EPA-600/R-9 5-090
July 1995
LUMPED-PARAMETER MODEL ANALYSES OF DATA
FROM THE 1992 NEW HOUSE EVALUATION PROJECT
FLORIDA RADON RESEARCH PROGRAM
Final Report
by
Kirk K. Nielson, Rodger B. Holt, and Vern C. Rogers
Rogers and Associates Engineering Corporation
P. O. Box 330, Salt Lake City, UT 84110-0330
EPA Interagency Agreement RWFL 933783-01
Florida Department of Community Affairs Contract 93RD-66-13-00-22-003
University of Florida Subcontract (Acct. 1506481-12)
U.S.EPA Project Officer: David C. Sanchez
Air and Energy Engineering Research -Laboratory
Research Triangle Park, NC 27711
DCA Project Officer: Mohammad Madani
2740 Centerview Drive
Tallahassee, FL 32399
Prepared for:
State of Florida
Florida Department of Community Affairs
2740 Centerview Drive
Tallahassee, FL 32399
and
U.S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
-------
. TECHNICAL REPORT DATA 11 iiinmini main in
{ _ — (Please read Instructions on the reverse before compkt III |||j II llllll 111 III llfl III III 1
1. REPORT NO, |2.
EPA-600/R~9 5-090
3,; iii mi ji iiiiii mill ¦mill in
;'v PB95_-24 3077 J
4, TITLE AND SUBTITLE
Lumped-parameter Model Analyses of Data from the
1992 New House .Evaluation Project—Florida Radon
Research Program
5. REPORT DATE
July 1995
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Kirk K. Nielson, Rodger B. Holt, and Vern C.
Rogers
8. PERFORMING ORGANIZATION REPORT NO.
RAE-9226/1-15R1
9, PERFORMING ORGANIZATION NAME AND ADDRESS
Rogers and Associates Engineering Corporation
P. O. Box 330
Salt Lake City, Utah 84110-0330
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO,
EPA IAG RWFL 933783"01
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air and Energy Engineering Research Laboratory*
Research Triangle Park, NC 27711
13, TYPE OF REPORT AND PERIOD COVERED
Final;
14. SPONSORING AGENCY CODE
EPA./600/13
15. supplementary notes AEERL project officer is David C. Sanchez, Mail Drop 54, 919/
541-2979. (*) Redesignated Air Pollution Prevention and Control Division.
» •,V ' 1
io. abstract The report documents analyses of Phase 2 data from the Florida Radon Re-
search Program's (FRRP's) New House Evaluation Project'fhat were performed
using a lumped-parameter model. The houses evaluated in Phase 2 were monitored
by the Florida Solar Energy Center (FSEC) and the University of Florida (UF). ) • ..
. Based on experience from Phase 1 of the NHEP, Phase 2 monitoring was aimed at
better isolating the effects of specified radon-resistant construction features. The
FSEC data included 15 houses, and the UF data included 14 houses. The lumped-
parameter analyses focused primarily on empirically characterizing the rado. resis-
tance of the house/soil interface for different foundation designs. .The analyses
were also aimed at comparing the effectiveness of active and passive radon protec-
tion features/. .
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b, IDENTIFIERS/OPEN ENDEDTERMS
c. COSATI Field/Group
Pollution Foundations
Radon
Emission
Residential Buildings
Mathematical Models
Construction
Pollution Control
Stationary Sources
13B
07B
14G
13 M
12 A
13. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report/
Unclassified
21. NO. OF PAGES
47
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify thai the contents necessarily
reflect the views and policy of the Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
ABSTRACT
The Florida Department of Community Affairs is developing radon-resistant building
construction standards under the Florida Radon Research Program (FRRP). For technical
support of the standards, the radon-resistance of certain construction features is being tested by
incorporating them into new houses and monitoring radon levels and related parameters. These
tests, first performed in 1991 under the FRRP's New House Evaluation Project (NHEP), were
analyzed previously using a lumped-parameter model to help identify the effectiveness of the
construction features. This report presents further lumped-parameter analyses of NHEP data
collected during 1992 by the Florida Solar Energy Center (FSEC) and by the University of
Florida (UF). The analyses are aimed at empirically characterizing the radon resistance of the
house-soil interface for different foundation designs, and at comparing the effectiveness of active
and passive radon protection features.
