(0.078)
0.09
0.42
1.7
ND(0.073)
ND(0.081)
ND(0 83)
ND(0 078)
1 7
0.1
ND(0.076)
0.48
026
0.65
ND(0.076)
0.22
008
0.11
051
0.08
02
009
ND(0 073)
009
ND(015)
ND(0. 1 7)
ND(0 078)
ND(0 078)
ND(0 078)
025
ND(0.14)
0.21
P[C]
grab
03/28/03
(ppbv)
ND(0 082)
ND(0.084)
ND(0.084)
ND(0.082)
ND(0 084)
ND(0 082)
0.13
070
0.09
0.09
0.36
1 4
ND(0.077)
ND(0.085)
0.38
ND(0 082)
2.4
0.12
ND(0.08)
0.66
027
0.69
ND(0.11)
0.25
ND(0 082)
033
22
012
0.36
0.13
ND(0 077)
0.15
ND(0 082)
0.31
ND(0 082)
ND(0 08)
013
0.64
ND(0.15)
026
sub-slab
mean
n=2
03/28/03
(ppbv)
ND(<0 080)
ND(<0.082)
ND(<0 082)
ND(<0.080)
ND(<0.082)
ND(<0.080)
<0.10
0.60
< 0.084
0.09
039
1.6
ND(<0 075)
ND(<0 083)
<061
ND(<0.080)
2.1
011
ND(<0 078)
0.57
0265
067
ND(<0.093)
024
ND(<0081)
0.22
1.4
01
028
0.11
ND(<0075)
012
ND(<012)
<0.24
ND(<0 080)
ND(<0.079)
<010
0.445
ND(<0.15)
024
sub-slab
stdev
n=2
03/28/03
(ppbv)
IND
IND
IND
IND
IND
IND
IND
0 15
IND
000
0.04
0.21
IND
IND
IND
IND
0.49
0.01
IND
0.13
0.01
003
IND
002
IND
0.16
1.2
0.03
0.11
003
IND
0.04
IND
IND
IND
IND
IND
0.28
IND
004
sub-slab
cov
n=2
03/28/03
(%)
IND
IND
IND
IND
IND
IND
IND
25
IND
0.0
11
14
IND
IND
IND
IND
24
13
IND
22
27
4.2
IND
9.0
IND
71
88
28
40
26
IND
35
IND
IND
IND
IND
IND
62
IND
15
bsmt/
sub-slab
ratio
(-)
IND
IND
IND
IND
IND
IND
IND
1 4E+00
IND
<9.6E-01
1 .OE+00
9.0E-01
IND
IND
IND
IND
1 3E+00
<7.4E-01
IND
9 5E-01
<1.6E-01
4 6E-01
IND
1 5E+00
IND
2.0E+00
9 2E-01
1 7E+00
1 .8E+00
1 .5E+00
> 1 .3E+00
7 1 E-01
IND
> 1 OE+00
IND
IND
>8.5E-01
7.8E+00
>2.2E+00
<34E-01
bsmt/
sub-slab
stdev
(-)
IND
IND
IND
IND
IND
IND
IND
3.6E-01
IND
IND
1 1E-01
1.2E-01
IND
IND
IND
IND
3.3E-01
IND
IND
2.1 E-01
IND
2 9E-02
IND
1 6E-01
IND
1 .4E+00
8.2E-01
4 7E-01
7 3E-01
3.7E-01
IND
2 6E-01
IND
IND
IND
IND
IND
5 1 E+00
IND
IND
IND = indeterminate
63
-------
respectively. Hydrocarbons, hexane, cyclohexane,
benzene, toluene, ethylbenzene, m/p-xylenes, o-
xylene, 1,2,4-trimethylbenzene, and 4-ethyltoluene
were detected in sub-slab air at concentrations up to
2.2 ppbv. Acetone, methyl ethyl ketone, methyl isobutyl
ketone, and methyl tertiary-butyl ether were detected at
maximum concentrations of 2.4, 0.66,0.27, and 0.69
ppbv, respectively. The compound, 1,3-butadiene, was
detected at a maximum concentration of 0.33 ppbv.
VOCs associated with sub-surface contamination were
not detected in one-liter Tedlar bags with detection
limits ranging from 2 to 5 ppbv.
Basement/sub-slab ratios of VOCs detected in
basement air and sampled using EPA Method TO-15
are summarized in Table 8. Basement/sub-slab air
concentration ratios for VOCs not associated with sub-
surface contamination ranged from less than 1.6E-01
(methyl isobutyl ketone) to 7.8E+00 (isopropyl alcohol).
Basement/sub-slab air concentration ratios were less
than 1 .OE+00 for eight compounds not associated
with sub-surface contamination demonstrating that
observation of a basement/sub-slab air concentration
ratio less than 1 .OE+00 does not necessarily indicate
vapor intrusion.
House G
There were several visible cracks in the slab and two
small diameter holes near an oil tank which serviced
an oil furnace. The basement wall consisted of
field stone. Concentrations of all VOCs detected
in basement and/or sub-slab air using EPA Method
TO-15 are summarized in Table 9a. The only VOC
associated with sub-surface contamination detected
in basement air was 1,1,1-TCA at a concentration
of 0.28 ppbv. The detection limit of other VOCs
associated with sub-surface contamination was 0.10
ppbv. Other chlorinated VOCs detected in basement
air were perchloroethylene and methylene chloride
at concentrations of 0.18 and 7.4 ppbv, respectively.
Freons, F-11 and F-12, were detected at concentrations
of 0.30 and 0.49 ppbv, respectively. Hydrocarbons,
hexane, benzene, toluene, ethylbenzene, m/p-
xylenes, o-xylene, 1,2,4-trimethylbenzene, 1,3,5-
trimethylbenzene, and 4-ethyltoluene were detected
in basement air at concentrations up to 42 ppbv.
Acetone, methyl ethyl ketone, and methyl tertiary-butyl
ether were detected at concentrations of 2.0, 0.81,
and 0.54 ppbv, respectively.
Five probes were installed for sub-slab sampling. Two
probes were sampled using EPA Method TO-15. All five
probes were sampled using one-liter Tedlar bags. As
indicated by Tables 9a and 9b, when sampling using
EPA Method TO-15, 1,1,1-TCA, 1,1-DCE, TCE, and
1,1 -DCA were detected at maximum concentrations
of 6.0, 0.75, 0.99, and 0.37 ppbv, respectively. Other
chlorinated VOCs perchloroethylene, chloroform,
and carbon tetrachloride were detected at maximum
concentrations of 0.20,0.73 and 0.09 ppbv, respectively.
Freons, F-11, F-12, and F-113, were detected at
maximum concentrations of 0.27,0.55, and 0.08 ppbv,
respectively. Hydrocarbons, hexane and toluene were
detected at maximum concentrations of 0.41 and 0.25
ppbv, respectively. Acetone, 2-hexanone, methyl ethyl
ketone, methyl isobutyl ketone, and methyl tertiary-
butyl ether were detected at maximum concentrations
of 2.5,0.12, 0.80, 0.30, and 0.090 ppbv, respectively.
As indicated by Table 9c, 1,1,1-TCA and TCE were
detected in one-liter Tedlar bag samples at maximum
concentrations of 7.5 and 2.4 ppbv, respectively.
Detection limits for 1,1-DCE and c-1,2-DCE were 3
to 5 ppbv. Radon was sampled at two probes with
64
-------
Table 9a. Basement and Sub-Slab Air Concentrations for VOCs Detected at House G Using EPA Method TO-15 During the March
2003 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
C-1.2-DCE
1,1 -DCA
1 ,2-DCA
PCE
CH2CI2
CHCI3
CCI4
CCI3F(F-11)
CCI2F2
(F-12)
CHBrCI,
vinyl chloride
CH3CH2CI
CCI3CF3
(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethyl benzene
m/p-xylenes
o-xylene
styrene
1,2,4-TMB
1,3,5-TMB
1,3-
butadiene
1 ,3-DCB
1,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
24-hr
03/27/03
(ppbv)
0.28
ND(0 10)
ND(010)
ND(0 10)
ND(010)
ND(0.10)
0.18
7.4
ND(0 10)
ND(010)
030
049
ND(0.096)
ND(011)
ND(1.1)
ND(010)
20
ND(0 098)
ND(0 10)
0.81
ND(0 092)
0.54
ND(0.10)
0.70
0.12
039
42
025
0.59
0.35
ND(0.096)
046
0.35
ND(0.20)
ND(010)
ND(010)
1.4
2.1
019
ND(0.098)
scaled
stdev
cov-6%
(ppbv)
002
ND
ND
ND
ND
IND
001
044
IND
IND
0.02
003
IND
IND
IND
IND
012
IND
IND
005
IND
003
IND
0.04
001
0.02
2.52
002
0.04
0.02
IND
003
0.02
IND
IND
IND
0.08
0.13
001
IND
P[A]
grab
)3/31/03
(ppbv)
6.0
0.21
0.990
ND(0 073)
037
ND(0.073)
020
ND(0.074)
073
ND(0 074)
0.25
0.55
0.11
ND(0 073)
ND(0.77)
ND(0 073)
25
010
ND(0071)
0.80
025
0070
ND(0071)
041
ND(0 073)
ND(0.074)
0.25
ND(0.073)
ND(015)
ND(0.074)
ND(0.069)
ND(0071)
ND(0.073)
ND(0 14)
ND(0 073)
ND(0.071)
ND(0.073)
016
ND(0.13)
ND(0.07)
ND( ) = Not detected (reporting limit)
P[C]
J3/31/03
(ppbv)
20
0.75
0.94
ND(0.071)
025
ND(0071)
ND(0071)
ND(0.073)
0.19
0.090
027
052
ND(0 067)
ND(0 074)
ND(0 76)
0.08
2.1
012
ND(0.07)
070
0.30
0090
ND(0 07)
0.17
ND(0.071)
ND(0 073)
0.25
ND(0.071)
ND(0.14)
ND(0.073)
ND(0 067)
ND(0.07)
ND(0071)
ND(0.14)
ND(0071)
ND(0.07)
ND(0071)
ND(0.13)
ND(013)
023
sub-slab
mean
n=2
03/31/03
(ppbv)
4.0
0.48
1.0
ND(<0.072)
0.31
ND(<0 072)
<0.14
ND(<0 074)
046
<0082
026
0.54
<009
ND(<0.074)
ND(<0 76)
<0.08
2.3
0.11
ND(<0.07)
0.75
028
0.08
ND(<0.07)
0.29
ND(<0.072)
ND(<0.074)
0.25
ND(<0.072)
ND(<0.15)
ND(<0.074)
ND(<0.068)
ND(<0.07)
ND(<0 072)
ND(<0 14)
ND(<0.072)
ND(<0.07)
ND(<0.072)
<0.15
ND(<0.13)
<0.15
sub-slab
stdev
n=2
)3/31/03
Ippbv)
28
0.38
0.04
IND
008
IND
IND
IND
0.38
IND
001
002
IND
IND
IND
IND
028
001
IND
007
0.04
001
IND
017
IND
IND
00
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
sub-slab
cov
n=2
03/31/03
(%)
71
80
37
IND
27
IND
IND
IND
83
IND
5.4
4.0
IND
IND
IND
IND
12
13
IND
9.4
13
18
IND
59
IND
IND
0.0
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
bsmt/
sub-slab
ratio
(-)
7 OE-02
<2.1E-01
<1.0E-01
ND
<3.2E-01
ND
>1 3E+00
> 1 .OE+02
<2.2E-01
IND
1 2E+00
9 2E-01
IND
IND
IND
IND
8.7E-01
<89E-01
IND
1.1E+00
<3.3E-01
6.8E+00
IND
2 4E+00
> 1 .6E+00
>53E+00
1.7E+02
>3.4E+00
>3.9E+00
>47E+00
IND
>65E+00
>4.8E+00
IND
IND
IND
>1.9E+01
> 1 4E+01
>1.5E+00
IND
bsmt/
sub-slab
stdev
(-)
5.0E-02
IND
IND
IND
IND
IND
IND
IND
IND
IND
9.3E-02
6 6E-02
IND
IND
IND
IND
1 2E-01
IND
IND
1 2E-01
IND
1 3E+00
IND
1 4E+00
IND
IND
1 OE+01
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND = indeterminate
65
-------
concentrations of 916 and 1035 pCi/l and detected
in basement air at a mean concentration of 2.3 pCi/l.
Results and statistical analysis of sub-slab sampling
for radon are summarized in Table 9d.
Figure 44 illustrates basement/sub-slab air
concentration ratios for VOCs and radon detected
in basement air. Radon was used as an indicator
compound because indicator VOCs, 1,1 -DCE, c-1,2-
DCE, and 1,1 -DCA, were not detected in basement
air. The results of Tedlar bag sampling were used
for hypothesis testing because sampling occurred
at five sub-slab vapor probes instead of at only two
probes for EPA Method TO-15 analysis. Since the
null hypothesis that the mean basement/sub-slab
air concentration ratio of 1,1,1 -TCA was equal to the
mean basement/sub-slab air concentration ratio of
radon was rejected using a one-tailed Approximate
t-Test at a level of significance less than or equal to
0.05 (p < 0.005), it was inferred that the presence of
1,1,1-TCA in basement air was not primarily due to
vapor intrusion at the time of sampling.
Tables 9b and 9c summarize basement/sub-slab air
concentration ratios determined using EPA Method
TO-15 and Tedlar bag sampling for VOCs associated
with sub-surface contamination. For EPA Method TO-
15 analysis, the basement/sub-slab air concentration
ratio of 1,1,1 -TCA was less than 7.0E-02. For Tedlar
bag sampling and on-site GC analyses, the basement/
sub-slab air concentration ratio of 1,1,1 -TCA was less
than 5.4E-02.
Table 9b. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House G Using EPA Method TO-15 During the March 2003 Sample Event
voc
1,1,1 -TCA
1,1 -DCE
TCE
c-1,2-DCE
1,1 -DCA
bsmt
24-hr
03/27/03^
(ppbv)
0.28
ND(0.10)
ND(0.10)
ND(0.10)
ND(0.10)
scaled
stdev
cov=6%
(ppbv)
0.02
IND
IND
IND
IND
P[A]
grab
03/31/03
(ppbv)
6.0
0.21
0.990
ND(0.073)
037
P[C]
grab
03/31/03
(ppbv)
2.0
0.75
0.94
ND(0.071)
0.25
ND( ) = Not detected (reporting limit), IND = indeterminate
sub-slab
mean
n=2
03/31/03
(ppbv)
4.0
0.48
1.0
ND(<0.072)
0.31
sub-slab
stdev
n=2
03/31/03
(ppbv)
2.8
038
0.04
IND
0.08
sub-slab
cov
n=2
03/31/03
(%)
71
80
3.7
IND
27
mean and standard deviation
bsmt/
sub-slab
ratio
(-)
7.0E-02
<2.1E-01
< 1 .OE-01
IND
<3.2E-01
<7.0E-02
bsmt/
sub-slab
stdev
(-)
5.0E-02
IND
IND
IND
IND
IND
Table 9c. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House G Using 1 -Liter Tedlar Bags and On-Site GC Analysis During the March 2003 Sample Event
VOC
1,1,1 -TCA
1,1 -DCE
TCE
C-1.2-DCE
bsmt
24-hr
03/27/03
(ppbv)
0.28
ND(0.10)
ND(0.10)
ND(0.10)
scaled
stdev
cov=6%
(ppbv)
0.02
IND
IND
IND
P[A]
grab
03/31/03
(ppbv)
7.5
ND(5.0)
1.1
ND(3.0)
P[B]
grab
03/31/03
(ppbv)
5.2
ND(5.0)
1.4
ND(3.0)
P[C]
grab
03/31/03
(ppbv)
2.6
ND(5.0)
1.4
ND(3.0)
ND() = Not detected (reporting limit), IND = indeterminate
P[D]
grab
03/31/03
(ppbv)
5.2
ND(5.0)
2.4
ND(3.0)
P[E]
grab
03/31/03
(ppbv)
5.2
ND(5.0)
2.4
ND(3.0)
sub-slab
mean
n=5
03/31/03
(ppbv)
5.1
ND(<5.0)
1.7
ND(<3.0)
sub-slab
stdev
n=5
03/31/03
(ppbv)
1.7
IND
0.6
IND
sub-slab
cov
n=5
03/31/03
{%)
34
IND
35
IND
mean and standard deviation
bsmt/
sub-slab
ratio
(-)
5.4E-02
IND
<5.9E-02
IND
<5.9E-02
bsmt/
sub-slab
stdev
(-)
1 .9E-02
IND
IND
IND
IND
66
-------
Table 9d. Basement/Sub-Slab Air Concentration Ratios of Radon in House G Using 48-hr Activated Carbon Canisters for
Basement Air Sampling (3/25-3/27/03) and Scintillation Cells for Sub-Slab Air Sampling During the March 2003 Sample Event
bsmt
48-hr
(pCi/L)
24
bsmt
48-hr
(pCi/L)
2.1
bsmt-48 hr
mean
n=2
3/27/2003
(pCi/L)
2.3
bsmt-48 hr
stdev
n=2
3/27/2003
(pCi/L)
0.21
bsmt-48 hr
cov
n=2
3/27/2003
(%)
94
P[B]
.grab
110A
3/31/2003
(PCi/L)
916
P[C]
grab
300A
3/31/2003
(pCi/L)
1035
sub-slab
mean
n=2
3/31/2003
(pCi/U)
976
sub-slab
stdev
n=2
3/31/2003
(PCi/L)
84.1
sub-slab
cov
n=2
3/31/2003
(%)
8.63
bsmt/
sub-slab
ratio
(-)
2.31 E-03
bsmt/
sub-slab
stdev
(-)
2.95E-04
ethylvinylacetate -
isopropylalcohol -
4-ethyltoluene -
1,3,5-TMB -
1,2,4-TMB -
o-xylene -
m/p-xylenes -
ethylbenzene -
toluene -
benzene -
cyclohexane -
hexane -
MTBE -
MIBK -
MEK -
2-hexanone -
acetone -
CCI2F2(F-12) -
CCI3F(F-11) -
CHCI3 -
CH2CI2 -
PCE -
1,1-DCA -
TCE -
1,1-DCE -
1,1,1-TCA -
radon -
io-3
w
W
H
10-2 10'1 10° 101
Basement/Sub-Slab Concentration Ratios
102
103
Figure 44. Basement/sub-slab concentration ratios using EPA Method TO-15 at House G during the March 2003 sample event.
Error bars represent one standard deviation. Arrows indicate greater than or less than values due to non-detection in basement or
sub-slab air.
