-------
It is important to understand that the soil gas concentration
referred to here is that found or calculated for the immediate vicinity
of the building (defined by Nazaroff, Jury, and Rogers (2) as a volume
extending 2 meters laterally from the basement walls and 1 meter below
the floor). It is not the soil gas concentration that would be calcu-
lated from Equation 1 of this Appendix.
A.3.2 Typical Simplified Assumptions Used
Estimations of indoor air concentrations are typically based on:
where,
Cjn = indoor air concentration
E = contaminant infiltration rate
Q = building ventilation rate
The assumptions used to derive both E and Q are important in
arriving at a realistic value of Cin. Note, for example, that the
equation predicts that as the building ventilation rate approaches zero
(minimum is the soil gas intrusion rate unless the building is complete-
ly sealed in which case the expression is indeterminant), the indoor air
concentration approaches the concentration in the soil gas.
As a first approximation, some modelers assume that soil gas
enters only by diffusion into the building and that
E = J»A*F (22)
where,
J = Contaminant flux estimated from source model (g/m2s)
(See Section 3.1)
A = Area of building floor (m2)
F = Fraction of floor through which soil gas can enter.
and
A-18
-------
Q = ACH .y (23;
3600
where,
ACH = building air changes per hour
V = volume of building (m3)
Mueller et al(15> reports that typical ACH for single family
residences range from 0.5 to 1.5, with new or retrofitted energy-
efficient structures generally ranging from 0.5 to 0.8 but with some as
low as 0.2. For general case application, choosing any reasonable value
in this range has only a modest effect on the final predicted indoor
contaminant concentration and is easily within the likely error for the
contaminant flux rate.
As a first assumption for E, it is frequently assumed that for
contaminant entry into buildings with dirt floors or ventilated crawl
spaces, a reasonable conservative assumption is that F = 1 (ie, all soil
gas enters the building). Measured data for radon(16) indicates 0.7 < F
> 1.0 for single family homes with ventilated crawl spaces.
For buildings with slab floors, a typical assumption mode is that
F = 0.001. Data used to support this is obtained from Carlos et al (17>,
who reports that the average California home has an open area of 2- to
10-cm2 per m2 of floor space and Grumund et al <18> who concluded that
the area of cracks to total floor space is 0.01 to 0.1 percent. Some
modelers assume soil gas enters only through an assumed 0.5 cm-wide
crack around the building perimeter at the slab/wall interface (note
that this approach results in decreasing F as the floor area increases).
The technical literature does not support the use of the assumptions
based on percent cracked area of the floor for the attenuation and their
use is not recommended. For screening purposes, it is recommended that
F be set to 1 in all cases.
With these assumptions, it is a simple matter to calculate Cjn
using the flux rate calculated by any method from Section A.I. Models
A-19
-------
for the more complex case in which pressure coupling to the building
must be considered are given in the following section.
A.3.3 Incorporating Building Impacts on Infiltration Rate
It is well recognized that buildings can develop negative pres-
sures (relative to ambient pressure) as a result of temperature differ-
ences and wind effects. The American society of Heating, Refrigerating,
and Air-Conditioning Engineering (ASHRAE) adopted equation 24
calculate this pressure difference, AP:
(19)
to
AP =
AP,
(24)
where,
PV
w' L
N
and P.,
where,
V
P
h
*K
static pressure over the building, Pa
pressure difference due to thermal gradient, Pa
area of building on windward and leeward side,
m2
empirical exponent = 0.65 (Wadden & Scheff,
1983}
0.6008 V2, V = wind velocity in meters per
second;
0.0342 Ph (Tr TJ/V,
wind velocity, M/S
atmospheric pressure, Pa
distance from neutral pressure plane, m
outdoor and indoor temperatures, respectively,
Evaluation indicates that pressure differences due to thermal
gradients are generally small (eg. about 1.5 Pa for a 25'F difference)
and that wind effects dominate in equation 24. Typical values of AP are
A-20
-------
1 to 10 Pa. It should also be recognized that there are many other
potential sources in building that can lead to negative (and, in some
cases, positive) pressures. Among those resulting in negative pressures
are bathroom and kitchen fans, attic fans, forced air combustion devices
(eg., oil and gas fired furnaces), fireplaces, and leaking air supply
ducts located in crawl spaces. Systems resulting in positive pressures
are typically found in certain types of commercial buildings, however,
it should be recognized that leaking air return ducts located in crawl
spaces or attics may cause the indoor pressure to be higher than
Equation 24 would predict.
Soil-gas entry due to depressurization of basements and crawl -
spaces has been experimentally demonstrated by a number of re-
searchers16'18'19'20. The current level of understanding is that both
diffusion and convection contribute to vapor intrusion and specific site
characteristics determine the significance of each. A number of
researchers <16.18«19.2°.21> nave attempted to model the transport for
radon. Johnson and Ettinger22 have adapted this work and extended it to
the case of chemical vapors. This model is described in this section.
For details of the derivation one should consult the original paper.
There are three basic derived equations for this model. They
correspond to:
1) Contaminant source is infinite (with respect to modeling
time of interest) and vapor infiltration is through
cracks/opening in the foundation
2) Soil gas transport into the building is substantially higher
through relatively permeable walls (eg.,concrete block
construction below grade) than through foundation cracks and
openings.
3) Contaminant is located near the building and decreases with
time (ie, this case provides an average when the contaminant
source would become depleted over the averaging period
used).
The model equation corresponding to case 1 above (section A.3.3.1)
is probably the most useful for general application. In this case it
would be assumed that the distance from the source to the building does
A-21
-------
not change with time (ie., Lt in Equation 25 is a constant) and does not
change in composition over the time of interest for the calculation.
The equation would be used for structures with crawl spaces and slab
floor construction with solid (eg., poured concrete) below grade walls.
The modeling equation corresponding to case 2 above (section
A.3.3.2) is applicable to construction which uses hollow concrete block
construction below grade (including cases where the large voids are
filled with poured concrete). This is a fairly common construction
technique for houses with basements. Soil gas entry into homes with
this type construction is discussed in EPA's guidance for radon reduc-
tion techniques'24'. The model formulation presented assumes an infinite
source at constant depth as does the first equation.
The model equation corresponding to case 3 above (section A.3.3.3)
is applicable to cases where a long term average concentration is needed
and it is reasonably expected that significant changes will occur in the
mass of chemical in the soil (and, therefore, the source to building
distance will change significantly) over the time period of interest.
The model should not be applied when the contaminant is near (within 2
meters) and intrusion into the building is dominated by convective flow.
One of the other two model equations should be used to make estimates
for the time this condition exists. This third model equation may then
be applied for the time period that intrusive flow is diffusion dominat-
ed.
The models below require an estimate of the source vapor concen-
tration, Csource< because what is calculated is the ratio of the concen-
tration inside the building to the soil gas concentration at the source
(ie the "«" in the equations). Two main approaches are used in vapor
transport modeling; in the first Csource is assumed to be proportional to
the residual level in the soil, and in the second Csource is independent
of the residual level, but is a function of composition. The former is
applicable in the limit of "low" residual levels where compounds are
sorbed to the soil, dissolved in the soil moisture, and present in the
vapor space; the latter is applicable for "high" residual levels where
free-phase liquid or precipitate is trapped in the soil interstices. It
is important to note that if one chooses an incorrect model for predict-
A-22
-------
ing Csource, then it is possible to over- or under-predict the actual
Csource va^ue by orders of magnitude. Section A.I provides a limited
discussion of models to estimate Csource for both cases.
In the following equations, the symbol "«" (alpha) is the ratio of
the contaminant concentration in the indoor air to the soil gas concen-
tration at the source (ie., oc = C
building' source
A.3.3.1 Model for Infinite Source, Vapor Infiltrating Through Cracks
The derived equation is:
x exp
| QHlllLcitC*
exp
«i^) . \ *"*• \ . [j^l Lp (_£«&-)_
:"*A««eJ [ObalUi^rl (Q.oilLT\ [ (DC"C*Acr«*)
(25)
where,
a =
DTeff -
AD
building
LT
Qsoil
crack
Dcrack
bui Iding' source'
overall effective diffusion coefficient, cm2/s
cross-sectional area through which contaminants
may pass (can be approximated by area of floor
and below grade walls), cm2
building ventilation rate, cm3/s
distance from contaminant source to building
foundation, cm
volumetric flow rate of soil gas into the build-
ing
thickness of foundation, cm
effective vapor-pressure diffusion coefficient
through the crack, cm2/s
area of cracks/openings through which vapors can
pass, cm2
Interesting features of the equation are:
A-23
-------
(a) (Qsoil Lcrack/DCraCk Acrack) " *• In this 1imit COnVBCtion is
the dominant transport mechanism through the basement (building) floor
and walls.
then,
neffA
L>T AB
effA
T Ag
If the source lies directly beneath the foundation (LT -* 0), then «-*
Qsoii/Qbuiiding' which 1S the proper result for convection-dominated
transport of a vapor stream with concentration Csource. If the source is
"far" from the basement (i.e., DTeff AB/QSOJI LT -*0), then transport is
limited by diffusion from the source to foundation, and « -* (DTEffAB/-
Qbuiiding ^T)- Note that these results are independent of the cracked area
of the floor and walls. This is because contaminant vapors are swept
into the building as fast as they are transported to the soil adjacent
to the floor and walls.
(b) (Qsoii Lcrack/DCrack Acrack) -0- In this limit diffusion 1s the
dominant transport mechanism through the basement floor and walls. When
Dreff AB Lcrack/DCrack Acrack LT * *» then diffusion through the floor and
walls is the rate limiting mechanism, and there is a vapor concentration
"buildup" below the building or basement.
(c) Q building "*^. This limit corresponds to a perfectly sealed
(no ambient air exchange) basement, Qsoil must also approach zero, and
the model predicts that « -»•!; this is, the indoor contaminant vapor
concentration approaches the contaminant vapor concentration in the soil
gas.
Sample calculations were presented for this equation using the
following parameters:
AB = 7m x 10m + 2(2m x 10m) = 138 m2 = 138 x 104 cm2
Lcrack - 6 in. = 15 cm
A-24
-------
^building = 7mxlOmx3mx0.5 volume exchanges/ h = 105 m3/h
= 2.9 x 104 cm3/s
Dair = 0.087 cm2/s (benzene)
DH2° = 0.087 1.0 x 10"5 cm2/s
H = 0.18 cm3 of H20/cm3 of air (benzene)
6m = 0.07 g of H20/cm3 of soil
£T = 0.38 cm3/cm3 of soil
Pb = 1.7 g/cm3
AP = 1.0 Pa = 10 g/cm-s2
and estimating Q soil from
Q.
soil
Hln[2ZcracJt/rcraeJt]
which is an analytical solution for flow to a cylinder of length Xcrack
and radius rcrack located a depth Zcrack below ground surface; this is an
idealized model for soil gas flow to cracks located at floor/wall seams.
ky is the soil permeability to vapor flow (cm2) and /t is the vapor
viscosity (g/cm-s). For this sample problem, Zcrack = 2 m, Xcrack is
taken to be the total floor/wall seam perimeter distance (34 m), and for
consistency rcrack is given by
rcrack ~ ^"B/*crack
where the ratio ;;= Acrack/AB, so that 0 < TJ< 1. For reference, T]= 0.01
corresponds to rcrack = 4.1 cm for the values of AB and Xcrack given above;
rcrack = 1 cm corresponds to rj = 0.0025.
The results in graphical form are:
A-25
-------
10 l« 10 10 10 10 10
Ifl".
Pcnne*bility(an:l
10"
LT =o
#"" W"* I04 M* 104
PtnnubQity [an1]
LT = 100 cm
n • O-Oi
H.CLOOI
to4 io4 W w4 w*
Penne»Wiiy[cnizJ
Ly = 1000 CHI
For a contaminant source adjacent to the building (LT =0) , « is
proportional to the soil permeability to vapor flow .K at ^ »10"*cm2
"7
'6
(permeable soils). In this example, 0.001 < « <0.01 for IO" < K^ tlO
which happens to fall in the range of values typically reported for
radon studies (see section A. 3.1). The results also predict that, for
practical purposes, it can be concluded that the effect of crack size on
contaminant vapor intrusion rates will be relatively insignificant in
the limit of convective-dominated transport.
For LT - 100 cm and 1000 cm, the dependence of « on ^ is a
sigmoidal -shaped curve, where « becomes independent of ^ for both
"large" and "small" soil permeabilities. For less permeable soils, soil
gas flow rates are so low that vapor intrusion is governed entirely by
the relative rates of diffusion through the soil and foundation. As the
soil becomes more permeable, the "sweeping" of contaminant vapors into
the building by soil gas flow increases the intrusion rate. At some
point, however, the ability of the contaminant to diffuse from the
contaminant vapor source to the region of soil gas flow limits the rate
of contaminant vapor transport into the building. For highly permeable
A-26
-------
soils, therefore, ex becomes independent of the soil permeability to
vapor flow and only weakly dependent on foundation properties. « and
the intrusion rate become less dependent on foundation properties as the
distance to the contaminant source increases. This can be seen from the
convergence of the curves at both high and low values of the soil
permeability, and the 77 = 0.01 and 77 = 0.001 predictions as LT increas-
es. That is, the further the source is from the building, the less it
matters what the building foundation is. In the limit of infinite
separation between the source and the building, diffusion through soil
becomes the limiting transport mechanism and the building substructure
does not matter at all.
A.3.3.2 Modeling Using Relatively Permeable Foundation Walls
Garbesi and Sextro(20) conducted a field study of a building with
basement walls constructed of hollow concrete blocks which were subse-
quently backfilled with concrete and coated with asphalt sealant on the
exterior. No evidence of cracks at the wall-floor interface was ob-
served. Their experimental work showed good pressure coupling between
the basement and soil which they interpreted to imply the entire wall
area had a measurable permeability and soil gas could infiltrate over
this entire area.
The Johnson and Ettinger model for this case is:
a =
exp
D
'ff
exp
exp -
-l
(26)
This equation is similar to equation 25 except Acrac(c is replaced
by the area of the basement walls and floor, AB, Dcrack is replaced by
Df, the effective diffusion coefficient through the porous foundation
crack
floor and walls, and Lc is replaced by the foundation/wall thickness,
LF. While eqs 26 and 25 appear similar, they can predict quite differ-
ent results. Equation 26 is independent of the area of cracks/openings
A-27
-------
because intrusion is assumed to occur uniformly over the floor/wall
area. For a given Qsoil, therefore, the soil gas velocity through the
floor/walls is lower for the permeable floor/wall case. The impact of
this is that eq 26 may predict that transport through the foundation is
diffusion dominated, while for the same conditions equation 25 would
predict that it is convection dominated.
A.3.3.3 Model for Source Depletion Over Time
Equation 25 provides a screening estimate of indoor vapor concen-
trations, but does not account for depletion of the contaminant vapor
source. This is reasonable when short-term exposures are being estimat-
ed and it provides a conservative (upper bound) estimate for long-term
exposures. It is not appropriate, however, when more realistic long-
term exposure estimates are desired and it is unlikely that the source
will remain constant for a long period of time. This model formulation
considers the depletion due to transport.
In this approach it is recognized that the separation between
contaminant source and the building increases with time due to source
depletion. It is assumed that the rate at which a steady-state vapor
concentration profile is established is much greater that the rate at
which depletion occurs. Implicit in this approach is the assumption
that depletion occurs first from the layers of contaminant closest to
the building floor and walls, and the mass of contaminant incorporated
in the soil disappears, beginning at the edge closest to the building.
This is a reasonable assumption for diffusion-dominated transport to the
building-soil interface, but not valid for convection-dominated trans-
port from contaminated soil adjacent to a building floor.
For time periods less than the time required for total depletion
of the contaminant, the average emission rate is given by:
[(p2
The corresponding long-term average attenuation coefficient <«> is then
A-28
-------
(a) = H^AtfeAfl I £j_ [(p2 + 2t|rt)i/2 _ p] (28)
^building^source^ \ ^"cj
While this approach is more sophisticated than equation 25,
increasing sophistication usually increases the amount of site-specific
information required. More sophisticated screening models are usually
also based on additional assumptions, and one must be careful to ensure
that these assumptions are valid for specific site characteristics.
For equation 27 and 28, the following definitions apply:
pb = soil density, g/cm3
CR = average residual contaminant level in soil, g/g
AHC = thickness of the contaminant layer, cm
T = time
L°T = initial contaminant-building floor separation, cm
= DgC.
source'
P =
exp -
A.3.4 Impact of Contaminated Outdoor Air
All of the preceding discussion in this chapter has assumed that
the outdoor air is contaminant free. To be conservative, we have also
ignored other building characteristics that would affect the estimated
indoor pollutant concentration. In the following, we attempt to remedy
both assumptions. This discussion is based on modeling equations
presented in EPA/400/3-91/003, "Introduction to Indoor Air Quality - A
Reference Manual." (This manual is a rich source of information.)
A-29
-------
The generalized mass balance equation for indoor air is:
dC,. = (1 - Fb) i/C0 + S - im/C,. - \_ - flFC,. (29)
dt kV kV kV
where,
C,. = indoor concentration (mass/volume);
Fb = fraction of outdoor concentration intercepted by the build-
ing envelope and not mixing with indoor air (dimensionless
fraction);
v = air exchange rate (I/time);
C0 = outdoor concentration (mass/volume);
S = indoor source generation rate (mass/time);
V = actual indoor volume;
kV = effective indoor volume where k is a dimensionless fraction;
A = decay rate (mass/time);
q = flow rate through air cleaning device (volume/time);
F = efficiency of the air cleaning device (dimensionless frac-
tion);
T = time; and
m = mixing factor.
Fb is included because not all outdoor air contaminants that move
into a structure reach the inside. Some fraction, Fb, is intercepted by
the cracks and crevices in the building envelope which decreases the
amount that actually reaches the indoor air.
KV, the effective indoor volume, is included because the actual
indoor volume available for contaminants dispersal depends on the degree
of air circulation. K is less than one when there is no forced mixing
and the degree of circulation depends on thermal gradients indoors.
X, the decay rate, is included to allow accounting for losses due
to indoor chemical reactions, and adsorption of contaminants on indoor
surfaces. Removal of contaminants by mechanical means through air
cleaning devices is dependent on flow through the device and F, the
fractional efficiency for removal of the particular contaminant.
A-30
-------
The mixing factor m is the ratio of the concentration of the
exiting air to the concentration of the indoor air. When the two
concentrations are the same, the air is said to be completely mixed, and
m equals one. When contaminants are exhausted directly from their
source, m will be >1. The complete mixing assumption is often used to
estimate average concentrations over time periods greater than 1 hour.
Equation 29 was developed for ambient air infiltration and
additional contaminant sources and sinks inside the structure. However,
it appears that, with little error, since soil gas comprises a very
small fraction of the indoor air, the source term S could be replaced
with the equivalent term from soil gas modeling.
For the simple case where X, F, and Fb equal zero and m = 1,
equation 29 reduces to :
dt
- c.)
kV
and the equilibrium concentration (when dC,./dt = 0) would be given by:
C, = C0 + S .
ki/V (30)
then if C0 = 0 and K = 1, equation 30 is identical to equation 21.
A.4 SOURCES OF DATA FOR MODEL PARAMETERS
The models presented in this Appendix require a variety of input
parameters for chemical and soil properties. Listed below are sources of
information for many of the more important parameters.
Model Parameter
Henry's Law constant
Diffusion Coefficient
Soil Porosity
Reference Number(s)
1, 25, 26, 27, 28
1, 5, 25, 29, 31
5, 30
A-31
-------
REFERENCES
1 Air/Superfund National Technical Guidance Study Series, Volume
II - Estimation of Baseline Air Emissions at Superfund Sites.
EPA-450/l-89-002a, August 1990.
2 Nazaroff, W.W., W.A. Jury, V.C. Rogers, "Enclosed-Space Vapor
Models - Technical Panel Report," 14 February 1991. Prepared for
Department of the Army as a review of the Rocky Mountain Arsenal
Human Health Exposure Assessment prepared by Ebasco Services, Inc.
under contract DAAA15-88-D-024, September, 1990.
3 Karickhoff, S.W., D.S. Brown and T.A. Scott. 1979. Sorption of
Hydrophobic Pollutants on Natural Sediments and Soils. Water
Research 13:241-248.
4 Karickhoff, S.W. 1981. Semi-Empirical Estimation of Sorption of
Hydrophobic Pollutants on Natural Sediments and Soils. Chemo-
sphere 10:833-846.
5 Superfund Exposure Assessment Manual, EPA 540/1-88/001, U.S. EPA
Office of Remedial Response, Washington, D.C.. April 1988.
6 The section is adapted from Appendix X of the Draft Remedial
Investigation Report for the Redwing Carriers, Inc-Saraland
Apartments Site, Alabama, February 1992.
7 Jury, W.A., W.F. Spencer, and W.J. Farmer. 1983. Behavior assess-
ment model for trace organics in soil. I. Model description. J.
Environ. Qual. 12(4):558-564.
8 U.S. Environmental Protection Agency (USEPA). Office of Research
and Development. Exposure Assessment Group. 1986. Development of
advisory levels for poly-chlorinated biphenyl (PCB) cleanup.
Washington, D.C. EPA/600/6-86-002.
9 Bomberger, D.C., J.L. Gwinn, W.R. Mabey, D. Tuse and T.W. Chou.
1982. Environmental Fate and Transport at the Terrestrial-Atmo-
spheric Interface. Symposium on models for predicting fate of
chemicals in the environment. American Chemical Society, Division
of Pesticide Chemistry 184th National Meeting, Kansas City, MO.
10 Nero, A.V., M.B. Schwehr, W.W. Nazaroff, and K.L. Revzan (1986).
Distribution of Airborne Radon-222 Concentrations in U.S. Homes,
Science. 234, 992-997.
11. Cohen, B.L. (1986). A national survey of 222Rn in U.S. homes and
correlating factors, Health Physics. 51, 175-183.
A-32
-------
12. Nero, A.J. Gadgil, W.W. Nazaroff, and K.L. Revzan (1990). "Indoor
Radon and Decay Products: Concentrations, Causes, and Control
Strategies", Report DOE/ER-0480P, U.S. Department of Energy,
Office of Health and Environmental Research, Washington, D.C.,
November 1990.
13. Alter, H.W. and R.A. Oswald (1987). Nationwide distribution of
indoor radon measurements: A preliminary data base, Journal of
the Air Pollution Control Association. 37, 227-231.
14. Cohen, B.L. (1987). "Radon: A Homeowner's Guide to Detection and
Control", Consumer Reports Books, Consumers Union, Mount Vernon,
New York, p. 61.