The lumped-parameter model used here was derived previously from sensitivity analyses
with the detailed Radon Emanation and Transport into Dwellings (RAETRAD) model and from
empirical definitions of typical house parameters. The lumped-parameter model explicitly
represents radon entry by pressure-driven advective flow through foundation cracks and by
diffusive movement through the cracks and through the intact concrete slab. The model
characterizes radon resistance by the ratio of net indoor radon concentration to sub-slab radon
concentration, CnJCmb. This approach normalizes the different radon source strengths for soils
under different houses to a common basis for comparison of the house radon resistance.
The FSEC houses included eight with floor slabs poured into hollow-block stem walls and
seven with monolithic poured-concrete slab and stem wall construction. The UF houses similarly
included nine houses with the slab poured into the stem wall (SSW) and five with monolithic
slabs. All of the houses had sub-slab ventilation (SSV) systems, and the houses were monitored
with the SSV systems in capped, passive, and in some cases, active (fan-ventilated) modes.
Indoor radon levels measured for each SSV mode were compared to capped-SSV sub-slab radon
levels for consistent comparisons of radon resistance. Indoor radon data were reduced by
estimated outdoor radon levels to obtain Cnct, the net soil-related component of the indoor radon
ii
-------
concentration. The outdoor levels were estimated from an empirical function of the sub-slab
radon concentrations.
Lumped-parameter model calculations for comparison with measured CBet/Csub ratios
utilized house parameters and surrogate measurements. House ventilation properties and air
pressures were estimated from blower-door test data. Concrete slab water/cement ratios were
estimated from reported values of the concrete slump. House dimensions were taken from direct
measurements, and soil water saturation fractions were estimated from soil moisture
measurements. Sub-slab ventilation effectiveness was estimated from changes in sub-slab radon
measurements under different SSV operating conditions.
The present analyses estimate more precisely the effectiveness of radon-resistant building
features than the previous NHEP data. They also suggest that the lumped-parameter model may
accurately predict CnJCsvb ratios when houses are built according to the FRRP construction
standard. The accuracy of the lumped-parameter model is suggested by a ratio of 1.01+0.16
for the calculated/measured geometric means of the Cnel/C5Ub ratios.
Several other important conclusions about radon resistance are suggested by the data
analyses. SSW construction, in accordance with the FRRP standard, reduces indoor radon to
about 0.09% of the sub-slab concentration (with an uncertainty of a factor of 2.2). Capping the
SSV system does not significantly alter its radon-resistance effectiveness compared to leaving it
in the passive mode. Monolithic slab construction may improve radon resistance by
approximately 33%, reducing indoor radon levels by a factor of 0.67 compared to SSW
construction. Activation of SSV systems with exhaust fans may improve radon resistance by
approximately 70%, reducing indoor radon levels to about 0.3 times the levels that occur when
the SSV system is in the passive or capped mode. The present data on active SSV systems are
sparse and uncertain, however, due to the small number of houses where the SSV systems were
activated. Future analyses should include more data on active SSV systems to better define their
effectiveness.
iii
-------
TABLE OF CONTENTS
Chapter Page No.
1 INTRODUCTION l-l
1.1 Background 1-1
1.2 Objective and Scope 1-2
2 THEORETICAL BASIS AND PARAMETER SENSITIVITY 2-1
2,1, The RAETRAD Model 2-1
2.2 Simplified Approximations 2-7
2.3 The Lumped-Parameter Model 2-9
3 HOUSE PARAMETERS AND RADON MEASUREMENTS 3-1
4 COMPARISONS WITH THE LUMPED-PARAMETER
MODEL 4-1
4.1 Definition of Measured Ratios 4-1
4.2 Definition of Lumped Parameters 4-4
4.3 Comparison of Measured and Calculated Radon Ratios 4-7
5 CONCLUSIONS AND RECOMMENDATIONS 5-1
6 LITERATURE REFERENCES 6-1
iv
-------
LIST OF FIGURES
Figure No. Page No.