House H
There were several visible one to two millimeter
wide cracks in the slab at House H. Basement
walls consisted of poured concrete. Concentrations
of all VOCs detected in basement and/or sub-slab
air using EPA Method TO-15 are summarized in
Table 10a. A replicate basement air sample was
collected at House H. The only VOC associated with
subsurface contamination detected in basement
air was 1,1,1-TCA at a concentration of 3.7 ppbv.
The detection limit for other VOCs associated with
sub-surface contamination was between 0.079 and
0.082 ppbv. Other chlorinated compounds detected
in basement air were perchloroethylene, methylene
chloride, chloroform, carbon tetrachloride, and 1.4-
dichlorobenzene at concentrations of 0.15, 270,
0.17, 0.090, and 36 ppbv, respectively. Freons,
F-11, F-12, and F-113, were detected in basement
air at concentrations of 1.8, 0.54, and 0.070 ppbv,
67
-------
respectively. Hydrocarbons, heptane, hexane,
cylcohexane, benzene, toluene, ethylbenzene, m/p-
xylenes, o-xylene, styrene, 1,2,4-trimethylbenzene,
1,3,5-trimethylbenzene, and 4-ethyltoluene were
detected at concentrations up to 14 ppbv. Acetone,
tetrahydrofuran, methyl ethyl ketone, methyl isobutyl
ketone, and methyl tertiary-butyl ether were detected
at concentrations of 3.0,0.82,1.1,0.17, and 1.0 ppbv,
respectively.
Four probes were installed for sub-slab sampling.
Two probes were sampled using EPA Method TO-15.
All four probes were sampled using one-liter Tedlar
bags. As indicated by Table 10b, when sampling
using EPA Method TO-15,1,1,1 -TCA, 1,1 -DCE, TCE,
c-1,2-DCE, and 1,1-DCA were detected in sub-slab
air at maximum concentrations of 46, 16, 24, 6.6,
and 9.7 ppbv, respectively. Other chlorinated VOCs
detected in sub-slab air were perchloroethylene,
methylene chloride, chloroform, carbon tetrachloride,
and 1,3-dichlorobenzene at maximum concentrations
of 0.44, 14, 4.5, 0.12, and 0.48 ppbv, respectively.
Freons, F-11, F-12, and F-113, were detected at
maximum concentrations of 1.1, 0.59, and 0.070
ppbv, respectively. Hydrocarbons, hexane, benzene,
toluene, ethylbenzene, m/p-xylenes, o-xylene, 1,2,4-
trimethylbenzene, 1,3,5-trimethylbenzene, and
4-ethyltoluene were detected at concentrations up
to 0.69 ppbv. Acetone, 2-hexanone, methyl ethyl
ketone, methyl isobutyl ketone, and methyl tertiary-
butyl ether were detected at concentrations up to 1.8
ppbv. Detection limits for other compounds varied
from 0.082 to 0.17 ppbv. As indicated by Table 10c,
when sampling with one-liter Tedlar bags, 1,1,1 -TCA,
1,1-DCE, TCE, and c-1,2-DCE were detected at
maximum concentrations of 61,18,24, and 6.0 pppv,
respectively. Radon was sampled at two probes with
concentrations of 406 and 343 pCi/l and detected in
basement air at a mean concentration of 0.8 pCi/l.
Results and statistical analysis of sub-slab sampling
for radon are summarized in Table 10d.
Figure 45 illustrates basement/sub-slab ratios for
VOCs and radon detected in basement air at House
H. The results of Tedlar bag sampling were used
for hypothesis testing because sampling occurred
at four sub-slab vapor probes instead of at only two
probes for EPA Method TO-15 analysis. Radon was
used as an indicator compound because indicator
VOCs, 1,1-DCE, c-1,2-DCE, and 1,1-DCA, were not
detected in basement air. Since the null hypothesis
that the basement/sub-slab air concentration ratio of
1,1,1-TCA was equal to the basement/sub-slab air
concentration ratio of radon could be rejected using a
one-tailed Approximate t-Test at a level of significance
less than or equal to 0.05 (p < 0.025), it was inferred
that the presence of 1,1,1-TCA in basement air was
not primarily due to vapor intrusion at the time of
sampling.
Tables 10b and 10c summarize basement/sub-slab
air concentration ratios determined using EPA Method
TO-15 and Tedlar bag sampling for VOCs associated
with sub-surface contamination. For EPA Method
TO-15 analysis, the basement/sub-slab concentration
ratio of 1,1-DCE was less than 4.7E-03. For Tedlar
bag sampling and on-site GC analyses, the basement/
sub-slab air concentration ratio of TCE was less than
5.2E-03.
68
-------
Table 10a. Basement and Sub-Slab Air Concentrations for VOCs Detected at House H Using EPA Method TO-15 During the
March 2003 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
c-1 ,2-DCE
1,1 -DCA
1,2-DCA
PCE
CH2CI2
CHCI3
CCI4
CCI3F(F-11)
CCI2F2
(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3
(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1,2,4-TMB
1,3,5-TMB
1,3-
butadiene
1,3-DCB
1,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
24-hr
03/25/03
(ppbv)
37
ND(0 082)
ND(0.082)
ND(0081)
ND(0.082)
ND(0081)
0 15
270
0 16
0.090
1.8
051
ND(0.076)
ND(0 084)
1 1
0.070
2.4
ND(0.077)
0.76
0.93
0.14
097
1.7
1.6
095
0.63
2.7
0.79
3.6
34
1 8
6.4
32
ND(0 16)
ND(0.081)
36
14
82
0.34
ND(0.077)
bsmt
24-hr
03/25/03
(ppbv)
36
ND(0.08)
ND(0.08)
ND(0 079)
ND(0.08)
ND(0.079)
0.15
250
0 17
0090
1.8
0.54
ND(0 074)
ND(0.082)
1.2
ND(0 079)
3.0
0090
082
1 1
0.17
1 0
1.7
1 5
1.0
0.68
2.8
0.79
39
36
1.9
69
34
ND(0.16)
ND(0 079)
30
13
6.5
0.33
ND(0.075)
bsmt
mean
n=2
03/25/03
(ppbv)
3.7
ND(<0.08)
ND(<0.08)
ND(<0 08)
ND(<0.08)
ND(<0081)
015
260
017
0.090
1 8
053
ND(<0075)
ND(<0083)
1 2
0.07
2.7
009
0.79
1 0
0 16
1.0
1 7
1 6
1 0
0.66
28
079
38
3.5
1 9
67
3.3
ND(<0.16)
ND(<0 080)
33
14
74
0.34
ND(<0.076)
ND( ) = Not detected (reporting limit)
bsmt
stdev
n=2
03/25/03
(ppbv)
007
IND
IND
IND
IND
IND
000
14.1
0.01
000
0.00
0.02
IND
IND
0.07
IND
0.42
IND
004
012
0.02
002
0.00
007
004
004
0.07
000
0.21
0.14
0.07
0.35
0.14
IND
IND
4.24
071
1 20
0.01
IND
P[B]
grab
03/27/03
(ppbv)
15
92
11
3.8
35
ND(<0.087)
0.22
77
1 1
ND(0 089)
0.63
054
ND(0 082)
ND(0091)
ND(0.92)
ND(0.087)
1 8
0.11
ND(0 086)
058
0.49
0.36
ND(0 086)
0.14
ND(0 087)
ND(0 089)
0.40
ND(0 087)
0.20
0.090
IMD(0 082)
ND(0 086)
ND(0 087)
ND(017)
ND(0.087)
ND(0 086)
0.15
0.20
ND(0.16)
ND(0 084)
P[C]
grab
03/27/03
(ppbv)
46
16
24
66
97
0.070
044
14
4.5
0 12
1 1
0.59
0.34
ND(0.12)
ND(1.2)
0070
26
0 13
ND(011)
072
019
033
ND(0.11)
0.20
ND(011)
0 14
0.46
010
018
0.17
ND(0.11)
069
0.25
016
048
ND(0.11)
029
0.33
0.10
ND(011)
sub-slab
mean
n=2
03/27/03
(ppbv)
31
13
18
52
6.6
<0079
0.33
11
28
<0.10
087
057
<021
ND(<011)
ND(<1.1)
< 0.079
2.2
0.12
ND(<010)
065
0.34
035
ND(<010)
017
ND(<010)
<011
0.43
<0.09
0.19
0.13
ND(<0 10)
<039
<0.17
<017
<0.28
ND(<0.10)
022
0.27
<013
ND(<0 09)
sub-slab
stdev
n=2
03/27/03
(PPbv)
22
48
9.2
2.0
4.4
IND
0 16
4.5
2.4
IND
033
0.04
IND
IND
IND
IND
0.57
001
IND
0.10
021
0.02
IND
004
IND
IND
004
IND
0.01
006
IND
IND
IND
IND
IND
IND
0.10
0.09
IND
IND
sub-slab
cov
n=2
03/27/03
(%)
72
38
53
38
66
IND
47
41
86
IND
38
63
IND
IND
IND
IND
26
12
IND
15
62
6
IND
25
IND
IND
10
IND
7.4
44
IND
IND
IND
IND
IND
IND
45
35
IND
IND
bsmt/
sub-slab
ratio
(-)
1 2E-01
<6.5E-03
<47E-03
< 1 6E-02
<1.2E-02
IND
4 5E-01
24E+01
5.9E-02
>87E-01
2 1 E+00
9.3E-01
IND
IND
>96E-01
>88E-01
1 2E+00
7 5E-01
>7.2E+00
1 6E+00
4 6E-01
2.9E+00
>1.5E+01
9 1 E+00
>8.9E+00
>6.0E+00
6.4E+00
>7.2E+00
2.0E+01
2.7E+01
> 1 7E+01
> 1 7E+01
> 1 9E+01
IND
IND
>30E+02
61E+01
2.8E+01
>3.0E+00
IND
bsmt
sub-slab
stdev
(-)
8.6E-02
IND
IND
IND
IND
IND
21E-01
9.9E+00
5 1 E-02
IND
8 OE-01
6 9E-02
IND
IND
IND
IND
3.7E-01
IND
IND
3.0E-01
2 9E-01
1.9E-01
IND
2 3E+00
IND
IND
6 5E-01
IND
1.8E+00
1 2E+01
IND
IND
IND
IND
IND
IND
2 8E+01
1.1E+01
IND
IND
IND = indeterminate
69
-------
Table 10b. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House H Using EPA Method TO-15 During the March 2003 Sample Event
voc
1,1,1-TCA
1,1-DCE
TCE
C-1.2-DCE
1,1 -DCA ^
bsmt
24-hr
03/25/03
(ppbv)
3.7
ND(0.082)
ND(0.082)
ND(0.081)
ND(0.082)
bsmt
24-hr
03/25/03
(PPbv)
3.6
ND(0 08)
ND(0.08)
ND(0 079)
ND(0 08)
bsmt
mean
n=2
^03/25/03
(ppbv)
3.7
ND(<0.08)
ND(<0.08)
ND(<0.080)
ND(<0.08)
bsmt
stdev
n=2
03/25/03
(PPbv)
0.07
IND
IND '
P[B]
grab
03/27/03
(PPbv)
15
9.2
11
IND 3.8
IND 1 3.5
PIC]
grab
03/27/03
(ppbv)
46
16
24
6.6
sub-slab I sub-slab
mean stdev
n=2 n=2
03/27/03 03/27/03
sub-slab
cov
n=2
03/27/03
bsmt/
sub-slab
ratio
bsmt/
sub-slab
stdev
(ppbv) '(ppbv) |(%) (-) (-)
31 22
13 48
18 1 9.2
5.2 2.0
97 !66 '4.4
ND( ) = Not detected (reporting limit), IND = indeterminate
72 1.2E-01
38 <65E-03
^53 "
38
66
mean and standard deviation
<4.7E-03
<1.6E-02
<1.2E-02
<4.7E-03
8.6E-02
IND
IND
IND
IND
IND
Table 10c. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House H Using 1 -Liter Tedlar Bags and On-Site GC Analysis During the March 2003 Sample Event
voc
1,1,1-TCA
1,1-DCE
TCE
C-1.2-DCE
bsmt
24-hr
03/25/03
(ppbv)
37
ND(0.082)
ND(0 082)
ND(0081)
bsmt
24-hr
03/25/03
(ppbv)
36
ND(0 08)
ND(0.08)
ND(0.079)
bsmt
mean
n=2
03/25/03
(ppbv)
3.7
ND(<0 08)
ND(<0 08)
ND(<0 08)
bsmt
stdev
n=2
03/25/03
(ppbv)
007
IND
IND
IND
P[A]
grab
03/27/03
(ppbv)
48
19
19
6.0
P[B]
grab
03/27/03
(ppbv)
22
11
10
36
P[C]
grab
03/27/03
(ppbv)
61
18
24
5.5
P[D]
grab
03/27/03
(ppbv)
22
11
98
33
ND( ) = Not detected (reporting limit), IND = indeterminate
sub-slab
mean
n=4
03/27/03
(ppbv)
38
sub-slab
stdev
n=4
03/27/03
(ppbv)
20
15 43
16 7.0
46 ;1.3
sub-slab
cov
n=4
03/27/03
(%)
51
29
45
29
mean and standard deviation
bsmt/
sub-slab
ratio
H "1
9.7E-02
<56E-03
<5.2E-03
<1.8E-02
<5.2E-03
bsmt/
sub-slab
stdev
(-)
4 9E-02
IND
IND
IND
IND
Table 10d. Basement/Sub-Slab Air Concentration Ratios of Radon in House H Using 48-hr Activated Carbon Canisters for
Basement Air Sampling (3/21-3/24/03) and Scintillation Cells for Sub-Slab Air Sampling During the March 2003 Sample Event
bsmt
48-hr
(pCi/L)
0.8
bsmt
48-hr
(pCi/L)
08
bsmt-48 hr
mean
n=2
03/24/03
(pCi/L)
0.8
bsmt-48 hr
stdev
n=2
03/24/03
(PCi/L)
00
bsmt-48 hr
cov
n=2
03/24/03
(%)
0.0
P[A]
03/27/03
(PCi/L)
406
P[C]
03/27/03
(pCi/L)
343
sub-slab
mean
n=2
03/27/03
(pCi/L)
375
sub-slab
stdev
n=2
03/27/03
(pCi/L)
44.5
sub-slab
cov
n=2
03/27/03
<%)
11.9
bsmt/
ratio
H
2.14E-03
bsmt/
stdev
(-)
2.54E-04
70
-------
"4i7^mi]|.:.: L_mim'
r
"' H74I
. . . nrn 1—i i i i MI 1—l ' , 11:i
ethylvinylacetate y—i -,-p-,-,-H— -,—-, y-rrfn! '—-[' -pi i
isopropylalcohol - - -"—'• J--i «-"-(•- _u.i-t-4.i-Ul| ~ ... i- •
4-ethyltoluene - -
1,4-DCB -
1,3.5-TMB -
1,2,4-TMB -
styrene -
o-xylene -
m/p-xylenes
ethylbenzene -
toluene-— -j—f-H'tH ~l '"1
benzene -- - .-- •. — 77) —'
cyclohexane -
hexane -
heptane -
MTBE -
MIBK -
MEK -
THF - - t
2-hexanone -
acetone -
CCI3CF3(F-113) -
CH3CH2CI -
CCI2F2(F-12) -
CCI3F(F-11) -
CCI4-
CHCI3 -
CH2CI2 -
PCE -
1,1-DCA -
c-1,2-DCE -
TCE -
1,1-DCE-- .--, , .- t
1,1,1-TCA -- -I- :-( ! Ml;
radon -
10-1 10o 101
Basement/Sub-Slab Concentration Ratios
Figure 45. Basement/sub-slab concentration ratios using EPA Method TO-15 at House H during the March 2003 sample event.
Error bars represent one standard deviation. Arrows indicate greater than or less than values due to non-detection in basement or
sub-slab air.
House I
There were several visible cracks in the slab which
varied in thickness between 1 to 2.5 millimeters.
Basement walls consisted of poured concrete.
Concentrations of all VOCs detected in basement
and/or sub-slab air using EPA Method TO-15 are
summarized in Table 11 a. 1,1,1 -TCA, 1,1 -DCE, TCE,
c-1,2-DCE, and 1,1-DCA were detected in basement
air at concentrations of 2.8, 2.0, 1.1, 0.54, and 0.46
ppbv, respectively. Other chlorinated compounds
detected in basement air were methylene chloride,
chloroform, and vinyl chloride at concentrations of 2.5,
0.15, and 0.17 ppbv, respectively. Freons, F-11 and
F-12, were detected in basement air at concentrations
of 0.42 and 1.0 ppbv, respectively. Hydrocarbons,
hexane, cyclohexane, benzene, toluene, ethylbenzene,
m/p-xylenes, o-xylene, 1,2,4-trimethylbenzene,
1,3,5-trimethylbenzene, and 4-ethyltoluene were
detected at concentrations up to 3.3 ppbv. Acetone,
tetrahydrofuran, methyl ethyl ketone, and methyl
tertiary-butyl ether were detected at concentrations
of 2.6, 1.1,1.1, and 3.5 ppbv, respectively.
Three sub-slab probes were installed at House I. Only
one probe was sampled using EPA Method TO-15.