15 Mueller Associates, Syscon Co., and Brookhaven National Laborato-
ry, "Handbook of Radon in Buildings," Hemisphere Publishing Co.,
New York, NY., 1988.
16 Nazaroff, W. W. and S. M. Doyle, "Radon Entry into Homes Having a
Crawl space," Health Physics. 48: 265-281.
17 American Society of Heating, Refrigerating and Air Conditioning
Engineering (ASHRAE), ASHRAE Handbook: 1981 Fundamentals. New
York, 1981.
18 Turk, B. H..; Prill, R. J.; Fisk, W. J.; Grimsrud, D. T.; Moed,
B. A.; Sextro, R. G. In Proceedings of the 79th Annual Meeting of
the Air Pollution Control Association; Minneapolis, MN; Air
Pollution Control Association: Pittsburgh, PA, 1986; Paper No.
86-43.2.
19 Nazaroff, W. W.; Lewis, S. R.; Doyle, S. M.; Moed, B. A.; Nero,
A. V., Environmental Science and Technology. 1987, 21, 459-466.
20 Garbesi, Karina, and R. G. Sextro, "Modeling and Field Evidence of
Pressure-Driven Entry of Soil Gas into a House through Permeable
Below-Grade Walls," Environmental Science and Technology. 23:
1481-1487.
21 Loureiro, C. 0.; Abriola, L. M.; Martin, J. E.; Sextro, R. G.
Environmental Science and Technology. 1990, 24: 1338-1348.
22 Johnson, P. C., and R. A. Ettinger, "Heuristic Model for Predict-
ing the Intrusion Rate of Contaminant Vapors Into Buildings,"
Environmental Science and Technology. 1991, 25: 1445-1452.
23 U. S. EPA, "Introduction to Indoor Air Quality - A Reference
Manual," EPA/400/3-91/003, July 1991.
24 U. S. EPA,"Radon Reduction Techniques for Detached Houses,
Technical Guidance," EPA/625/5-87/019, January 1988.
A-33
-------
25 U. S. EPA, "Hazardous Waste Treatment, And Disposal Facilities
(TSDF) -- Air Emission Models. EPA-450/3/87-026.
26 Thibodeau, L.J., and S.T. Hwang, "Landfarming of Petroleum
Wastes - Modeling the Air Emission Problem. Environmental
Progress, l(l):42-46, 1982.
27 Macay, D., and P.J. LeinOnen, "Rate of Evaporation of Low-
Solubility Contaminants from Water Bodies to Atmosphere.
Environmental Science and Technology, 9(13): 1178-1180,
1975.
28 Mackay, D. and W.Y. Shiu, "A review of Henry's Law Constants
for Chemicals of Environmental Interest". Journal of Physi-
cal Chemistry Reference Data,10(4):1175-1199, 1981.
29 Perry, R.H., and C.H. Chilton, Chemical Engineer's Handbook.
Sixth Edition, McGraw-Hill Book Company, Inc. New York, NY.
30 Brady, N. C. The Nature and Properties Of Soils. Eighth
Edition, McMillian Publishing Company, Inc., New York.
31 Lyman, W.J., Reehl, W. F., Rosenblatt, D. H., Handbook of
Chemical Property Estimation Methods. McGraw-Hill Book
Company, Inc. New York. 1982
A-34
-------
APPENDIX B. MONITORING METHODS
-------
TABLE OF CONTENTS
APPENDIX B. MONITORING METHODS
B.I Methods for Use Outside Buildings B-l
B.I.I Organics in Ambient Air B-l
B.I.2 Particulars B-4
B.I.3 Organics in Soil Gases B-14
B.I.4 Soil Permeability to Gas Flow B-16
B.2 Methods for Use Inside Buildings B-17
B.2.1 Pollutant Measurement Methods B-17
B.2.2 Air Infiltration Rates B-19
B.3 Indoor Air Pollutants B-21
-------
APPENDIX B. MONITORING METHODS
In this section information is provided on monitoring methods
relevant to an indoor air impact assessment and types of compounds
typically found in indoor environments. These methods include those
applicable to measuring pollutants outside the building which may enter,
as well as indoor monitoring methods. The general applicability and
procedures are described. References are provided for details of the
methods.
The information provides only a summary of the relevant methods.
Successful application requires skilled professionals and rigorous QA/QC
programs.
B.I METHODS FOR USE OUTSIDE BUILDINGS
Pollutants outside a building may enter through many openings in
the structure. This includes organics and particulates in the ambient
air and gases in the soil in the immediate vicinity of the building.
Outdoor measurements needed to obtain data required in the various
approaches of Section 2 include:
• Temperature
• Wind velocity
• Barometric pressure
• Organics concentrations in ambient air
• Particulate matter in ambient air
• Chemical concentrations in soil gases
• Soil permeability to gas flow
Methods for the latter four are described here. It should be noted
that in some cases it may be preferable to use an indoor method outdoors
to achieve consistency in measurements.
B.I.I Organics in Ambient Air
Because buildings have a relatively high rate of air exchange with
the outside air (air exchange rates of 0.5 to 1 building volume per hour
are typical for a building with windows and doors closed), pollutant
concentrations in ambient air can be significant in an assessment.
B-l
-------
Ambient air may contain many pollutants not associated with the nearby
Superfund site (e.g., auto exhaust, lawn chemicals, industrial emis-
sions, etc.). It is important to recognize that, unless an assessment
is to be made for background risks also, the sampling method and
analytical finish need only address known or suspected site related
pollutants.
Whichever of the following sampling methods is selected, the
equipment should be located so as to minimize collection of pollutants
from nearby sources such as automobiles, gasoline powered equipment, and
oil storage tanks. It should be located on the windward side of the
building away from windshields such as trees and bushes. The intake
should be about 5 feet above ground level (approximately at the midpoint
of the ground level story of the building) and 5 to 15 feet away from
the building.
Ambient air sampling should begin between 1- and 2-hours before
indoor air sampling begins. This is recommended because concentration
fluctuations during this period will be reflected inside the building
for several hours since the building effectively acts as an equalization
chamber. Likewise, for monitoring periods exceeding 4 hours, consider-
ation may be given to terminating ambient air sampling up to 30 minutes
before terminating indoor sampling.
Very complex monitoring situations can occur if ambient air
transport of Superfund site emissions is of concern. In these cases
additional sources emitting compounds of interest may exist upwind of
the Superfund site as well as between it and the building being investi-
gated. In these cases multiple monitoring stations and site specific
modeling could be required to distinguish the impact of the site from
the other sources. Information on various monitoring techniques for
these complex situations can be found in Section 4.3 of EPA 450/1-89-
002a (volume II of the Air/Superfund NTGS Series).
As a special case, it should be recognized that if the ambient air
pathway is the only pathway of concern for Superfund site impact from
non-particulates on the indoor air quality, there is no reason to
conduct indoor monitoring. That is because the average long-term
concentration of Superfund site related chemicals (except possibly
B-2
-------
participates) would be the same as the outdoor average at the building
location.
B.I. 1.1 Methods
In general EPA Methods TO-1, TO-14, IP-1A, or IP-IB will encompass
the types of organics normally associated with a Superfund site.
However, certain specific types of compounds (e.g., pesticides) may
require specific methods. All "TO-x" methods can be found in "Compendi-
um of Methods for the Determination of Toxic Organic Compounds in
Ambient Air," EPA 600/4-84-041 and supplements EPA 600/4-87-006 and EPA
600/4-89-018. All "IP-x" methods can be found in "Compendium of Methods
for the Determination of Air Pollutants in Indoor Air," EPA 600/4-90-
010. It should be noted that the methods in the latter reference are
not currently certified and should not be regarded as officially
recommended or endorsed by EPA. One source of these and other methods
is the U.S. EPA Air Methods Database prepared by the EPA Environmental
Response Team. Information about this PC-based software package can be
obtained by calling 1-800-999-6990.
TO-14 and IP-1A Methods
Both of these methods use pre-evacuated SUMMA® passivated stainless
steel containers to collect whole air samples. The canisters can be -
used to collect grab samples or time integrated samples. For most
efforts related to indoor air monitoring, time integrated samples are
needed.
For time integrated samples a flow restricting device must be used
to control sampling rate. A mass flow controller is preferred for this
use because a critical orifice flow restrictor results in decreasing
flow rate into the canister as the canister pressure approaches atmo-
spheric. Greater sampling volumes can be obtained by using a metal
bellows type pump to boost final canister pressures to as high as 30
psi.
This method allows convenient integration of samples over a
specified time period, remote sampling and central analysis, ease of
shipping and storing samples, unattended sample collection, and collec-
tion of sufficient volumes to allow assessment of measurement precision
and/or analysis of samples by several analytical systems. Contamination
B-3
-------
is a critical issue with canister-based systems because the canister is
the last element in the sampling train. Thus, as with all sampling
systems, care must be used in the selection, cleaning, and handling of
both canisters and associated sampling apparatus.
Although the method has been used for many VOCs and SVOCs (i.e.,
compounds with vapor pressures greater than 10"7 mm Hg at 25'C), it has
been validated for less than 40 compounds (See Table B.I). Furthermore,
minimal documentation is currently available demonstrating VOC stability
in subatmospheric pressure canisters.
One of the primary advantages of the canister methods is that the
sample volume that can be collected is not limited by the "break
through" capacity as are methods using solid sorbents. This is particu-
larly advantageous when sampling in areas of unknown concentrations or
where concentrations are varying during sampling.
Canister samples are analyzed using a high resolution gas chromato-
graph coupled to one or more appropriate detectors. Although non-
specific detectors such as flame ionization (FID), electron capture
(ECO), or photoionization (PID) may be less expensive per analysis and
in some cases more sensitive than specific detectors, they suffer from
variable specificity and sensitivity, non-positive compound identifica-
tions, and lack of resolution of co-eluting compounds. It is strongly
recommended that specific detectors be used for positive identification
and primary quantification. The recommended detectors are mass spec-
trometer operating in either the selected ion monitoring mode (for use
when a specific set of compounds are to be determined) or the SCAN mode
for identification of all compounds.
Methods TO-1 and IP-IB
These methods are based on the collection of VOCs on Tenax® solid
adsorbent. Integrated sampling over periods up to 12 hours can be
performed. The air to be sampled is drawn through the Tenax® loaded
cartridge by small, portable pumps. As with any adsorption method,
knowledge of flow rate and volume sampled are important data.
B-4
-------
TABLE B.I
VOLATILE ORGANIC COMPOUND DATA SHEET8
COMPOUND (SYNONYM)
Freon 12 (Oichlorodifluororaethane)
Methyl chloride {Chloromethane)
Freon 114 (1.2-Oichloro-l.l .2,2-
tetrafluoroethane)
Vinyl chloride (Chloroethylene)
Methyl bromide (Bromomethane)
Ethyl chloride (Cnloroethane)
Freon 11 (Trichlorofluoromethane)
Vtnylldene chloride (1,1-Oichloroethene)
Dichlororaethane (Methylene chloride)
Freon 113 (1.1.2-TrJchloro-1.2,2-
trifluoroethane)'
1,1-Olchloroethane (Ethylidene chloride)
cis-l,2-0ich1oroethylene
Chloroform (Trichloronethane)
1,2-Olchloroethane (Ethylene dichloride)
Methyl chloroform (1,1.1-Trichloroethane)
Benzene (Cyclohexatriene)
Carbon tetrachloride (Tetrachlorcmethane)
1,2-Oichloropropane (Propylene
dichloride)
Trichloroe thy lene (Trichloroethene)
cl s-l,3-01chloropropene (cis-1,3-
dichloropropylene)
trans-l.3-D1chloropropene (cls-1,3-
Oichloropropylene)
1,1.2-Trichloroethane (Vinyl trichloride)
Toluene (Methyl benzene)
1,2-Olbromoethane (Ethylene dibromide)
Tetrachloroe thy lene (Perch loroe thy lene)
Chlorobenzene (Phenyl chloride)
Ethyl benzene
n-Xylene (1,3-Oimetnylbenzene)
•p-Xylene (1,4-Oimethylxylene)
Styrene (Vinyl benzene)
•1,1,2,2-Tetrachloroethane
o-Xylene (1.2-Oiraethylbenzene)
.1.3,5-Trimethylbenzene (Mesitylene)
1,2,4-Trinethylbenzene (Pseudocumene)
.Bi-01 Chlorobenzene (1,3-Oichlorobenzene)
Benzyl chloride («-Chlorotoluene)
o-01chlorobenzene (1,2-Dichlorobenzene)
p-Oichlorobenzene (1,4-Olchlorobenzene)
1,2,4-Trichlorobenzene
Hexachlorobutadiene (1.1,2,3.4,4-
Hexach1oro-l,3-butadiene)
FORMULA
C12CF2
CH3C1
C1CF2CC1F2
CH2-CHC1
CH38r
CH3CH2C1
CC13F
C2H2C12
CH2C12
CF2C1CC12F
CH3CHC12
CHC1-CHC1
CHC13
C1CH2CH2C1
CH3CC13
CC14
CH3CHC1CH2CI
C1CH«CC12
CH3CC1-CHC1
C1CH2CH-CHC1
CH2C1CHC12
BrCH2CH2Br
C12C-CC12
C6HSC1
1.3-(CH3)2C6H4
1,4-(CH3)2C6H4
CHC12CHC12
1.3.S-(CH3)3C6H6
1.2,4-(CH3)3C6H6
1.3-C12C6H4
1,2-C12C6H4
1.4-C12C6H4
1,2.4-C13C6H3
MOLECULAR
WEIGHT
120.91
50.49
170.93
62.50
94.94
64.52
137.38
96.95
84.94
187:38
98.96
96.94
119.38
98.96
133.41
78.12
153.82
112.99
131.29
110.97
BOILING
POINT (*C)
-29.8
-24.2
4.1
-13.4
3.6
12.3
23.7
31.7
39.8
47.7
57.3
60.3
61.7
83.5
74.1
80.1
76.5
96.4
87
76
KLTING
POINT CC)
-158.0
-97.1
-94.0
-1538.0
-93.6
-136.4
-111.0
-122.5
-95.1
-36.4
-97.0
-80.5
-63.5
-35.3
-30.4
5.5
-23.0
-100.4
-73.0
CAS
NUMBER
74-87-3
75-01-4
74-83-9
75-00-3
75-35-4
75-09-2
74-34-3
67-66-3
107-06-2
71-55-6
71-43-2
56-23-5
78-87-5
79-01-6
110.97
112.0
133.41
92.15
187.88
165.83
112.56
106.17
106.17
106.17
104.16
167.85
106.17
120.20
120.20
147.01
126.59
147.01
147.01
181.45
113.8
110.6
131.3
121.1
132.0
136.2
139.1
138.3
145.2
146.2
144.4
164.7
169.3
173.0
173.3
180.5
174.0
213.5
•36.5
•95.0
9.8
•19.0
•45.6
•95.0
•47.9
13.3
•30.6
36.0 '
•25.2
•44.7
•43.8
24.7
39.0
17.0
53.1
17.0
79-00-5
108-88-3
106-93-4
127-18-4
108-90-7
100-41-4
100-42-5
79-34-5
108-67-8
95-63-6
541-73-1
100-44-7
95-50-1
106-46-7
120-82-1
a - EPA 600/4-90-010
B-5
-------
The method offers some advantages over the canister methods but has
significant limitations and overall is complex and difficult to use.
The primary, if not only, advantage is that the method has been validat-
ed for more compounds than the canister methods (See Table B.2). Note
that although small, portable (even personal sampling) pumps may be
used, mass flow controllers are recommended to ensure flow stability. A
10% deviation in initial and final flow rates may invalidate data
collected using cartridges with that pump.
Limitations of the method include:
• Breakthrough volumes of some compounds, such as vinyl chloride,
are very low. The lowest breakthrough volume limits the volume
of sample that can be collected. (See Table B.3).
• High benzene backgrounds from Tenax® are common
• Sensitive to high humidity
• Contamination with compounds of interest (e.g., benzene,
toluene, chloroform, methylene chloride, etc.) commonly encoun-
tered problem
• Variable desorption rates for compounds adsorbed on adsorbent
• Necessary to know approximate range of contaminate concentra-
tions
• Overall accuracy of method is unknown.
The analytical finish for this method is similar to that for the
canister method above. The primary differences are that the compounds
must be thermally desorbed from the Tenax® before injecting it into the
gas chromatograph and that each cartridge can be analyzed only once. It
is strongly recommended, therefore, that the mass spectrometer be
operated in the SCAN mode.
B-6
-------
TABLE B.2
COMPOUNDS IDENTIFIED AND QUANTIFIED BY AUTOMATED
GC-MS-DS PROCEDURE WITH TYPICAL DETECTION LIMITS IN
FULL SCAN MODE8
Quantitatlon Detection Limits
Compound Mass fm/z) fnq)
perfluorotoluene (internal standard)* 217 0.3
benzene* ' 78 2.6
methyl benzene* 91 2.0
1,2-dimethylbenzene* 106 0.5
1,3,5-trimethylbenzene 120 2.5
ethylbenzene* 91 1.6
ethylbenzene* 104 1.7
(1-methylethyl) benzene 105 1.1
butylbenzene 91 0.7
l-methyl-4-(l-methylethyl) benzene 119 4.0
chlorobenzene* 112 1.7
bromobenzene 156 14.1
1,2-dichlorobenzene* 146 12.4
l-ethenyl-4-chlorobenzene 138 2.0
trichloromethane . 83 2.7
tetrachloromethane* 82 2.1
bromochloromethane* 130 2.1
bromotrichloromethane* 163 1.6
dibromomethane* 174 4.5
tribromomethane* 171 8.5
1,1-dichloroethane* 63 5.7
1,2-dlchloroethane 62 3.8
1,1,1-trichloroethane* 99 1.7
1,1,2-trichloroethane* 85 2.1
1,1,1,2-tetrachloroethane 31 0.9
1,1,2,2-tetrachloroethane 83 6.5
pentachloroethane* 167 1.8
1,1-dichloroethane* 961 6.9
trichloroethene* 132 0.8
tetrachloroethene 166 2.6
bromoethane* .108 7.8
1,2-dibromoethane* 107 3.3
1-chloropropane* 42 1.7
2-chloropropane* 43 3.4
1,2-dichloropropane 63 4.0
1,3-dichloropropane 76 9.6
1,2,3-trichloropropane 753 4.7
l-bromo-3-chloropropane 158 1.6
3-chloro-l-propene 41 1.6
B-7
-------
Compound
1,2-dibromopropane*
2-chlorobutane
1,3-dichlorobutane
1,4-dichlorobutane
2-3-dichlorobutane*
l,4-dichloro-2-butane (cis)
3,4-dichloro-l-butane
tetrahydrofuran
1,4-dioxane
l-ch!oro-2,3-epoxypropane
2-chloroethoxyethene
benzaldehyde*
acetophenone
benzonitrile
ISOMER GROUPS
TABLE B.2 -- continued
Quantitation
Mass (m/z)
121
57
55
55
90
752
75
72
88
71
631
77
105
103
1,3- & OR 1,4-dimethylbenzene
1,2- & OR 1,3-dichlorobenzene*
2- & OR 3- & OR 4-chloro-l-
methylbenzene*
SURROGATE GROUPS AND INTERNAL STANDARDS
4-bromofluorobenzene
chlorobenzene-ds
1,4-dichlorobenzene
1,4-di f1uorobenzene
(BFB)
106
146
126
95
117
150
114
Detection Limits
(ng)
14.4
3.5
0.5
8.2
5.1
1.9
6.5
1.2
3.9
8.1
8.2
5.9
2.9
1.3
0.5
1.3
0.5
* Compounds used to calibrate GC-MS-DS on a daily basis either by direct
injection or on spiked adsorbent tubes.
a - EPA 600/4-90-010
B-8
-------
TABLE8 B.3
BREAKTHROUGH VOLUMES6 AND SAFE SAMPLING VOLUMES6 FOR
TENAX-GC AND TENAX-TA
Acetaldehyde
Acrolein
Acrylonitrile
Ally! chloride
Benzene
Benzyl chloride
Bromobenzene
Carbon tetrachloride
Chlorobenzene
Chloroform
Chloroprene
Cresol
p-Dichlorobenzene
1,4-Dioxane
Ethylene dibromide
Ethylene dichloride
Ethylene oxide
Formaldehyde
Hexachlorocyclo-
pentadiene
Methyl bromide
Methyl chloroform
Methylene chloride
Nitrobenzene
Perch!oroethylene
Phenol
Propylene oxide
Trichloroethylene
Vinyl chloride
Vinylidene chloride
Xylene
Tenax-GC
Tenax-TA
breakthrough breakthrough
volume6 volume6
38*C
0.6
4
-
-
19
300
300
8
150
8
-
440
510
-
60
-
-
-
-
0.8
-
3
-
-
-
3
21
0.6
-
200
20*C
0.6
5
8
8
36
440
"27
184
13
26
570
820
58
77
29
0.5
0.6
2000
0.8
9
5
520
100
300
3
45
.06
4
177
35'C
0
2
3
3
15
200
13
75
5
12
240
330
24
35
12
0.3
0.2
900
0.4
4
2
240
45
140
1
17
.03
2
79
Tenax-GC
safe sampling
volumec
38'C
0.3
1.7
8.2
130
130
3.5
6.5
4
191
221
87
26
0.4
1.5
1.5
8.5
.03
89
Tenax-TA
safe sampling
volume0
20'C 351C
2
3
3
14
175
11
5
5
10
230
290
23
30
12
800
3
2
200
40
120
1
18
2
70
1
1
6
80
5
2
2
5
95
130
10
14
5
360
2
95
18
55
7
32
"EPA 600/4-90/010
Breakthrough volumes expressed as liters/gram of sorbent.
°Safe sampling volume = {[Breakthrough volume (L/g)]/1.5) x 0.65 grams of
sorbent.
dBreakthrough volumes for other chemicals can be extrapolated on the basis of
boiling points for chemicals in the same chemical class.
B-9
-------
Other Methods for Specific Organic Classes
Other site related organic compounds of potential interest include
aldehydes and ketones, pesticides, and polynuclear aromatic hydrocar-
bons.
Aldehyde and ketone concentrations may be determined by either EPA
Methods TO-5 or IP-6A. The methods have different sampling techniques
but similar analysis techniques (both'use high pressure liquid chroma-
tography (HPLC)). In Method TO-5, ambient air is drawn through a midget
impinger sampling train containing 10 ml of a 2N HC1/0.05% 2,4-dini-
trophenylhydrozine (DNPH) reagent. Aldehydes and ketones react with the
DNPH to form stable derivatives which can be separated using HPLC.