Two-dimensional grid and boundaries used to define slab and
soil regions for indoor radon and air entry calculations 2-6
Comparison of calculated and measured C^IC^ ratios for
capped, passive, and active SSV systems in SSW and
monolithic-slab houses 4-10
v
-------
LIST OF TABLES
Table No. Page No.
1 FSEC house parameters 3-2
2 UF house parameters 3-3
3 FSEC house ventilation and soil measurements 3-4
4 UF house ventilation and soil measurements 3-5
5 FSEC house pressure and ventilation measurements 3-7
6 UF house pressure and ventilation measurements 3-8
7 FSEC sub-slab and indoor radon measurements 3-10
8 UF sub-slab and indoor radon measurements 3-11
9 Lumped-parameter values calculated for FSEC houses 4-2
10 Lumped-parameter values calculated for UF houses 4-3
11 Comparison of measured and calculated C^/C^ ratios for
slab-in-stem-wall houses 4-8
12 Comparison of measured and calculated C^IC.^ ratios for
monolithic-slab houses 4-9
vi
-------
1. INTRODUCTION
1.1 BACKGROUND
Inhalation of indoor radon (222Rn) and its decay products dominates exposures to
natural radiation in the U.S. population (Ner88). Radon causes 7,000 to 30,000 lung cancer
fatalities annually from chronic exposure (EPA92a). Indoor radon comes mainly from decay
of naturally occurring radium (226Ra) in underlying soils, although contributions from water,
building materials, and outdoor air may also be important (EPA92b). Radon enters buildings
through cracks and pores in their floors and foundations, and its rate of accumulation
depends on the competing rates of entry and of dilution by outdoor air. Indoor radon levels
therefore vary significantly with time due to pressure-related changes in entry rate (from
wind, temperature, and air-handler changes) and pressure- and occupant-related changes in
dilution rates (e.g., from the pressure changes, door and window openings, ventilating
appliances, fireplace or flue openings). Since radon-related health risks accumulate over
years or even decades, hourly or daily variations are relatively unimportant except as they
affect the long-term average occupant exposure rate or the results of short-term radon
measurements. The U.S. Environmental Protection Agency (EPA) recommends remedial
action where long-term average radon levels are 4 picocuries per liter (pCi L*1) or higher
(EPA92b). Indoor radon levels in the United States average about 1.25 pCi L"1, and about
1% of all homes have levels that exceed 8 pCi L"1 (EPA92a).
The Florida Department of Community Affairs (DCA) is developing radon-protective
building standards to help reduce radon-related health risks (San90, DCA91). The standards
and their technical basis are being developed under the Florida Radon Research Program
(FRRP), which has studied building designs, materials, dynamics, basic processes, and radon
source potentials. The FRRP also has evaluated various radon-resistant construction features
by incorporating them into new houses under its New House Evaluation Program (NHEP).
Under this program, test houses with radon-resistant features are monitored to assess each
feature's effectiveness.
1-1
-------
The effectiveness of radon-resistant construction features has been difficult to estimate
because of the complexity of radon entry and accumulation processes, and because of
uncontrolled differences among the houses. These differences include varying soil radon
potentials at the different sites, differences in house pressure and ventilation characteristics,
and differences in the coupling of radon potentials with house dynamics. Soil radon
potentials depend on soil radium concentration, radon emanation coefficient, moisture, air
permeability, diffusion coefficient, and density. Indoor air pressures affect both radon entry
rates and house ventilation. House floor and foundation properties also affect radon entry
rates for a given soil radon potential. Although the effects of these parameters can
potentially be separated by sophisticated mathematical models, the models usually require
more detailed data than are available from the NHEP projects.
To deal with the complexity and variability of radon entry, a simplified, lumped-
parameter model was developed (Nie93a) to help interpret the NHEP data by accounting for
the uncontrolled differences among the houses. The lumped-parameter model was developed
from numerous sensitivity analyses with a detailed numerical model (Nie94), and from
analyses of empirical data on house ventilation rates and concrete slab properties. In its
initial comparisons with NHEP data (Nie93a), the lumped-parameter model suggested
relatively large uncertainties in the performance of the radon-resistant construction features.