All three probes were sampled using one-liter Tedlar
bags. Using EPA Method TO-15, 1,1,1-TCA, 1,1-
DCE, TCE, c-1,2-DCE, and 1,1-DCA were detected
at 320,390,200,74, and 73 ppbv, respectively. Other
chlorinated compounds detected in sub-slab air were
perchloroethylene, methylene chloride, chloroform,
71
-------
Table 11a. Basement and Sub-Slab Air Concentrations for
VOCs Detected at House I Using EPA Method TO-15 During
the March 2003 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
c-1 ,2-DCE
1,1 -DCA
1 ,2-DCA
PCE
CH2CI2
CHCI3
CCI4
CCI3F(F-11)
CCI2F2
(F-1 2)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3
(F-1 13)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1 ,2,4-TMB
1,3,5-TMB
1 ,3-butadiene
1,3-DCB
1 ,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
24-hr
03/26/03
(ppbv)
2.8
20
1.1
054
0.46
ND(0.12)
ND(0.12)
2.5
0.15
ND(0.12)
042
1.0
ND(0.11)
0.17
1.2
ND(012)
2.6
ND(011)
1.1
1 1
ND(0.11)
3.5
ND(0.11)
1.6
0.87
0.65
3.3
0.71
2.1
0.79
ND(0.11)
0.65
019
ND(0.23)
ND(0.12)
ND(0.11)
0.62
ND(0.21)
3.8
ND(0.11)
scaled
stdev
cov = 6%
(ppbv)
0.17
0.12
0.07
0.03
0.03
IND
IND
0.15
001
IND
0.03
0.06
IND
0.01
0.07
IND
0.16
IND
0.07
0.07
IND
0.21
IND
0.10
005
0.04
0.20
004
0.13
0.05
IND
0.04
0.01
IND
IND
IND
004
IND
023
IND
ND( ) = Not detected above reporting limits
P[A]
grab
03/28/03
(ppbv)
320
390
200
74
73
ND(0.17)
0.87
0.19
0.970
ND(0.17)
0.26
054
ND(0.16)
0.27
ND(1 8}
ND(0.17)
2.9
0.14
ND(0.16)
0.52
0.21
0.26
ND(0.16)
ND(017)
ND(0.17)
ND(017)
0.3
ND(0.17)
ND(0 33)
ND(0.17)
ND(0.16)
ND(0.16)
ND(0.17)
ND(0.33)
ND(0.17)
ND(0.16)
ND(0.17)
ND(0.30)
0.36
ND(0.16)
IND = indeterminate
and vinyl chloride at concentrations of 0.87, 0.19, 0.97,
and 0.27 ppbv, respectively. Freons, F-11 and F-12, were
detected in sub-slab air at concentrations of 0.26 and
0.54 ppbv, respectively. The only hydrocarbon detected
in sub-slab air using EPA Method TO-15 was toluene at
0.30 ppbv. Acetone, 2-hexanone, methyl ethyl ketone,
methyl isobutyl ketone, and methyl tertiary-butyl ether
were detected at concentrations of 2.9,0.14,0.52,0.21,
and 0.26 ppbv, respectively. As indicated by Table 11 b,
when sampling with one-liter Tedlar bags, 1,1,1-TCA, 1,1-
DCE, TCE, and c-1,2-DCE were detected at maximum
concentrations in Probe [B| at 430, 320, 194, and 77
pppv, respectively. Radon was sampled at only one
probe with a concentration of 1295 pCi/l and detected
in basement air at a mean concentration of 13.0 pCi/l.
Figure 46 illustrates basement/sub-slab air concentration
ratios for sub-slab samples collected in 1-liter Tedlar
bags and analyzed on site. Since the null hypotheses
that the basement/sub-slab air concentration ratios of
1,1,1-TCA and TCE were equal to the basement/sub-
slab air concentration ratios of 1,1 -DCE and c-1,2-DCE
could not be rejected using a one-tailed Approximate
t-Test at a level of significance less than or equal to 0.05
(p > 0.1), it was inferred that the presence of 1,1,1-TCA,
TCE, 1,1 -DCE, and c-1,2-DCE in basement air was due
to vapor intrusion at the time of sampling.
Table 11b summarizes basement/sub-slab air
concentration ratios determined using Tedlar bag
sampling for VOCs associated with sub-surface
contamination. Use of basement/sub-slab air
concentration ratios for 1,1,1-TCA, 1,1-DCE, TCE,
and c-1,2-DCE resulted in computation of an average
basement/sub-slab air concentration ratio of 8.9E-03.
Coefficients of variation in sub-slab air concentration
ranged from 50 to 59%.
72
-------
Table 11b. Summary of Basement/Sub-Slab Concentration Ratios of VOCs Associated with Sub-Surface Contamination in House I
Using 1-Liter Tedlar Bags and On-Site GC Analysis During the March 2003 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
C-1.2-DCE
bsmt
24-hr
03/26/03
(ppbv)
2.8
2.0
1.1
0.54
scaled
stdev
cov = 6%
(ppbv)
0.17
0.12
0.07
0.03
P[A]
grab
03/28/03
(ppbv)
380
317
173
58
P[B]
grab
03/28/03
(ppbv)
430
320
194
77
P[C]
grab
03/28/03
(ppbv)
131
103
52
18
sub-slab
mean
n=3
03/27/03
(ppbv)
314
247
140
51
sub-slab
stdev
n=3
03/27/03
(ppbv)
160
124
76.6
30
sub-slab
cov
n=3
03/27/03
(%)
51
50
55
59
mean and standard deviation
bsmt/
sub-slab
ratio
(-)
8.9E-03
8.1E-03
7.9E-03
1.1E-02
8.9E-03
bsmt/
sub-slab
stdev
(-)
4.6E-03
4.1E-03
4.3E-03
6.3E-03
2.5E-03
C-1.2-DCE -- — 1~
TCE
1,1-DCE-
1,1,1-TCA-
io-3
-r4
I I
H -r
-r— —
4-- I U ! 4- i
I I I I I I
H-
i i !
I-2
10
Basement/Sub-Slab Concentration Ratios
Figure 46. Basement/sub-slab concentration ratios using one-liter Tedlar bags and on-site GC analysis at House I during the
March 2003 sample event. Error bars represent one standard deviation.
73
-------
House J
There were several visible cracks in the concrete slab
in the basement of House J. Basement walls consisted
of poured concrete but were covered with sheet rock.
Concentrations of all VOCs detected in basement and/or
sub-slab air using EPA Method TO-15 are summarized in
Table 12a. 1,1,1 -TCA, 1,1 -DCE, and TCE were detected
in basement air at concentrations of 0.44, 0.20, and
0.18 ppbv, respectively. Other chlorinated compounds
detected in basement air were perchloroethylene,
methylene chloride, and chloroform at concentrations of
0.10,15, and0.18ppbv, respectively. Freons, F-11 and
F-12, were detected in basement air at concentrations of
1.4 and 0.48 ppbv, respectively. Hydrocarbons, heptane,
hexane, cylcohexane, benzene, toluene, ethylbenzene,
m/p-xylenes, o-xylene, 1,2,4-trimethylbenzene, 1,3,5-
trimethylbenzene, and 4-ethyltoluene were detected at
concentrations up to 23 ppbv. Acetone, tetrahydrofuran,
methyl ethyl ketone, and methyl tertiary-butyl ether were
detected at concentrations of 4.2, 2.4, 0.86, and 12
ppbv, respectively. The compound, 1,3 butadiene, was
detected at 0.48 ppbv.
Four sub-slab probes were installed at House J. Only
one probe was sampled using EPA Method TO-15
(sampled twice). All four probes were sampled using
one-liter Tedlar bags. Using EPA Method TO-15,1,1,1-
TCA, 1,1-DCE, TCE, C-1.2-DCE, and 1,1-DCA were
found at concentrations of 36, 26, 15, 4.6, 12 ppbv,
respectively. Other chlorinated compounds detected
in sub-slab air were perchloroethylene, methylene
chloride, and chloroform at concentrations of 0.20, 1.1,
and 1.3 ppbv, respectively. Freons, F-11, F-12, and F-
113, were detected in sub-slab air at concentrations of
0.36, 0.55, and 0.08 ppbv, respectively. Hydrocarbons,
hexane, benzene, toluene, ethylbenzene, m/p-xylenes,
o-xylene, 1,2,4-trimethylbenzene, and 4-ethyltoluene
were detected at concentrations up to 1.6 ppbv. Acetone,
tetrahydrofuran, methyl ethyl ketone, methyl isobutyl
ketone, and methyl tertiary-butyl ether were detected at
concentrations of 2.8, 0.30, 0.85, 0.21, and 0.54 ppbv,
respectively.
Table 12a. Basement and Sub-Slab Air Concentrations of
VOCs at House J Using EPA Method TO-15 During the March
2003 Sample Event
voc
1,1,1 -TCA
1,1 -DCE
TCE
C-1.2-DCE
1,1 -DC A
1 ,2-DCA
PCE
CHA
CHCI3
CCI4
CCI3F(F-11)
CCI2F2
(F-1 2)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3
(F-1 13)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1,2,4-TMB
1 ,3,5-TMB
1,3-
butadiene
1 ,3-DCB
1 ,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
24-hr
03/24/03
(ppbv)
044
0.20
018
ND(014)
ND(0.14)
ND(0.14)
010
15
018
ND(0 14)
1 4
048
ND(013)
ND(0.14)
ND(1 5)
ND(0 14)
42
ND(0.13)
24
0.86
ND(013)
12
37
64
1.8
3.9
23
2.5
8.6
2.7
ND(013)
1.7
0.45
0.48
ND(014)
ND(014)
20
0.63
3.9
ND(013)
scaled
stdev
cov = 6%
(ppbv)
003
001
001
IND
IND
IND
0.01
090
0.01
IND
008
003
IND
IND
IND
IND
0.25
IND
0 14
005
IND
072
0.22
038
0.11
023
1 38
015
052
0.16
IND
0.10
003
003
IND .
IND
0 12
0.04
023
IND
P[A]
grab
03/26/03
(ppbv)
33
26
14
42
12
ND(0 11)
0.20
092
1 2
ND(012)
033
0.52
ND(011)
ND(012)
ND(1 2)
ND(0 11)
24
ND(0.11)
030
0.63
021
0.54
ND(011)
058
ND(0.11)
052
1.9
0 19
0.58
022
ND(011)
018
ND(011)
ND(0 22)
ND(0.11)
ND(011)
ND(011)
ND(0 21)
038
013
P[A]
grab
03/26/03
(ppbv)
36
24
15
4.6
9.7
ND(011)
0 19
1 1
1 3
ND(0.12)
036
055
ND(011)
ND(012)
ND(1 2)
0.08
2.8
ND(0.11)
ND(0.11)
0.85
0.17
050
ND(0.11)
0.64
ND(011)
048
1 6
0.16
047
018
ND(0 11)
016
ND(011)
ND(0.23)
ND(0.11)
ND(011)
017
ND(021)
047
013
P[A]
mean
n=2
03/26/03
(ppbv)
35
25
15
4.4
11
ND(011)
0.20
1 0
1 3
ND(012)
0.35
054
ND(0 11)
ND(0.12)
ND(1.2)
<010
26
ND(011)
<021
0.74
019
0.52
ND(0 11)
0.61
ND(011)
050
1.8
018
053
0.20
ND(0.11)
0.17
ND(0 11)
ND(0 23)
ND(011)
ND(011)
<0 14
ND(021)
0.43
0.13
ND = Not detected (reporting I mit)
74
-------
As indicated in Table 12b, when sampling with one-
liter Tedlar bags, 1,1,1-TCA, 1,1 -DCE, TCE, and c-
1,2-DCE were found at maximum concentrations in
Probe [A] at 43, 26, 14, and 3.0 ppbv, respectively.
Radon was sampled at two probes with concentrations
of 1029 and 1253 pCi/l and detected in basement air
at a mean concentration of 3.0 pCi/l. Results and
statistical analysis of sub-slab sampling for radon are
summarized in Table 12c.
Table 12b. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House J Using 1-Liter Tedlar Bags and On-Site GC Analysis During the March 2003 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
c-1 ,2-DCE
bsmt
24-hr
03/24/03
(ppbv)
0.44
0.20
0.18
ND(0.14)
scaled
stdev
cov=6%
(ppbv)
0.03
0.01
0.01
IND
P[A]
grab
03/26/03
(ppbv)
43
26
14
3.0
P[B]
grab
03/26/03
(PPbv)
8.5
5.2
2.6
ND(3.0)
P[C]
grab
03/26/03
(ppbv)
17
9.2
4.6
ND(3.0)
ND = Not detected (reporting limit), IND = indeterminate
P[D]
grab
03/26/03
(ppbv)
8.5
5.2
2.4
ND(3.0)
sub-slab
mean
n=4
03/26/03
(ppbv)
19
11
5.9
<3.0
sub-slab
stdev
n=4
03/26/03
(ppbv)
16 _,
9.9
5.5
IND
sub-slab
cov
n=4
03/26/03
(%)
85
87
93
IND
mean and standard deviation
bsmt/
sub-slab
ratio
(-)
2.3E-02
1.8E-02
3.1E-02
IND
2.4E-02
bsmf
sub-slab
stdev
(-)
1 .9E-02
1.5E-02
2.8E-02
IND
1 .3E-02
Table 12c. Basement/Sub-Slab Air Concentration Ratios for Radon in House J Using 48-hr Activated Carbon Canisters for
Basement Air Sampling (3/21-3/24/03) and Scintillation Cells for Sub-Slab Air Sampling During the March 2003 Sample Event
bsmt
48-hr
(pCi/L)
2.90
bsmt
48-hr
(PCi/L)
3.00
bsmt
mean
n=2
(pCi/L)
3.0
bsmt
stdev
n=2
(pCi/L)
0.07
bsmt
cov
n=2
(%)
2.4
P[A]
110A
03/26/03
(pCi/L)
1029
P[B]
110A
03/26/03
(pCi/L)
1253
sub-slab
mean
n=2
03/26/03
(pCi/L)
1141
sub-slab
stdev
n=2
03/26/03
(PCi/L)
158.4
sub-slab
cov
n=2
03/26/03
(%)
13.88
bsmt/
sub-slab
ratio
(-)
2.59E-03
bsmt/
sub-slab
stdev
(-)
3.64E-04
Figure 47 illustrates basement/sub-slab air
concentration ratios for radon and sub-slab samples
collected in one-literTedlar bags and analyzed on site.
Sub-slab air samples for VOCs from one-liter Tedlar
bags were used for statistical testing because only
one probe was sampled using EPA Method TO-15.
At House J, the basement/sub-slab air concentration
ratios for both radon and the indicator VOC, 1,1-
DCE, were available. The null hypothesis that the
basement/sub-slab air concentration ratio of radon
was equal to the basement/sub-slab air concentration
ratio of 1,1 -DCE could be rejected using a two-tailed
Approximate t-Test at a level of significance of 0.1.
The null hypotheses that the basement/sub-slab air
concentration ratios for 1,1,1 -TCE and TCE were equal
to the basement/sub-slab air concentration ratio of
1,1 -DCE using a one-tailed Approximate t-Test could
not be rejected at a level of significance less than or
equal to 0.05 (p > 0.1). The null hypotheses that the
basement/sub-slab air concentration ratios of 1,1,1-
TCA and TCE were equal to the basement/sub-slab
air concentration ratio of radon could be rejected using
a one-tailed Approximate t-Test at a significance level
of 0.1 but not at a significance level of 0.05. Since
the rejection criteria for the null hypothesis is a level
of significance less than or equal to 0.05, use of
both radon and the indicator VOC, 1,1-DCE, led to
a consistent finding that the presence of 1,1,1-TCA,
75
-------
1,1 -DCE, and TCE in basement air was due to vapor
intrusion at the time of sampling. However, visually
and statistically, there was more consistency in the
use of 1,1 -DCE as an indicator compound compared
to radon.
Table 12b summarizes basement/sub-slab air
concentration ratios determined using one-liter
Tedlar bag sampling for VOCs associated with sub-
surface contamination. Use of basement/sub-slab
concentration values for 1,1,1 -TCA, 1,1 -DCE, and TCE
resulted in computation of an average basement/sub-
slab ratio of 2.4E-02. The overall basement/sub-slab
air concentration ratio for House J appears high relative
to other testing locations. However, the basement/
sub-slab concentration ratio for 1,1 -DCE, a VOC only
associated with sub-surface contamination, was 1.8E-
02. Also, throughoutthis investigation, when TCE was
detected in basement air, it was always associated
with sub-surface contamination. At House J, TCE
had a basement/sub-slab air concentration ratio of
3.1 E-02 which was similar to 1,1 -DCE.
TCE -
1,1-DCE
1,1,1-TCA -
radon -
10-2
Basement/Sub-Slab Concentration Ratios
10-
Figure 47. Basement/sub-slab concentration ratios using one-liter Tedlar bags and on-site GC analysis at House J during the
March 2003 sample event. Error bars represent one standard deviation.
76
-------
House K
There were several visible cracks and holes in the
concrete slab in the basement of House K. The slab
was less than 2.5 centimeters thick. Basement walls
consisted of poured concrete. Concentrations of all
VOCs detected in basement and/or sub-slab air using
EPA Method TO-15 are summarized in Table 13a. All
VOCs associated with sub-surface contamination were
detected in basement air. 1,1,1 -TCA, 1,1 -DCE.TCE, c-
1,2-DCE, and 1,1 -DCA were detected in basement air
at concentrations of 1.1,1.3,0.54,0.23, and 0.31 ppbv,
respectively. Other chlorinated compounds detected
in basement air were perchloroethylene, methylene
chloride, and 1,4-dichlorobenzene at concentrations of
0.10,3.3, and 0.19 ppbv, respectively. Freons, F-11 and
F-12, were detected in basement air at concentrations
of 0.38 and 0.54 ppbv, respectively. Hydrocarbons,
heptane, hexane, benzene, toluene, ethylbenzene,
m/p-xylenes, o-xylene, 1,2,4-trimethylbenzene,
1,3,5-trimethylbenzene, and 4-ethyltoluene were
detected at concentrations up to 3.6 ppbv. Acetone,
tetrahydrofuran, methyl ethyl ketone, and methyl
tertiary-butyl ether were detected at concentrations
of 3.3, 1.6, 2.1, and 0.52 ppbv, respectively.
Three sub-slab probes were installed at House K. Sub-
slab air was not sampled using EPA Method TO-15. All
three probes were sampled using one-liter Tedlar bags.
As indicated in Table 13b, maximum concentrations
of 1,1,1-TCA, 1,1-DCE, TCE, and c-1,2-DCE were
found at Probe B at 947, 933, 440, and 190 ppbv,
respectively. Radon was sampled at two probes with
concentrations of 142 and 1144 pCi/L and detected
in basement air at a mean concentration of 3.2 pCi/L.
Results and statistical analysis of sub-slab sampling
for radon are summarized in Table 13c.