Method IP-6A uses a prepackaged silica gel cartridge coated with
acidified DNPH. The relatively high pressure drop across the cartridges
(»30 inches of water) limits flow rates to about 1.5 liters/minute.
Some experimentation with HPLC operating conditions is necessary for
good compound separation. Table B.4 gives sensitivities versus sampled
volume.
Pesticides may be determined using EPA Methods 608 (EPA 600/4-82-
057) or Method IP-8. The latter is preferred for the current purpose
and is briefly described. Ambient air is drawn through a cartridge
filled with polyurethane foam (PUF) at a rate of 1 to 5 liters/minute.
Concentrations of 0.01 to 50 /ig/m3 can be determined with sampling
periods of 4 to 24 hours. The PUF is recovered and extracted with an
ether/hexane mixture. The extract is concentrated and analyzed using
gas-liquid chromatography with electron capture detector. Compounds
listed in Table B.5 have been determined with this method.
Polynuclear aromatic hydrocarbons (PAHs) may be determined using
EPA Method TO-13 or IP-7. Method TO-13 is a high volume method applica-
ble only to outdoor environments. IP-7 is a low volume method but has
detection capabilities adequate for indoor air impact assessments. The
method is applicable to PAHs adsorbed on particulates as well as in the
vapor phase. It also can be adapted for use with a PM-10 cut-point
sampling nozzle.
In Method IP-7, air is drawn through a filter backed by either an
XAD-2 or PUF filled cartridge. Up to 30m3 can be sampled at a rate of
B-10
-------
TABLE B.4
SENSITIVITY (ppb, v/v) OF SAMPLING/ANALYSIS USING
ADSORBENT CARTRIDGE FOLLOWED BY HPLC
Sample Volume, I
Compound
Formaldehyde
Acetaldehyde
Acrolein
Acetone
Propionaldehyde
Crotonaldehyde
Butyraldehyde
Benzaldehyde
Isovaleraldehyde
Valeraldehyde
o-tolualdehyde
tn-tolualdehyde
p-tolualdehyde
Hexanaldehyde
2.5-dimethyIbenzaIdehyde
10
20
30
40
50
100 200
300
400 500 1000
1.45
1.36
1.29
1.28
1.28
1.22
1.21
1.07
1.15
1.15
1.02
1.02
1.02
1.09
0.97
0.73
0.68
0.65
0.64
0.64
0.61
0.61
0.53
0.57
0.57
0.51
0.51
0.51
0.55
0.49
0.48
0.45
0.43
0.43
0.43
0.41
0.40
0.36
0.38
0.38
0.34
0.34
0.34
0.36
0.32
0.36
0.34
0.32
0.32
0.32
0.31
0.30
0.27
0.29
0.29
0.25-
0.25
0.25
0.27
0.24
0.29
0.27
0.26
0.26
0.26
0.24
0.24
0.21
0.23
0.23 •
0.20
0.20
0.20
0.22
0.19
0.15
0.14
0.13
0.13
0.13
0.12
0.12
0.11
0.11
0.11
0.10
0.10
0.10
0.11
0.10
0.07
0.07
0.06
0.06
0.06
0.06
0.06
0.05
0.06
0.06
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.03
0.03
0.03
0.04
0.03
0.04
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.02
0.03
0.03
0.03
0.03
0.03
0.02
0.02
0.02
*0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
Note: ppb values are measured at 1 atra and 25'C; sample cartridge Is eluted with 5 ml acetonitrile,
and 25 ml are injected onto HPLC column.
Note: Maximum sampling flow through a DNPH-coated Sep-PAK* cartridge is about 1.5 L per minute.
B-ll
-------
TABLE B.5
PESTICIDES DETERMINED BY GAS CHROMATOGRAPHY/
ELECTRON CAPTURE DETECTOR (GC-ECD)
Aldrin
BHC (a-and /3-Hexa-
chlorocyclohexanes)
Captan
Chlordane, technical
Chlorothalonil
Chlorpyrifos
2,4,-D esters
fi.fi,-DDT
fi.fi,-DDE
Dieldrin
Dichlorvos (DDVP)
Dicofol
2,4,5-Tri chlorophenol
Folpet
Heptachlor
Heptachlor epoxide
Hexachlorobenzene
Lindane (r-BHC)
Methoxychlor
Mexacarbate
Hirex
trans-Nonachlor
Oxychlordane
Pentachlorobenzeiie
Pentachlorophenol
Ronnel
B-12
-------
up to 20 liters per minute without significant breakthrough. Because
some of the PAH collected by the filter may volatilize and be collected
on the adsorbent, the two should not be analyzed separately. The filter
and adsorbent are extracted in Soxhlet extractors, cleaned up with
silica gel column chromatography and analyzed using gas chromatography
(with FID or MS detector) or by HPLC. In general, MS operated in SCAN
mode is preferred because of the positive compound identifications.
B.I.2 Particulates
Methods are available to determine total suspended participates and
respirable (<10/xm) particles. In many cases only the latter will be of
concern. However, in some cases consideration may be given to the
larger particulate. Those particles that penetrate the building
envelope will rapidly deposit on indoor surfaces. Exposure to these
particulates may occur through dermal contact, ingestion, or inhalation
if resuspended by occupant traffic or during such normal activities as
cleaning.
Total suspended particles may be determined by the High-Volume
Method given in Section 2.2 of "Quality Assurance Handbook for Air
Pollution Measurement System: Volume II - Ambient Air Specific Meth-
ods," EPA 600/4-77-027a. Ambient air is drawn into a covered housing
through a filter (nominal 8 x 10 inch), with a collection efficiency of
at least 99 percent for particle size 0.3pi, at 1.1 to 1.7 m3/min.
Particles up to 50 jwn (aerodynamic diameter) collect on the filter
surface. Air containing up to 750 #g/m3 may be sampled for up to 24
hours at this rate. Particulate concentration is determined by dividing
air volume sampled into the weight of particulate collected by the
filter. If should be noted that because of the high air volume sampled,
the method is not applicable to indoor air sampling.
Method IP-10A can be used to determine the concentration of
particulate in both the <2.5/im and 2.5 to 10/wn ranges. The method
contains both a fixed site monitor and a personal sampler. Only the
fixed site monitor is recommended here for outdoor ambient air monitor-
ing. The monitor operates on the principal of impaction. The horizon-
tal slotted inlet prevents very large particles from entering. Either a
one-stage or two-stage impactor assembly can be used to trap particles
B-13
-------
above 10/im or 2.5/zm, respectively. Participate passing the impactor
section is collected on a filter for subsequent weighing. Note that
particles in both size ranges cannot be collected simultaneously with a
single monitor. It is possible to connect two sampling heads to the
same pump assembly to simultaneously collect in both size ranges. The
limit of detection for a 12 hour sampling period at the specified
sampling rate of 4 liters per minute is 4/ig/m3.
The sampler does not use a critical orifice to control flow rate.
Thus, a feedback control from a mass flow meter is required to vary pump
speed in order to maintain flow constant with ±5 percent.
B.I.3 Organics in Soil Gases
Knowledge of the chemical concentrations in soil gases near a
building will be important when site impact via the underground pathway
is of concern. These situations can arise when soil gases migrate
directly from the site (e.g., landfills), the nearby soil is contaminat-
ed, or the underlying groundwater is contaminated. As discussed in
Section 2, it is important in these cases to determine concentrations in
soil gas very near the building (i.e., in the region between 0.5 and 1
meter from the underground walls and floor). Gases in this soil region
are most likely to diffuse through building cracks or be drawn into the
building as a result of building underpressurization.
Because soil properties are inhomogeneous, both vertically and
horizontally, it is expected that concentrations will vary around the
building. Therefore, soil gas concentrations must be measured at
several locations around the building. Following the approach used in
Appendix C of Volume II of the Air/Superfund NTGS Series (EPA-450/1-89-
002a), it is estimated that a minimum of two sampling points should be
used on each side of the building (typical residential building). It is
preferred that one additional sample be collected from directly under
the building (approximate center). It may not be practical to collect
this sample for slab-on-grade or basement constructions because a hole
would have to be made in the floor. Efforts to collect this sample
should be made, if possible. Buildings of this type generally have a
very permeable zone (1 to 4 inches in depth) immediately below the slab
B-14
-------
due either to the use of a gravel bed underlying the slab or gaps formed
by soil not making continuous contact with the slab.
The preferred technique for collecting soil gas samples is the soil
vapor (ground) probe. The installation of probes is simple. A small
diameter pipe (=0.5 inch OD) is inserted into the ground to the desired
depth either by hammering or driving it down a slightly undersized pre-
augured hole. The use of pre-augured holes is recommended if soil
permeabilities, described below, are to be measured. Retractable
centering probe tips should be used to keep soil out of the probe. The
probe should be withdrawn a minimum of two inches to expose soil at the
probe tip.
A minimum of two probe volumes of soil gas should be extracted
before samples are taken. Because soil gas concentrations must be
fairly elevated if indoor air impact is to be of significant concern,
low detection limits are not typically required. (Soil gas intrusion
rates into a building are typically less than 5 percent of ambient air
infiltration rates.) However, positive compound identification and
accurate concentrations relative to other measured compounds are needed.
It is preferred that integrated samples be collected over at least
one hour when indoor air monitoring is not being also conducted. If
indoor air monitoring is also being conducted, integrated samples over
the same time period should be collected. (Although soil gas concentra-
tions should not vary significantly over this time period, some varia-
tion is possible and it is best to eliminate this uncertainty.)
In most cases, canister sample collection and analysis procedures
such as those given in EPA Methods TO-14 or IP-1A are preferred because
prior knowledge of actual concentrations is not required and concerns
over breakthrough volumes are eliminated. However, professional
judgement is required in the selection of technique based on known or
suspected contamination at the site. For example, it may be possible to
adapt IP-7 for pesticides or IP-8 for PAHs if soil contamination in the
immediate vicinity of the building is known or suspected. Sampling flow
rates, and, thus, sample volumes, would need to be kept low to reduce
concern about drawing in surface air.
B-15
-------
B.I.4 Soil Permeability to Gas Flow
The permeability of soil to gas flow can be an important indicator
of the rate at which soil gas can enter a building. If the permeability
is low (i.e., below 10~8cm2), it is unlikely building underpressures
have much effect on the rate of intrusion and gases will enter primarily
by diffusion. In this case, soil gas concentrations below the building
may be higher than estimated from modeling equations. As the permeabil-
ity increases, however, building pressure becomes more important and may
result in gases several meters from the building being drawn inside.
This effect reduces the soil gas concentration (relative to model
assumptions) at the soil-building interface but increases the intrusion
rate thus resulting in higher indoor concentrations. In essence, this
is the reason predictive models assume a soil gas concentration of zero
at the basement floor-soil interface.
Permeabilities are easy to measure using the soil gas probes
discussed above. Permeabilities should not be measured until after soil
gas samples have been obtained because the injected gas will distort
concentration measurements. Permeabilities are determined by measuring
the gas pressure in the probe as a metered flow of air is passed through
the probe into the soil. A cylinder of compressed air, flow meters with
ranges from about 5 to 500 cm3/min., and a differential pressure gauge
with a range of 0 to 125 pascals are needed. Probe pressure should be
measured at three or more flow rates to obtain a good average. The
permeability is calculated, assuming Darcy flow, from
KV = Q u
4wr2Pa
where,
KY = Permeability, m2
Q = Air Flow rate, m3/s
H = Viscosity of air, 1.83 x 10s Kg/m-s
r = Internal radius of probe, M
Pa = Probe pressure in pascals
The probe tip internal diameter should be measured to within ±0.005
inches before inserting into the soil and after extraction. Method
reference is contained in DMSA Action LTD., "Review of Existing Informa-
tion and evaluation for possibilities of research and development of
B-16
-------
instrumentation to determine future levels of radon ^at a proposed
building site.", Report INFO-0096, Atomic Energy Control Board, Ottawa,
Canada (1985).
B.2 METHODS FOR USE INSIDE BUILDINGS
Critical measurements to be made indoors include those to determine
pollutant concentrations and those to determine air exchange rates.
Both are discussed here.
B.2.1 Pollutant Measurement Methods
The air inside a typical building may contain hundreds of compounds
(over 3800 compounds have been identified in tobacco smoke alone). To
assess the potential impacts from a nearby contaminated site, it is only
necessary to monitor for those compounds identified with the site.
Because these compounds are also common in building materials and
consumer products, their detection in the indoor air does not necessari-
ly indicate their origin.
Discussion of appropriate sampler locations is included in Section
2. Principal considerations in locating samplers are:
• Air exchange rates in the area should be large compared to
sampling volume.
• Samplers should be located in the normal breathing zone (i.e.,
3 to 5 feet above floor level).
• At least one sampling location should be established on each
level of the building.
• Samplers should not be located near obstructions, air supply or
return registers, or local pollutant sources.
• Exterior windows and doors should be kept closed as much as
possible. All indoor doors should be open.
• Sampling is best conducted under conditions that yield building
underpressurizations of several pascals. Low rates of air
infiltration and soil gas convective intrusion may result from
low building underpressurizations.
In general it is recommended that indoor air monitoring use methods
listed in the "Compendium of Methods for the Determination of Air
B-17
-------
Pollutants in Indoor Air," EPA-600/4-90/010. In particular it is
recommended that:
• Method IP-1A, Appendix B, use of portable gas chromatograph, be
used for initial screening investigations and to assist in
calculating proper sample size for methods based on use of
solid adsorbents.
• Method IP-1A, canister, be used for collection of most target
VOCs. Analytical finish by GC-MS in SIM or SCAN mode.
• Method IP-IB, Tenax® cartridge, be used only for compounds not
quantifiable by IP-1A. Analytical finish by GC-MS in SCAN
mode.
• Method IP-6A, solid adsorbent, be used for aldehydes/ketones.
Method TO-5 may also be used.
• Method IP-7, PUF/XAD solid adsorbents, be used for PAHs. Note
that if the PUF adsorbent is used, the method may also be used
to determine pesticides. GC-MS in SIM or SCAN mode is the
preferred analytical finish.
• Method IP-8, PUF adsorbent, be used for pesticides. The low
sample volume collected in this method may be inadequate for
PAH quantitation in indoor and ambient air. GC-MS in SIM or
SCAN mode is the preferred analytical finish.
• Method IP-lOa, size specific impactor, be used for airborne
particulate matter. The single stage impactor collecting
particles up to 10/fm aerodynamic diameter is preferred. The
stationary microenvironmental exposure monitor (MEM) is pre-
ferred. Collected particulates may be subjected to additional
analysis if desired.
Specific circumstances may also warrant collection of deposited
dusts/chemicals. The OSHA Method, described in 29 CFR 1910.132(a), may
be used to collect deposited material for characterization. The method
cannot be used for air inhalation estimates. Collection procedure is
simple. A Whatman 42 filter is moistened with an appropriate solvent
and used to wipe a known area (generally about 100 cm2). The filter is
folded, with exposed side in, and folded again to form a 90-degree angle
at the center of the filter. The filter is placed in a clean vial for
transport to the laboratory. A blank filter that is also moistened with
B-18
-------
the solvent should be submitted in a separate vial. Clean gloves,
impervious to the solvent used and potential contaminants, should be
worn during all phases of collection.
B.2.2 Air Infiltration Rates
Knowledge of air infiltration rates is not necessary to determine
site related impacts on indoor air quality for the time period over
which sampling occurred. However, if any estimate is to be made of
likely concentrations at other times, knowledge of the infiltration rate
and building air changes per unit time during the sampling period are
required.
Preferred techniques to make these measurements are IP-4A and IP-4B
which are both tracer gas methods. Indirect measurements can be made
using the fan pressurization/depressurization method (ASTM-779).
Because the ASTM method significantly affects building air infiltration
rates, it cannot be used immediately before (=12 hours) or during indoor
monitoring.
Method IP-4A can be used to obtain the integrated air exchange rate
over the monitoring period. This method uses constant emitting perfluo-
carbon (PFT) sources which must be placed in the building at least 8
hours before monitoring begins. One source is required for about every
500 ft2 of living area. They are temperature sensitive and care must be
used in placement. Samples are collected on activated charcoal
spherules inside capillary adsorption tubes over the entire monitoring
period. The tubes are analyzed by GC-ECD. The method includes cleanup
for other compounds that respond to the ECD detector.
The method is applicable to small and large buildings and can be
used to determine interzonal flows as well as exchange rates. The
method assumes the effective exchange volume of the building is equal to
the physical volume. Poor mixing within the building may effect this
assumption.
Method IP-4B is a tracer concentration decay method. In this
method, a tracer, such as SF6, is injected into the structure, thorough-
ly mixed and its concentration measured over time. The concentration
decreases due to air exchange with outside air. Tracer concentrations
can be measured on-site with a GC-ECD or SF6 specific detectors. Grab
B-19
-------
samples or time-integrated samples for off-site analysis can also be
obtained.
Adequate mixing in the indoor environment is critical to the
success of this method. Local fans and/or heating system fans should be
used to obtain tracer concentration throughout the structure within 5
percent of the average. It is preferred that measurements be made in
several locations on each floor of the building. Tracer injections can
be made at about one-hour intervals to obtain a series of infiltration
measurements. If sufficient difference exists in meteorological condi-
tions during these measurements, it is possible to characterize the
effects of temperature, wind, and structural factors on the infiltration
rates.
Tracer concentration is assumed to follow
C = C0 exp (-IT)
where,
C = Concentration at time T
C0 = Concentration at T = o
I = Air exchange rate
T = Time
The air exchange rate can be obtained from the slope of a LnC versus
time plot.
When using either Method IP-4A or 4B, measurements should be made
of indoor and outdoor temperatures, wind velocity, barometric pressure,
and building underpressurization.
In the ASTM Method, a large fan is mounted on a door connected to
the outside. The fan is used to draw air out of and force air into the
building at several differential pressures. Flow rates are determined
from a fan calibration curve. The effective leakage area is calculated
from
L - Q/(2APp)-°-5
where,
L = Leakage area
Q = Air flow rate
AP = Pressure difference across building shell
p = Density of air
B-20
-------
Infiltration rates and air exchange rates can then be estimated
using
Q = L(AAT + B/i2)0"5 from page 8 of EPA 400/3-91/003
where,
Q = Air infiltration rate, CFM
L = Effective leakage area, in2
AT = Indoor-outdoor temperature difference, °F
/x = Wind speed, mph
A = Stack coefficient
B = Wind coefficient
Values for A and B are in the above document.
B.3 INDOOR AIR POLLUTANTS
Literally hundreds of compounds have been detected in indoor air.
Over 3,800 compounds have been detected in tobacco smoke alone. The
sources of the compounds are many and include building materials,
furnishings, cleaners and waxes, paints, pesticides, hobby supplies,
combustion devices, and personal care products. It has been estimated
(Tancrede et al, "The Carcinogenic Risk of Some Organic Vapors Indoors:
A Theoretical Study", Atmospheric Environment, Vol. 21, No. 10, 1987)
that cancer risks from exposure to indoor air probably well exceeds 1 x
lO'4.
Many of the compounds typically associated with a contaminated site
are also present in typical non-impacted indoor environments. This
significantly complicates monitoring efforts to differentiate site
contributions to indoor air concentrations. The data currently avail-
able indicates that compounds detected and concentrations of those
compounds is a function of many factors including building type (e.g.,
office, residence, school, etc.), age, and location.
Although many studies of indoor air exposures have been made (most
notable is The Total Exposure Assessment Methodology (TEAM) Study - see
EPA 600/6-87-002Q for summary and analysis), it does not appear to be
feasible at this time to tabulate the averages and ranges for individual
compounds in all types of potentially impacted structures in a way that
would be useful for determining site impacts from indoor air concentra-
tions. The TEAM studies do, however, have a substantial amount of
B-21
-------
information on averages and ranges for many compounds in a number of
different building types. This may be considered the best currently
available source of information on typical concentrations.
Examples of some specific compound measurements are contained in
EPA 400/3-91-003, "Introduction to Indoor Air Quality: A Reference
Manual." EPA 600-R-92-02, "Indoor Air Quality Data Base for Organic
Compounds," provides the best current summary of data on organics from
studies over the last decade. This data base contains information on
over 220 organic compounds. The compounds most frequently reported in
the studies reviewed were: formaldehyde, tetrachloroethylene, 1,1,1-
trichloroethane, trichloroethylene, benzene, p-dichlorobenzene, toluene,
ethylbenzene, xylene, decane, and undecane. It should be noted that not
every study monitored for the same compounds or used similar methods.
B-22
-------
APPENDIX C. CASE STUDIES
-------
TABLE OF CONTENTS
APPENDIX C. CASE STUDIES
C.I Buildings on Landfill Sites C-l
C.I.I Quality Assurance/Qua!ity Control . . . C-5
C.I.2 Results C-7
C.I.3 Conclusion Reached C-9
C.I.4 Comments on The Study C-9
C.2 Buildings Near Landfill C-ll
C.2.1 Nature and Extent of the Problem .... C-13
C.2.2 Prevention of Gas in Structures .... C-20
C.2.3 Comments on the Study C-20
C.3 Buildings Over Groundwater Plume (A) C-21
C.3.1 Nature and Extent of the Problem .... C-22
C.3.2 Indoor Air Modeling Results C-30
C.3.3 Comments on Case Study C-30
C.4 Buildings Over Groundwater Plume (B) C-35
C.4.1 Nature and Extent of Problem C-37
C.4.2 Conclusions C-41
C.4.3 Comments on Case Study C-42
C.5 Buildings Over Groundwater Plume (C) C-44
C.5.1 Nature and Extent of the Problem .... C-44
C.5.2 Results C-48
C.5.3 Conclusion C-50
C.5.4 Comments on Study C-51
C.6 Buildings Near Sludge Disposal Pit C-53
C.6.1 Nature and Extent of the Problem .... C-53
C.6.2 Residential Air Monitoring C-57
C.6.3 Comments on Indoor Air Effort C-58
C.7 Fugitive Dust Case Study C-60
C.7.1 Nature and Extent of the Problem .... C-60
C.7.2 Comments on the Study C-66
C.8 Buildings Over Waste Oil Pit C-67
C.8.1 Nature and Extent of Problem C-69
-------
APPENDIX C
CASE STUDIES
The potential impacts of Superfund sites on the indoor air quality
of nearby buildings have been investigated in a number of cases. This
Appendix documents a sample of such cases. The examples cover a range
of site conditions, potential types of indoor air impacts, phases in the
Superfund program, and judgements of investigating personnel.