1-2 OBJECTIVE AND SCOPE
This report analyzes the second phase data from the FRRP NHEP using the lumped-
parameter model. The houses evaluated in this phase were monitored by the Florida Solar
Energy Center (FSEC)
-------
2. THEORETICAL BASIS AND PARAMETER SENSITIVITY
The lumped-parameter model was derived to represent long-term average radon levels
in slab-on-grade houses with well-mixed interior air volumes and constant rates of ventilation
by outside air. Radon entry into the houses involves the dominant mechanisms characterized
by the RAdon Emanation and TRAnsport into Dwellings (RAETRAD) model (Nie93b),
including diffusion through the floor slab, advection (with pressure-driven air flow) through
floor cracks, and diffusion through the floor cracks. The lumped-parameter model is based
on simplified,, empirical approximations of RAETRAD sensitivity analyses to represent
various soil properties, house sizes, floor cracks, and indoor air pressures. An empirical
approach reduced the complex, interacting equations of the RAETRAD model to the simpler,
parametric form of the lumped-parameter model.
2.1 THE RAETRAD MODEL
The differential equations describing radon emanation and transport involve solid,
liquid, and gas phases in the simultaneous, interactive processes of diffusion, advection,
absorption, and adsorption. Radon moves primarily by diffusion and advection. Diffusion,
driven by radon concentration gradients, is important in the liquid as well as gas phases
because of frequent intermittent blockages of soil pore segments by water. Advection,
resulting from pressure-driven air flow, carries radon at the interstitial soil gas velocity.
Both mechanisms establish along the transport route new equilibria of radon concentrations
with local aqueous and solid phases in a chromatograph-like process.
Three coupled differential equations characterizing radon concentration changes with
time in the solid, liquid, and gas phases give the complete description of radon transport.
With appropriate parameter definitions, these equations are reduced to a single, multi-phase
differential equation (Rog91a) for radon in the air phase, as it commonly is measured. For
steady-state conditions representing long-term averages, the equation that RAETRAD solves
is written as:
2-1
-------
v-faDV(cyfs) - v-KK/pXcyfg) vp] - + rpxe = o q>
where
V = gradient operator (cm"3)
fa = £(l-S+SkH)
e = soil porosity (dimensionless: cm3 pore space per cm3 bulk space)
S = soil water saturation fraction (dimensionless)
kjj = 222Rn distribution coefficient (water/air) from Henry's Law
(dimensionless)
D = diffusion coefficient for 222Rn in soil pores (cm2 s"1)
Cb = fgCa = 222Rn concentration in bulk soil space (pCi cm"3)
Ca = 222Rn concentration in air-filled pore space (pCi cm*3)
fs = £(l-S+SkH)+pka
p = soil bulk density (g cm"3, dry basis)
ka = ka0 exp(-bS)
kao = dry-surface adsorption coefficient for 222Rn (cm3 g"2)
b = adsorption-moisture correlation constant (g cm"3)
K = bulk soil air permeability (cm2)
p = dynamic viscosity of air (Pa s)
VP = air pressure gradient (Pa cm"1)
X = 222Rn decay constant (2,lxl0"6 s"1)
R = soil 226Ra concentration (pCi g"1)
E = total 222Rn emanation coefficient (air + water) (dimensionless).
This equation applies to gas-phase advective transport of radon, and to combined gas-
phase and liquid-phase diffusive transport of radon. The four respective terms defining radon
concentration changes in equation (1) represent radon diffusion, advective transport of radon
2-2
-------
by soil gas movement, radon decay, and radon generation by emanation from soil minerals.