Table 13a. Basement Air Concentrations for VOCs
at House K Using EPA Method TO-15 During the
March 2003 Sample Event
voc
1,1,1 -TCA
1,1-DCE
TCE
C-1.2-DCE
1,1 -DCA
1,2-DCA
PCE
CHA
CHCI3
CCI4
CCI3F(F-11)
CCI2F2(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1 ,2,4-TMB
1,3,5-TMB
1 ,3-butadiene
1,3-DCB
1 ,4-DCB
4-ethyltoluene
isopropyl alcohol
ethyl/vinyl acetate
CS2
bsmt
24-hr
03/24/03
(ppbv)
1.1
1.3
0.54
0.23
0.31
ND(0.15)
0.10
3.3
ND(0.15)
ND(0.15)
0.38
0.54
ND(0.14)
ND(0.15)
ND(1 .6)
ND(0.15)
3.3
ND(0.14)
1.6
2.1
ND(0.13)
0.52
0.40
0.38
ND(0.15)
0.32
3.0
1.4
3.6
0.65
ND(0.14)
1.6
0.51
ND(0.29)
ND(0.15)
0.19
1.3
0.39
6.2
ND(0.14)
scaled
stdev
cov = 6%
(ppbv)
0.07
0.08
0.03
0.01
0.02
IND
0.01
0.20
IND
IND
0.02
0.03
IND
IND
IND
IND
0.20
IND
0.10
0.13
IND
0.03
0.02
0.02
IND
0.02
0.18
0.08
0.22
0.04
IND
0.10
0.03
IND
IND
0.01
0.08
0.02
0.37
IND
ND( ) = Not detected (reporting limit)
IND = indeterminate
77
-------
Table 13b. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House K Using 1-Liter Tedlar Bags and On-Site GC Analysis During the March 2003 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
c-1 ,2-DCE
bsmt
24-hr
03/24/03
(ppbv)
1.1
1.3
0.54
0.23
scaled
stdev
cov=6%
Ippbv)
0.07 |
0.08
0.03
0.01
P[A]
grab
03/26/03
(ppbv)
530
513
209
82
P[B]
grab
03/26/03
(ppbv)
947
933
440
190
P[C]
grab
^03/26/03
(PPbv)
500
513
210
84
ND = Not detected (reporting limit), IND = indeterminate
sub-slab
mean
n=3
03/26/03
(ppbv)
659
653
286
119
sub-slab
stdev
n=3
03/26/03
(ppbv)
250
242
133
62
sub-slab
cov
n^3
03/26/03
(%)
38
37
46
52
mean and standard deviation
bsmt/
sub-slab
ratio
(-)
1 7E-03
2.0E-03
1 9E-03
1 9E-03
1 9E-03
bsmt/
sub-slab
stdev
(-)
6.4E-04
7.5E-04
8.8E-04
1 .OE-03
4.2E-04
Table 13c. Basement/Sub-Slab Air Concentration Ratios for Radon in House K Using 48-hr Activated Carbon Canisters for
Basement Air Sampling (3/21-3/24/03) and Scintillation Cells for Sub-Slab Air Sampling During the March 2003 Sample Event
bsmt
48-hr
(pCi/L)
3.0
bsmt
48-hr
(pCi/L)
34
bsmt
mean
n=2
03/24/03
(pCi/L)
32
bsmt
stdev
n=2
03/24/03
(pCi/L)
0.28
bsmt
cov
n=2
03/24/03
(%)
8.8
P[A]
110A
03/26/03
(pCi/L)
142
P[B]
110A
03/26/03
IpCVL)
1144
sub-slab
mean
n=2
03/26/03
(pCi/L)
643
sub-slab
stdev
n=2
03/26/03
(pCi/L)
709
sub-slab
cov
n=2
03/26/03
(%)
110
bsmt/
sub-slab
ratio
(-)
4.98E-03
bsmt/
sub-slab
stdev
(-)
5.50E-03
Figure 48 illustrates basement/sub-slab air
concentration ratios for radon and VOCs collected
in one-liter Tedlar bags and analyzed on site. The
null hypotheses that the basement/sub-slab air
concentration ratio of radon was equal to the basement/
sub-slab concentration ratios of indicator VOCs, 1,1-
DCE and c-1,2-DCE, could not be rejected using a
two-tailed Approximate t-Test at a level of significance
less than 0.1. This finding was in part due to the large
standard deviation associated with the basement/
sub-slab air concentration ratio of radon (standard
deviation larger than mean). The null hypotheses
that the basement/sub-slab air concentration ratios
of 1,1,1 -TCA and TCE were equal to the basement/
sub-slab air concentration ratios of indicator VOCs,
1,1 -DCE and c-1,2-DCE, could not be rejected using a
one-tailed Approximate t-test at a level of significance
less than or equal to 0.05 (p > 0.1) inferring that the
presence of 1,1,1-TCA, TCE, 1,1-DCE, C-1.2-DCE,
and 1,1 -DCA (indicator VOC) in basement air was
all due to vapor intrusion at the time of sampling.
The null hypotheses that the basement/sub-slab air
concentration ratios of 1,1,1 -TCA and TCE were equal
to the basement/sub-slab concentration ratio of radon
could not be rejected using a one-tailed Approximate
t-Test at a level of significance less than or equal to
0.05 (p > 0.1). This provided a consistent finding with
indicator VOCs that the presence of 1,1,1-TCA and
TCE in basement air was due to vapor intrusion at
the time of sampling.
Table 13b summarizes basement/sub-slab air
concentration ratios determined using Tedlar
bag sampling for VOCs associated with sub-
surface contamination. Use of basement/sub-slab
concentration values for 1,1,1-TCA, 1,1-DCE, TCE ,
and 1,2-DCE resulted in computation of an average
basement/sub-slab ratio of 1.9E-03. Coefficients of
variation in sub-slab air concentration ranged from
38 to 52%.
78
-------
c-1,2-DCE-
TCE
1,1-DCE
1,1,1-TCA--
radon - -
-i t--t
-j. f_|_
i 1__L_1J_L_
i I I
rrrn4
J L..l_i_
1 I I 1
i I
—t-i-H-
.._L_i_ LJ..LLU
i
-^~
-t--M-*-
1 i i I ! I
J L_I_1JLLL
! I M i 1 i
t i i 1 i i I
_( _t-..1_^_H.
i i i i i I i
i i f i I M
i ! I I M I
__!_ L_L LI LI.
i I I M I l
i i i I I I I
-t--1--t-!"t-t t
10-4
10-3 10-2
Basement/Sub-Slab Concentration Ratios
10-
Figure 48. Basement/sub-slab concentration ratios using one-liter Tedlar bags and on-site GC analysis at House K during the
March 2003 sample event. Error bars represent one standard deviation.
House L
Basement walls at House L consisted of poured
concrete. The basement was partitioned into finished
and unfinished areas. Concentrations of all VOCs
detected in basement and/or sub-slab air using EPA
Method TO-15 are summarized in Table 14a. VOCs
associated with sub-surface contamination, 1,1,1-TCA,
1,1-DCE, and TCE, were detected in basement air at
concentrations of 0.27,0.24, and 0.20 ppbv, respectively.
The only other chlorinated VOC detected in basement air
was methylene chloride at a concentration of 1.1 ppbv.
Freons, F-11, and F-12, were detected in basement air
at concentrations of 0.28 and 0.67 ppbv, respectively.
Hydrocarbons, heptane, hexane, benzene, toluene,
ethylbenzene, m/p-xylenes, and o-xylene were detected in
basement air at concentrations up to 2.0 ppbv. Acetone,
tetrahydrofuran, methyl ethyl ketone, methyl isobutyl
ketone, and methyl tertiary-butyl ether were detected in
basement air at concentrations of 1.9, 0.38, 1.7, 0.82,
and 0.23 ppbv, respectively.
Three sub-slab vapor probes were installed at House L.
Sub-slab air was sampled at only one probe using EPA
Method TO-15. All probes were sampled using one-liter
Tedlar bags. Using EPA Method TO-15, 1,1,1-TCA,
1,1 -DCE, TCE, c-1,2-DCE, and 1,1-DCA were detected
in sub-slab air at concentrations of 170, 140, 120, 48,
and 43 ppbv, respectively. Other chlorinated VOCs
detected in sub-slab air using EPA Method TO-15 were
perchloroethylene and chloroform detected at 0.44 and 1.1
ppbv, respectively. Freons, F-11 and F-12, were detected
at 0.26 and 0.53 ppbv, respectively. Hydrocarbons,
hexane and toluene were detected at concentrations of
0.19 and 0.37 ppbv, respectively. Acetone, methyl ethyl
ketone, methyl isobutyl ketone, and methyl tertiary-butyl
ether were detected at concentrations of 2.8, 0.81, 0.28,
and 0.17 ppbv, respectively.
79
-------
Table 14a. Basement and Sub-Slab Air Concentrations for
VOCs at House L Using EPA Method TO-15 During the March
2003 Sample Event
VOC
1,1,1-TCA
1,1-DCE
TCE
c-1,2-DCE
1,1 -DCA
1 ,2-DCA
PCE
CHA
CHCI3
CCI4
CCI3F(F-11)
CCI2F2
(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3
(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1,2,4-TMB
1,3,5-TMB
1 ,3-butadiene
1,3-DCB
1,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
24-hr
03/26/03
(ppbv)
027
0.24
0.20
ND(0 10)
ND(0.10)
ND(0.10)
ND(010)
1.1
ND(0.10)
ND(0.10)
028
0.67
ND(0.096)
ND(0.11)
ND(1 1)
ND(0.10)
1.9
ND(0 098)
0.38
1.7
0.82
0.23
0.75
0.23
ND(0.10)
0.28
20
0.19
0.57
0.15
ND(0.096)
ND(0.10)
ND(010)
ND(0.20)
ND(0.10)
ND(0.10)
ND(010)
0.30
12
ND(0.098) .
scaled
stdev
cov = 6%
(ppbv)
0.02
0.01
0.01
IND
IND
IND
IND
007
IND
IND
0.02
0.04
IND
IND
IND
IND
0.11
IND
0.02
0.10
0.05
0.01
0.05
0.01
IND
0.02
0 12
0.01
0.03
001
IND
IND
IND
IND
IND
IND
IND
0.02
0.72
IND
ND( ) = Not detected (reporting limit)
P[B]
grab
04/01/03
(ppbv)
170
140
120
48
43
ND(0 17)
0.44
ND(0.18)
1.1
ND(0.18)
0.26
0.53
ND(0.17)
ND(017)
ND(1 9)
ND(0.18)
2.8
ND(0.17)
ND(0.17)
0.81
0.28
0.17
ND(0.17)
0.19
ND(0.18)
ND(0.18)
037
ND(0.18)
ND(0.35)
ND(0.18)
ND(0.17)
ND(0.17)
ND(0.18)
ND(0.35)
ND(018)
ND(0.17)
ND(0.18)
0.84
ND(0 32)
ND(0.17)
IND = indeterminate
As indicated in Table 14b, Probe A was sampled
sequentially five times using one-liter Tedlar bags to
assess the impact of extraction volume on sample
results. These results were previously discussed. As
indicated in Table 14c, when sampling with one-liter
Tedlar bags, 1,1,1-TCA, 1,1-DCE, TCE, and c-1,2-DCE
were found in sub-slab air at maximum concentrations
of 210, 141, 123, and 43 ppbv, respectively. Radon
was sampled at three probes with concentrations of
695, 567, and 387 pCi/L and detected in basement
air at a mean concentration of 2.6 pCi/L. Results and
statistical analysis of sub-slab sampling for radon are
summarized in Tables 14cl and 14e.
Figure 49 illustrates basement/sub-slab air
concentration ratios for radon and VOCs collected
in one-liter Tedlar bags and analyzed on site.
The null hypothesis thai the basement/sub-slab
air concentration ratio of radon was equal to the
basement/sub-slab air concentration ratio of the
indicator VOC, 1,1-DCE, could be rejected using a
two-tailed Approximate t-Test at a significance level
less than or equal to 0.1. The null hypotheses that
the basement/sub-slab air concentration ratios of
1,1,1-TCA and TCE were equal to the basement/
sub-slab concentration ratio of 1,1 -DCE could not be
rejected using a one-tailed Approximate t-Test at a
significance level less than or equal to 0.05 (p > 0.1).
This inferred that the presence of 1,1,1 -TCA, TCE, and
1,1-DCE in basement air was due to vapor intrusion
at the time of sampling. Since the basement/sub-slab
air concentration ratio for radon was greater than
basement/sub-slab air concentration ratios for 1,1,1 -
TCA and TCE, use of radon as an indicator compound
also inferred that detection of 1,1,1-TCA and TCE in
basement air were due to vapor intrusion at the time
of sampling. As visually illustrated in Figure 49, there
was greater consistency in basement/sub-slab air
80
-------
concentration ratios of VOCs associated with vapor
intrusion than between VOCs associated with vapor
intrusion and radon.
Table 14c summarizes basement/sub-slab air
concentration ratios determined using Tedlar bag
sampling for VOCs associated with sub-surface
contamination. Use of basement/sub-slab air
concentration ratios for 1,1,1 -TCA, 1,1 -DCE, and TCE
resulted in computation of an average basement/sub-
slab ratio of 1.7E-03.
Table 14b. Results of Sequential Sub-Slab Air Sampling Using 1-Liter Tedlar Bags and On-Site GC Analysis at House L During the
March 2003 Sample Event
voc
1,1,1 -TCA
1,1 -DCE
TCE
c-1,2-DCE
P[A]-01
grab
04/01/03
(ppbv)
210
145
122
44
P[A]-02
grab
04/01/03
(PPbv)
210
138
122
43
P[A]-03
grab
04/01/03
(PPbv)
204
138
123
43
P[A]-04
grab
04/01/03
(ppbv)
207
141
124
43
P[A]-05
grab
04/01/03
(PPbv)
207
141
124
44
P[A]
mean
n=5
04/01/03
(ppbv)
207.6
140.6
123
43.4
P[A]
stdev
n=5
04/01/03
(ppbv)
2.51
2.88
1.00
0.55
P[A]
cov
n=5
04/01/03
(%)
1.21
2.05
0.81
1.26
Table 14c. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House L Using 1-Liter Tedlar Bags and On-Site GC Analysis During the March 2003 Sample Event
VOC
1,1,1 -TCA
1,1-DCE
TCE
C-1.2-DCE
bsmt
24-hr
03/28/03
(PPbv)
0.27
024
0.20
ND(0.10)
scaled
stdev
cov=6%
(ppbv)
0.02
0.01
0.01
IND
ND = Not detected(reporting limit)
P[A] mean
grab
04/01/03
(ppbv)
208
141
123
43
P[B]
grab
04/01/03
(ppbv)
210
131
109
38
P[C]
grab
04/01/03
(ppbv)
184
100
104
31
IND = indeterminate
sub-slab
mean
n=3
04/01/03
(ppbv)
201
124
112
37
sub-slab
stdev
n=3
04/01/03
(PPbv)
14
21
10
6
sub-slab
cov
n=3
04/01/03
(%)
7.2
17
8.8
17
mean and standard deviation
bsmt/
sub-slab
ratio
(-)
1 .3E-03
1 9E-03
1 .8E-03
< 2.7E-03
1 .7E-03
bsmt/
sub-slab
stdev
(-)
1.3E-04
3.5E-04
1 .9E-04
IND
1 .4E-04
Table 14d. Summary of 48-Hour Indoor Air Measurements for Radon Using Activated Charcoal at House L
Location
1st floor
1 st floor
bsmt
bsmt
bsmt-bar
bsmt-bar
Start Date
03/26/03
03/26/03
03/26/03
03/26/03
03/26/03
03/26/03
End Date
03/28/03
03/28/03
03/28/03
03/28/03
03/29/03
03/29/03
Cone.
(pCi/L)
1.1
0.9
3.2
33
1.9
1.9
Table 14e. Basement/Sub-Slab Air Concentration Ratios for Radon in House L Using 48-hr Activated Carbon Canisters for
Basement Air Sampling (3/26-3/28/03) and Scintillation Cells for Sub-Slab Air Sampling During the March 2003 Sample Event
bsmt 48-hr
mean
(pCi/L)
2.6
bsmt
stdev
n=4
03/29/03
(PCi/L)
0.78
bsmt
cov
n=4
03/29/03
(%)
30.3
P[A]
110A
04/01/03
(pCi/L)
695
P[B]
110A
04/01/03
(pCi/L)
567
P[C]
110A
04/01/03
(PCi/L)
387
sub-slab
grab mean
n=3
(pCi/L)
550
sub-slab
grab stdev
n=3
(PCi/L)
155
sub-slab
grab cov
n=3
(%)
28.1
bsmt/
sub-slab
ratio
(-)
4.68E-03
bsmt/
sub-slab
stdev
(-)
1 .94E-03
81
-------
c-1,2-DCE -
TCE -
1,1-DCE -
1,1,1-TCA -
radon -
10-3
102
Basement/Sub-Slab Concentration Ratios
Figure 49. Basement/sub-slab concentration ratios using one-liter Tedlar bags and on-site GC analysis at House L during the
March 2003 sample event. Error bars represent one standard deviation. Arrows indicate less than values due to non-detection in
basement air.
House M
Basement walls at House M consisted of field stone
and were covered with sheet rock and wood framing.
Concentrations of all VOCs detected in basement
and/or sub-slab air using EPA Method TO-15 are
summarized in Table 15a. 1,1,1-TCA and 1,1-DCE
were detected in basement air at concentrations of
0.14 and 0.12 ppbv, respectively. The only other
chlorinated compound detected in basement air was
methylene chloride at a concentration of 0.20 ppbv.
Freons, F-11 and F-12, were detected in basement air
at concentrations of 0.27 and 0.47 ppbv, respectively.
Hydrocarbons, heptane, hexane, benzene, toluene,
m/p-xylenes, o-xylene, 1,2,4-trimethylbenzene, 1,3,5-
trimethylbenzene, and 4-ethyltoluene were detected
at concentrations up to 0.66 ppbv. Acetone, methyl
ethyl ketone, and methyl tertiary-butyl ether were
detected at concentrations of 2.2, 0.30, and 0.44
ppbv, respectively.
Three sub-slab vapor probes were installed at House
M. Sub-slab air was sampled at only one probe using
EPA Method TO-15. This probe was sampled twice.
All three probes were sampled using one-liter Tedlar
bags. UsingEPAMethodTO-15,1,1,1 -TCA, 1,1 -DCE,
TCE, c-1,2-DCE, and 1,1-DCA were detected at 6.4,
4.3, 5.2, 1.2, and 2.2 ppbv, respectively. The only
other chlorinated compounds detected in sub-slab
air were methylene chloride and 1,4-dichlorobenzene
at concentrations of 2.9 and 1.2 ppbv, respectively.
Freons, F-11 and F-12, were detected in sub-slab air
at concentrations of 0.29 and 0.54 ppbv, respectively.
Hydrocarbons, hexane, benzene, toluene, m/p-
xylenes, o-xylene, 1,2,4-trimethylbenzene, 1,3,5-
trimethylbenzene, and 4-ethyltoluene were detected at
82
-------
concentrations up to 1.2 ppbv. Acetone, 2-hexanone,
tetrahydrofuran, methyl ethyl ketone, methyl isobutyl
ketone, and methyl tertiary-butyl ether were detected
at concentrations of 6.2, 0.28, 0.30, 1.7, and 0.36
ppbv, respectively.