It should be noted that, in most cases, additional work beyond that
presented here has been conducted. Therefore, this Appendix should be
read in the context that it presents partial results of investigations
and should not be construed as the final analysis of the potential site
impact. The purpose is to provide an overview of the modeling and
monitoring techniques that have been used and the relative success of
these techniques.
It is not intended as a revie of all possible case types. It is
also not intended as documentation that indoor air impacts have been
shown to occur. It is hoped that sufficient detail is provided to allow
site investigations to avoid some of the major pitfalls in assessing
indoor air impacts.
C.I BUILDINGS ON LANDFILL SITES
The Landfill extends over approximately ten acres of land. It
received both municipal and industrial wastes from about 1910 to 1967.
The wastes included drummed industrial wastes and uncontained liquid
wastes. Prior to 1964, open burning took place at the site. After
closure, when the wastes were covered by approximately two feet of soil,
the property was subdivided and sold for both residential and commercial
development. Based on verbal reports of methane buildup in buildings
currently located on the site, a study was conducted in three commercial
buildings in late 1985. The results of the study indicated the presence
C-l
-------
of methane in one of the buildings at levels well below the lower
explosive limit for methane in air, even in the floor cracks where
levels were most highly concentrated. However, the concern existed that
the methane entering buildings acts as a carrier gas for toxic species.
Preliminary soil gas analysis conducted by the regulated party's
contractor utilizing a portable chromatograph indicated the presence of
eleven toxic chemical species.
Therefore, indoor air quality sampling for volatile organic
chemicals was performed at the site. It took place between approximate-
ly 9 AM and 5 PM on a day in September, 1990.
The purpose of this study was to collect data of a quality suffi-
cient to document the presence of any toxic volatile organic chemicals
in buildings constructed on the former landfill site which may pose a
health risk to their occupants.
The buildings sampled were selected on the basis of an on-site
reconnaissance visit conducted jointly by EPA and State personnel as
well as existing data. They are single-family residences referred to
here as Building A and Building B. Both homes are of split-foyer design
and have garages adjacent to finished portions of their lower levels.
Aside from automobile storage, miscellaneous small gasoline engines (and
the fuel for such) are commonly kept in such indoor areas. The garage in
Building A, in particular, smelled strongly of fuels and solvents (HNu
readings were 2 ppm). Virtually all of the tentatively identified
compounds detected are known components of gasoline. Both homes are
inhabited by cigarette smokers. Although no smoking took place during
the sampling period, information on the prior 24 hours was not obtained.
Sampling and analytical methodologies were tailored for purposes of
identifying and quantifying the compounds listed in Table C-l.
Compounds of interest were selected based on preliminary data developed
during the soil gas sampling conducted in November and March and April
1989. EPA personnel performed both the field sampling activities and
the analytical work.
C-2
-------
TABLE C-l
Targeted Compounds
A - Tenax
Benzene
Ethylbenzene
Tetrachlorocthylene
Toluene
Trichloroethylene
1,1, 1-trichloroethane
m, p-xylene
o-xylene
B - Spherocarb
Trans-1, 2-dichloroethylene
Vinyl Chloride
Sampling and analysis of indoor air conformed to Method TO-1,
Method for the Determination of Volatile Organic Compounds in Ambient
Air using Tenax Adsorption and Gas Chromatography/Mass Spectrometry
(GC/MS) and Method TO-2, Method for the Determination of Volatile
Organic Compounds in Ambient Air by Carbon Molecular Sieve Adsorption
and GC/MS, from the Compendium of Methods for the Determination of Toxic
Organic Compounds in Ambient Air, EPA-600/4-84-041, May 1987.
One sampling station was located on each of the lower two floors of
each building. Each station on the lower floors consisted of primary
and secondary tubes in series plus a duplicate (of each type of tube).
Each station on the upper floors, consisting of single tubes of each
type, was located upstairs on the main living level in the kitchen. The
lower-level station in each building was expected to produce the highest
readings of any of the stations. Because, the possibility of break-
through of the species through the cartridges, with resultant loss of
accuracy, was greatest at this point, the series tube configurations
were utilized there to gauge breakthrough. A single cartridge of each
C-3
-------
type was also set up outdoors to monitor ambient concentrations, for
comparison with those measured indoors.
Samples were collected over an 8-hour period, utilizing personal
constant-flow pumps to draw ambient air through cartridges containing
adsorbents. Certain nonpolar volatile organic compounds having boiling
points in the range of approximately 25 to 95 *C were captured on Tenax
GC (poly 2,6-Diphenyl phenylene oxide). Other more volatile compounds
having boiling points ranging from -26 to +50'C (which will pass through
Tenax adsorbent) were collected on a carbon molecular sieve (CMS)
adsorbent, Spherocarb.
A sample volume requirement of approximately 10 liters dictated a
flow rate of approximately 21 ml/min over the 8-hour sampling period.
Sampling cartridges were positioned in the "breathing zone" (approxi-
mately three feet above floor level).
Occupants were requested to eliminate, insofar as is possible, the
opening of any doors or operation of ventilation/exhaust fans (clothes
driers act as exhaust fans) for a minimum of 24 hours preceding the
sampling period. Windows remained closed for 24 hours preceding
sampling. Door openings were kept to a minimum during the sampling
period.
During the eight hour period while the pumps were running, methane
levels were measured utilizing an Organic Vapor Analyzer (OVA) in the
gas chromatograph mode, with the output going to a strip-chart recorder.
An Hnu (which does not respond to methane) was used for total non
methane readings. The highest methane readings obtained were 8 ppm
(directly above a cat litter pan).
Temperature, barometric pressure and relative humidity (RH) were
measured periodically inside the buildings throughout the eight-hour
sampling period. Temperature readings were obtained with an Ertco
thermometer, pressure with a Taylor aneroid barometer and relative
humidity with a Bacharach Instrument sling psychrometer. Inside
temperatures were fairly constant at 22-23'C and RH was constant at 42-
44 percent. The test report did not give barometric pressure.
C-4
-------
C.I.I Quality Assurance/Quality Control
Cartridge Conditioning
Cartridges were thermally conditioned in a Tekmar Model 5100
Thermal Conditioner two days prior to sampling with a nitrogen purge
rate of approximately 100 ml/min. Tenax cartridges were conditioned for
14 hours at 250°C. Spherocarb cartridges were conditioned for 15 hours
at 399° C.
After the Tenax cartridges were conditioned, they were injected
with 5 micro!iters of a surrogate solution, composed of deuterated
Benzene (d6), Toluene (d8) and bromofluorobenzene in methanol, using the
flash vaporization technique. The cartridges were then refrigerated
prior to field sampling activities.
Pump Calibration
Prior to sampling, the pumps were calibrated using a Mini-Buck
commercial primary standard flow meter with optical sensing of bubble
passage. Following sample collection, calibrations were again checked
with the Mini-Buck. The criterion for the difference between the
initial and final flow rates is no more than 15%. Where the difference
exceeds 15%, the data is reported as an estimated value. Four pumps
exhibited a flow rate difference of more than 15%.
Sample Storage and Transportation
- Cartridges were stored in individual containers following
conditioning for transportation and storage. The individual containers
were sealed in jars containing granulated charocal to minimize the
possibility of cross-contamination.
- Cartridges were stored in a clean laboratory freezer maintained
at approximately -20 to 0* C following conditioning, and were returned
to said freezer following sample collection while awaiting analysis.
- Cartridges were transported to and from the field in an ice-
filled cooler.
C-5
-------
Blanks
One cartridge of each type used in the sampling process accompanied
the sampling cartridges to the field and was handled in precisely the
same manner as the sample cartridges, except that they remained in their
containers to serve as field blanks.
A laboratory blank of each type used remained in the freezer and
was subject to the same criteria as the field blanks.
The results of the analysis of the Tenax trip blank showed it to be
contaminated with two compounds, 1,1,1-trichloroethane and benzene. The
Tenax laboratory blank was found to be contaminated with four compounds,
1,1,1-trichloroethane, benzene, toluene and 1, 2, 3-trichlorobenzene.
All of the sample cartridge data for benzene, with two exceptions, met
the criterion and were considered valid from the standpoint of blank
contamination. All of the Tenax cartridge data for toluene met the
criterion for blank contamination.
No contamination was found on either the trip or laboratory
Spherocarb cartridges.
Breakthrough
Breakthrough of compounds through the adsorption medium was
assessed by a comparison of the levels found on the primary-secondary
cartridge pairs. Secondary, or backup cartridges must meet the criteri-
on of containing less than 20% of the amount of each species found on
the primary cartridge of each pair. At Building B, two compounds failed
the criterion. At Building A, all the components failed the criterion
with the exception of benzene. None of the Spherocarb primary-secondary
pairs exhibited any breakthrough.
Precision
Each sampling event consisted of the collection of an additional
set of parallel samples collected simultaneously at different flow
rates. Agreement between parallel samples should generally be within ±
25% if the concentration is at least 10 times greater than the minimum
detection level. None of the targeted Tenax compounds detected at this
level failed the criterion. Methylene chloride was the only compound
detected on Spherocarb which failed the criterion.
C-6
-------
C.I.2 Results
Much of the chemical concentration data from this study were
qualified because of:
• Four pumps not meeting flow-rate acceptance criterion
• Poor recovery on two cartridges
• Blank contamination for 1,1,1-Trichloroethane and benzene
• Breakthrough on primary cartridges using Tenax (precision
criteria were met however).
Summarized results are in Table C-2.
C-7
-------
TABLE C.2
Maximum Concentrations of Targeted Compounds Found (1) (PPBV/V)
A - Tenax
benzene
ethylbenzene
tetrachloroethylene
toluene
trichloroethylene
1,1, 1-trichloroethane
m,p-xylene
Building B
5.9
2.6
0.34
19*
0.04*
3*
8.4®
Building A
17*
7.7*
0.52
56*
ND
5.2*
25®*
Ambient
ND(3)
ND
ND
ND
ND
ND
ND
* - Estimated Value
B - Spherocarb
trans- 1 ,2-dichloroethylene
vinyl chloride
ND
ND
ND
ND
ND
ND
(1) - 18 other compounds were detected; the highest was 15 ppb V/V for isopropylbenzene
(2) - Reported as total o,m,p-xylenes
(3) - ND - not detected in any sample
C-8
-------
C.I.3 CONCLUSION REACHED
The test report stated,
"In summary, all eight of the targeted compounds detected in the
homes studied here have potential sources apart from the underlying
landfill and are, in fact, not uncommon in a household environment.
Therefore, it is impossible to determine with complete certainty which
compounds are actually migrating into the homes from past landfill ing
and which are present merely as a result of typical household activi-
ties.
Finally, whether or not these levels constitute a health hazard
must be determined by a risk analysis."
The complete test report was submitted to the residents along with
a cover letter that states, in part:
For Building A;
"The levels of chemicals detected in your home were comparable to
those found in the average home.
At this time we do not know the origin of these chemicals. As we
are planning to complete the Remedial Investigation of the landfill in
the summer of 1992, we hope, at that time to determine the source of
these indoor pollutants."
For Building B;
"The levels of chemicals detected in your home were found to be
slightly higher than those normally found in an average home. However,
the levels detected in your home do not pose an immediate health threat.
At this time we do not know the origin of these chemicals. As we
are planning to complete the Remedial Investigation of the landfill in
the summer of 1992, we hope, at that time to determine the source of
these indoor pollutants.
The chemicals identified in your home at slightly elevated levels
are: benzene, ethylbenzene, toluene, and xylenes. Apart from the
landfill, possible sources of these indoor air pollutants are numerous."
C.I.4 Comments On The Study
This case provides a good example of some of the pitfalls of indoor
air monitoring. The study met its primary objective which was to
"document the presence of any toxic volatile organic chemicals in
C-9
-------
buildings which may pose a health risk to their occupants." However,
because the indoor air concentrations of chemicals of the same identity
as those associated with the Superfund site were similar to typical
indoor concentrations, no conclusive statements about the source of the
chemicals could be made. This would be true even if all sampling and
analysis had gone perfectly.
In retrospect, because the study design was such the contribution
of site chemicals to the indoor air concentrations could not be deter-
mined, it would have been preferable to state the purpose as: Determine
whether or not indoor air contaminants were of types and concentrations
to suggest significant intrusion of site related chemicals. With this
as the purpose, it would have been easier to develop a risk communica-
tion plan that had specific follow-up steps depending on the contaminant
concentrations determined. Building occupants could, then, have been
advised before monitoring took place what actions would be taken
depending on the concentrations found in this screening study.
C-10
-------
C.2 BUILDINGS NEAR LANDFILL
The Landfill is in South King County, Washington. Puget Sound is
slightly more than a mile to the west. Residential areas surround the
site, with the exception of a commercial strip along Highway 99 to the
west and a stand of second-growth mixed conifer-hardwood forest on the
north. Two elementary schools and a community college are within one
mile of the site. Interstate 5 (1-5) borders the site on the east.
Approximately one mile east of 1-5 is the Green River, which meanders
north, becomes the Duwamish River, and enters Puget Sound. Figure C-l
shows the location of the landfill and the landmarks in its vicinity.
From 1945 to 1966, the site was operated as a gravel pit. The pit
originally was adjacent to a peat bog lake located northeast of the
center of the present landfill. As the pit was mined, water was drawn
from the lake to wash silt and clay from the gravel and sand, then the
water was returned to the lake. Silt and clay built up on the lake
bottom. Near the end of the operation of the gravel mine the barrier
between the lake and the gravel pit was broken, allowing the silty lake
water to flow into the gravel pit. As a result, a clay/silt layer
underlies much, but not all, of the landfill.
In January 1966, the 60-acre site was leased and used as a landfill
for nonputrescible waste, which includes organic material that decompos-
es slowly, such as demolition debris and wood wastes. The landfill
received demolition debris from commercial haulers and wood wastes and
yard trimmings. Records beginning in 1980 indicate that some industrial
wastes also were deposited. Information included in EPA's Emergency and
Remedial Response Information System (ERRIS) files indicates that the
landfill may have received industrial liquid and sludge wastes before
1980.
C-ll
-------
LANDFILL
SCALE w FEET
500 1000
FIGURE C-1
Location Map
C-12
-------
During the course of operations at the landfill, an estimated 3
million cubic yards of solid waste were deposited covering approximately
40 acres up to about 130 feet deep in places. The east side of the
landfill rises above the adjacent property; the landfill surface slopes
downward to the northwest corner. The landfill was closed in October
1983. The entire site was covered with 6 to 24 inches of a silt/sand
material when operations ceased; it now appears as an open grassy area
with scattered shrubs and a few areas of exposed soil. Approximately
102 acres of land east of 1-5 currently drain directly into the solid
waste.
C.2.1 Nature And Extent Of The Problem
Potential problems related to solid waste landfills similar to the
Landfill mainly derive from the combustible gas that is a byproduct of
decomposition, and leachate, i.e., water or other liquid that has come
into contact with the waste. Methane, carbon dioxide, and small amounts
of hydrogen sulfide are normally generated by the decomposition of
landfill wastes. Small amounts of toxic volatile organics are commonly
found as well. These gases can be released to the atmosphere or migrate
offsite by means of underground pathways. Leachate can flow or seep
offsite to contaminate surface water or it can infiltrate the groundwa-
ter underlying the landfill if no artificial or natural barriers exist.
The Landfill is a natural drainage basin from which no surface water
exits. However, depending on subsurface conditions, groundwater
contamination is possible.
The landfill was closed in the fall of 1983 and extensive testing
of gas and water in the landfill and its vicinity began. Samples of
leachate and groundwater from monitoring wells in and around the
landfill and gas samples from gas probes indicated the presence of
organic and inorganic contaminants with a high potential for offsite
migration. In May 1986, the EPA placed the site on its National
Priority List for cleanup and a remedial investigation was initiated.
Initial remedial measures were taken to control the offsite
migration of gas and prevent possible emissions from the landfill
C-13
-------
surface; gas extraction wells were installed offsite and gas migration
control wells were installed on the perimeter of the landfill.
Combustible Gas
Combustible gas, primarily methane (CH4), was detected at concen-
trations up to 10,000 ppm in the basements of homes near the Landfill in
1985. A series of gas probes were installed around the site perimeter
to monitor gas concentrations. Gas was found in many of the probes.
Measures were taken to seal cracks in building foundations. Numerous
shallow and some deep probes were installed offsite. A program to
monitor homes for gas was established. Data from the offsite probes and
the monitoring program indicated that the gas had migrated up to
approximately 2,600 feet from the landfill and was seeping into homes at
that distance. Additional probes were installed farther from the site.
The sealing programs within homes and businesses was expanded, and some
families were evacuated.
In September 1985, and again in September 1986, a series of onsite
gas migration control wells around the perimeter of the landfill were
installed as initial remedial measures. Designed to prevent offsite gas
migration by withdrawing and venting the gas, these wells are connected
by manifold piping to motor blowers and flares. As part of final
landfill closure, portions of this system will be replaced by permanent
equipment that will remain in operation for the indefinite future.
Leachate
It is estimated that over 50 million gallons per year of surface
water (drainage and direct precipitation) enter the landfill. Measure-
ments of water levels in leachate monitoring wells indicate that
stormwater discharge from drainage pipes produces rapid and significant
increases in water levels within the solid waste. Since there is no
surface runoff from the landfill, leachate must eventually enter the
groundwater system if it does not remain in the landfill. Leachate
flows to the southeast corner of the landfill and then downward and then
east and west away from a mound located beneath the Sand Aquifer sink.
Based on chloride concentrations, leachate is calculated to be indistin-
guishable from background groundwater at a maximum of 3,000 feet from
C-14
-------
the landfill boundary to the east and southeast and 1,500 feet from the
landfill boundary to the west.
Offsite Gas Monitoring
Data recorded from the monitoring probes were computerized and used
to generate gas concentration isopleths (maps showing gradients of gas
concentrations) for the areas surrounding the Landfill.
The isopleths for February 1986, shown in Figures C-2 and C-3,
represent the status of landfill gas migration just after the Phase I
onsite gas migration control system was put into continuous operation.
Methane was detected northwest of the landfill in concentrations over
75%, north in concentrations over 25%, and east and southwest in
concentrations over 40%. Methane was detected in the deep soil zone in
all directions from the landfill and in concentrations over 90% in some
areas.
C-15
-------
LANDFILL
A* M« Of protttl (no
»*»» may not M shown
0 SCO 1000
FIGURE C-2
Shallow Gas Concentration
Isopleth, February, 1986
C-16
-------
LANDFILL
Note AJ s«ts of (xo«$ and
wefts may not be shown
SCALE IN FEET
0 500 1000
FIGURE C-3
Deep Gas Concentration
Isopleth, February, 1986
C-17
-------
Field measurements were taken using the following instruments:
• Hnu PI-101 photoionization detector (total non-methane volatile
organic compounds)
• MSA 361 portable detector (combustible gas, hydrogen sulfide,
and oxygen)
• Fyrite carbon dioxide analyzer
Gas samples were collected in Tedlar bags and shipped. Gas samples were
analyzed for Hazardous Substances List volatile organic compounds (HSL
VOC) by gas chromatography/mass spectrometry (GC/MS) in accordance with
USEPA Method 624 guidelines.
This analysis found that subsurface gas collected from the onsite
gas extraction wells and flare manifolds contained a wide variety of
substances, including 23 USEPA HSL VOCs. The compounds found most
frequently and in the highest concentrations in the onsite subsurface
gas included ethylbenzene, vinyl chloride, total xylenes, toluene, and
benzene. The maximum concentrations of these compounds were in the low
parts-per-million (ppm) range.
TABLE C-3
ESTIMATED ONSITE/OFFSITE ATTENUATION OF PRIMARY
HSL VOCS DETECTED IN SUBSURFACE GAS SAMPLES
CAS#
100-41-4
1 33O-20-7
108-88-3
71-43-2
1 00-42-5
75-01-4
Target
Compound
Ethylbenzene
Xylenet (total)
Toluene
Benzene
Styrene
Vinyl Chloride
MAXIMUM CONCENTRATION
On-Site
(ppb)
16,610
29.195
24,044
1,384
508
31,215
Off-
Site
(ppb)
127
106
68
185
134
275
Atten.
(%)
99.2
99.6
99.7
86.6
73.6
99.1
MEAN CONCENTRATION
On-Site
(ppb)
2.825
3.419
1.920
318
41
2.607
Off-
Site
(ppb)
44
19
18
28
18
35
Atten.
(%)
98.4
99.4
99.1
91.2
56.1
98.8
Attenuation (%) = Onsite Concen. (Dob) - Offsite Concen. (ppb) x 100
Oniite Concen. (ppb)
C-18
-------
Vinyl chloride was frequently associated with the BTX-group
compounds in onsite subsurface gas, but during sampling offsite, it was
found at only two of the gas monitor probe locations, both south of the
landfill.
The analysis concluded that offsite migration of at least some of
the contaminants consistently identified in landfill gas has occurred,
possibly in all directions away from the landfill. The basis for this
conclusion is the presence of the BTX-group compounds in gases found
onsite and offsite. However, offsite concentrations of most BTX-group
compounds were attenuated by more than 98 percent (Table C-3). Further-
more, no BTX compounds were found beyond 2,300 feet from the landfill
perimeter. Because BTX-group compounds are also present in gas from
other sources, their presence in a particular sample of offsite gas is
not conclusive evidence that the gas originated from the landfill; it is
only further evidence that the gas may have originated in the landfill.
Mechanisms for Gas Movement
Where gas migration potential exists, the inherent slight pressure
of gas being generated by decomposition, fluctuations in barometric
pressure, and fluctuations in water table may create mechanisms for gas
movement.
The low gas generation pressures in the landfill initially move the
gas away from the points of origin through the paths of least resis-
tance. The generation pressures are estimated to be in the range of two
inches of water column, based upon measurements in other landfills.
Changes in atmospheric pressure of -9.5 inches of water column are
common during the winter months as high pressure ridges give way to low
pressures during storms. This can create a relatively strong pressure
differential between the air or gas within the soil or landfill and the
air at the ground surface.
When barometric pressure is falling, the air or gas within the soil
or landfill tends to flow toward the ground surface through the paths of
least resistance. When barometric pressure is rising, air from the
ground surface tends to move down into the soil both vertically and
C-19
-------
horizontally. The combined effect of the generation pressures and the
barometric changes is a pumping action within the soils.