The factor fa physically corresponds to the effective porosity in which radon is distributed,
including the gas-phase and liquid-phase components. The factor fs similarly represents the
effective porosity containing radon, but includes also an equivalent pore volume for radon
adsorbed on solid pore surfaces. For developing the lumped-parameter model, the radon
adsorption characteristics of the soils were ignored by setting the value of ka equal to zero
in equation (1),
Combined-phase diffusive transport of radon is characterized by appropriate moisture-
and porosity-dependent values of the pore-average diffusion coefficient, D (Rog91b). This
approach is important to correctly characterize radon diffusion in unsaturated soil pores that
may have small intermittent water blockages, but that still may transmit significant radon
flux (Nie84, Rog89). Liquid-phase advective transport of radon is not addressed because it
typically is negligible. The radon fluxes between different soil layers and at the top surface
are calculated as:
The first (diffusive) term of equation (2) results from Fick's law, which describes radon flux
in a single phase (Fic55, Cra75). The second (advective) term represents the radon flux from
bulk air movement in the air-filled pore space.
The soil air velocities, (K/p)VP in equations (1) and (2), are calculated in the
RAETRAD model from corresponding air pressure and flow equations that are applied to the
same regions as the radon equations. The differential equation that is solved to characterize
steady-state, pressure-driven air flow is obtained from the equation of continuity and the
equation of state for gases under isothermal expansion (Yua81):
F = -D fa VCa + (K/ji) VP Ca
(2)
where
F
bulk flux of 222Rn (pCi cm"2 s"1).
3(P>P0)/3t = V-(K/p)VP = 0,
(3)
2-3
-------
where
3(P/P0)/3t = time derivative of the fractional change in soil air pressure (s"1).
Air velocities are computed between soil or interface boundaries from Darc/s law as:
v = -K/[pe(l-S)] VP (4)
where
v = soil gas velocity in the air-filled pore space (cm s"1).
Several simplifying surrogate parameters and predictive correlations already are
available to simplify equations (1) through (4) to more readily useable forms. Soil porosities
for use in equations (1) through (4) are generally calculated from specified soil densities as;
e = 1 - p / pg (5)
where
e = total soil porosity
q
p = soil bulk dry density (g cm" )
pg = soil specific gravity (nominally 2.7 g cm" ),
The conversion between water contents on a weight-percent basis, a volume-percent basis,
and fraction of saturation basis utilizes the relationships:
100S = MJt = pMJc (6)
soil water content (volume percent)
soil water content (dry weight percent).
where
My
Mw
-------
Diffusion coefficients for radon in soils for use in equations (1) and (2) can be
measured by standard laboratory methods (Nie82), or can be estimated for modeling purposes
from specified values of the soil water content and the soil porosity (Rog91b):
D = D0e exp(-6fiS - 6S14e) (7)
where
D = diffusion coefficient for 222Rn in soil pores (cm2 s"1)
D0 = diffusion coefficient for 222Rn in air (0.11 cm2 s*1).
This correlation is based on 1073 laboratory measurements of radon diffusion in recompacted
soils at moistures ranging from dryness to saturation (Rog91b). The soil textures ranged
from sandy gravels to fine clays, and their densities covered a range typical of Florida soils.
Soil air permeabilities for use in equations (1) to (4) also can be measured by standard
field methods (Nie89), or can be estimated for modeling purposes from specified values of the
soil water content, porosity, and average particle diameter as:
K = 104 (e/500)2 d4/3 exp(-l2S4), (8)
where
d = arithmetic mean soil particle diameter, excluding >#4 mesh (m).
This correlation was based on more than one hundred in-situ field measurements of soil air
permeability, many of which were made in Florida (Kog91b).
RAETRAD computes radon entry into a house using an elliptical-cylindrical form of
equations (1) through (4) (2-dimensional gradient operators, Nie93b). With this
computationally efficient approach, 2-dimensional arrays of properties represent the house
foundation and its vicinity soils for use in finite difference calculations. The arrays are
oriented in the vertical and radial dimensions about the house center of symmetry, as
2-5
-------
illustrated in Figure 1. The elliptical-cylindrical geometry represents the footprint of a
rectangular house by an ellipse of equal area.