Table 15a. Basement and Sub-Slab Air Concentrations for
VOCs at House M Using EPA Method TO-15 During the March
2003 Sample Event
VOC
1,1,1-TCA
1,1 -DCE
TCE
01,2-DCE
1,1 -DCA
1,2-DCA
PCE
CH2CI2
CHCI3
CCI4
CCI3F(F-11)
CCI2F2
(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CC13CF3
(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1,2,4-TMB
1,3,5-TMB
1,3-
butadiene
1 ,3-DCB
1,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
24-hr
03/24/03
(ppbv)
0.14
0.12
ND(0.12)
ND(0.12)
ND(0.12)
ND(012)
ND(0.12)
020
ND(012)
ND(012)
027
0.47
ND(0.11}
ND(0 13)
ND(1.3)
ND(0.12)
22
ND(012)
ND(012)
030
ND(011)
0.44
0 19
033
ND(012)
0.32
0.66
ND(0 12)
035
0.15
ND(011)
0.42
015
ND(0.24)
ND(012)
ND(0 12)
029
ND(0 22)
1 5
ND(0.12)
scaled
stdev
cov = 6%
(ppbv)
001
001
IND
IND
IND
IND
IND
0.01
IND
IND
0.02
003
IND
IND
IND
IND
0.13
IND
IND
002
IND
003
001
0.02
IND
0.02
004
IND
002
001
IND
0.03
001
IND
IND
IND
002
IND
0.09
IND
ND = Not detected (reporting limit)
P[A]-01
grab
03/27/03
(ppbv)
6.4
43
5.2
1.2
22
ND(0 24)
ND(0.24)
2.9
ND(0.24)
ND(0 24)
0.30
054
ND(0 22)
ND(0.25)
ND(2 5)
ND(0 24)
6.20
028
030
1 7
036
036
ND(0 23)
0.58
ND(0 24)
0.28
1 20
ND(0 24)
066
027
ND(0.22)
0.53
ND(0 24)
ND(0.47)
083
ND(0 23)
073
064
22
ND(0.23)
P[A]-02
grab
03/27/03
(ppbv)
2.8
1.7
36
0.88
1.0
ND(021)
ND(0.21)
1.30
ND(0.21)
ND(0.21)
028
0.53
ND(0.20)
ND(0.22)
ND(2.2)
ND(0.21)
5.10
0.28
ND(0.21)
1 6
027
0.33
ND(021)
0.53
ND(021)
027
0.93
ND(0.21)
092
028
ND(0 20)
0.64
022
ND(0 42)
ND(021)
1.2
076
054
1 7
ND(0.20)
P[A]
mean
n=2
03/27/03
(ppbv)
4.6
30
44
1.0
1.6
ND(<0.23)
ND(<0.23)
2 1
ND(<0.23)
ND(<023)
029
0.54
ND(<0.21)
ND(<024)
ND(<2.4)
ND(<023)
5.65
0.28
<0.23
1.65
0315
0345
ND(<022)
0.555
ND(<0.23)
0.275
1 065
ND(<0.23)
0.79
0275
ND(<021)
0.585
<023
ND(<0.45)
<0.52
<0.72
075
059
20
ND(<022)
IND = indeterminate
As indicated in Table 15b, Probe B was sampled
sequentially five times using one-liter Tedlar bags to
assess the impact of extraction volume on sample
results. These results were previously discussed. As
indicated in Table 15c, when sampling with one-liter
Tedlar bags, maximum concentrations of 1,1,1-TCA,
1,1-DCE, TCE, and c-1,2-DCE were detected in
Probe B at 542, 480,189, and 46 ppbv, respectively.
Radon was sampled at two probes with concentrations
of 732 and 766 pCi/L and detected in basement air
at a mean concentration of 2.4 pCi/L. Results and
statistical analysis of sub-slab sampling for radon are
summarized in Table 15d.
Figure 50 illustrates basement/sub-slab air
concentration ratios for radon and VOCs collected
in one-liter Tedlar bags and analyzed on site. The
null hypothesis that the basement/sub-slab air
concentration ratio of radon was equal to the basement/
sub-slab air concentration ratio of the indicator
VOC, 1,1-DCE, could be rejected using a two-tailed
Approximate t-Test at a significance level less than
or equal to 0.1 (p < 0.025). The null hypothesis that
the basement/sub-slab air concentration ratio of
1,1,1-TCA was equal to the basement/sub-slab air
concentration ratio of 1,1-DCE could not be rejected
using a one-tailed Approximate t-Test at a significance
level less than or equal to 0.05 (p > 0.1) inferring that
the presence of 1,1,1 -TCA and 1,1 -DCE in basement
air was due to vapor intrusion at the time of sampling.
Since the basement/sub-slab air concentration ratio
of radon was greater than the basement/sub-slab air
concentration ratio of 1,1,1-TCA, use of radon as an
indicator compound also inferred that detection of
1,1,1 -TCA in basement air was due to vapor intrusion
at the time of sampling.
83
-------
Table 15c summarizes basement/sub-slab air
concentration ratios determined using Tedlar bag
sampling for VOCs associated with sub-surface
contamination. Use of basement/sub-slab air
concentration ratios for 1,1,1-TCA and 1,1 -DCE
resulted in computation of an overall basement/sub-
slab ratio of 6.3E-03.
Table 15b. Results of Sequential Sub-Slab Air Sampling Using 1 -Liter Tedlar Bags and On-Site GC Analysis at House M During
the March 2003 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
C-1.2-DCE
P[B]-01
grab
03/27/03
(ppbv)
548
483
183
44
P[B]-02
grab
03/27/03
(ppbv)
541
484
P[B]-03
grab
03/27/03
(PPbv)
533
475
189 190
46 46
P[B]-04
grab
03/27/03
(ppbv)
540
480
190
46
P[B]-05
grab
03/27/03
(ppbv)
550
478
192
46
P[B]
mean
n=5
04/01/03
(ppbv)
542
480
189
46
P[B]
stdev
n=5
04/01/03
(ppbv)
6.80
367
3.42
0.89
P[B]
cov
n=5
04/01/03
(%)
1.25
0.77
1.81
1.96
Table 15c. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House M Using 1 -Liter Tedlar Bags and On-Site GC Analysis During the March 2003 Sample Event
VOC
1,1,1-TCA
1,1 -DCE
TCE
c-1,2-DCE
bsmt
24-hr
03/24/03
(ppbv)
0.14
0.12
ND(0.12)
ND(0.12)
scaled
stdev
cov = 6%
(PPbv)
001
001
IND
IND
P[A]
grab
03/27/03
(PPbv)
52
31
31
9.5
P[B] mean
grab
03/27/03
(ppbv)
542
480
P[C]
grab
03/27/03
(ppbv)
76
64
189 J17
46 1 .4
ND( ) = Not detected (reporting limit), IND = indeterminate
sub-slab
mean
n=3
03/27/03
(PPbv)
223
192
79
19
sub-slab
stdev
n=3
03/27/03
(ppbv)
276
250
[95
24
sub-slab
cov
n=3
03/27/03
(%)
124
131
121
125
mean and standard deviation
bsmt/
sub-slab
ratio
>)~
6.3E-04
6.3E-04
<1.5E-03
<6.4E-03
6.3E-04
bsmt/
sub-slab
stdev
(-)
7 8E-04
8 2E-04
IND
IND
5 6E-04
Table 15d. Basement/Sub-Slab Air Concentration Ratios of Radon in House M Using 48-hr Activated Carbon Canisters for
Basement Air Sampling (3/22-3/24/03) and Scintillation Cells for Sub-Slab Air Sampling During the March 2003 Sample Event
bsmt
48-hr
(pCi/L)
2.4
bsmt
48-hr
(pCi/L)
2.4
bsmt
mean
(pCi/L)
2.4
bsmt
stdev
(pCi/L)
0.00
bsmt
cov
(%)
0.0
P[A]
03/27/03
(PCi/L)
732
P[C]
03/27/03
(pCi/L)
766
sub-slab
mean
n=2
03/27/03
(PCi/L)
749
sub-slab
stdev
n=2
03/27/03
(PCi/L)
12.0
sub-slab
cov
n=2
03/27/03
(%)
1.60
bsmt/
ratio
<-)
3.20E-03
bsmt/
stdev
(-)
5.14E-05
84
-------
C-1.2-DCE
TCE-
1,1-DCE-
1,1,1-TCA-
radon -
10-3
Basement/Sub-Slab Concentration Ratios
i i s
i i i
i i i
i i i
~i -t—t •
I ! I
irj2
Figure 50. Basement/sub-slab concentration ratios using one-liter Tedlar bags and on-site GC analysis at House M during the
March 2003 sample event. Error bars represent one standard deviation. Arrows indicate less than values due to non-detection in
basement air.
House N
Basement walls at House N consisted of poured
concrete. The basement was equipped with a sub-slab
depressurization system which had not been operated
for over 3 years. Concentrations of VOCs detected
in basement and/or sub-slab air using EPA Method
TO-15 are summarized in Table 16a. The only VOC
associated with subsurface contamination detected in
basement air was 1,1,1 -TCA at a concentration of 0.10
ppbv. The detection limits of other VOCs associated
with subsurface contamination ranged from 0.092 to
0.094 ppbv. Other chlorinated compounds detected
in basement air were perchloroethylene, methylene
chloride, and chloroform at concentrations of 0.11,
0.54, and 0.11 ppbv, respectively. Freons, F-11 and
F-12, were detected in basement air at concentrations
of 0.28 and 0.45 ppbv, respectively. Hydrocarbons,
hexane, cyclohexane, benzene, toluene, m/p-
xylenes, o-xylene, and 1,3,5-trimethylbenzene were
detected at concentrations up to 4.1 ppbv. Acetone,
tetrahydrofuran, methyl ethyl ketone, and methyl
tertiary-butyl ether were detected at concentrations
of 3.3, 0.67, 1.1, and 0.56 ppbv, respectively.
Three sub-slab probes were installed at House N.
Two probes were sampled using EPA Method TO-15.
All three probes were sampled using one-liter Tedlar
bags. As indicated in Table 16b, using EPA Method
TO-15, 1,1,1-TCA, 1,1-DCE, TCE, c-1,2-DCE, and
1,1-DCA were found at maximum concentrations in
Probe B at 32, 42,12, 9.9, and 20 ppbv, respectively.
Other chlorinated compounds detected in sub-slab
air were perchloroethylene, methylene chloride, and
85
-------
Table 16a. Basement and Sub-Slab Air Concentrations for VOCs at House N Using EPA Method TO-15
During the March 2003 Sample Event
voc
1,1,1-TCA
1,1 -DCE
TCE
c-1 ,2-DCE
1,1 -DCA
1 ,2-DCA
PCE
CHA
CHCI3
CCI4
CCI3F(F-11)
CCI2F2
(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3
(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1,2,4-TMB
1 ,3,5-TMB
1,3-
butadiene
1 ,3-DCB
1,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
03/28/03
(ppbv)
0.10
ND(0 094)
ND(0 094)
ND(0.092)
ND(0 094)
ND(0.092)
011
0.54
011
ND(0 094)
028
0.45
ND(0.086)
ND(0 095)
ND(0.97)
ND(0 092)
33
ND(0 088)
067
1 1
ND(0 083)
0.56
ND(0.09)
1 4
058
054
4.1
0.17
054
0 19
ND(0.086)
ND(0.09)
010
ND(018)
ND(0 092)
ND(0.09)
0.22
07
20
ND(0.088)
scaled
stdev
cov 6%
(ppbv)
0.01
IND
IND
IND
IND
IND
001
003
001
IND
002
0.03
IND
IND
IND
IND
020
IND
0.04
007
IND
0.03
IND
0.08
0.03
003
025
001
003
001
IND
IND
001
IND
IND
IND
0.01
004
0.12
IND
P[B]
grab
03/28/03
(ppbv)
32
42
12
99
20
ND(0.098)
026
024
029
ND(0 10)
027
056
ND(0 092)
ND(010)
ND(1.0)
ND(0.098)
2.2
011
ND(0 096)
T/3
016
050
ND(0 096)
025
ND(0 098)
0 14
0 18
ND(0 098)
ND(0 20)
ND(0.10)
ND(0.092)
ND(0 096)
ND(0 098)
ND(0 19)
ND(0 098)
ND(0.096)
ND(0 098)
ND(0 18)
031
ND(0 094)
ND( ) = Not detected (reporting limit)
P[C]
03/28/03
(ppbv)
97
5.3
25
ND(1 0)
69
ND(1.0)
ND(1 .0)
ND(1 0)
1 0
ND(1 0)
ND(1 0)
ND(1 0)
ND(0 96)
ND(1 1)
ND(11)
ND(1.0)
91
ND(0 98)
ND(1.0)
3.7
ND(0 92)
ND(0 98)
ND(1.0)
1 9
14
ND(1 0)
ND(1 0)
ND(1.0)
ND(2 0)
ND(1 0)
ND(0 96)
060
ND(1 0)
ND(2 0)
ND(1 0)
ND(1 0)
ND(1 0)
ND(1.9)
30
ND(0 98)
sub-slab
mean
n=2
03/28/03
(ppbv)
21
24
7.3
<55
135
ND(<0.55)
<063
<062
0.6
ND(<0.55)
<064
<078
ND(<0 53)
ND(<0.55)
ND(<6 0)
ND(<0 55)
57
<056
ND(<055)
22
<054
<074
ND(<0.55)
1.1
<70
<057
<0.59
ND(<0 55)
ND(<1.1)
ND(<0.55)
ND(<0.53)
<035
ND(<055)
ND(<1 1)
ND{<0 55)
ND(<0.55)
ND(<055)
ND(<1.0)
1 7
ND(<0 54)
sub-slab
stdev
n=2
03/28/03
(ppbv)
16
26
67
IND
93
IND
IND
IND
0.5
IND
IND
IND
IND
IND
IND
IND
49
IND
IND
21
IND
IND
IND
1 2
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
1 9
IND
sub-slab
cov
n=2
03/28/03
(%)
76
110
93
IND
69
IND
IND
IND
78
IND
IND
IND
IND
IND
IND
IND
86
IND
IND
bsmV
sub-slab
ratio
(-)
4.8E-03
<40E-03
< 1 3E-02
IND
<7.0E-03
IND
>1.7E-01
>8.6E-01
1.7E-01
IND
> 4 4E-01
>5.8E-01
IND
IND
IND
IND
5 8E-01
IND
> 1 2E+00
95 5.0E-01
IND IND
IND !>7.6E-01
IND
109
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
115
IND
IND
1 .3E+00
>83E-02
>95E-01
>69E+00
>3.1E-01
>49E-01
>35E-01
IND
IND
> 1 8E-01
IND
IND
IND
>40E-01
>7.2E-01
1 2E+00
IND
bsmt/
sub-slab
stdev
(-)
3 6E-03
IND
IND
IND
IND
IND
IND
IND
1 3E-01
IND
IND
IND
IND
IND
ND
-------
chloroform detected at concentrations of 0.24, 0.24,
and 1.0 ppbv, respectively. Freons, F-11 and F-12,
were detected in sub-slab air at concentrations of 0.27
and 0.56 ppbv, respectively. Hydrocarbons, hexane,
cyclohexane, benzene, and toluene were detected
at concentrations of 1.9, 14, 0.14, and 0.18 ppbv,
respectively. Acetone, 2-hexanone, methyl ethyl ketone,
methyl isobutyl ketone, and methyl tertiary-butyl ether
were detected at concentrations of 9.1,0.11,3.7,0.16,
and 0.50 ppbv, respectively.
The results of sequential sampling using one-liter Tedlar
bags at Probe A at House N to assess the impact of
extraction volume on sample results are presented in
Table 16c. These results were previously discussed.
using EPA Method TO-15 for sub-slab air analysis. The
basement/sub-slab air concentration ratio of radon was
used to assess vapor intrusion since indicator VOCs,
1,1 -DCE, c-1,2-DCE, and 1,1 -DCA, were not detected
in basement air. A basement/sub-slab air concentration
ratio of c-1,2-DCE was not calculated using EPA Method
TO-15 analysis because c-1,2-DCE was detected in only
one probe. The null hypothesis that the basement/sub-
slab air concentration ratio of 1,1,1-TCA was equal to
the basement/sub-slab air concentration ratio of radon
could not be rejected using a one-tailed Approximate
t-Test at a level of significance less than or equal to
0.05 (p>0.1) inferring that the presence of 1,1,1-TCA
in basement air was due to vapor intrusion at the time
of sampling.
As indicated in Table 16d, when sampling with one-liter
Tedlar bags, 1,1,1 -TCA, 1,1 -DCE, TCE, and c-1,2-DCE
were found at maximum concentrations of 47,6.3,14,
and 6.4 ppbv, respectively. Radon was sampled at two
probes with concentrations of 197 and 409 pCi/l and
detected in basement air at a mean concentration of
3.1 pCi/L. Results and statistical analysis of sub-slab
sampling for radon are summarized in Table 16e.
Figure51 illustrates basement/sub-slab airconcentration
ratios for VOCs and radon detected in basement air
Table 16b summarizes basement/sub-slab air
concentration ratios determined using EPA Method
TO-15 and Tedlar bag sampling for VOCs associated
with sub-surface contamination. For EPA Method TO-
15 analysis, the basement/sub-slab air concentration
ratio of 1,1,1-TCA was 4.8E-03. Table 16d summarizes
basement/sub-slab air ratios using Tedlar bags
for sampling along with on-site GC analyses. The
basement/sub-slab air concentration ratio of 1,1,1-TCA
was 4.0E-03.