C.2.2 Prevention Of Gas In Structures
Several residences adjacent to control well installations typically
recorded levels of methane above 10,000 ppm despite repeated efforts to
seal the foundations of those structures until the control wells were
turned on. According to gas monitoring data, all gas was effectively
removed from these structures within one day of the startup of the
offsite control wells. Further, gas has remained out of structures in
the vicinity of the landfill since the implementation of the offsite
control wells. Combustible gas above 100 parts per million has not been
recorded in a structure in the vicinity since November 1986.
C.2.3 Comments On The Study
The study demonstrated that landfill gas migrated from the site
independently from contaminated groundwater. It demonstrates the diffi-
culty in preventing soil gas intrusion by retrofit patching of existing
buildings and the effectiveness of soil gas extraction wells. The data
suggests that vinyl chloride is migrating in the direction of groundwa-
ter flow rather than with near surface gases.
Because of the high concentrations of methane detected in the
initial phase of the investigation, few data were obtained related to
other gases that might also be migrating. This initial oversight, while
understandable, significantly reduced the data needed for an accurate
baseline assessment.
C-20
-------
C.3 BUILDINGS OVER GROUNDWATER PLUME (A)
The site is located at the southern end of San Francisco Bay. In
1982 and 1983, it was discovered that three facilities (identified here
as A, B, and C) in the city had leaking underground tanks contaminating
soil and groundwater. All three were placed on the National Priorities
List (NPL). Facility C has since been removed from the NPL and is being
regulated under the Resource Concentration and Recovery Act (RCRA).
Although each site has its own source of pollution, the off-site
groundwater pollution areas have merged and the sites and are being
treated as one unit.
The area is heavily populated with commercial, light industrial,
and residential use. According to the 1980 Census information, there
are approximately 4,000 people living in the general area of the off-
site groundwater contamination:
The off-site area is determined by the extent of the groundwater
contamination of volatile organic compounds. The plume of contaminated
groundwater extends approximately one and three-quarters mile north of
the site. This area is composed of light industry and residential
housing. The former junior high school is located west to northwest of
the site.
The closest residential neighborhood begins about 1/4 mile north of
the site. According to the 1980 Census, of the approximately 1,500
housing units, 54% are single family homes and 24% have 10 or more
units. The area consists of modest single family homes interspersed
with small apartment complexes. Within 3/4 mile of the site is a mobile
home park which is primarily occupied by senior citizens.
A former elementary school houses approximately 200 children per
day. These children attend either a state child development program, a
state preschool, or a YMCA day-care program and a Head Start Program,
all operating at the school facility. There is at least a half acre of
grassy land on the side of the facility that is not included in the
fenced off playground. This is used as a neighborhood playground.
C-21
-------
C.3.1 Nature and Extent of Problem
Chlorinated volatile organic compounds (VOCs) are the major
contaminants tested for and found in soils and groundwater. Highest
concentrations were found around the leaking underground tanks. The
tanks and much of the contaminated soil were removed from the sites in
1983 and 1984. Soils with a total VOC concentration as high as 15,700
parts per million (PPM) were excavated. Soil boring samples were
obtained in 1988 at the A and B Facilities. The data are in Table C-4.
Groundwater
The VOC contamination in groundwater is currently monitored by 29
wells on the A site and 30 wells on the B site, in addition to 83 off-
site wells. There are 7 extraction wells on the A site and 6 extraction
wells on the B site. There are 23 off-site extraction wells. The on-
site extraction wells have been in operation since 1985, and the off-
site wells began operation in 1986 and 1988.
Three major water-bearing zones (aquifers) - defined as the A-,
B-, and C-aquifer zones - exist at the site. The A-aquifer is the
shallowest and the C-aquifer is the deepest of these three zones. The
B-aquifer is further divided into the Bl through B5 zones. The approxi-
mate depths (below ground level) at which these zones occur at the A
site are as follows: A: 10-28 ft; Bl: 28-50 ft; B3: 70-90 ft; B4: 90-
110 ft; and B5: 110-123 ft. A regional aquitard, the B-C aquitard, is
reportedly located at the depth range from 100-150 feet below ground
level. The C-aquifer, which supplies most of the municipal water in the
region, is located below the regional aquitard.
The aquifer zones appear to consist of mostly discontinuous layers
and lenses of fine to coarse sand, gravels, and often a substantial
proportion of clay and silt. These predominantly discontinuous layers
and lenses are separated and/or isolated by low-permeability clays and
silts (aquitards).
C-22
-------
o
CO
CO
en
o
ts>
00
o
o
<
I-H
t— 1
o
2
"§
a
c
o
re
4->
0)
o
o
U
'6
w
o
c
'5
m
a>
TJ
"to
c.
0
C
jo
0
T>
1
o
k.
TO
'5
CT
U
re
t
M
3
W
Aquitard
o
o
epnsqns
TJ
C
TJ
CN
CO
TJ
C
TJ
c
0>
TJ
J5
6
_c
•*
10
CO
CO
TJ
C
CO
CN
1
0
o
d
„.
0
o
d
TJ
c
•Dichloro-
e
31
u o>
TJ
C
CO
LO
O
d
CN
CO
o
TJ
CN
CO
CN
O
d
•oethylene
o
^
_o
'C
1-
c
CO
d
TJ
c
T>
C
0
c
0)
c
loroethyle
.c
u
re
4-»
0)
I-
n
c
CO
TJ
TJ
C
CO
CN
TJ
CN
"^
CO
o
d
V
®
richloroethan
i-
i
*
-
CD
0
0
O
TJ
C
8
o
TJ
C
05
CN
r—
TJ
C
C
«
c
^
,2-Dichloroett
U)
C
re
4-*
LO
O
cp
TJ
C
(^
CN
T—
d
i
TJ
C
O
d
CO
TJ
CO
CO
1
o
d
V
oethylene
•r
^
(j
H
if)
CN
0
O
0
TJ
C
^.
8
d
i
TJ
C
O
in
CO
TJ
c
loroethylene
JZ
u
k.
V
4-»
4^
C
CN
1
TJ
C
hlorobenzene
o
0
1
s
c
o
•a
o
•8
S -o
S £
o «
•
C-23
-------
Based on the well monitoring results at the site, it appears the
bulk of the VOC contamination is confined to the A-, Bl- B2-, and B3-
aquifers. However, low concentrations of VOCs have been detected in the
B4-aquifer. On-site VOC concentrations in the fourth quarter of 1990 at
the A and B Facilities are given in Table C-5 and C-6.
Off-site VOC concentration in groundwater for the fourth quarter
of 1990 are given in Table C-7. The data were from predetermined wells
that were not necessarily in the middle of the plume. The maximum
concentration of trichloroethylene, 1, 2-dichloroethylene, 1, 1-dichlor-
oethane, and tetrachloroethylene exceeded drinking water standards.
Soil Gas
On March 12, 1991, the soil surface emission of organic vapors was
analyzed using a flux chamber at off-site locations just to the north of
Facility A and near the former Junior High School. The soil moisture
content and ambient temperature during this sampling period were not
typical for this area and would tend to under-represent typical organic
vapor emission from the soil. Another sampling round was conducted on
April 28, 1991. One sampling location north of Facility A was repeated
while the rest of the samples were taken on the former elementary school
property. The data are given in Table C-8. [Note: The ATSDR draft
report on this site states that data collected using the neutral
pressure isolation flu chamber may be inappropriate for estimating flux
into buildings.]
Indoor Air Monitoring
Indoor air monitoring has been conducted only at Facility B. On
May 16, 1991, a crude analysis for organic compounds was made using a
hand-held detector. All readings were below the limit of detection (0.5
to 1.0 ppm). On May 18,1991, sampling was conducted inside and outside
the building. One-hour samples were collected at each location five
feet above ground. The results are in Table C-9.
C-24
-------
U
LU
_1
OQ
cc
UJ
Q
Z
D
O
CC
(D
Z
CO
Z
o
<
cc
UJ
o
Z
o
o
o
o
UJ
t
CO
I
O
O
ration
c
c
o
o
o
o
'5
CT
S
w
00
m
0}
? I- "°
.= C o
•^ 5 -5
5*1
w
q
c-i
8
01
i
g
i
u>
00
i
§
| Dichloromethane
'
?
i
o
i
R
i
1
o
c
a
Dichloroflourome
o
i-
?
in
0
e
1 ,2-Dichloroethai
'
-------
(O
I
U
LU
_l
m
cc
LU
<
Q
Z
D
O
CC.
CO
•z.
g
i-
cc
LU
O
•z.
o
o
o
o
LU
CO
I
o
CO
U
2
ncentration (ppbl
0
o
Q>
'5
CT
m
5
<
M
III
a *
CO
•0
c
o
1
TD
O
CM
•o
C
o
o
CM
,1,1 -Trichloroethane
r-
0
in
to
i
c
O
CM
•6
c
•O
6
o
CM
r—
CO
T—
o
V
k>»
U.
c
o
o
CO
c
CO
CO
c
in
d
inyl chloride
>
CD
•6
C
0
O
q
CN
in
•6
c
o
CO
OJ
•6
o
T) •
C-26
-------
UJ
_i
CD
f
cc.
UJ
<
Q
Z
D
O
CC
05
Z
O
<
cc
LU
U
Z
o
o
U
o
111
u.
LL
o
c
o
'+-•
2
*->
CO
u
c
o
0
CO
*»•
cr
<
3
CD
2
CN
CO
»—
CO
<
V)
0 w "S
111
i« ^i n
Q > 2
u >
^f
r»'
i
TJ
C
TJ
C
O
«—
TJ
C
O3
op*
TJ
C
9
•o
b
t
omethane
c
_o
u
0
^t
o
TJ
C
CO
o
TJ
C
IT)
CO
TJ
C
CO
CN
TJ
|
b
O
f—
§
o
s-
o
o
sz
o
TJ
C
TJ
C
TJ
C
0
*—
i
TJ
C
CO
tN
TJ
jj
tn
hloroethane
o
5
*
r-
*t
cp
TJ
C
TJ
C
fc^
r^*
i
•D
C
0
o
r^
TJ
c
(^
TJ
U
O
o
CN
richloroethane
H-
r—
LT>
? —
TJ
O
O
CN
-ri
W
TJ
C
O
CO
1
TJ
C
CO
CN
TJ
C
b
o
CN
T"
n
c
0
(U
til
CO
CN
TJ
C
TJ
C
TJ
C
tn
CM
TJ
c
^
«
TJ
u
CO
hloroethylene
o
b
T—
't
T—
TJ
C
TJ
C
T)
C
O
CO
1
T)
C
co
TJ
ja
CO
hloroethylene
u
b
i
CNJ
00
CO
TJ
C
o
o
»r
CN
TJ
C
O
o
t
CN
TJ
C
O
o
*f
CM
TJ
C
O
CO
up
TJ
C
U1
oethylene
U-
O
JT
o
*c
(-
TJ
C
TJ
TJ
C
T)
C
CN
TJ
C
u
UD
loroethylene
^
u
TO
*-»
^
>
0
>
G
s.
8
*
0)
o
i
c
re
i:
*•"
w
*~ - j
eg o
2 c a)
*•; 5 .c
Jn — ~
« JO
0) -C O
TJ 'C •*-
5^d d
2S5 5
ro re «
s - 1 1
3-1 .2-j £
« " -5 " -"5
5 S o 5 o
a ffl TJ « TJ
o £ SI £ S
c o o o o
„ S: ca *: Q.
11 « 2 « S
0 O Q.CJ Q.
Z « ^ U -Q
C-27
-------
cc
o
0_
op
6
LU
_l
m
LU
H
op
LL
LL
O
(O
y
LU
I
U
y
z
DC
O
c
o
o
a
CO
u.
co
a
c
CO
£
CO
V)
o
2
c
0)
u
c
o
u
en
o
CD
m
|~
'o
re
LL
£
0)
X
er Junior
School
E
0
LL
CO
£
'o
re
LL
C
c
c
c
ichloromethane
Q
c
c
c
c
o
ichlorodiflouromethan
Q
c
c
•o
c
c
,1-Dichloroethane
t—
00
co
o
00
d
CO
ID
d
CO
r-
d
•6
CO
CO*
6
,1,1 -Trichloroethane
i—
iD
co
CO
d
in
en
d
CO
TJ
O)
CD
d
d
CO
o
c
c
c
•a
,1-Dichloroethane
r-
C
T3
C
co
U)
d
•o
Q)
s-1 ,2-Dichloroethylen
'o
c
•o
c
•o
•O
0)
C
ans-1 ,2-Dichloroethyl
4-1
O
CO
o
TJ
C
•D
•a
richloroethylene
1—
c
c
•a
c
c
etrachloroethylene
H-
c
c
en
d
•6
•o
c
,2-Dichlorobenzene
T-
2
E
c
g
o
4-<
T3
(D
= not detected abovi
= not tested
•0 4-
c c
C-28
-------
CD
1
0
LU
i
m
^
^~
CJ
Lt_
LU
I
H
^
QC
1-
LU
CD
2
^
Z
(/)
^
O
2
LU
I
O
u
z
(3
OC
0
c
o
ss
5
Q)
Q.
U)
C
ro
a.
c
a>
'&
0)
ro
U)
c
o
4^
2
4->
C
0)
c
0
u
^
a
c
'J5
'5
CO
0)
2
<2
D
O
a
c
'•6
(i
c
•o
"Z
(O
LU
LU
at
c
13
u
c
u.
or
u
c
a>
£
o
it
O
Q>
a>
2
o
4-<
CO
C
o
a
a
3
CO
c
0
1
TJ
8
CL
O
o
oo
c
c
c
H
s
c
Dichloromethane
i
c
4-<
4-*
C
€
c
Dichlorodiflouro-
methane
o
CO
c
c
4-*
C.
c
c
c
1 ,2-Dichloroethane
IT)
in
CO
d
CO
o
CO
o
CM
CO
0
CO
6
1,1,1-Trichloro-
ethane
i
c
4-»
C
4^
4->
«
CO
o
0)
LL
IT)
T3
C
C
C
•o
c
•o
c
•o
c
Vinyl chloride
,
•o
c
•a
c
T>
•o
C
•o
•o
1 ,2-Dichloroethylene
0
CO
CM
CM
O
O>
O
0
6
tn
o
6
V
Trichloroethylene
OT
T3
CM
O
0
T5
1
r—
o
Tetrachloroethylene
o
o
CO
•o
c
•a
c
•o
c
c
•o
•o
c
Dichlorobenzene
13
E
c
o
u
M
V
•o
0)
8
C
J "O
D Q)
U W
•^ 09
3 o
c C
II II
3 4^
4)
c
0)
N
C
0)
8
5
^:
u
'•5
*-«'?4
u
T3
a>
a.
a
=re£
O .
C-29
-------
C.3.2 Indoor Air Modeling Results
Estimates of indoor air concentrations were made, using models,
for current residences, the former elementary school, and possible
future apartment complexes. The Farmer model (see Appendix A of this
report) was used to estimate emission flux at ground surface. Based on
assumptions about building construction (area of the flooring through
which gas could infiltrate is the dominant factor) and building air
exchange rates, the indoor air concentrations were then calculated. The
modeling results are given in Tables C-10 through C-12. [The tables show
that identical groundwater concentrations and contaminant vapor phase
concentrations were used for the average and plausible worst cases.]
Interpretation of Results
The Agency for Toxic Substances and Disease Registery (ATSDR)
reviewed the above data (as well as other data not presented here) and
estimated cancer and non-cancer risks based on their standard methodolo-
gy. The analysis indicated that only current off-site residences (not
including the former elementary school) have any increased risk from
exposure to chemicals volatilizing from the groundwater plume. The
upperbound lifetime excess cancer risk was estimated to be 1 x 10"4 (low
increased risk) for average case conditions, and 2 x 10~3 (moderate
increased risk) for the plausible maximum conditions. The summary of
the ATSDR report states that the site is an indeterminate public health
hazard because of potential human exposure to concentrations of chemi-
cals that upon long-term exposure could cause adverse health effects.
C.3.3 Comments on Case Study
1. As noted by ATSDR, use of the flux chamber to check emission
flux predictions of the model at ground surface may be valuable, but
flux chamber data may be inappropriate as flux source data for infiltra-
tion into a building due to negative pressure in the building.
2. The modeling report states that convective flow due to
building vacuum (ie. stack effect) is generally more important than
simple diffusion into the building. Only diffusion into the buildings
C-30
-------
U
LU
CQ
U
o
LL
O
Z
g
^
cc
\-
z
LU
o
z
o
0
cc
J<
cc
o
o
o
z
LL
o
z
o
h-
LU
CO
0
cc
LL
r-
LU
Q
CC
(D
Z
O
Q
CO
LU
U
Z
LU
Q
CO
LU
CC
(J
z
Q 0
UJ CJ
5 <
P £E
) O
UJ o
Q
Z
LU
OC
X
U.
O
5
UJ
I
O
0
0
UJ
w
I
Q.
OC
|
(J
O
o
oc
UJ
z
0
oc
o
sis
'SEE
• s ?
c s -
I!
*§.*
g X j|
C f 3
? 2
5 £
< s
•* § -
m
! s
i -J
s1!
jr ^
J S PJ
H**
E
1 !
Z
2c~
(if
o
0
UJ
I
9
UJ
,J
flD
T—
24E-00
"*
?
UJ
CO
10
LO
01
9
LU
*~
CM
9
UJ
CM
71 E-02
CM
CJ
UJ
0
"
9
UJ
10
*"
P)
P5
,1-Dichloro-
thylene
«- 0>
9
UJ
r»
CO
27E-01
CM
O
UJ
CO
o>
CM
LU
r»
n
9
Ul
CO
97E-03
^
-309
LO
CO
9
UJ
LO
N
0)
9
UJ
CO
00
9
UJ
n
37E-02
t
o
9
UJ
CM
CO
P)
9
UJ
O
09
03
O
CM
O
CM
richloro-
thylana
H- «
9
UJ
CD
n
n
CM
9
UJ
S3
9
UJ
LO
cn
CM
9SE-03
CM
9
UJ
8
"
P)
9
UJ
o
LO
O
^
--
ans-1,2-Di-
hloroethylene
^ o
CM
9
UJ
r~
CO
68E-01
*
CO
9
UJ
o
o
09
9
LU
0
35E-01
P)
o
9
Ul
8
0)
9
UJ
O
0)
ri
CM
CM
P)
i
LL
§
UJ
p^
n.
n
9
UJ
CO
09
9
Ul
LO
^
0
UJ
CD
CM
n
9
LU
CM
P)
9
UJ
CM
fN
O
9
Ul
O
03
^
CM
9
Ul
o
03
CM
P>
<6
n
o
,1,1-Trl-
hloroathana
•- o
C-31
-------
O
LU
CO
f
O
o
U-
O
Z
o
ENTRATI
O
Z
0
o
cc
cc
0
0
Q
z
LL
0
Z
o
\-
F
LU
1
SCHOOl
^.
CC
^
^—
Z
UJ
UJ
UJ
cc
UJ
cc
o
LL
d
Q 0
UJ U
t—
ESTIW
INDOOR >>
UJ
i
x
u.
<
I
O
o
o
u
Vl
VAPOR PH
O
z
o
u
cc
UJ
GROUNOW/
S E
£ ? *2
lit
n
IK
= 2
? vJ
5 I
< 3
In
il«
II
5 >-5
Q •£
- =s
* « ^
1 -^ o E
o*
E
1 I
I11
_J
O
2
Ul
U
9
UJ
q
r>
eo-308
^
.58E-O9
ID
CO
9
Ul
V
fM
9
Ul
(S|
2.71E-02
in
9
UJ
O
*~
9
UJ
to
*~
2
r)
o
«- j:
*" *
n
1.61E-0
18E-O4
"*
O
Ul
CT)
CO
(SI
_
UJ
JJ
"'
n
9
UJ
r»
a>
1.97E-03
9
UJ
CD
CO
(Si
9
Ul
n
4.37E-02
0
9
UJ
(N
CO
P)
9
UJ
O
CO
00
o
CN
O
(S|
richloro-
thylene
1- 0
9
UJ
CD
CD
6OE-O4
r*
O
UJ
*~
UJ
ID
0)
9
UJ
(SI
CO
ui
6.82E-O4
c
o
0
u.
o
9
UJ
CSI
03
61E-03
^"
.15E-09
"
O
UJ
CD
O)
CM
r)
9
Ul
(SI
7.21E-O3
CD
9
UJ
8
N
(S
9
UJ
O
CD
(SI
r>
CD'
r>
CD
I
,1,1-Trl-
iloroetha
•- o
C-32
-------
CN
i
u
LU
CO
1-
CJ
o
LL
O
Z
0
r-
^
CC
f—
Z
LU
O
Z
o
u
cc
<
cc
o
o
Q
Z
LL
O
Z
o
h-
co
LU
U
Z
LU
Q
(/O
LU
CC
LU
r-
w
1
Z
0
LU
CC
D
H
D
LL
LU
o
Q 0
uj u
<£
5 <
w o
UJ O
Z
p
^
cc
X
u.
O
5
UJ
X
0
(J
Z
o
o
Ul
CC
O
Q.
>
O
O
(J
cc
UJ
0
Z
o
cc
o
i E -
S 3 CO
2 g >
C 5 3
01 CO
s 1
*E?
"S c CM
5 x E
1? * C»
CL 2 —
* "S
ft f^
« E
< S
1 i co
si£
«I«
1 co
1 1
< ^
i •*
-• Si
t j a E o
I " 1 'e g
i o S
E
i 1
•g
U!
o
0
UJ
I
o
o
UJ
CO
CD
,_
CM
O
Ul
•9
"*
§
Ul
CO
CM
*-
o>
o
Ul
0)
co
CM
O
Ul
03
en
in
S
Ul
CD
cn
8
Ul
0
r-
O
UJ
in
in
cri
in
en
1,1-Dichloro-
ethylena
CM
o
Ul
in
fx
CM
CO
o
Ul
CM
^
8
Ul
in
CD
•-
o
Ul
01
*
s
111
CM
CM
8
UJ
CM
CN
8
Ul
o
ls-
CM
O
UJ
0
CM
2
CO
Tetrachloro
ethylene
o
Ul
CO
o
in
O
Ul
O
•~
8
Ul
03
co
8
Ul
in
rx
CO
r-
O
Ul
S
CM
O
Ul
3
8
Ul
CM
CO
8
Ul
o
0)
03
O
CO
in
o
CD
in
Trichloro-
ethylena
o
Ul
0
CM
CM
8
Ul
o
in
8
Ul
CD
*-
O
Ul
CO
03
CO
8
Ul
co'
8
Ul
CD
8
Ul
fx,
0
*•
O
Ul
o
en
CD
O
CM
O
CM
Vinyl chloride
0
LU
in
-------
was considered. No explanation for not considering convection was
given.
3. Whereas gas was considered to infiltrate buildings through
the entire floor area for current private residences, infiltration
through only a 0.5 cm crack around the perimeter was allowed for the
former elementary school and future residences. For example, for the
school, this assumes only 0.08 percent of the soil gas flux enters. For
future residences the assumption is 0.2 percent flux entry. The
assumption that gas entry rates are a direct function of the percent
cracked area is not supportable and could result in significant under-
predictions of indoor air levels.
4. The modelling approach made no attempt to correct for the
capillary fringe above the groundwater. While this is appropriate for
an initial simple screening approach, it should be recognized that this
could lead to large overestimations of contaminant flux to the surface.
5. Incorrect Henry's Law Constants were used leading to a
potential overestimation of flux by a factor of about 50. (Non-dimens-
ional values used in a dimensional equation).
Because of the opposing influence of item 2 through 5, the
relevance of the model predictions is uncertain. Recognizing that the
initial modeling approach was simplistic, in retrospect it would have
been preferred to conduct more refined modeling before releasing the
data or conducting on-site monitoring.
C-34
-------
C.4 BUILDINGS OVER GROUNDWATER PLUME (B)
The site (Figure C-4), a former printed circuit board manufactur-
ing facility, is located at the southern end of San Francisco bay. The
facility operated from 1970 to 1985. The U.S. Environmental Protection
Agency (EPA) placed the site on the National Priorities List (NPL) on
February 1, 1990.
The primary site of both wet and dry manufacturing processes was
the "wet floor" building where waste waters containing heavy metals and
organic compounds were released to the "wet floor" and then discharged
to the neutralization sump. Other solvents and wastes were placed in
drums and disposed of off-site. It is still unclear exactly how the
chemicals discharged to the "wet floor" or the neutralization sump were
released to the soil and then the ground water.
In 1985 manufacturing stopped. As part of the plant closure
process, soil and ground-water investigations were initiated. The
investigation showed chlorinated volatile organic compounds (VOCs) and
metals (copper, lead and nickel) are the major contaminants found in the
soil and ground-water samples. The highest concentration of organic
contaminants in the soils was found around the sump and the "wet floor".
The sump and surrounding contaminated soil and the contaminated soil
beneath the "wet floor" were removed.
Seven wells on the site and 24 off-site wells (Figure C-4)
currently monitor the VOC contamination in the groundwater. The ground
water monitoring indicates that the plume extends north underneath
Interstate 101 (1/8 mile) and west of the site approximately a quarter
mile. Remediation of the groundwater began around 1986. Currently
there are 3 extraction wells on the site and 4 extraction wells off-
site.
The site consists of a number of high-ceiling one-story buildings
now occupied by other businesses. Light industry and businesses also
rent the other four buildings that were once part of the manufacturing
complex.
C-35
-------
KEY
• Monliorinf well
A Extraction weU
Conumintted pound wttcr pturac
Figure 1. Contaminated ground-water plume.
FIGURE C-4
C-36
-------
According to 1990 census information, approximately 1000 people
live in the 10 block area surrounding the plume of groundwater contami-
nation connected with the site. The bulk of the population resides on
the south side of Highway 101, with 1990 census information showing only
10 people living on the north side of Interstate Highway 101.
Land in the area surrounding the site has been used for residen-
tial, commercial, and agricultural purposes for many years. The areas
surrounding the site are zoned for a mixture of multiple family and two
family residential units, and general and limited industrial use.
Within the four block vicinity of the site, there are 409 housing units
with many of these being 14 unit complexes.
The area north of Highway 101 is zoned light industrial and
planned community designation. This type of zoning is designed for
research and development, office space, and light manufacturing. No new
housing is allowed under this zoning. Within this area approximately 20
acres of land are used for commercial farming. Two schools are located
northwest within a half mile of the site. The nearest city park is one-
half mile southwest of the site.
Four major water-bearing zones (aquifers)-defined as the A-, B-,
intermediate - and C- zones-exist at the site. The A-aquifer is the
shallowest and the C-aquifer is the deepest of these three zones. The
approximate depths (below ground level) at which these zones occur at
the site are as follows: A: 10-20 feet; B: 30-40 feet; and intermediate:
60-75 feet. A regional impermeable zone, the B-C aquitard, is reported-
ly located at a depth range from 100-150 feet below ground level. The
C-aquifer, which supplies most of the municipal water in the region, is
located below the regional aquitard.
C.4.1 Nature and Extent of Problem
Between August 1985 and August 1986, the neutralization sump and
some surrounding soil (70 cubic feet) was excavated and sequentially
backfilled to grade level with pea gravel, well-compacted clay, and
asphalt. In September 1986, about 255 cubic yards of soil were excavat-
ed under the "wet floor".
In May 1987, soil samples were taken at five feet depth from
borings at the sump and soil around it and analyzed for organic contami-
C-37
-------
nants {"wet floor" soil was not analyzed). Results are shown in Table
C-13. Results of the post-remediation analyses showed the presence and
levels of various organic contaminants not originally found in the
neutralization sump area.
TABLE C-13
ORGANIC CONTAMINANTS IN ON-SITE SUBSURFACE SOIL
1, 1-Dichloroethane
1 , 1 , 1-Trichloroethane
1 , 1-Dichloroethylene
t-l,2-Dichloroethylene
Trichloroethylene
Sump
Before
(4/85)
nd
10
7.3
nd
nd
Sump
After
(5/87)
nd-24
nd-69
nd
nd-110
41-220
" Wet "Wet
Floor" Floor"
Before After
(10/86)
na
na
na
na
nd-100
na
na
na
na
na
nd = not detected above detection limits
na = not analyzed
Since 1986, 7 on-site monitoring wells have been installed to
characterize the vertical and horizontal extent of contamination. The
most contaminated wells are located near to or downgradient, north to
northwest, from the sump excavation. The migration of contaminants
through the major water-bearing aquifers is being monitored. The two
shallowest aquifers (A and Bl) are contaminated with dichloroethane,
1,1,1-trichloroethane, 1,1-dichloroethylene, trans-1, 2-d-
ichloroethylene, and trichloroethylene at concentrations exceeding state
or federal drinking water standards (Table C-14).
Three extraction wells were installed on-site: two draw from the
A-aquifer and one draws from the B-aquifer. The untreated ground water
is discharged by permit into the sanitary sewer system.
C-38
-------
TABLE C-14
ORGANIC CONTAMINANTS IN ON-SITE MONITORING WELLS
Data are taken from the quarterly monitoring well report, fourth
quarter, 1990. The historical range of contaminant concentrations for
each aquifer is given with the highest concentration currently measured
in that aquifer given in parenthesis.
1,1- & 1,2-Dichloroethane
1,1,1 -Trichloroethane
1 ,1-Dichloreothyiene
t-1 ,2-Dichloroethyiene
Trichloroethylene
Drinking Water
Standards
0.5'
200'-"
6"
10C
5'-b
Aquifer Concentration (ppb)
A B Intermediate
nd- 13 (4)
nd- 45 (6.2)
nd- 17(1.4)
nd- 36(7.03)
nd-1 22(20)
nd- 20(nd)
nd-1 3,000(1. 8)
nd- 1 ,000 (nd)
nd- 14(4.3)
1.3-7,500(26)
nd
nd
nd
nd
nd
nd = not detected above detection limits
•California MCL
"Federal MCL
'proposed California MCL
Starting in late 1986, 24 off-site monitoring wells were installed to
assess the extent and severity of the migration of contaminants from the
site. The two shallowest aquifers (A and B) were found to be contami-
nated with organic compounds. Six chlorinated organic compounds
identical to those found at the site, have been detected in the ground
water (Table C-15). Concentrations of dichloroethane, 1,1,1-trichloro-
ethane, 1,1-dichlorethylene, trans-l,2-dichloroethylene, and trichloro-
ethylene, exceed drinking water standards.
Four extraction wells were installed off-site: two draw from the
A-aquifer and two draw from the B-aquifer. The untreated ground water
is discharged by permit into the City sanitary sewer system.
C-39
-------
TABLE C-15
ORGANIC CONTAMINANTS IN OFF-SITE MONITORING WELLS
Data are taken from the quarterly monitoring well report, fourth
quarter, 1990 (10). The historical range of contaminant concentrations
for each aquifer is given with the highest concentration currently
measured in that aquifer given in parenthesis.
1,1- & 1,2-Dichloroethane
1,1,1 -Trichloroethane
1 , 1 -Dichloreothylene
t, 1 ,2-Dichloroethyien8
Trichloroethylene
Drinking Water
Standards
0.5'
200'*
6'
10e
5'-k
Aquifer Concentration (ppb)
A B intermediate
nd- 330(64)
nd- 2500(1 40)
nd- 420(47)
nd- 580(580)
nd- 1599(520)
nd- 310(310)
nd- 1 100(12)
nd- 350(140)
nd- 550(550)
nd- 320(200)
nd
nd-1.0(nd)
nd
nd
nd
nd = not detected above detection limits
•California MCL
"Federal MCL
"proposed California MCL
The concentrations of site-related contaminants in ground water
are at levels that would be of potential public health concern if
domestic use of the contaminated ground water occurred. No municipal
wells currently exist near the site. There are several private wells in
the area or directly within the contaminated ground-water plume, but
most of these are not active. Of those wells that are still active, the
water pumped from these wells is not currently used for domestic
purposes. At this time, sufficient water for municipal use is available
from aquifers having higher quality water and water yield. Additional-
ly, regulatory barriers exist to prohibit installation of shallow
private wells.
A theoretical model (Farmer-See Appendix A) was used to estimate
the contaminant soil gas flux and the air concentrations within single-
family residences located above the contaminated plume. Air concentra-
tions for an "average case scenario", "plausible maximum scenario", and
a "most plausible case" were calculated. These scenarios use different
C-40
-------
assumptions for area of infiltration (crawlspace or crack around
perimeter of building), the fraction of air that infiltrates from the
crawlspace, and/or the air exchange rate of the home or business.
Based on the concentrations of organic contaminants accumulating
in a house as developed by this model, the risk of developing cancer
from inhaling volatilized organic contaminants for current residents
living above the contaminated ground water was calculated. The exposure
was assumed to be 24 hours per day, 365 days per year for 40 years .for
residents. The exposure was assumed to be 8 hours per day, 260 days per
year for 40 years off-site workers.
The lifetime excess cancer risk from the inhalation of volatiliz-
ing organic contaminants from the shallow ground water was estimated
from the addition of the individual risks to the three potential
carcinogens, 1,1-dichloroethylene, 1,1-dichloroethane and trichloroeth-
ylene.
Lifetime excess cancer risk for off-site residents ranged from 1.8
x 10"8 ("no increased risk") for the least conservative estimate and 2.9
x 10"5 ("no apparent increased risk") for upperbound worst-case condi-
tions. Lifetime excess cancer risk for off-site workers ranged from 2.1
x 10"9 ("no increased risk") for the least conservative estimate and 3.5
x 10"6 ("no increased risk") for upperbound worst-case conditions.
Potential risks were also assessed for the noncarcinogens (trans-
1,2-dichloroethylene and 1,1,1-trichloroethane) for residents above the
contaminated ground-water plume. The analyses indicated that the
inhalation of organic contaminants from off-site ground water would not
result in adverse noncarcinogenic effects.
C.4.2 Conclusions
Based on information reviewed, ATSDR and CDHS concluded that the
site is not an apparent public health hazard. As noted in the human
exposure pathways section above, off-site resident and worker exposure
is predicted by an air-model but the exposure is at a level below that
of health concern.
Future significant exposure to ground-water contaminants is
unlikely if the ground-water extraction and treatment system reduces
concentrations of site related contaminants to below levels of health
C-41
-------
concern, no wells currently in place are used for domestic purposes, and
future drinking water wells are not placed in areas of known contamina-
tion if ground-water remediation does not clean up contaminants to
drinking water standards.
C.4.3 Comments on Case Study
As we reviewed this case study, we were struck by the similarity
to the preceding case study. The two sites are both located at the
southern end of San Francisco Bay and are only a few miles apart. They
appear to have similar hydrology and geology. Based on the reported
chemical concentrations in the ground water plumes for the two sites,
the groundwater appears to be substantially more contaminated at the
site discussed in this section. Based on the relative lifetime excess
cancer risks for current off-site residents for the two cases, it
appears that for this second case indoor air model calculation used a
diffusion only method and a flux attenuation factor based on an infil-
tration area represented by a 0.5 cm crack around the building perime-
ters. We note that this is extremely more restrictive (yields lower
predicted indoor concentrations) than the assumptions used for current
residents in the preceding example but is similar to that used for the
former elementary school and future residences in that case example.
This assumption is likely overly restrictive and insupportable in both
case examples.
For illustration purposes only, we have calculated indoor air
concentrations for off-site structures above the ground water plume for
the current site using the same parameters used for current off-site
residences in the previous case example. We also ignore corrections for
the capillary fringe (as did the study authors) because only a screening
level result is desired. [This correction should be made if a better
estimate is needed.] Because we did not have access to sufficient data
to calculate the geometric mean concentrations in the ground water, as
was used in the previous case, we have used one-half the maximum values
C-42
-------
TABLE C-16
HYPOTHETICAL CALCULATIONS FOR INDOOR AIR
IN STRUCTURES ABOVE THE GROUND WATER PLUME
Compound
1 , 1 -Dichloroethylene
Trichloroethylene
1 , 1 -Dichloroethane
Trans 1 ,2-Dichloroethylene
1,1,1 -Trichloroethane
Sunnyvale
Off-site cone.
(max) in
ground water
4.3
120
2.5
11
6.3
Mountain
Viow"1
Off-site Cone.
(max) in
ground water
47
520
64
580
140
Modeled Indoor Air Concentrations (ug/m3)
Sunnyvale
AVG
4.24
5.50
0.087
0.46
0.87
Plausible
Max
16.4
21.3
0.34
1.77
3.37
Mountain View
AVG
23
12
1.1
12
9.7
Plausible
Max
90
46
4.3
47
37
(1) - For calculation purposes, one-half this value was used in model
C-43
-------
given in Table C-15. The results are in Table C-16. These calculated
results indicate that under these modeling assumptions, indoor air near
the site would be substantially more contaminated than in the previous
case. Thus, it is likely these estimates are much too high.
C.5 BUILDINGS OVER GROUNDWATER PLUME (C)
The site is located in a suburb of Casper, Wyoming. A residential
subdivision is located approximately one-quarter mile northeast of an
industrial park. An oil refinery is located west-southwest of the
subdivision on the opposite side of the North Platte River.
In the mid-1980's, a contaminated water plume was discovered
underlying part of the residential subdivision. Because the residences
all use wellwater, an indoor air study comparing periods with and
without showers operating was conducted.
Air sampling was performed in five residences with contaminated
wells, five homes with uncontaminated wells and five Casper area homes
as controls. Each home was sampled during a 5-hour period with the
shower operated at least 10 minutes and on a subsequent day, another 5-
hour period without shower operation. In addition, 5-hour basement and
5-hour outdoor air samples were collected. Samples were obtained with
low-flow air pumps and Tenax collectors and were analyzed for volatile
organics. The study was designed to compare shower vs. non-shower
concentrations and to compare results among the three study groups to
see if significant differences existed in indoor concentrations.
C.5.1 Nature and Extent of Problem
Organic contaminants found in groundwater samples were used to
divide the residents into two groups, contaminated (Group 1) and
uncontaminated (Group 2). In the residential air study, these two
groups were compared to background homes in the Casper area (Group 3).
Residential sampling included indoor air with shower (IAS), indoor
air without shower (IA), basement air (BA), and outdoor air (OA)
collected in the backyard of each home. In addition, a questionnaire
was administered at each home to identify potential alternative sources
of airborne contaminants.
C-44
-------
The air/Tenax analysis was designed for quantitation of HSL
volatile compounds. Tenax tubes were used for sample collection. The
Tenax tubes can retain most volatile HSL compounds (lowest boiling,
i.e., chloroethane excluded), but do not adequately retain many light
hydrocarbons and low boiling compounds, i.e., saturated hydrocarbons
-------
O3
C
cc
-z.
UJ
O
•z.
o
o
r-
U
UJ
_i
CO
o
U
y
z
CC
o
UJ
z
cc
o
CQ
CC
O
Z
o
cc.
t
o
UJ
UJ
CD
Z
a
o
0
a
cS
Outdoor
c
E
o
V)
CD
S
C 0
O .C
W yj
*
c S
o o
« £
f -
°- i
0
TJ
3
O
Basement
o
c o
o ^
2 to
o
Q.
C $
0 0
111 -C
£-
°- $
<*J
co" *
CO —
3
1 I
m
^
^
rZ «
Benzene
CO
• in
1 I
t 1
1°
! 1
i !
. .
1 i
o
•o
•c
z
Carbon Tetrac
I i
1 I
1 I
1 1
1 I
in
d *•;
3
: i
; i
0
c
.c
1,1-Dichloroe
1 I
1 1
1 1
1 1
1 1
§3
l i
i :
Ethylbenzene
I i
I I
1 I
1 !
1 I
1 1
! i
*~
2-Hexanone
6 •*:
7 o
3
I !
1 1
1 1
1 1
3
co •-;
3
i°"
Methylene
Chloride
I i
I !
v—
^2
3
O <">
3
1 1
23
f J-
3
in" >"
n —
3
0)
3
Tetrachloro-
ethylene
: ;
(N
0 CO
3
i°'
! 1
0)
"° 9
3
CO
' ui
3
CO
"9 «•>
3
Trichloro-
olhylene
?5
3
i :
f*s
in «
o> ro
3
en
6 °\
3
«2
in
in
iri PI
O) »-
CO t
'
1 1
(*^
!*• "?
3
P>
r-.'
CM
en i;
CO (si
O
to
d
r^
en ^;
^ *~
in en
3
Toluene
C-46
-------
T3
03
C
O
o
O
LLJ
CO
cn ^~
(j M
C
O
•£ ^
'•6 ^j
3
(0
Q.
3
O
<5
o
o
•5
3
0
1
isement
m
$
c o
o - to
d
(O
4 oi
Benzene
1 I
d "
3
9 d
3
° d
3
3
O
T3
•c
O
Carbon Tetrach
! 1
I 1
I 1
1 1
: :
i ;
c
03
1,1-Dichloroeth
1 i
1 I
1 I
1 I
! !
1 1
Ethylbenzene
1 i
1 I
1 i
CO ^j
3 *~
i !
cn ^_
3 *~
2-Hexanone
m "
n
in •
3 *~
: i
i i
CO
r^" ^
CO 01
3
in
co q
3
Methylene
Chloride
I I
I I
D "J
i0'
in
6
-------
In order to calculate airborne concentration, the flow rates were
adjusted to conditions of standard temperature and pressure. The
compound weights were blank corrected by subtracting mean weights of all
field blank analytes from the reported field sample value prior to
dividing by standard liters of air to calculate the flow rate.
Descriptive statistics, including mean and range of concentration
were calculated for each sample. When calculating mean values, one-half
the instrument detection limit was used for non-reported values. The
data were lognormally distributed and samples were randomly selected.
Data were analyzed statistically using the Mann-Whitney (MW) U-
Test for independent groups. The data were then logarithmically
transformed and analyzed using the more powerful t-test for basement
samples and Analysis of Variance (ANOVA) for indoor and outdoor samples.
One tailed probability values (p) are reported as an indication of
statistical significance.
The MW test converts the original data to ranks and compares two
independent groups of data. Ranks are used due to the fact that the
small size and high variability of the data do not met the assumptions
of normality or homogeneity of variance necessary to use parametric
statistical analysis. The MW test closely approximates the parametric
t-test. It is less powerful than the t-test in finding a difference
between two means if one exists.
A questionnaire was completed by each participant to identify the
following potential sources of airborne organic contaminants: cigarette
smoking, use of water operated appliances, use of organic chemicals,
house cleaning, and open windows during sampling. The results indicated
that Group 2 had more smokers, greater use of appliances, more house
cleaning, and more open windows during sampling. Participants were
asked not to use chemicals during the sampling period.
C.5.2 RESULTS
Shower Compared to Non-Shower Exposure
Samples collected with and without a shower from homes in Group 1
were compared to determine whether groundwater contaminants entered
indoor air via showering. As expected, PCE, TCE, and TCA levels were
C-48
-------
slightly higher in IAS samples compared to IA samples. A statistical
comparison of the data indicated that the results were not significant
at the (ANOVA) p<0.10 level. The highest significance level was found
for toluene at (MW) p<0.34. IA and IAS samples also contained virtually
the same amounts of contaminants when comparisons were made for Groups 2
and 3.
Airborne Contaminant with Shower Exposure in the Three Groups
Airborne contaminant exposure in IAS samples were assessed in the
three study groups to determine the contribution of contaminants from
drinking water to residential air. Mean values were higher in Group 1
homes, compared to Groups 2 and 3, for PCE, TCE, and TCA. However, the
values were not found to be statistically significant at the (ANOVA)
p<0.10 level. The highest level of significance was found for TCE
between Groups 1 and 2 (MW) (p<0.20). Significance levels for other
comparisons ranged from (MW) p<0.27 to (MW) p<0.42. The lowest levels
of PCE and TCA were found in Group 3 homes. TCE levels were very
similar in Groups 2 and 3.
Basement Air Samples
None of Group 2 homes had basements, consequently, comparisons
were made between Groups 1 and 3. PCE concentration was higher in the
Group 1 samples at a significance level of (t-test) p<0.05. Mean TCE
concentration was higher in Group 1, but not significantly (t-test
p<0.17). The other contaminants were not tested statistically due to
the small difference between mean values.
Ambient Air Samples
Outdoor air concentrations of all contaminants were consistently
lower in Group 1 compared to Group 3 indicating that they did not
contribute to indoor levels in Group 1. TCE and PCE were not found in
Group 1 outdoor samples, but were found in Group 3 outdoor samples.
Toluene and benzene levels were lower in Group 1 samples at significance
levels of (MW) p<0.26 and (MW) p<0.20, respectively. TCA levels were
slightly lower in Group 1 samples compared to Group 3 samples.
C-49
-------
C.5.3 Conclusion
Contaminants found in previous well water samples from homes in
Group 1 were also present in indoor air. Mean levels of PCE, TCE, and
TCA were higher in Group 1 homes compared to Group 2 and 3 homes (not
statistically significant at p<0.1 level). The fact that outdoor
concentrations of these contaminants were relatively low, even though
the primary wind direction was found to be from the southwest emanating
from the industrial park toward the subdivision, suggests an indoor
source of contaminants. Showering was investigated as a potential
source. TCE, PCE, and TCA mean concentrations were found to be slight-
ly, but not significantly higher in IAS as compared to IA samples
collected in Group 1 homes.