Table 16b. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House N Using EPA Method TO-15 During the March 2003 Sample Event
voc
1,1,1 -TCA
1,1 -DCE
TCE
c-1 ,2-DCE
1,1 -DCA
bsmt
24-hr
03/28/03
(ppbv)
0.10
ND(0.094)
ND(0.094)
ND(0.092)
ND(0.094)
scaled
stdev
cov = 6%
(ppbv)
0.01
IND
IND
IND
IND
P[B]
grab
03/28/03
(ppbv)
32
42
12
99
20
P[C]
grab
03/28/03
(PPbv)
9.7
5.3
2.5
ND(1 0)
6.9
ND( ) = Not detected (reporting limit), IND = indeterminate
sub-slab
mean
n=2
03/28/03
(ppbv)
21
24
73
<5.5
13.5
sub-slab
stdev
n=2
03/28/03
(ppbv)
16
26
6.7
IND
9.3
sub-slab
cov
n=2
03/28/03
(%)
76
110
93
IND
69
mean and standard deviation
bsmt/
sub-slab
ratio
(-)
4.8E-03
<4.0E-03
<1.3E-02
IND
<7.0E-03
4.8E-03
bsmt/
sub-slab
stdev
(-)
3 6E-03
IND
IND
IND
IND
3.6E-03
87
-------
Table 16c. Results of Sequential Sub-Slab Air Sampling Using 1 -Liter Tedlar Bags and On-Site GC Analysis at House N During
the March 2003 Sample Event
voc
1,1,1-TCA
1,1-DCE
TCE
C-1.2-DCE
P[A]-01
grab
03/31/03
(ppbv)
13
5.8
25
ND(3.0)
P[A]-02
grab
03/31/03
(ppbv)
14
6.3
2.5
ND(3.0)
P[A]-03
grab
03/31/03
(ppbv)
15
6.5
2.5
ND(3.0)
P[A]-04
grab
03/31/03
(ppbv)
16
6.6
2.5
ND(3.0)
P[A]-05
grab
03/31/03
(ppbv)
15
6.4
2.7
ND(3.0)
P[A]
mean
n=5
03/31/03
(ppbv)
15
6
3
ND(3.0)
P[A]
stdev
n=5
03/31/03
(ppbv)
1.14
0.31
0.09
IND
P[A]
cov
n=5
03/31/03
(%)
781
4.93
3.52
IND
ND = Not detected (reporting limit), IND = indeterminate
Table 16d. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House N Using 1 -Liter Tedlar Bags and On-Site GC Analysis During the March 2003 Sample Event
VOC
1,1,1-
TCA
1,1-DCE
TCE
c-1,2-
DCE
bsmt
24-hr
03/28/03
(ppbv)
0.10
ND(0 094)
ND(0 094)
ND(0.092)
scaled
stdev
cov = 6%
(ppbv)
0.01
IND
IND
IND
P[A]
mean
grab
03/31/03
(ppbv)
15
63
25
ND(3 0)
P[B]
grab
03/31/03
(ppbv)
47
4.6
14
6.4
P[C]
grab
03/31/03
(ppbv)
13
4.8
3.9
ND(3.0)
ND( ) = Not detected (reporting limit), IND = indeterminate
sub-slab
mean
n=3
03/28/03
(ppbv)
25
5.2
68
<4.1
sub-slab
stdev
n=3
03/28/03
(ppbv)
19
0.9
6.3
IND
sub-slab
cov
n=3
03/28/03
(%)
77
18
92
IND
mean and standard deviation
bsmt/
sub-slab
ratio
(-)
4.0E-03
< 1 8E-02
<1.4E-02
IND
4.0E-03
bsmt/
sub-slab
stdev
(-)
3.1E-03
IND
IND
IND
3.1E-03
Table 16e. Basement/Sub-Slab Air Concentration Ratios of Radon in House N Using 48-hr Activated Carbon Canisters for
Basement Air Sampling (3/25-3/27/03) and Scintillation Cells for Sub-Slab Air Sampling During the March 2003 Sample Event
bsmt
48-hr
(pCi/L)
3.3
bsmt
48-hr
(pCi/L)
2.9
bsmt
mean
n=2
03/27/03
(pCi/L)
3.1
bsmt
stdev
n=2
03/27/03
(pCi/L)
0.28
bsmt
cov
n=2
03/27/03
(%)
9.1
P[A]
300A
03/31/03
(pCi/L)
197
P[B]
330A
03/27/03
(pCi/L)
409
sub-slab
mean
n=2
03/31/03
(pCi/L)
303
sub-slab
stdev
n=2
03/31/03
(pCi/L)
150
sub-slab
cov
n=2
03/31/03
(%)
495
bsmt/
ratio
<-)
1 02E-02
bsmt/
stdev
<-)
5.15E-03
88
-------
ethylvinylacetate -
isopropylalcohol -
4-ethyltoluene -
1,3,5-TMB -
o-xylene -
m/p-xylenes -
ethylbenzene -
toluene -
benzene -
cyclohexane -
hexane -
MTBE -
MEK -
THF -
acetone -
CCI2F2(F-12) -
CCI3F(F-11) -
UnUI3 -
CH2CI2 -
PCE -I
1,1 -UUA -
1 L/t
1,1,1-1 U A
radon -
1(
' ' ' } '" ' A t ._ T^fc, '' ''
f * _- .. • J: i '.':'•<
: . . '• , , , : , * ; , r^ L ' ' ','.''
,- , -,T - ' • ; l ^^ , i • I ; • •
• •"'• ' • -^
it I , , ' I , ?"'. ' ; P'T " ' J-P* ; ' • '
• ,,, -^
- ;'!'''- r ' ' . • r ' ' i • , * ->•* • ' ' i ' s l 1
, • i . • •• , --^* • , : ; ' , ' ; '• , :
1 ,',,,„ A.I - •
1 ' ' '1 • ' * ! ' F • •
• • - ,','',•!' --> • i ' , ;
| . . A < 1 '• • ^ •' ' ' l
1 W • 1 '
1*1 ....-,.
r w i
• ~^»
A • ">"»*, '
• - • j- * j>>' •
M— ^* A
^.rtT^ „ , , A
_r-^" ! « A
1 , , A , 1 '.<...
| V |
1 1 1 1 ' ~ ~"
3-3 1C-2 10-1 10° 101 1C
Basement/Sub-Slab Concentration Ratios
Figure 51. Basement/sub-slab concentration ratios using EPA Method TO-15 at House N during the March 2003 sample event.
Error bars represent one standard deviation. Arrows indicate greater than or less than values due to non-detection in basement or
sub-slab air.
House O
The slab appeared relatively intact with few cracks.
Basement walls consisted of poured concrete.
Concentrations of all VOCs detected in basement
and/or sub-slab air using EPA Method TO-15 are
summarized in Table 17a. No VOCs associated with
subsurface contamination were detected in basement
air. The detection limits of VOCs associated with
sub-surface contamination ranged from 0.097 to
0.099 ppbv. Other chlorinated compounds detected
in basement air were methylene chloride, chloroform,
carbon tetrachloride, and 1,4-dichlorobenzene at
concentrations of 1.2, 0.22, 0.17, and 0.16 ppbv,
respectively. Freons, F-11 and F-12, were detected
in basement air at concentrations of 0.74 and 0.52
ppbv, respectively. Hydrocarbons, heptane, hexane,
cyclohexane, benzene, toluene, ethylbenzene,
m/p-xylenes, o-xylene, 1,2,4-trimethylbenzene,
1,3,5-trimethylbenzene, and 4-ethyltoluene were
detected at concentrations up to 21 ppbv. Acetone,
tetrahydrofuran, methyl ethyl ketone, and methyl
tertiary-butyl ether were detected at concentrations of
19,11,1.1, and 19 ppbv, respectively. The compound,
1,3-butadiene, was detected in basement air at a
concentration of 0.80 ppbv.
Three sub-slab probes were installed at House O.
Two probes were sampled using EPA Method TO-
15. All three probes were sampled using one-liter
Tedlar bags. As indicated by Table 17b, 1,1,1-TCA,
1,1 -DCE, TCE, c-1,2-DCE, and 1,1-DCA were
found at maximum concentrations in Probe C at 7.2,
2.2, 1.6, 0.080, and 1.3 ppbv, respectively. Other
89
-------
Table 17a. Basement and Sub-Slab Air Concentrations for VOCs at House O Using EPA Method TO-15 During the March 2003
Sample Event
voc
1,1,1-TCA
1,1-DCE
TCE
0-1,2-DCE
1,1 -DCA
1,2-DCA
PCE
CH.CI,
CHCI3
CCI4
CCI3F(F-11)
CCI2F2
(F-12)
CHBrCI2
vinyl chloride
CH3CH2CI
CCI3CF3
(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1 ,2,4-TMB
1 ,3,5-TMB
1,3-
butadiene
1 ,3-DCB
1,4-DCB
4-ethyl-
toluene
isopropyl
alcohol
ethyl/vinyl
acetate
CS2
bsmt
24-hr
03/26/03
(ppbv)
ND(0.097)
ND(0 099)
ND(0.099)
ND(0 097)
ND(0 099)
0080
ND(0.092)
1.2
022
017
0.74
050
ND(0 092)
ND(010)
1 5
ND(0 097)
19
ND(0 094)
11
1.1
ND(0.088)
19
72
89
0.94
34
21
25
11
36
ND(0 092}
4.3
1.2
0.80
ND(0 097)
0.16
39
0.36
ND(018)
ND(0 094)
scaled
stdev
cov=6%
(ppbv)
IND
IND
IND
IND
IND
000
IND
007
0.01
001
0.04
003
IND
IND
009
IND
1 14
IND
0.66
007
IND
1 14
043
053
006
020
1.26
015
0.66
022
IND
0.26
007
005
IND
0.01
023
0.02
IND
IND
P[A]
grab
03/26/03
(ppbv)
56
1 1
056
ND(0.07)
0930
ND(0 07)
034
0 10
0.82
ND(0071)
030
057
ND(0.066)
ND(0 073)
ND(0 74)
0.080
1.9
0080
ND(0 068)
046
0.17
0080
ND(0.068)
0.11
ND(0 07)
ND(0071)
0 17
ND(0.07)
ND(0.14)
ND(0071)
ND(0 066)
ND(0 068)
ND(0 07)
ND(014)
ND(0 07)
ND(0.068)
ND(0 07)
ND(0.13)
ND(0.13)
ND(0 067)
ND( ) = Not detected (reporting limit)
P[C]
grab
03/26/03
(ppbv)
7.2
22
1.6
0080
1.3
ND(0071)
033
ND(0,073)
10
ND(0 073)
029
0.13
ND(0 067)
ND(0 074)
ND(0 75)
0.090
20
010
ND(0.07)
0.80
015
0.12
ND(0 07)
019
ND(0071)
ND(0 073)
023
ND(0071)
ND(0 14)
ND(0 073)
ND{0.067)
ND(0 07)
ND(0071)
ND(0.14)
ND(0.071)
ND(0.07)
ND(0.071)
ND(013)
ND(013)
ND(0.069)
sub-slab
mean
n=2
03/26/03
(ppbv)
6.4
1 7
1.1
<0.08
1 1
ND(<0.07)
0.34
<009
54
ND(<0 072)
030
0.35
ND(<0 067)
ND(<0.074)
ND(<0.75)
009
2.0
0.09
ND(<0.07)
063
016
010
ND(<0 07)
0.15
ND(<0 07)
ND(<0 072)
0.20
ND(<0.07)
ND(<0 14)
ND(<0.072)
ND(<0.067)
ND(<0.07)
ND(<0071)
ND(<0 14)
ND(<0 07)
ND(<0 07)
ND(<0.07)
ND(<013)
ND(<0.13)
ND(<0 068)
sub-slab
stdev
n=2
03/26/03
(ppbv)
1 13
078
0.74
IND
0.26
IND
001
IND
649
IND
0.01
0.31
IND
IND
IND
001
0.07
001
IND
0.24
001
003
IND
0.06
IND
IND
004
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
sub-slab
cov
n=2
03/26/03
(%)
18
47
68
IND
23
IND
2.1
IND
120
IND
2.4
89
IND
IND
IND
83
36
16
IND
38
88
28
IND
38
IND
IND
21
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
bsmt/
sub-slab
ratio
bsmt/
sub-slab
stdev
(-) (-)
<1 5E-02 IND
<59E-02
<90E-02
IND
<87E-02
>1 1E+00
<2.9E-01
> 1 .3E+00
4.1E-02
?23E+00
2 5E+00
1 4E+00
IND
IND
>2.0E+00
< 1 1 E+00
9 7E+00
<1.1E+00
> 1 6E+02
1 7E+00
<5.5E-01
1 9E+02
>1.0E+02
5 9E+01
>1.3E+01
>4.9E+01
1.1E+02
>36E+01
>79E+01
^5.1E+01
IND
>6.1E+01
> 1 7E+01
>57E+00
IND
>23E+00
>56E+01
>2.8E+00
IND
IND
IND
IND
IND
IND
IND
IND
>ND
4 9E-02
IND
I 6E-01
I 3E+00
IND
IND
IND
IND
6 8E-01
IND
IND
6 7E-01
IND
5 5E+01
IND
2 3E+01
IND
IND
2 3E+01
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND = indeterminate
90
-------
chlorinated compounds detected in sub-slab air
were perchloroethylene, methylene chloride, and
chloroform at concentrations of 0.34, 0.10, and 10
ppbv, respectively. Freons, F-11 and F-12, were
detected in sub-slab air at concentrations of 0.30
and 0.57 ppbv, respectively. Hydrocarbons, hexane
and toluene were detected at concentrations of 0.19
and 0.23 ppbv, respectively. Acetone, 2-hexanone,
methyl ethyl ketone, methyl isobutyl ketone, and methyl
tertiary-butyl ether were detected at concentrations of
2.0, 0.10,0.80,0.17, and 0.12 ppbv, respectively. As
indicated by Table 17c, when sampling with one-liter
Tedlarbags, 1,1,1-TCA, 1,1-DCE, and TCE were found
at maximum concentrations of 9.3,2.1, and 1.4 ppbv,
respectively. Radon was sampled at two probes with
concentrations of 948 and 958 pCi/l and detected in
basement air at a mean concentration of 4.8 pCi/L.
Results and statistical analysis of sub-slab sampling
for radon are summarized in Table 17d.
Figure 52 illustrates basement/sub-slab air
concentration ratios for VOCs and radon detected in
basement air. Table 17b summarizes basement/sub-
slab air concentration ratios determined using EPA
Method TO-15. Use of the lowest basement/sub-slab
air concentration ratio for a VOC associated with
sub-surface contamination (1,1,1-TCA) resulted in
computation of a basement/sub-slab air concentration
ratio of less than 1.5E-02. Table 17c summarizes
basement/sub-slab air concentration ratios using
Tedlar bags for sampling along with on-site GC
analyses. Use of a basement/sub-slab concentration
value for 1,1,1-TCA resulted in computation of a
basement/sub-slab ratio of less than 1.1E-02.
Table 17b. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House O Using EPA Method TO-15 During the March 2003 Sample Event
VOC
1,1,1-TCA
1,1-DCE
TCE
C-1.2-DCE
1,1 -DCA
bsmt
24-hr
03/26/03
(ppbv)
ND(0.097)
ND(0.099)
ND(0.099)
ND(0.097)
ND(0.099)
scaled
stdev
cov = 6%
(ppbv)
IND
IND
IND
IND
IND
P[A]
grab
03/26/03
(ppbv)
5.6
1 1
0.56
ND(0.07)
0.930
P[C]
grab
03/26/03
(ppbv)
7.2
2.2
1.6
0.080
1.3
ND( ) = Not detected (reporting limit), IND = indeterminate
sub-slab
mean
n=2
03/26/03
(PPbv)
6.4
1.7
1 1
<0.08
1.1
sub-slab
stdev
n=2
03/26/03
(ppbv)
1.13
0.78
0.74
IND
0.26
sub-slab
cov
n=2
03/26/03
(%)
18
47
68
IND
23
mean and standard deviation
bsmt/
sub-slab
ratio
(-)
< 1 .5E-02
<5.9E-02
<9.0E-02
IND
<8.7E-02
< 1 .5E-02
bsmf
sub-slab
stdev
(-)
IND
IND
IND
IND
IND
IND
Table 17c. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House O Using 1-Liter Tedlar Bags and On-Site GC Analysis During the March 2003 Sample Event
VOC
1,1,1-TCA
1,1-DCE
TCE
c-1 ,2-DCE
bsmt
24-hr
03/26/03
(PPbv)
ND(0.097)
ND(0.099)
ND(0.099)
ND(0.097)
scaled
stdev
cov = 6%
(ppbv)
IND
IND
IND
IND
P[A]
grab
03/31/03
(ppbv)
7.6
ND(5.0)
ND(1.2)
ND(3.0)
P[B]
grab
03/31/03
(ppbv)
9.3
ND(5.0)
ND(1,2)
ND(3.0)
P[C]
grab
03/31/03
(ppbv)
8.6
2.1
1.4
ND(3.0)
ND( ) = Not detected (reporting limit), IND = indeterminate
sub-slab
mean
n=3
03/31/03
(ppbv)
8.5
<4.0
<1.3
ND(3.0)
sub-slab
stdev
n=3
03/31/03
(ppbv)
0.85
IND
IND
IND
sub-slab
cov
n=2
03/31/03
(%)
10
IND
IND
IND
mean and standard deviation
bsmV
sub-slab
ratio
(-)
< 1.1 E-02
IND
IND
IND
<1 1E-02
91
-------
Table 17d. Basement/Sub-Slab Air Concentration Ratios for Radon in House O Using 48-hr Activated Carbon Canisters for
Basement Air Sampling (3/25-3/27/03) and Scintillation Cells for Sub-Slab Air Sampling During the March 2003 Sample Event
bsmt
48-hr
(pCi/L)
4.9
bsmt
48-hr
(pCi/L)
4.6
bsmt
mean
n=2
03/24/03
(pCi/L)
4.8
bsmt
stdev
n=2
03/24/03
(pCi/L)
0.21
bsmt
cov
n=2
03/24/03
(%)
4.5
P[B]
300A
03/31/03
(pCi/L)
948
P[C]
110A
3/27/03
(pCi/L)
958
sub-slab
mean
n=2
3/31/03
(pCi/L)
953
— !
sub-slab
stdev
n=2
3/31/03
(PCi/L)
7.1
sub-slab
cov
n=2
3/31/03
(%)
0.74
bsmt/
sub-slab
ratio
(-)
5.0E-03
bsmt/
sub-slab
stdev
(-)
2.26E-04
isopropylalcohol -
4-etnyltoluene -
1 A RPR -
I ,^-UOD
1 3-butadisn6 -
1,3,5-TMB-
*l *> A TMR -
I 1^,*+" 1 IVID
o-xylene -
m/p-xylenes -
ethylbenzene -
toluene -
benzene -
cyclohexane -
hexane -
heptane -
MTBE-
MIBK-
MEK-
THF-
i ni
2-hexanone -
acetone .
CCI3CF3(F-113)-
CH3CH2CI-
CCI2F2(F-12)-
CCI3F(F-11)-
CCI4-
CHCI3-
CH2CI2-
PCEJ
1,2-DCA-
1,1-DCA-
TCE-
1,1-DCE-
1,1,1-TCA-
radon -
:, : ":.,' , : ':!' ' •— ^ * > ...
•i i • i ^TT»
~l 1 U^.M.MJ^.
• -.:,.:.., ->_, .• . ,
* _l?l£*^
• ~7^r
l , . .^-^
, , , ^ ' ' ' ' ^_ ^^
. : ' ' • ; .- , ' ^ — ....• ^
1 L ' ' , • ' ' L i ' ' • HH
,,,,,, , • i ' • ; • ^ ;,..>-, ^^.
, , ' , 1 0 , . 1>. ' , , • ,
1. ^ j
, , , ^ ^
' ' ' ' i [ ^ )
-«^ , „', , g|
| 0_j
• •*--,
— ^-
tt
*7 — ^*
14
•- — ^»
•— - — (
* — ' — ~^~
'•^c ' — •
• ;S»
-------
Location P
Location P is a commercial building. It has a concrete
slab approximately 1.6 meters below ground surface.
At the time of sub-slab probe installation, the slab
appeared relatively intact with few cracks. Basement
walls consist of field stone and mortar.