PCE concentrations in basement samples were significantly higher
(t-test; p<0.05) in Group 1 homes compared to Group 3 homes. The source
of PCE in basements may have been vapor directly from contaminated
groundwater or from appliances that use water, such as hot water heaters
or washing machines, located in the basement. If groundwater were the
source, TCE levels would also be expected to be elevated. TCE values
were higher in Group 1 homes, but not at as significant a level (t-test
p<0.17) as PCE. A possible explanation for this result is that PCE
partitions from water to air more readily than TCE and, consequently,
may volatilize from groundwater and accumulate in residential basements
at a higher rate than TCE. The air/water partitioning of the two
contaminants was determined using Henry's Law constants at constant
atmosphere: PCE, 1.1 x 103; TCE, 5.5 x 102. The larger of the Henry's
Law constant, the greater the equilibrium concentration of the compound
in air compared to its concentration in water. It should also be noted
that the sample size for the basement sample comparison is quite small,
three homes per group. Very small sample size makes a statistical
comparison of the data difficult and consequently, makes the data
comparison inconclusive.
The influence of other potential sources of airborne contaminants
were investigated. Prior to sampling, all participants agreed to avoid
use of household or other chemical during the sampling period. Group 2
homes had increased numbers of cigarette smokers, use of water operated
C-50
-------
appliances, use of organic chemicals and house cleaning activity during
sampling. These factors would be expected to increase the airborne
levels of the same volatile organic contaminants found in well water, as
well as increased volatile organics from cigarette smoke and organic
cleaners and solvents. However, the measured indoor air contaminant
levels were not consistently elevated in Group 2 homes compared to
Groups 1 and 3.
Cigarette smoking may not have been an important factor in
contributing to the airborne contaminant levels presented in Table 2-17.
According to the Surgeon General Report, 1981 (Wynder, E. and Hoffman,
D.), none of the airborne contaminants identified in the study were
found as "major toxic agents" in cigarette smoke. The report does say
that the list presented was incomplete and adds that cigarette smoke may
contain "such carcinogens as volatile chlorinated olefins". This group
may contain PCE and TCE.
The fact that more windows were open during sampling in Group 2
homes may have lowered airborne indoor contaminant concentration. This
would negate the effects of the previously discussed factors that tend
to increase contaminant levels.
The levels of contaminants found in homes in all three study
groups were typical of indoor concentrations found in enclosed living
spaces.
Outdoor mean contaminant levels in the three groups and 24-hour
samples were also found to be lower than those in the urban areas of
four major U.S. cities. (Houston, St. Louis, Denver, and Riverside).
C.5.4 Comments on Study
This study was fairly complex in that it monitored at three type
homes, with and without showers operating, as well as ambient air. It
comes to the conclusion that the indoor air quality in all cases is
within the "typical" range for residences. It should be noted that
because none of the Group 2 homes had basements, whereas group 1 and
Group 3 did, and the possibility of soil gas intrusion from the contami-
nated groundwater plume existed for Group 1 homes, the Group 2 homes
would not appear to be an adequate control group. However, the design
C-51
-------
has left open the question whether or not soil gas is infiltrating the
homes even at these low levels by presenting an inconclusive statistical
result indicating a significant difference from the controls.
Extraction wells were later installed at this site to control
groundwater migration.
C-52
-------
C.6 BUILDING NEAR SLUDGE DISPOSAL PIT
The site is located in a rural section of Brunswick County, North
Carolina (Figure C-5).
The site was used by sludge hauling and oil spill cleanup compa-
nies for the disposal of septic tank sludge, oil sludge, and other waste
materials from 1969 to 1976. These wastes were disposed in shallow
(approximately two-to eight-feet deep) unlined pits or directly on the
land surface at the site. A residential subdivision now exists in the
area of the former disposal areas. Typical dwellings in the subdivision
consist of manufactured homes situated on one or two-acre lots, each
having a private domestic well. There are approximately 150 residential
lots in the area, 70 of which were occupied in 1990.
In May 1976, the owner was informed that an oil disposal pit
violated North Carolina statutes and must be cleaned up immediately.
This pit was approximately 60 feet long, 20 feet wide, and two to four
feet deep. At the time, it was estimated that approximately 2,000 to
3,000 gallons of black oil were contained within the pit. The owner
pumped the oil from the pit and then covered the pit with soil.
Documentation pertaining to the chemical composition of materials
disposed in the pit, the fate of the liquid removed from the pit, and
the quantities and characteristics of the material buried on site is not
available.
C.6.1 Nature and Extent of Problem
In August 1976, the failure of an earthen berm allowed approxi-
mately 20,000 gallons of black oil to escape from an unlined pit on the
property and flow into an on-site creek. The oil remaining in the pit
was pumped and transferred to tank trucks and hauled away. The bottom
sludges and some oil-stained soils were excavated and disposed in the
County Landfill. The remaining thicker sludges, which could not be
C-53
-------
SITE LOCATION MAP
SOURCE: USCS OUAtXUNGLE ACME.
NOflTH CAROCXA 1514.
CONTOURS AND ELEVATIONS
IN METERS
SCALE
FIGURE C-5
C-54
-------
pumped, were mixed with sand and buried on site; however, the burial
location was not documented.
In 1980, the property changed hands through foreclosure and
subdivided for residential development. Family A purchased two lots in
the subdivision in 1982 and 1983 and in July 1983, uncovered creosote,
oil spill residue, tank bottom sludges, and septic tank sludges in the
yard.
State of North Carolina personnel sampled and analyzed soils at
the site in response to the discovery. The investigation confirmed the
presence of these materials in soil close to the surface at locations on
all sides of the house. The well was contaminated with phenols (100
mg/1), dimethyl benzene, methyl phenol, one unknown organic compound,
creosote, and septic tank sludge. The shallow well was subsequently
condemned and the house was connected to a neighbor's well system.
In September 1983, EPA and the Region IV Field Investigation Team
(FIT) conducted an investigation of the property. The investigation
included an electromagnetic survey of the area; initial ambient air
monitoring under the home; and collection and laboratory analysis of
surface water and sediment samples, five subsurface soil samples, and
six groundwater samples from off-site and on-site wells. All readings
of ambient air were negative. Based on surface water and sediment
sampling and analysis, no contamination of Chinnis Branch was detected.
Groundwater analysis for inorganic compounds indicated the presence of
seven priority pollutant constituents. Groundwater analyses for organic
compounds (purgeable, extractable, pesticide/PCBs, and other chlorinated
compounds) indicated the presence of 81 different compounds, many of
which are commonly associated with creosote or coal-tar derivatives.
Soils analyses identified a total of 24 inorganic constituents in the
subsurface soil samples, 10 of which are target compound list (TCL)
parameters. Soils analyses also indicated the presence of 46 organic
compounds in the subsurface samples, 14 of which are TCL parameters, and
many of which are polynuclear aromatics associated with asphaltic and
coal-tar derivatives.
In February 1984, EPA used ground penetrating radar (GPR) to
further delineate the site boundaries. The GPR survey revealed two
C-55
-------
anomalous areas that roughly correspond to the location of two surface
impoundments illustrated on historical maps. The pits were estimated to
be six feet deep toward the center and three feet deep toward the edges.
In March 1984, an Immediate Removal Action (IRA) at the site was
requested by the EPA. The IRA at the property consisted of the excava-
tion and removal of approximately 1,770 tons of oil sludge and soils
which were transported to a hazardous waste landfill in Pinewood, South
Carolina. Soil cleanup activities were completed in April 1984.
In May 1984, EPA installed nine groundwater monitoring wells at
the site and conducted sampling and analysis of groundwater samples for
volatile organic compounds. Relatively high concentrations (in the ppm
range) of benzene, ethyl-benzene, toluene, and xylenes (BETX) were
detected in the groundwater samples from the shallow aquifer. Groundwa-
ter sampling indicated that the deep aquifer had not been affected by
the contamination in the shallow zone. Neither the upgradient nor down-
gradient deep wells contained detectable volatile organic compounds.
EPA recommended that these monitoring wells be regularly monitored.
In 1988, the groundwater monitoring wells were resampled and
samples analyzed for purgeable organics and some inorganic parameters
(metals, nutrients, etc.). As was the case with the EPA 1984 data, the
1988 monitoring well data indicated that gasoline or waste petroleum
product (especially benzene, toluene, and xylenes) were still prevalent
at the site. However, the concentration of purgeable organics detected
during the two sampling episodes varied significantly. In addition, the
1988 data indicated the possibility of low level benzene, ethyl-benzene,
and xylenes in a deep well which would indicate that an aquifer underly-
ing the surficial aquifer had now been affected.
In 1989, EPA determined that the level and extent of on-site
contamination warranted a more thorough investigation and assessment.
Consequently, a Remedial Investigation (RI) was undertaken through a
CERCLA action. The purpose of the initial RI was to assess the nature
and distribution of contaminants at the site and to provide the data
necessary for developing a Feasibility Study (FS) and ultimately
conducting a Remedial Action (RA).
C-56
-------
The field investigation of the initial RI was conducted from
January 1990 through April 1990, and included soil gas surveys, surface
and subsurface soil sampling, monitor well installation, groundwater
sampling (of monitor wells and residential wells), hydraulic conductivi-
ty testing, and identification of potential ecological and human
receptors. Three separate study areas were addressed during the initial
RI. Area 1 was comprised of the Family A Property, which is the
original site. Area 2, located approximately 1.5 miles from Area 1 in a
similar rural residential area, was identified during the document
review. Historical records indicated that a waste oil disposal pit
existed within Area 2. Area 3, located adjacent to Area 1, was identi-
fied in historical records as a potential waste disposal site based on
historical aerial photographs of the area. Area 2 was not well docu-
mented and could not be located even after extensive interviewing and a
thorough reconnaissance of the area. Therefore, no field investigation
was initiated. A majority of the field investigation activities
subsequently centered in Areas 1 and 3.
The initial RI report concluded that the extent of contamination
which poses a risk to human health or potential ecological receptors is
limited to the area in the vicinity of the Family A residence. Contami-
nation has impacted surface and subsurface soils, the shallow aquifer,
and creek sediments in this area. The predominant media and compounds
contributing to public health risk include: benzene and lead in ground-
water, and carcinogenic polynuclear aromatic compounds (PAHs) and lead
in surface soils.
C.6.2 Residential Air Monitoring
Air samples were collected in February 1990 in the crawl space and
inside the residences of Family A and Family B. Sampling was conducted
at these two residences since they are situated on or near the former
waste disposal pits. Unlike the Family A residence, the Family B
residence is not located above or adjacent to identified contaminated
areas.
A total of five residential air samples were collected using EPA
method TO-14 from within the crawl spaces and interiors of the two
C-57
-------
homes. The air samples were analyzed for the 34 TCL VOCs listed in
Table 2.18. Only three compounds, chloromethane, methylene chloride,
and 1,1,1-trichloroethane were detected. Methylene chloride was
detected inside the Family B residence at a concentration of 11 ppbv.
Low levels of chloromethane (16 ppbv) and 1,1,1-trichloroethane (1.5
ppbv) were detected in the crawl space beneath the Family B residence.
No VOCs were detected within or beneath the Family A residence.
Of these three VOCs, 1,1,1-trichloroethane was the only compound
detected in sampling in the vicinity of the site and it was only
detected in low//g/kg concentrations in background soil samples. None
of these VOCs were detected in known contaminated areas. These
observations suggest that the source(s) of VOCs in the Family B resi-
dence are not related to the site.
C.6.3 COMMENTS ON INDOOR AIR EFFORT
The fact that no VOCs were detected in or under the Family A resi-
dence is counter to published information on typical residential indoor
air quality. These data are suspect.
Perhaps of more significance is what was not sampled and analyzed
for. The soils analysis demonstrated the site is contaminated with
creosote, polynuclear aromatic hydrocarbons, phenols, and other hazard-
ous compounds. Even low concentrations of some of these can result in
substantially elevated risk numbers. Vapor pressures are high enough to
consider their transport.through soil gases especially considering
contamination was found in the immediate vicinity of the residence.
C-58
-------
TABLE C.18
TARGET COMPOUNDS
COMPOUNDS
CHLOROMETHANE
BROMOMETHANE
VINYL CHLORIDE
CHLOROETHANE
METHYLENE CHLORIDE
1,1-DICHLOROETHENE
1,1-DICHLOROETHANE
CHLOROFORM
1,2-DICHLOROETHANE
1,1,1-TRICHLOROETHANE
1.1.1 -TRICHLOROETH ANE
CARBON TETRACHLORIDE
DIBROMOCHLOROMETHANE
1 ,2-DICHLOROPROPANE
TRANS-1 ,3-DICHLOROPROPENE
TRICHLOROETHENE
BENZENE
DIBROMOCHLOROMETHANE
1 ,1 ,2-TRICHLOROETHANE
CIS-1 ,3-DICHLOROPROPENE
BROMOFORM
1,1,2.2-TETRACHLOROETHANE
TETRACHLOROETHENE
TOLUENE
CHLOROBENZENE
ETHYL BENZENE
TOTAL XYLENES
ACETONE
CARBON DISULFIDE
VINYL ACETATE
METHYL ETHYL KETONE
METHYL ISOBUTYL KETONE
METHYL BUTYL KETONE
STYRENE
1,2-DICHLOROETHENE (TOTAL)
C-59
-------
C.7 FUGITIVE DUST CASE STUDY
Primary zinc smelting operations at this facility (Figure C.6)
since the turn of the century emitted large quantities of zinc, cadmium,
lead, and copper into the atmosphere in the vicinity of the plant.
Significant concentrations of these heavy metals have been measured in
the soil within a large area surrounding the plant. The plant area is
currently a Superfund and a RCRA site. The facility is currently
operating under new ownership as a secondary zinc smelter. A NAMS
reference monitor is actively monitoring the area for attainment of the
lead NAAQS. The facility has been in compliance with the standard for
the past several years.
C.7.1 Nature and Extent of Problem
Investigations were conducted at two homes by the State in
response to citizen requests. Two dust sampling methods used were based
on protocols of the Center for Disease Control, Lead Poisoning Preven-
tion Branch, Division of Environmental Hazards and Health Effects,
Center for Environmental Health and Injury Control (CDC). The first
method used a vacuum pump and filter cassette to pick up dust from a one
square foot area. A plastic template was used to measure the area to be
sampled - except on window sills where the area was estimated. The
second method, based on the Kellogg Dust Protocol Number 2, was wipe
sampling. A Whatman Number 42 filter paper dampened with isopropyl
alcohol was used, although the original method specified alcohol swabs.
Field blanks were taken at each residence by momentarily opening
the cassette. The wipe sample field blank filter was removed from the
transport tube, moistened with alcohol, then immediately returned to the
transport tube. A new disposable latex glove was used to handle each
filter.
C-60
-------
PREVAILING WIND
X = AIR MONITORS
MINE TAILINGS
1 SMELTER
X COMPLEX
LOCATIONAL MAP
FIGURE C-6
C-61
-------
The U.S. Public Health Service/EPA Toxicological Profile Document
for lead uses a value of 300 /jg/g lead in household dust, as a baseline-
value to estimate overall population lead exposure from various media.
The urban Particulate Standard 1648 from the National Institute for
Standards and Technology (NIST), which is a composite urban dust sample,
has values of 6,550 jig/g for lead, 75 /ig/g for cadmium and 4,760 /zg/g
for zinc. The lead values obtained from the household dust in the two
homes were well above the national averages.
An indoor monitoring study was developed to determine the extent
of exposure to lead and heavy metals inside the residents homes.
Twenty-five homes were selected randomly in areas where there is a large
probability of deposition of airborne lead. Atmospheric dispersion
modeling with ISCLT was used to determine these areas. The homes were
vacuumed sampled with mini rams outfitted with HEPA filters.
Results from the indoor study demonstrated that most of the homes
in the area had very high levels of lead. However, since a health
threat had not yet demonstrated to exist due to the high levels of lead
found in the residential homes, ATSDR was contacted to determine whether
the environmental contamination is a public health hazard for the
residents.
ATSDR proposed an exposure study to determine the extent of the
health hazard, if any, in the town. Although the study has been
completed, the results have not yet been reported.
The study:
• Measured blood lead and urinary cadmium levels among a
representative group of residents
• Compared the levels of lead and cadmium found in residents
to levels found in a comparison community while controlling
for other risk factors known to influence exposure to these
heavy metals
• Performed a standardized panel of medical tests on both
groups
A Superfund Removal Action was scheduled to be performed at those
homes in with elevated lead levels. The objective was to clean the
homes and eradicate them of lead, thus reducing the probable health
C-62
-------
threat. However, Superfund is reluctant to perform this Remedial Action
until adequate evidence is available to identify the source of lead dust
currently infiltrating the homes so that recurrence can be prevented.
Data obtained in the 25 home sampling program showed that lead levels
inside the homes were much higher than in the outside soil. Although
this suggests an airborne pathway for the indoor air contamination, the
data did not conclusively establish whether the contamination was from
soil erosion, the smelter, or other sources.
The problem, therefore, has become one of determining the actual
current source of lead dust infiltrating the homes. Based on current
information, it can be demonstrated with reasonable scientific certainty
that:
1. Past primary Zn smelting practices have contributed to the
burden of hazardous substances in soil and dust
2. Present secondary Zn recovery processes are adding to the
burden of hazardous substances in soil and dust
3. Both 1. and 2. are "non-de minimus"
4. Other common anthropogenic sources are responsible for little
of the hazardous substance burden
The problem, thus, becomes one of determining whether the lead dust
is from the old primary smelter operations and is being tracked into
homes or carried in by wind erosion, or whether it is from wind dispen-
sion of the current secondary smelting operations.
There are five basic identification methods that can be used:
geostatistical, elemental composition coupled with multi-variate analy-
sis, chemical speciation, individual particle analysis, and lead isotope
ratios. In the geostatistical approach, the spatial variability of the
substance of interest is determined and displayed on isoplethic plots.
while this provides an easily understandable presentation from which one
can infer source locations, a large number of highly representative
samples are needed.
The elemental composition method requires determination of multiple
elements from both potential sources and receptors. The data are then
analyzed using multi-variate techniques such as ratio discriminant
C-63
-------
functions and factor analysis. While one can potentially learn a lot
from a few samples and infer sources by comparison of source and
receptor patterns, the method is not sensitive for subtle sources and
post-release chemical transformation can pose analysis difficulties.
Also, the method depends on having a fairly complete picture of all
potential sources.
Chemical speciation methods depend on determining the actual
chemical compound or "species" in source and receptors. Very selective
analytical techniques are required. The method suffers significantly if
post release transformations take place. For lead, many source com-
pounds may weather to lead sulfate making source identification diffi-
cult or impossible.
Individual particles may be analyzed microscopically and by various
analytical techniques to determine elemental and chemical composition.
This approach appears to combine the best features of the chemical
speciation and multi-variate methods. For example, particles from lead-
based paint would be associated with carbonates, chromates, titanium,
and zirconium whereas lead particles from batteries would contain
antimony.
The use of lead isotope ratios can potentially be a very powerful
technique, especially for the current problem. Lead ores from different
geological ages contain different ratios of the four lead isotopes
206Pb", 207Pb, 208Pb to the minor isotope 204Pb. Because of this, different
economically important lead ore bodies have different lead isotope
fingerprints as shown in Table 2.19. These ratios can easily be
determined using either thermal ionization mass spectroscopy (MS) or
inductively coupled argon plasma MS.
C-64
-------
TABLE C.19
LEAD ISOTOPE RATIOS IN WORLD-WIDE LEAD ORES
Source
Rosetta, S. Africa
Ivigtut, Greenland
Broken Hill, Australia
Bunker Hill, Idaho
Coeur D'Alene Group
Pine Point, NWT
Tintic, Utah
Casapalca, Peru
Metalline Falls, WA
Missouri Leads
206/204
12.5
14.8
16.0 -
16.2 -
16.1 -
18.2 -
18.5 -
18.6 -
19.5
Ratio
16.2
16.4
16.7
18.6
18.7
19.0
21.0 -24.0
It is suspected that lead ores used during primary smelter
operations were from a limited number of geologically well-defined
sources and the secondary smelting operations process lead from a highly
average group of world-wide sources. Thus, the lead isotope ratios will
be substantially different from these operational periods. If this is
the case, then the isotope ratios in household dusts and collected
human body fluids will provide convincing evidence for the current
source of the lead dust. Furthermore, it is likely the primary Zn
smelter emissions were reflective of sulfide ore feedstocks containing
Zn, Cd Pb, As, Se, Cu, Ga, In, Ag, and the slag was rich in iron and
manganese. Secondary zinc smelter operations, on the other hand, likely
reflect compositions of diverse ferrous alloy feed materials and contain
high levels of elements not associated with primary zinc smelting such
as Be, V, Co, Ni, Mo, Pd, Pt, and lanthanides.
Based on the scientific evidence and available analytical methodol-
ogies, the investigative course of action to be taken is:
1. Review existing site characterization data; identify source areas;
review site history (what ores were smelted). Review other informa-
tion such as RCRA records, Air Enforcement records, OSHA records -
(the smelter has been cited in the past for worker exposure in the
plant and subsequent high blood lead levels).
C-65
-------
2. Develop piggyback sampling strategy for residential areas; obtain
samples of sources (slag, primary smelter emissions, secondary feed
materials, waste piles).
Types of Samples Desired
Sources
1. Primary Zn smelter stack emissions
2. Slag piles
3. Pre-1980 air filters
4. Post-1980 (present) air filters
5. Fugitive sources of dust from secondary Zn plant (piles, bins,
etc.)
6. Raw materials received
7. Products made
8. Present day stack emissions, baghouse dust, etc.
Receptors
1. House dust
2. Soil from yards, public areas, etc.
3. Hillside soil
4. Park, etc. dust
3. Apply a combination of the following approaches:
a. Elemental composition + MVA
b. Lead isotope ratio analysis
c. Individual particle analysis
C.7.2 Comments on Study
The investigative procedures outlined are currently underway. The
value of this case example is that it illustrates that investigative
techniques are frequently available to distinguish impacts from specific
current sources from other current and past emission sources. Although
this is a very specific case involving smelters, the fingerprinting
investigative approach may be applicable to other investigations of
Superfund site impacts on indoor air quality.
C-66
-------
C.8 BUILDINGS OVER WASTE OIL PIT
The Site is located in Mobile County, Alabama. The 5.1 acre site,
about 11 miles north of Mobile, Alabama, is bounded to the east by U.S.
Highway 43, by a gas pipe line easement to the north, an undeveloped lot
to the west, and a residential development to the south.
According to available records, in March 1961 a trucking terminal
was constructed for intrastate and interstate commercial trucking opera-
tions. The Site was used for parking, maintaining, and cleaning trucks
and trailers. According to a salesman at the Site, from May 1966 until
late 1969, tank trailers were regularly and routinely washed at the
Site. The washwater was then discharged in the rear of the terminal
property where the ground level was low.
The salesman states that he was told that in late 1965 (before he
worked at the Site) levees were built around the terminal to contain
materials and an aboveground asphalt tanker on wheels was placed at the
back of the property for storing asphalt products to be returned to
customers.
In February 1971, the terminal operator was encouraged to move its
operations. In March 1971, an offer to buy the property was accepted.
Figure C-7 shows the terminal layout as of May 1972, including the
configuration of a dike built in 1971. In September 1971, six geotechn-
ical borings, were drilled at the Site to depths ranging from 40 feet to
55 feet. The data collected from these borings were used to design the
foundations of the Site. Logs of these borings were compiled on a
drawing that was part of the building architectural plans. These boring
logs show the stratigraphy underlying the Site consists of a sand
stratum, up to 7.5 feet thick, overlying a silty clay stratum, which
attains a thickness of up to 34 feet. Underlying the silty clay is
another sand stratum of which the thickness was not determined by the
borings. To date, a drawing showing the locations of these borings has
not been found.
C-67
-------
C-68
-------
In 1973, the property was sold to the property's current owner
which constructed an apartment complex (see Figure C-8) consisting of 13
buildings on the Site. Approximately 160 people live in the apartment
complex.
C.8.1 Nature and Extent of Problem
In 1977, the owner first became aware of tar-like substances
seeping to the surface. The Department of Housing and Urban Development
(HUD), which inspected the apartments on an annual basis, inspected the
seepage and recommended to the managing agent that the seepage be
excavated to a depth of 1^ feet and the area be filled in with sand.
This was done periodically as the seepage appeared. HUD continued to
inspect the apartments on an annual basis. The managing agent reported
the presence of the tar-like substance to the Department of Health.
They continued to excavate the property and fill with sand when any
seepage occurred. Continued appearance of the tar-like substance
prompted the manager of the apartment complex to report the matter to
the State in the fall of 1984.
The State investigated the Site and reported the matter to the U.S.
Environmental Protection Agency (USEPA) during the same year. The State
identified 18 individual seeps where a tar-like substance was oozing to
the surface and described the material as a tar or asphalt with a
vitreous luster. The material was described as very sticky with a
petroleum odor and at cooler temperatures was very plastic and viscous.
The investigation included six hand augured holes ranging from 1.1 feet
to 5.0 feet in depth. Samples were taken from two holes in the interior
grassy courtyard area of the apartments and one hole in the southwest
corner of the complex in a grassy area inside the paved driveway. One
analysis was taken from these three soil samples although it is unclear
C-69
-------
: - 0
FIGURE
C-8
C-70
-------
if this was an analysis of a composite sample. The sample contained the
following compounds and concentrations:
Compounds Concentration (ug/g)
Naphthalane 135.4
Acenaphthene 101.9
Fluorene 128.6
Phenanthrene 33.9
Anthracene 53.7
Benzene 0.05
Preliminary sampling by the USEPA in April and May of 1985 detected
concentrations of naphthalene, 1,2,4 trichlorobenzene and other com-
pounds in the tar-like material and soil. In February 1990, the USEPA
added the Site to the NPL with an MRS score of 30.83.
Compounds reported to have been detected in the subsurface 4 feet
from the southeast corner of the courtyard area included butyl ate
(20,000 ppb), vernolate (90,000 ppb) and a mixture of compounds normally
found in coal-tar products. Compounds reported to have been detected in
the subsurface 4 feet from the east-central courtyard included butyl ate
(2,000,000 ppb), vernolate (300,000 ppb), eptam (30,000 ppb), and coal-
tar products at a total concentration of (70,000 ppb).
No organic compounds were reported to have been detected in a
composite surface sample of scrapings of white material from nine areas
in the complex. However, this sample was reported to contain 37,000,000
ppb sodium, with a Ph of 8.7.
Samples SA-17, SA-18, and SA-20 were surface samples of tar-like
seep material. Sample SA-17 from the east side of the Site was reported
to contain vernolate (1,000,000 ppb), 1, 2, 4-trichlorobenezene
(9,300,000 ppb), coal tar products (total concentration of 380,000 ppb),
carbon disulfide (5,000-ppb) and two unidentified terpenes (20,000 ppb).
A petroleum based product was also reported to have been detected in
this sample. Sample SA-18 in the southeastern corner of the courtyard
was reported to contain butylate (70,000 ppb) and vernolate (300,000
ppb) along with some coal products (total concentration of 760,000 ppb).
A petroleum based product was also reported to have been detected in
this sample. East of the ditch by the playground, Sample SA-20 was
C-71
-------
reported to contain coal-tar products at a total concentration of
(2,160,000 ppb), xylene (2,000 ppb) and a petroleum product. The USEPA
samples of April and May of 1985 found no contamination at the sampling
points located outside or downstream of the reported position of the
former levee.
A Phase I RI was conducted in February and March of 1991. Phase II
was conducted in September of 1991. The investigation was quite
thorough and reporting all details is beyond the present scope.
Given in Table C-20 is a summary of the number of compounds found
on-site.
TABLE C-20
SUMMARY OF NUMBER OF CONSTITUENTS DETECTED IN EACH MEDIUM
Media Type/
Sample Location
Alluvial Groundwater
Surficial Groundwater
Soils
Ditches
Tar-like Material
Total
Detected
Constituents
41
66
110
68
69
Total
Constituents
with
Unqualified
Detections
27
37
40
20
34
Total
Constituents
with More than
One Unqualified
Detection
15
24
32
13
19
Table C-21 gives organic chemical concentration data from the
monitoring wells at the Site.
C-72
-------
r-
1
O
UJ
03
CD
'I
5
c
Q
II
_
"o
a
c
o
"c
o
5
II
5
5
VI
c
o
CO
£
-Q
"el <
gj =
II
0
o
0}
c
o
3
a
_J
Q
—
c
o
o
03
o
cc
II
UJ
cc
c
n
a
c
0
o
0}
c
o
a
0)
II
f2
u
'5
0
Z
o
J—
ui "J
0^
II) *-
ti U ^J
Q 2 (n
O ~™N
U
O O O O O O «— O ^" O 00 O •— LOroOOOOOOOOT-OOOO^l"O
•— OOOOcoOO*.OOCoOOOOtOOOOOMCScr)
co co ro co J CS «— «~ CJ »— ^-r-CO
--
a.
O
o
•z.
o
m
cc
o
o
cc
a
UJ
x
a.
cc.
O
X
UJ
X
uj X
O uj
iu —
O w
< CQ
CL
O
$5
_i ca
< cc.
x <
H O
X O
a. cc
X X
ui
X
X
a.
X
UJ
x a x
||H|§S
UJ CC Q UJ O O
CM O 7 CM Z I-
^ J " a £ S
ca > <
o w x
< ca u
g_
X
UJ
I
Q «V
V M
•
H X J
£?£
X
X
Q.
UJ
1 fc
i£ 2
: -7 N
X 5 . X S
UJ ', UJ UJ .
X Ci Z X -J
-J •. uj -JO
(E X
3x
5^
O
z
"I I £i fc
"t < " 3
CM u ca ca
o
ui
<
O
u ca
o. uj
UJ >
0
a
a
c
a
o
6
UJ UJ
OL CC
UJ UJ
cc cc
pooppopooqoqpoppooooopcorococococofpco
C-73
-------
•a
0)
c
o
o
o
UJ
_l
QQ
o
0
=
«5
o
o
II
—
o
£
o
c
o
'c
o
5
II
5
M
C
o
s
o
.O
"°l<
ll
.2
Q
II
_J
Q
c
o
i
X
o
ir
II
UJ
CC
0
JD
(0
r
o
o
$
GO
C
o
0
0
w
II
w
5
1
o
0
OOOfOOOOOOOOOOOOOOOOOOOOOOOO
pr-f-;(^qoqoqq'-;'-_oqqoqqqqqqoqqq
O)OU1
T 00
TJ- CQ CO
C-74
-------
Indoor Air Modeling
No actual monitoring was conducted inside the structures. All
estimates are based on modeling. [The modeling approach for soil gas
intrusion is described in Appendix A, Sections A.1.5 and A.1.6].
In this case, the modeling approach to estimate soil gas flux was
based on the assumption that the contaminants are initially uniformly
distributed vertically and horizontally throughout the soil. The model
then allows equilibrium to be established among chemicals adsorbed on
soil, chemicals dissolved in soil water, and chemicals in the soil gas.
The soil gas diffuses to the surface resulting in an ever-expanding zone
of clean soil from the soil surface downward. It appears that the model
assumptions result in an average flux over a ten year period. These
source modeling equations were based on and conform to those given in
EPA/600/6-86-002, "Development of Advisory Levels for Poly-chlorinated
Biphenyl (PCB) Cleanup."
The above calculation yields only the flux (mass of chemical
exiting the soil surface per unit area per unit of time, eg. g/m2-s).
To calculate the rate of entry into the structures, it was necessary to
make estimations of:
• the ratios of diffusive to convective flow of soil gas,
• the area of the structure through which soil gas could enter,
• the air exchange rate for the structure.
The ratios of convective to diffusive flow was calculated from the
Peclet number, a dimensionless parameter, that considers the diffusiv-
ity, soil permeability, and vacuum created by the structure. The
calculation indicated the diffusion mechanism dominated (see comments
section below). The area through which gas could enter the structure
was estimated as 0.1 percent of the floor area, based on literature data
for homes on slabs. The structure air exchange rate was estimated to be
about 0.7 air exchanges per hour, based on average wind speeds and
indoor-outdoor temperature difference.
In addition, outdoor air concentrations can contribute to indoor
air concentrations. Since the buildings are surrounded by the Grassy
Area, indoor air concentrations can be affected by vapors from the
Grassy Area. Thus, the total indoor air concentration was estimated by
C-75
-------
summing the modeled indoor air and 100% of the Grassy Area air concen-
trations (indoor exposure, excluding seeps). Seeps of tar-like material
may also contribute to the indoor air concentrations. Thus, total
indoor air concentration including vapor emissions from seeps was
estimated by summing modeled indoor air concentrations, Grassy Area air
concentrations (99.66%), and emissions from seeps of tar-like material
(0.34%) (indoor exposure, including seeps). This conservatively assumes
that all tar seeps are located in the Grassy Area and contribute 100% to
indoor air. The results, excluding contribution from seeps, are given
in Table C-22.
Data from all sources was used to complete a risk assessment. The
results for the current exposure are summarized in Tables C-23 and C-24.
(Risks for future exposure scenarios, which assume use of on-site wells,
are not included in the table, but are discussed below.
Total current cancer risks including exposure to seeps were
estimated to be between 7 x 10"5 and 2 x 10"4 for the various receptor
populations. The most important pathway contributing to the risk is
indoor inhalation of vapors containing PAHs from seeps of tar-like
materials. In particular, benzo(a)anthracene, benzo(b)fluoranthene, and
chrysene contribute about 90% of the risk from indoor inhalation.
Under the Future I scenario including seeps, the cancer risks range
between 4 x 10"4 and 2 x 10~3. Ingestion of water from the alluvial
aquifer accounts for most of the cancer risk, approximately 75% of the
total risk. Arsenic and beryllium constitute the greatest proportion of
the risk, with aldrin, BEHP and dieldrin also contributing to the total
risk.
The Future II scenario including seeps poses risks ranging from 3 x
10"3 to 1 x 10"2. Inhalation of indoor air vapors during showering
account for approximately 100% of the risk, and the risk is primarily
due to chloroform from the on-site surficial aquifer.
C-76
-------
CO
z
O
h-
cc
l-
z
UJ
o
(N Z
eg o
i °
LLJ £
CO
oc
o
o
Q
Z
Q
LU
_l
UJ
Q
O
c
a
i:
V
u
c
o
u
"o
9 £ i-0
- (3^-2
c —
l}|i
ujS
Oi on
||
a
III
m
a en
_ 3
^
l=\
a "~
si
i!li
u -^.
Itf
- • 1.
Contaminant!
?
O
in
1
s
•
3
ff>
§
CM
O
§
6.248-06
45.50471
6.078-02
S
id
a
ACENAPHTHENE
?
S
CM
1
O
ft
CM
O
e
to
CM
O
O
3
1.248-05
0.130183
8.588-04
o
§
Si
ACETONE
CM
§
n
(D
8
O
8
|
8
S
4.508-06
1
1
5
O
CM
CO
in
ALDRIN
?
to
CM
§
01
CM
O
O
p
8
O
O
s
5.778-06
128.2497
S
CO
S
o
CO
N
ANTHRACENE
^
CO
5
n
c\j
8
O
CM
in
8
i
(vi
O
8
s
in
1811.575
4.830-05
in
o
e
in
8
BENZO(A)ANTHRACENE
CM
£
CO
CM
m
p
CM
O
C
CO
CM
O
O
r-
to
S
c
in
CM
0
c
o
3
0
0
CT
ri
4.806-06
9384 613
3.948-O4
S
CM
0
CHLORDANE. ALPHA
CM
Ol
tn
CO
CM
*
5
fe
o
V
0
o
4.808-06
295.7S52
3.948-04
o
c
p
0
CHLORDANE, GAMMA
o
n
i-
0
I
8
«
8
CO
8
7.306-06
3.14816
9
in
CM
O
§
s
CHLOROBENZENE
?
to
o
1
o
at
q
O
ri
8
Cvi
1.048-05
2.073532
oj
S
CO
CM
O
CHLOROFORM
C-77
-------
n
OJ
D
,C
C
o
O
s^
CJ
«N
1
O
LU
m
t-
o
f-
oc
f— •
2
LU
O
Z
o
o
cc
^
oc
0
o
Q
^
Q
LU
_J
LU
Q
O
C
0
C
tJ
3
- 0 * ~
c *—
1 fl"e
UJ -2
&«
It
•'-3
=? • <
J| ^ W
H S 3.
I"5
a en
3~
f-T
1 c'V
a
•i
, |j
i a
i"f
"•^
If!
j
2
9
u
*-
c
*
^-
&
CM
3
2
8
•
2i
CM
8
X
^
9
fi
n
1811575
1
8
£
CM
§
1 CHRYSENE
2
$
0
CM
*
r~
0
*
CO
8
3
A
*"
?
3
^
1
9
3
2
*
8
P
c
X
"*•
i
5
9
in
0
r*.
••^
3
%
8
8
&
*
3
1
ri
E9GICTO
I
n
0
^
n
«
| DICAMBA
CM
tn
0
$
•^
8
e>
8
4
Q
«
8
5
a
9
r?
"
414.8168
1
3
fi
to
8
V
(M
*
0
1
•n
8
X
Ol
ri
8
8
f?
•*
9
1!
*r
1129.943
3
2
a
8
g
03
9
$
rj
CM
o
£
CM
s
<^
n
72.12009
?
1
CM
S
g
Ol
S
1 FLUORENE
^
i
CM
^
c?
*-
8
rj
8
^
y
•»
8
:?
n
9
1
»
43.51676
8
£
CO
m
s
1
in
5
CC
n
HEPTACHL
-
m
rj
,_
a
8
§
8
*
8
9
5
*
5.174339
|
3
0
K
§
.
Cj
cc
O
I
C-78
-------
TJ
0)
a
.£
•*-*
c
o
O
"^
C\J
OJ
O
LU
m
j_
O
H
oc
H
LU
0
•z.
o
0
DC
DC
O
O
Q
Z
Q
LU
|
LU
Q
O
C
c
o
• c
(5
o
1M||
-
§^_
•
3II-E
SlCji
ii
• "S •=
i ; fi
t- S ~-
t»
a en
^~
I"-*
= e1^
a
=f
= Ii
i-t
|ii
c
c
E
c
,3
OJ
C
C4
(M
4
8
i
8
+
§
*
8
$
*
?
S
in
232625
2.400-04
S
X
o
8
OL
m
CM
CN*
"
O
W7
8
V
8
+
£
•»
8
$
0
§
c
in
SZ9TCZ
3.050-05
8
X
a
m
o
m
csj
to
n
1
r-
8
a
8
O
8
9
tn
23.2625
\
8
2
n
a
(UNOANE)
S
5?
m
r>
f,
"
n
«
8
S
8
+
*
tfl
D
8
9
in
14389.85
2.800-O6
8
*
to
UJ
CDJPYREN
| INDENO(1.2,3-
0
t~-
to
o
i
o
o
V
S
^
?
*
0
(9
*
«
£
a
38.39342
2.280-01
S
&
«">
S
§
1 METHOXYCHt
^
1
CO
,_
i
*
8
*
8
o
8
Si
h-
«
0
to
0243466
1.910-03
0
§
o
R
i
!->
METHYL ETW
o
CT
0
3
8
S
m
9
Z
"
0
1.310-01
8
*
r-
3
UJ
o
1
UJ
S
._
*
r>
_
5
o
«
8
o
o
$
rj
9
§
in
0.764199
0.000057
8
S
1
MOLINATE
ft
i
f-
C?
0
2
n
o
-i*
to
09
S
+
in
•"
CM
O
N
rJ
9
1
in
10.42455
^
S
5
a
|| NAPHTHALEN
ft
(M
O
OJ
„
3
*
g
•*•
a
*~
o
fi
9
•8
in
I
6
g
S
g
PEBULATE
_
£
-------
TJ
Q)
a
_c
•4-*
C
o
O
CM
I
O
UJ
_J
m
cc
h-
2
HI
O
2
O
o
DC
CC
O
O
Q
2
Q
111
_J
HI
Q
O
1
•o
m
Ja
•
u
3
"o
(A
8 - • §•£
15 sf t
- 0 * —
ill!
>->,§£
?f
U
111
311
•
11
fi
II!
a
s!
• ?-s-
^$1*
-51
S5
|-g
• - T
Ift
•t
C
a
E
a
3
9
K
h~
?
2
rs
3
+
S
S
•t-
§
n
8
o
7.009-06
I
S
1.12» + 03
8
UJ
2
a
fc
C-80
-------
TABLE C-23
SUMMARY OF PATHWAY-SPECIFIC CARCINOGENIC RISKS
Exposure Scenario
RME Receptor (Wortt-Caee Scenario)
Sum of Adult
Cancer Rltk
Sum of
9-Year-OW
Cancer Rltk
Sum of
4-Year-OW
Cancer Rltk
Sum of 9-and
4-Year-OW
Cancer Risk
CURRENT EXPOSURE: No known documented cases of exposure at these levels
Target Area E
Ingestion of soil
Dermal contact with soil
Inhalation of vapor
Inhalation of particulates
Grassy Area
Ingestion of soil
Dermal contact with soil
Inhalation of vapors
Total Indoor Exposure
Inhalation of vapors (including seeps)
Inhalation of vapors (excluding seeps)
Ditch
Ingestion of sediments
Dermal contact with sediments
7x10"'
7 X 1 O'6
6 x 10'*
1 x 10"*
1 X 10'7'
1 X 10-"
9 X 10-"
5X 10*
2X10"7
0
0
2 X 108
2 X 10'6
6 X 10'"
1 x 10'9
6X 10"'
7 X 10"'
1 X 1 0"'
5X 10*
2X10'7
2X 10'7
3X 10"s
5 X 108
2 X 1 0'6
7 x 1 0'9
2x10'
5 X 1 0'7
2 X 10'*
1 X 10"
6X 10*
2 X 10'7
6 X 10'7
2 X 20'*
6x10-'
4 x 1 0'6
1 x 1 0'6
3 x.10-'
6 X 1 0'7
3 X 1 0'*
2X10"'
1 X 10-*
3X10'7
8 X 10'7
5X 10'*
EXPOSURE TO SEEPS OF TAR-LIKE MATERIAL: A««ume« no removal of teep* for RME Receptor*
Ingestion of tar-like material
Dermal contact with tar-like material
Inhalation of vapors
TOTAL CURRENT EXPOSURE
(EXCLUDING SEEPS)
TOTAL CURRENT EXPOSURE
(INCLUDING SEEPS)
4X 10'*
7X 107
5 X 10'*
7 X 106
5X 10"'
1 X 10*
5 X 10'*
3 X 1 0'6
9 X 10'6
2X 10'7
2 X 10*
5 X 10'9
4 X 10'6
1 X 10"4
3 X 1 0'7
3X 10-*
1 X 10-5
7 X 10-*
2 X 10"4
C-81
-------
TABLE C-24
SUMMARY OF PATHWAY-SPECIFIC TOTAL HAZARD INDICES
Exposure Scenario
RME Receptor (Worst-Case Scenario)
Adult
Nine-Year Old
Four- Year Old
CURRENT EXPOSURE: No known documented case* of exposure at theee level*
Target Area E
Ingastion of soil
Dermal contact with soil
Inhalation of vapor
Paniculate inhalation
Grassy Area
Ingestion of soil
Dermal contact with soil
Inhalation of vapor
Paniculate inhalation
Northern Ditch
Ingestion of sediments
Dermal contact with sediments
Indoor Exposure
Inhalation of vapor*, total indoor (including seep*)
Inhalation of vapor*, total indoor (excluding *eep*)
4 X 10 J
3 x 10'
2 X 10J
9 X lO*
3 X 10J
6 X 10-3
5 X 10'
0
0
1 X 10*
1 X10J
1 X 10'1
1 X 10°
3 X 10J
1 X 10'
3 X 1 0 3
7X 10-3
9X10''
3 X 10J
1 X 10'
2 X 10*
2X 10J
4x 10'
2 X 10°
4 X 10J
2 X 10 '
2 X 10'
2X 10'1
1 X 10^
7 X 104
1 X 10'1
2 X 10*
3X 10J
EXPOSURE TO SEEPS OF TAR-LIKE MATERIAL: A«ume« no removal of seep* for RME Receptor
Ingestion of tar-like material
Dermal contact with tar-like material
Inhalation of vapor*
TOTAL CURRENT EXPOSURE
(EXCLUDING SEEPS)*
TOTAL CURRENT EXPOSURE
(INCLUDING SEEPS)2
9 X 10'
7X10^
1 X 10J
4X 10'1
4X10'1
2X 10-4
2X 10"3
2X 103
1X10°
1X10°
9 X 10"4
3 X 10J
2X 10-3
2 X 10°
2 X 10°
C-82
-------