Concentrations of VOCs detected in basement and/or
sub-slab air using EPA Method TO-15for sampling and
analysis are summarized in Table 18a. The only VOC
associated with sub-surface contamination detected
in basement air was 1,1,1 -TCA at a concentration of
0.12ppbv. Detection limits for other VOCs associated
with sub-surface contamination were 0.09 ppbv. Other
chlorinated compounds detected in basement airwere
methylene chloride, chloroform, carbon tetrachloride,
and 1,4-dichlorobenzene at concentrations of 0.55,
0.49, 0.17, and 0.12 ppbv, respectively. Freons,
F-11 and F-12, were detected in basement air at
concentrations of 0.23 and 0.46 ppbv, respectively.
Hydrocarbons, hexane, cyclohexane, benzene,
toluene, ethylbenzene, m/p-xylenes, o-xylene, styrene,
1,2,4-trimethylbenzene, 1,3,5-trimethylbenzene, and
4-ethyltoluene were detected at concentrations up 2.7
ppbv. Acetone, tetrahydrofuran, methyl ethyl ketone,
methyl isobutyl ketone, and methyl tertiary-butyl ether
were detected at concentrations of 7.7, 11, 33, 3.4,
and 0.37 ppbv, respectively.
Four sub-slab probes were installed at location R
Two probes were sampled using EPA Method TO-15.
All probes were sampled using one-liter Tedlar bags.
As indicated in Table 18b, using EPA Method TO-15,
1,1,1-TCA, 1,1-DCE, TCE, c-1,2-DCE, and 1,1-DCA
were found at maximum concentrations in Probe [B]
at 250,100, 92,18, and 54 ppbv, respectively. Other
chlorinated compounds detected in sub-slab airwere
perchloroethylene and chloroform at concentrations
of 2.8 and 1.9 ppbv, respectively. Freons, F-11 and
F-12, were detected in sub-slab air at concentrations
of 0.37 and 0.52 ppbv, respectively. Hydrocarbons,
hexane, toluene, m/p-xylenes, 1,2,4-trimethylbenzene,
and 4-ethyltoluene were detected at concentrations
of 0.19, 0.19,0.16, 0.16, and 0.15 ppbv, respectively.
Acetone, 2-hexanone, methyl ethyl ketone, and methyl
isobutyl ketone were detected at.concentrationsof 2.6,
0.17, 1.0, and 0.20 ppbv, respectively. As indicated
in Table 18c, when sampling sub-slab air with one-
liter Tedlar bags, 1,1,1-TCA, 1,1-DCE, TCA, and c-
1,2-DCE were found at maximum concentrations in
Probe [B] at 273, 92, 77, and 15 ppbv, respectively.
Radon was sampled at two probes with concentrations
of 691 and 1258 pCi/l and detected in basement air
at a mean concentration of 3.8 pCi/l. Results and
statistical analysis of sub-slab sampling for radon are
summarized in Table 18d.
Figure 53 illustrates calculated basement/sub-slab
ratios for radon and VOCs using EPA Method TO-15
analysis. Since indicator VOCs were not detected in
basement air, the basement/sub-slab air concentration
ratio of radon was used to assess vapor intrusion.
The null hypothesis that the basement/sub-slab air
concentration ratio of 1,1,1-TCA was equal to the
basement/sub-slab air concentration ratio of radon
could not be rejected using a one-tailed Approximate
t-Test at a level of significance less than or equal to
0.05 (p > 0.1) inferring that the presence of 1,1,1 -TCA
in basement air was due to vapor intrusion at the time
of sampling.
Table 18b summarizes basement/sub-slab air
concentration ratios determined using EPA Method
93
-------
TO-15. Only the upper limit of basement/sub-slab air
concentration ratios could be calculated for 1,1 -DCE,
TCE, c-1,2-DCE, and 1,1-DCA. The basement/sub-
slab air concentration ratio of 1,1,1 -TCA was 7.2E-04.
Table 18c summarizes basement/sub-slab ratios
using Tedlar bag and on-site GC analyses for sub-slab
sampling. The basement/sub-slab air concentration
ratio of 1,1,1 -TCA was 1 .OE-03.
Table 18a. Basement and Sub-Slab Air Concentrations for VOCs at House P Using EPA Method TO-15 During the March 2003
Sample Event
voc
1,1,1-TCA
1,1-DCE
TCE
c-1 ,2-DCE
1,1 -DCA
1,2-DCA
PCE
CHA
CHCI3
CCI4
CCI3F(F-11)
CCI2F2(F-12)
bsmt
24-hr
03/26/03
(ppbv)
012
ND(0.092)
ND(0 092)
scaled
stdev
cov-6%
(PPbv)
001
IND
IND
ND(0 09) IND
ND(0.092) IND
ND(009) IND
ND(0 09) ' IND
0.55
049
003
003
0.17 001
0 23 0 01
0.46
CHBrCI2 ND(0 076)
vinyl chloride ND(0 094)
CH3CH2CI ' ND(0 96)
CCI3CF3(F-113)
acetone
2-hexanone
THF
MEK
MIBK
MTBE
heptane
hexane
cyclohexane
benzene
toluene
ethylbenzene
m/p-xylenes
o-xylene
styrene
1,2,4-TMB
1,3,5-TMB
1 ,3-butadiene
1,3-DCB
1,4-DCB
4-ethyltoluene
isopropyl alcohol
ethyl/vinyl acetate
CS2
ND(0.09)
77
ND(0 087)
11
33
3.4
037
ND(0 089)
033
034
0.26
55
0.83
2.7
057
013
0.85
031
ND(018)
ND(0 09)
012
0.85
47
6.7
ND(0 087)
003
IND
IND
IND
IND
0.46
IND
066
1.98
020
0.02
IND
0.02
002
002
0.33
005
0.16
003
0.01
005
002
IND
IND
0.01
005
0.28
040
IND
P[B]
grab
04/01/03
(ppbv)
250
100
92
18
54
ND(014)
28
ND(0 14)
1 9
ND(0.14)
0.37
050
ND(0.082)
ND(014)
ND(1 5)
ND(0.14)
26
017
ND(0 14)
1 0
020
ND(0.13)
ND(0 14)
0.19
ND(0 14)
ND(0 14)
0 19
ND(0 14)
0.16
ND(014)
ND(0.13)
0.16
ND(0.14)
ND(0 28)
ND(0 14)
ND(0 14)
015
ND(0 25)
037
ND(0.13)
ND( ) = Not detected (reporting limit)
P[C]
grab
04/01/03
(ppbv)
83
22
15
0.36
52
ND(0 16)
1.7
ND(0 16)
045
ND(016)
025
0.52
034
ND(0 16)
ND(1 6)
ND(016)
25
ND(015)
ND(015)
076
015
ND(0.15)
ND(015)
ND(0.16)
ND(0.16)
ND(016)
0 12
ND(0.16)
ND(031)
ND(0.16)
ND(0 15)
ND(0.15)
ND(0 16)
ND(0.31)
ND(0 16)
ND(0 15)
ND(0 16)
ND(0.28)
030
ND(0.15)
sub-slab
mean
n=2
04/01/03
(ppbv)
167
61
54
92
30
ND(<015)
23
ND(<0 15)
1.2
ND(<015)
031
051
<021
ND(<015)
ND(<1.6)
ND(<0 15)
26
<0.16
ND(<015)
0.88
0 18
ND(<014)
ND(<0.15)
<0.18
ND(<0 15)
ND(<015)
016
ND(<0 15)
<024
ND(<0.15)
ND(<0 14)
<0 16
ND(<0.15)
ND(<0 30)
ND(<0 15)
ND(<0 15)
<016
ND(<0 27)
034
ND(<014)
sub-slab
stdev
n=2
04/01/03
(ppbv)
118
55
54
12
35
IND
078
IND
1.0
IND
0.08
0.01
IND
IND
IND
IND
007
IND
IND
0.17
0.04
IND
IND
IND
IND
IND
005
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
005
IND
sub-slab
cov
n=2
04/01/03
(%)
71
90
102
136
117
IND
35
IND
87
IND
27
28
IND
IND
IND
IND
2.8
IND
IND
19
20
IND
IND
IND
IND
IND
32
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
bsmt/
sub-slab
ratio
(-)
7 2E-04
< 1 .5E-03
< 1 7E-03
<98E-03
< 3 OE-03
IND
<4.0E-02
>3.4E+00
4 2E-01
> 1.1 E+00
7 4E-01
9 OE-01
IND
IND
IND
IND
3 OE+00
IND
>69E+01
3.8E+01
1 9E+01
>2.3E+00
IND
>1 8E+00
> 2 1 E+00
>1.6E+00
3 5E+01
>52E+00
>1 1E+01
>3 6E+00
>8 1E-01
>53E+00
> 1 9E+00
IND
IND
>75E-01
>53E+00
>1.7E+01
2 OE+01
IND
bsmV
sub-slab
stdev
(-)
5 1 E-04
IND
IND
IND
IND
IND
IND
IND
3 6E-01
IND
2 1E-01
6.0E-02
IND
IND
IND
IND
2.0E-01
IND
IND
7.6E+00
4.1 E+00
IND
IND
IND
IND
IND
1 2E+01
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
IND
3.2E+00
IND
IND = indeterminate
94
-------
Table 18b. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House P Using EPA Method TO-15 During the March 2003 Sample Event
voc
1,1,1-TCA
1,1-DCE
TCE
C-1 ,2-DCE
1,1 -DCA
bsmt
24-hr
03/26/03
(ppbv)
0.12
ND(0.092)
ND(0.092)
ND(0.09)
ND(0.092)
scaled
stdev
cov = 6%
(ppbv)
0.01
IND
IND
IND
IND
P[B]
grab
04/01/03
(ppbv)
250
100
92
18 J
54
P[C]
grab
04/01/03
(ppbv)
83
22
15
0.36
5.2
ND( ) = Not detected(reporting limit), IND = indeterminate
sub-slab
mean
n=2
04/01/03
(ppbv)
167
61
54
9.2
30
sub-slab
stdev
n=2
04/01/03
(ppbv)
118
55
54
12
35
sub-slab
cov
n=2
04/01/03
(%)
71
90
102
136
117
mean and standard deviation
bsmt/
sub-slab
ratio
(-)
7.2E-04
< 1 .5E-03
< 1 .7E-03
<9.8E-03
<3.0E-03
7.2E-04
bsmV
sub-slab
stdev
(-)
5.1E-04
IND
IND
IND
IND
IND
Table 18c. Summary of Basement/Sub-Slab Air Concentration Ratios of VOCs Associated with Sub-Surface Contamination in
House P Using 1 -Liter Tedlar Bags and On-Site GC Analysis During the March 2003 Sample Event
voc
1,1,1-TCA
1,1-DCE
TCE
c-1 ,2-DCE
bsmt
24-hr
03/26/03
(ppbv)
0.12
ND(0.092)
ND(0 092)
ND(0.09)
scaled
stdev
cov = 6%
(ppbv)
0.01
IND
IND
IND
P[A]
grab
04/01/03
(ppbv)
37
84
4.6
ND(3.0)
P[B]
grab
04/01/03
(ppbv)
273
92
77
15
P[C]
grab
04/01/03
(ppbv)
117
28
12
ND(3.0)
P[D]
grab
04/01/03
(ppbv)
38
78
49
ND(3.0)
ND( ) = Not detected (reporting limit), IND = indeterminate
sub-slab
mean
n=4
04/01/03
(ppbv)
116
34
25
<6.0
sub-slab
stdev
n=4
04/01/03
(ppbv)
111
40
35
IND
sub-slab
cov
n=4
04/01/03
(%)
95
117
142
IND
mean and standard deviation
bsmf
sub-slab
ratio
(-)
1 OE-03
<2.7E-03
<3.7E-03
IND
1 .OE-03
bsmt/
sub-slab
stdev
(-)
9.9E-04
IND
IND
IND
9 9E-04
Table 18d. Basement/Sub-Slab Air Concentration Ratios of Radon in House P Using 48-hr Activated Carbon Canisters for
Basement Air Sampling (3/26-3/28/03) and Scintillation Cells for Sub-Slab Air Sampling During the March 2003 Sample Event
bsmt
48-hr
(pCi/L)
3.8
bsmt
48-hr
(pCi/L)
38
bsmt
48-hr
(pCi/L)
1.4
bsmt
mean
n=3
03/28/03
(pCi/L)
3.0
bsmt
stdev
n=2
03/28/03
(pCi/L)
1.4
bsmt
cov
n=2
03/28/03
(%)
46
P[A]
300A
04/01/03
(pCi/L)
691
P[C]
300A
04/01/03
(PCi/L)
1258
sub-slab
mean
n=2
04/01/03
(pCi/L)
975
sub-slab
stdev
n=2
04/01/03
(PCi/L)
401
sub-slab
cov
n=2
04/01/03
(%)
41.1
bsmt/
sub-slab
ratio
(-)
3.08E-03
bsmt/
sub-slab
stdev
(-)
1 .90E-03
95
-------
li 1-1 I'll Mil I
isopropylalcohol - — ~~''~^—
ethyl vinylacetate — '-! l^J
4-eth^ltoluene - • -j-j Iftllih;
1,3',5-TMB i- U-UUlt--
1,2,4-TMB —t l-frrrm-
styrene T ~ ,-rfnt-
o-xylene - - U LUill,
m/p-xylenes -(— I
10-5
ethylbenzene - - ;—:-, p-rn-f— -, -p, rny.i— r-r , ,-n-rt
toluene - • -L-l J-UliJf---J~l uiiisl- iJ-iUUi
benzene - -
cyclohexane - -?- -, TTi (Tn-- -
hexane- -1 \ldM- -
MTBE 1 f-ftiiHl
MIBK- -r -,-rfl,-
MEK- -~1 "-•"'-•
THF - - - t
acetone -
CCI2F2(F-12)-
CCI3F(F-11) -- i-i [HHf —
cS8!:- ii-ii::LnniipDiflii!| j
CH2CI2 - -f -i fHflfr- -H-t-ttw- + i *-H^ii -f
.^i.jipiiJ-jraijLuaiiii
C-1.2-DCE - -i i rii'rft -i- i-ritttt
TCE1 ~; iTMTlTt-T "
1,1-DCE
1,1,1-TCA-
radon -
Basement/Sub-Slab Concentration Ratios
Figure 53. Basement/sub-slab concentration ratios using EPA Method TO-15 at House P during the March 2003 sample event.
Error bars represent one standard deviation. Arrows indicate greater than or less than values due to non-detection in basement or
sub-slab air.
6.4 Results of Radon Testing to Assess
Vapor Intrusion
Figure 54 provides a comparison of basement/sub-
slab air concentration ratios for radon and indicator
VOCs associated with vapor intrusion. At locations
J, L, and M, the null hypothesis that the basement/
sub-slab air concentration ratio of radon was equal
to the basement/sub-slab air concentration ratio of
the indicator VOC, 1,1 -DCE, could be rejected using
a two-tailed Approximate t-Test at a significance level
less than 0.1. At location K, the null hypotheses
that the basement/sub-slab air concentration ratio
of radon was equal to the basement/sub-slab air
concentration ratios of the indicator VOCs, 1,1 -DCE
and c-1,2-DCE, could not be rejected using a two-
tailed Approximate t-Test at a significance level less
than 0.1. As illustrated in Figures 47,48,49, and 50,
for locations J, K, L, and M, respectively, there was a
visual dissimilarity between the basement/sub-slab
air concentration ratio of radon and VOCs associated
with vapor intrusion. This is in contrast to visual and
statistical (levels of significance always greater than
0.1) similarity of basement/sub-slab concentration air
ratios for indicator VOCs and other VOCs associated
with vapor intrusion illustrated in Figures 36, 38, 46,
47, 48, 49, and 50 at locations B, C, I, J, K, L, and M,
respectively. The internal consistency of basement/
sub-slab air concentration ratios of 1,1-DCE, c-1,2-
DCE, TCE, and 1,1,1-TCA (when associated with
vapor intrusion) relative to the basement/sub-slab
concentration ratio of radon when compared to VOCs
associated with sub-surface contamination, indicates
that for this investigation, the use of indicator VOCs
96
-------
10-
.0
"•1— •
03
c
o
03
O
c
o
O
.a
JD
CO
CO
c
CD
£
CD
CD
io
-3-
-5 .
10
-I 0-6
I
• Bsmt/Sub-Slab Ratio - Radon
o Bsmt/Sub-Slab Ratio - Indicator VOCs
M
Location
Figure 54. Comparison of basement/sub-slab air concentration ratios for radon and indicator VOCs associated with vapor
intrusion. Samples for VOCs collected in one-liter Tedlar bags with on-site GC analysis.
was preferable over the use of radon as an indicator
compound to assess vapor intrusion. However, data
for comparison of radon with indicator VOCs were
available at only four locations. Further research
is needed at sites containing conservative VOCs to
assess the usefulness of radon as a conservative
compound.
Figure 55 illustrates COVs of VOCs associated with
vapor intrusion and radon as a function of location.
COVs for VOCs are from Tedlar bag sampling and
on-site GC analysis. Sub-slab air concentrations
of radon generally do not appear to be more or less
variable than VOCs associated with vapor intrusion.
Thus, the number of probes used to estimate mean
sub-slab air radon concentration should be equivalent
to the number of probes used to estimate mean sub-
slab air concentrations of VOCs associated with vapor
intrusion.
97
-------
ItU
120 -
100 -
80 -
-s-
> 60 -
O
O
40 -
20 -
0 -
V
o
A
D
¥
o
s
A
V
A V
8
A
I I I I 1
J K L M N
o
A
o 1,1,1-TCA
V 1,1-DCE
n TCE
O C-1.2-DCE
A radon
1 '
P
Location
Figure 55. Coefficient of Variation (COV) as a function of location and compound for VOCs detected in basement air as a result of
vapor intrusion.
6.5 Summary of Basement/Sub-Slab
Concentration Ratios
Figure 56 summarizes the overall basement/sub-slab
concentration ratios for VOCs associated with vapor
intrusion at each location tested during the July 2002
and March 2003 sample events. Basement/sub-slab
concentration ratios of 1,1,1-TCA were removed
from consideration of overall basement/sub-slab
concentration ratios when statistical testing supported
a difference between indicator VOCs and 1,1,1 -TCA.
While building construction and slab conditions were
generally similar, basement/sub-slab concentration
ratios varied significantly.
In this investigation, basement/sub-slab concentration
ratios were utilized to determine whether or not VOCs
detected in basement air during the time of sampling
were due to vapor intrusion. Causative factors for
basement/sub-slab concentration ratio variation were
not investigated. However, variation in basement/
sub-slab concentration ratios in this investigation
indicates that it would have been unwise to select
a generic basement/sub-slab concentration ratio
such as 0.01 or 0.02 and measure only sub-slab air
concentrations to assess risk. This approach would
not have been conservative at House J where an
overall basement/sub-slab concentration ratio of 2.4E-
02 ± 1.3E-02 was determined for VOCs associated
with vapor intrusion. This approach would have been
borderline at Houses B and I where basement/sub-
slab concentration ratios of 8.3E-03 ± 5.3E-03, and
8.9E-03 ± 2.5E-03, respectively, were determined
for VOCs associated with vapor intrusion. Also,
98
-------
basement/sub-slab concentration ratios determined
during this investigation were specific to the time
of sampling. Basement and sub-slab sampling at
House B occurred during the summer (July) when
air exchange would be expected to be greater than
fall or winter months.
Finally, statistical testing sometimes resulted in a
finding thatthe presence of 1,1,1-TCA in basement air
was not due to vapor intrusion despite being found in
sub-slab air, soil gas, and ground water in the vicinity of
a building. Also, statistical testing sometimes resulted
in afinding that the presence of 1,1,1-TCAin basement
air was likely due to vapor intrusion despite being
found in outdoor air at concentrations comparable to
basement air. When 1,1-DCE, TCE, and 1,1-DCA
were detected in indoor air, their presence was always
due to vapor intrusion.
CD
O
O
P -
O -
N -
M -
L -
K -
J -
I -
H -
G -
E -
D -
C -
B -
A -
10-6
•' I •—I .,
'I -«—I ;
. w'- ,-
, i .:,•
10°
•I 0-5 1Q-4 10-3 10-2 10'1
Overall Basement/Sub-Slab Concentration Ratios
Figure 56. Summary of average basement/sub-slab concentration ratios for VOCs present in basement air due to vapor intrusion
using one-liter Tedlar bags and on-site GC analysis. Arrows indicate less than values due to non-detection in basement air.
99
-------
7.0 Summary
There were three primary objectives in this
investigation. The first objective was to develop a
method of sub-slab probe installation and sampling.
The second objective was to develop a method of
assessing vapor intrusion using basement and sub-slab
air samples that would be appropriate for building-to-
building investigations and sites containing petroleum
hydrocarbons. The third objective was to directly assist
EPA's New England Regional Office in evaluating
vapor intrusion at 15 homes and one business near
the RaymarkSuperfund Site in Stratford, Connecticut.
Sub-slab air sampling offers an opportunity to collect
samples directly beneath the living space of a building
and thereby eliminate uncertainty associated with
interpolation or extrapolation of soil-gas and/or ground-
water concentrations from monitoring points distant
from a building. Sub-slab sampling also provides
an opportunity to evaluate the validity of claims that
petroleum hydrocarbons degrade prior to vapor entry
into sub-slab material.
In this investigation, a VOC detected in basement
air was considered due to vapor intrusion if: (1) the
VOC of interest was detected in ground water and/or
soil-gas in the "vicinity" (e.g., 30 meters) of the house,
and (2) the results of statistical testing indicated
that the presence of a VOC in indoor air was due
to vapor intrusion. Statistical testing consisted of
evaluating the null hypothesis that the indoor/sub-
slab concentration ratio of a VOC of interest was
equal to the indoor/sub-slab concentration ratio of
an "indicator" VOC at the time of sampling. A one-
tailed Approximate t-test for Independent Samples of
Unequal Variance was used with rejection of the null
hypothesis at a level of significance (p) less than or
equal to 0.05. An indicator VOC was defined as a VOC
detected in sub-slab air and known to be associated
only with subsurface contamination. The VOCs
1,1-dichloroethylene and 1,1-dichloroethane were
considered indicator VOCs in this investigation because
they are degradation products of 1,1,1 -trichloroethane
and not associated with commercial products. The
VOC cis-1,2-dichloroethylene was considered an
indicator VOC because it is a degradation product
of trichloroethylene and not commonly associated
with commercial products. The variance associated
with each basement/sub-slab concentration ratio was
calculated using the method of propagation of errors
which incorporated the variance associated with both
basement and sub-slab air measurement. An average
or overall basement/sub-slab concentration ratio was
computed at each location using concentration ratios
of all VOCs associated with vapor intrusion. The
method of propagation of errors was then used to
calculate the variance associated with the average
basement/sub-slab concentration ratio.
Statistical testing sometimes resulted in a finding that
the presence of 1,1,1-trichloroethane in basement
air was not due to vapor intrusion despite being
100
-------
found in sub-slab air, soil-gas, and ground water in
the vicinity of a building. Also, hypothesis testing
sometimes resulted in a finding that the presence
of 1,1,1-trichloroethane in basement air was due to
vapor intrusion despite being found in outdoor air at
a concentration comparable to basement air. When
1,1-dichloroethylene, trichloroethylene, and 1,1-
dichloroethane were detected in basement air, their
presence was always due to vapor intrusion.
The usefulness of radon as an indicator compound in
assessing vapor intrusion was evaluated by statistically
comparing basement/sub-slab concentration ratios
for radon and indicator VOCs. However, the data set
for indicator VOCs versus radon comparison was
limited, consisting of testing at only four locations. At
three locations, the null hypothesis that the basement/
sub-slab air concentration ratio of radon was equal to
the basement/sub-slab air concentration ratio of the
indicator VOC, 1,1 -DCE, was rejected using a two-
tailed Approximate t-Test at a significance level less
than 0.1. There was a visual dissimilarity between
the basement/sub-slab air concentration ratio of
radon and VOCs associated with vapor intrusion.
This is in contrast to visual and statistical (levels
of significance always greater than 0.1) similarity
of basement/sub-slab concentration air ratios for
indicator VOCs and other VOCs associated with vapor
intrusion. The internal consistency of basement/
sub-slab air concentration ratios of 1,1-DCE, c-1,2-
DCE, TCE, and 1,1,1-TCA (when associated with
vapor intrusion) relative to the basement/sub-slab
concentration ratio of radon when compared to VOCs
associated with sub-surface contamination, indicate
that for this investigation, use of indicator VOCs were
preferable to use of radon as an indicator compound
to assess vapor intrusion. Further research is needed
at sites containing conservative VOCs to assess the
usefulness of radon as a conservative compound.
A number of specific recommendations regarding sub-
slab probe installation and sampling are provided in
this report. A design for a sub-slab vapor probe was
presented which allows for multiple use and "floats" in
a slab to enable air sample collection from sub-slab
material in direct contact with a slab or from an air
pocket directly beneath a slab created by subsidence.
It was demonstrated that probe materials used in this
investigation did not serve as a source of VOCs.
A method of drilling through a concrete slab is presented
where a rotary hammer drill was used to create an
"inner" and "outer" diameter hole in a concrete slab for
probe installation. Initial depth of penetration of the
"outer" diameter hole was equivalent to the length of
brass couples to ensure that probes were flush with the
upper surface of the slab. The "inner" diameter hole
fully penetrated the slab and extended into sub-slab
material to create an open cavity to prevent potential
obstruction of probes during sampling. Aquick-drying,
lime-based cement which expanded upon drying (to
ensure a tight seal) was mixed with tap water to form a
slurry and tapped into the annular space between the
probe and inside of the "outer" diameter hole using a
small metal rod. Using this procedure, 3 probes were
typically installed in less than 2 hours. Schematics
illustrating the location of sub-slab probes and other
slab penetrations (e.g., suction holes for sub-slab
permeability testing) were prepared for each buildingto
document sample locations, interpret sample results,
and design corrective measures.
Basement air samples were collected in six-
liter SilcoCan canisters and analyzed by EPA's
101
-------
New England Regional Laboratory using EPA Method
TO-15. One-hour samples were collected during
the July 2003 sample event while 24-hour samples
were collected during the March 2003 sample event.
Sub-slab air samples were collected in evacuated
six-liter SilcoCan canisters using EPA Method TO-
15 and in one-liter Tedlar bags using a peristaltic
pump. The canisters were provided and analyzed
by EPA's New England Regional Laboratory. Tedlar
bags were stored in a cooler without ice (to avoid
condensation) and analyzed for target VOCs on-site
by EPA's New England Regional Laboratory within 24
hours of sample collection. Canister samples were
collected by using a brass NPT to Swagelok union
fitting to connect vapor probes to a "T" fitting made of
a stainless steel flexible line and an in-line valve. A
portable vacuum pump was then used to purge vapor
probes and sampling lines. Samples were collected
by closing the in-line valve on the pump end of the
"T" fitting and opening a valve for entry into a six liter
SilcoCan canister. A particulate filter was attached
to the inlet port. Samples were collected by simply
opening the canister valve and waiting until canister
pressure approached atmospheric pressure (grab
sampling). This took approximately two minutes.
Tedlar bag samples were collected using threaded
brass or plastic nipples, a peristaltic pump, and Tygon
and Masterflex tubing. All tubing was disposed of after
sampling at each probe to avoid cross contamination.
Tedlar bags were filled in about one minute resulting
in an average flow rate of 1 SLPM.
Open-faced activated charcoal canisters were used
to measure indoor radon gas concentrations in
accordance with sampling procedures outlined in EPA
402-R-93-004. Canisters were placed with the open
side up 1.2 to 1.5 meters above a floor in a central
location with unimpeded airflow and left undisturbed for
a period of 48 hours. Sub-slab sampling and analysis
using scintillation cells were conducted in accordance
with the Grab Radon/Scintillation Cell Method outlined
in EPA 402-R-93-003. Tygon tubing was attached to
the sub-slab probes using threaded barbed nipples.
A peristaltic pump was used to create a vacuum in
the probe for sample collection and circulation of sub-
slab airthrough scintillation cells. Barbed fittings were
used to connect Tygon to Masterflex tubing used for
the peristaltic pump. A flow meter was placed on the
outlet side of a scintillation cell to ensure a flow rate
of approximately 1 SLPM and to determine when 10
cell volumes were exchanged in each cell. The outlet
end of the flow meter was vented outside each house.
A particulate filter was placed on the inlet side of the
scintillation cell. Quick-connect assemblies were
used for connection of Tygon tubing to scintillation
cells. Samples were analyzed within four hours as
recommended in EPA 402--R-93-003.
Air permeability testing in sub-slab media was
conducted to support corrective action (sub-slab
depressurization) and air flow simulations used to
help assess potential limits to extraction volume
and flow rate during sampling and purging. A small
regenerative blower was used to extract air from sub-
slab material. A variable-area flowmeter was used to
measure flow rate. Air pressure was measured with
magnehelic gauges and a digital manometer. Radial
and vertical air permeability of sub-slab media was
estimated using Baehr and Joss's (1995) analytical
solution for two-dimensional, axisymmetric, steady-
state gas flow in asemi-confined domain. Estimates of
radial permeability, vertical permeability, and recharge
at the top of sub-slab material along with streamline
computation and particle tracking were used to support
air flow simulations.
102
-------
Three methods were used to evaluate infiltration of
indoor air flow into a sample container during air
extraction (purging + sampling). The first method
entailed sequentially collecting five one-liter Tedlar
bag samples at a flow rate of 1 SLPM and monitoring
vapor concentration as afunction of extraction volume.
This was carried out in three homes with little effect on
sample concentration indicating insignificant infiltration.
Concrete slabs at these three buildings consisted of
approximately 2-4" of concrete and were relatively
intact (few cracks). Similar testing was conducted
with evacuated canisters representing extraction
volumes of 5 to 9 and 10 to14 liters at two homes with
similar results. A second method was then employed
which utilized a simple mass balance equation and
sub-slab and basement air concentrations. When the
sensitivity of the test permitted assessment, infiltration
of basement air into an evacuated canister was less
than 1%. However, use of this method to assess
infiltration during sampling required detection of fairly
high levels of VOCs not associated with subsurface
contamination in basement air and low levels or low
detection limits for these compounds in sub-slab air.
Sensitivity could be increased by enclosing a wide area
around a probe with a chamber during air extraction
and injecting a compound not present in sub-slab or
basement air (tracer) for a specified period of time.
However, the tracer concentration would have to be
held constant during the application period, and air
permeability testing and flow analysis would have
to be conducted to estimate the potential area of
infiltration during testing. A third method of evaluating
infiltration of basement air into a sampling vessel
during air extraction involved simulating streamlines
and particle transport during air extraction using
estimated permeability parameters. Simulations
indicated that between 5% and 10% of air extracted
during purging and sampling could have originated
as basement air when extracting up to 12 liters of air.
However, if there was subsidence of sub-slab material
below a concrete slab, most of this airflow would have
been lateral flow directly beneath the slab. Overall,
the extraction volume used in this investigation had
little effect on sample results. However, the impact
of large extraction volumes was not evaluated, and
results of this investigation do not justify use of large
extraction volumes.
The impact of rate-limited mass transport was
evaluated during air extraction by comparing sequential
sample results with airflow simulations using estimated
permeability and recharge parameters. At a sampling
rate of 1 SLPM, constant concentration in sequential
samples indicated an absence of rate-limited mass
transport during air extraction. The California
Environmental Protection Agency in conjunction with
the California Department of Toxic Substances and the
Los Angeles Regional Water Quality Control Board (Cal
EPA, 2003) recently published an advisory on soil-gas
sampling specifying a maximum flow rate of 0.1 to 0.2
SLPM during sampling. Given simulations presented
here, this recommendation appears reasonable.
To evaluate a necessary equilibration time after probe
installation for sampling, advective air flow modeling
with particle tracking was used to estimate maximum
radii of perturbation during probe installation occurring
over a period of one hour at a pressure differential of 15
Pa (highest pressure differential or most conservative
value used in EPAs vapor intrusion guidance). These
radii were then used with a spherical diffusion model
to estimate time to reach 99% of a steady-state
concentration at a probe given an initial concentration in
the modeled domain of zero (most conservative case).
103
-------
At homes near the former Raymark site, sub-slab and
underlying soils underlying each building consisted
of relatively dry sand and gravel. Little or no sorption
would be expected in this material, and volumetric
water content was relatively low. Simulations indicated
that under these conditions, equilibration would occurin
less than 2 hours. Sub-slab probes in this investigation
were allowed to equilibrate for 1 to 3 days. For sub-
slab material consisting of silt or clay, simulations
indicated that an equilibration time of approximately 10
hours would be necessary. However, most sub-slab
material consists of a mixture of sand and gravel or
sand even for homes built directly on clay. Thus, an
equilibration time of two hours should be conservative
for most cases.
coefficients of variation in sub-slab air exceeded
100% indicating substantial spatial variability in sub-
slab air concentration and the need for placement of
multiple probes. In this investigation, 55 probes were
installed in 16 buildings which, on average, resulted
in placement of one probe every 220 ft2.
In conclusion, this work provides an extensive
analysis of sub-slab sampling and supporting data-
interpretation techniques. It represents an important
first step in this area. Further research needs to be
conducted to evaluate the use of radon as an indicator
compound to assess vapor intrusion.
A simple mass-balance equation was used to estimate
the purging requirement priorto sampling. Simulations
indicated that collection of 5 purge volumes should
ensure that the exiting vapor concentration is 99%
of the entering concentration even when vapor
concentration inside the sample system has been
reduced to zero priorto sampling (most conservative
case). A purge volume for the sample train used in
homes near the former Raymark site was typically
less than 10 cm3.
Generally, during this investigation, one sub-slab vapor
probe was centrally located while two or more probes
were placed within one ortwo meters of basement walls
in each building. This was done to ensure detection
of vacuum throughout the entire sub-slab during sub-
slab depressurization testing. In this investigation,
there appeared to be little correlation between probe
placement and VOC concentration. That is, placement
of a probe in a central location did not ensure detection
of the highest VOC concentrations. At several houses,
104
-------
References
Baehr, A.L. and C.J. Joss. 1995. An updated model
of induced airflow in the unsaturated zone. Water
Resources Research 31, (2): 417-421.
California Environmental Protection Agency. 2003.
Advisory - Active soil gas investigations. http://www.
dtsc.ca.gov/PolicyandProcedures/SiteCleanup/
SMBR_ADV_activesoilgasinvst.pdf
Crank, J. 1975. The Mathematics of Diffusion, 2nd
Ed., Oxford University Press, New York, NY.
Hartman, B. 2004. Howto collect reliable soil-gas data
for risk-based applications-specifically vapor intrusion,
Part 3 - Answers to frequently asked questions.
LUSTLine Bulletin 48, November, 2004.
Millington, R.J. and J.R Quirk. 1961. Permeability
of porous solids. Trans. Faraday Society 57: 1200-
1207.
Tetra Tech NUS, Inc. 2000. Draft final remedial
investigation, Raymark-OU-Groundwater, Stratford,
CT, Response action contract (RAC), Region I, EPA
contract No. 68-W6-0045, EPA work assignment no.
029-RICO-01H3
U.S. Environmental Protection Agency. 2002a. Draft
guidance for evaluating the vapor intrusion to indoor
air pathway from groundwater and soils (subsurface
vapor intrusion guidance), http://www.epa.gov/
correctiveaction/eis/vapor.htm. Office of Solid Waste
and Emergency Response, Washington, D.C.
U.S. Environmental Protection Agency. 2002b. Air
sample analysis for volatile organic compounds.
Internal report, Feb. 12, 2002. USEPA New England
Regional Laboratory, Lexington, MA.
U.S. Environmental Protection Agency. 2001.
Development of recommendations and methods to
support assessment of soil venting performance
and closure, http://www.epa.gov/ada/download/
reports/epa_600_r01_070.pdf. EPA/600/R-01/070.
Office of Research and Development, National Risk
Management Research Laboratory, Ada, OK.
U.S. Environmental Protection Agency. 1999.
Compendium of methods for determination of toxic
organic compounds in ambient air, Determination of
volatile organic compounds (VOCs) in air collected
in specially-prepared canisters and analyzed by
gas chromatography/ mass spectrometry (GC/MS).
http://www.epa.gov/ttn/amtic/files/ambient/airtox/to-
15r.pdf. EPA/625/R-96/010b. Office of Research and
Development, National Risk Management Research
Laboratory, Cincinnati, OH
U.S. Environmental Protection Agency. 1998.
Pressurized canisters for clean certification, standard
operating procedure. EPA-REG1 -OEME/CANISTER-
PREP-SOP Revision 2, April 1998. USEPA New
England Regional Laboratory, Lexington, MA.
U.S. Environmental Protection Agency, 1996. Canister
evacuation standard operating procedures. EPA-
REG1-OEME/CAN-EVACUATION-SOP May 1996.
USEPA New England Regional Laboratory, Lexington,
MA.
U.S. Environmental Protection Agency. 1994. Sampling
of trace volatile organic compounds using summa
polished stainless steel canisters. EPA-REG1-ESD/
CAN-SAM-SOR Revision 1, March 1994. USEPA New
England Regional Laboratory, Lexington, MA
U.S. Environmental Protection Agency. 1993. Protocols
for radon and radon decay product measurements
in homes, http://www.epa.gov/radonpubs/homprot1.
html. EPA 402-R-93-003, June, 1993. Office of Air and
Radiation, Washington D.C.
105
------- |