Because of the independence of soil air pressures from soil radon concentrations, the
solution to Equation (1) is computed in two steps. First, the pressure gradients required for
Equation (1) are computed by separately solving the air flow profiles with equation (3), Then,
equation (1) is solved similarly by substituting into it the computed velocities. Boundary
conditions for the finite-difference numerical calculations are: constant air pressure and radon
concentration at the top surface of the house floor; constant air pressure and radon
concentration (but different numerical value) at the top surface of the soil outside the house;
and zero air velocity and radon flux at the center of symmetry, at the outer radial limit of the
finite-difference grid, and at the bottom of the finite-difference grid.
Figure 1. Two-dimensional grid and boundaries used to define slab and soil
regions for indoor radon and air entry calculations.
2-6
-------
2.2 SIMPLIFIED APPROXIMATIONS
The detailed RAETRAD numerical model has shown previously that the primary radon
entry routes and mechanisms are diffusion through the concrete floor slab and advection and
diffusion through cracks in the concrete floor (Nie91). Concrete air permeabilities are so low
that air flow through intact regions of the slab is negligible (Rog94). Diffusion and advection
through cracks are grouped together here for simpler empirical analyses of the effects of cracks
on radon entry. Cracks, as used here, refer to all openings in the floor slab and foundation stem
wall, regardless of their particular geometry. The starting point for developing the lumped
parameter model is the radon entry rate, grouped into two categories: slab entry and crack entry.
Thus, the radon entry rate is represented by:
Qua = Qslb + Qc.fc (9)
where
Qtol = total radon entry rate from soil (pCi s"1)
Qslb = radon entry rate through intact slab (pCi s"5)
Qcit = radon entry rate through slab cracks (pCi s'1).
The radon entry rates are related to indoor radon concentrations for a well-mixed interior
volume by the relation:
C* = (C,* + 3.6 QJW / (1 + W*t), (10)
where
Cjn = indoor radon concentration (pCi L"1)
= outdoor radon concentration (pCi L1)
3.6 = unit conversion (pCi L"1 h 1 per pCi m"3 s"1)
^ = rate of house ventilation by outdoor air (h1)
Vb = interior house volume (m3)
2-7
-------
= radon decay constant (0.0075 h"1).
Since radon decay is nearly always negligible compared to ventilation by outdoor air
the soil-related indoor radon concentration can be written as a simple function of the soil radon
entry rates:
On " Ow = 3.6 (Qslb + Q.J / >„Vh. (11)
Since radon entry rates through both the slab and the crack are proportional to their
respective cross-sectional areas and to the sub-slab radon concentrations (Nie91), they can be
normalized by their areas and sub-slab concentrations to obtain more constant, lumped parameters
that have the units of radon entry velocities. Thus, the diffusive and advective components of
radon flux in equation (2), which dominate the respective slab and crack areas, can be expressed
as equivalent velocities:
vsib = Qsib! (Asib CSJ, (12)
for the intact slab area, and as
Vcrk Qcit ^ (^cit C sub), (13)
for the crack area, where
vslb = equivalent velocity for radon diffusion through the slab (mm s"1)
Aslb = slab area (m2)
Csub = area-weighted average sub-slab radon concentration (pCi L')
vcric = equivalent velocity for diffusive and advective radon entry through the crack
(mm s1)
Acric = crack area (m2)
C'sub = sub-slab radon concentration under the crack (pCi L'1).
2-8
-------
Combining equations (12) and (13) with equation (11), and defining the sub-slab
ventilation factor, the crack area fraction, and the crack radon ratio gives the following
expression for the net, soil-related indoor radon:
^net ~ ^in " ^out ^sub ^ssv^'slb + ^c%nucrk^ ^ (14)
net indoor radon concentration resulting from soil and slab sources
(pCi L'1)
sub-slab ventilation factor (1 - 0,01 • radon reduction percentage)
Agj^/Agib = crack area fraction (dimension!ess)
C' sui/Cgub = crac^ radon ratio (dimensionless)
mean height of the interior volume of the house (m).
Equation (14) illustrates that the sub-slab radon concentration can be utilized as a
surrogate of soil radon source potential. The net indoor radon concentration is further
affected primarily by the house ventilation rate, its interior height, its crack area fraction,
and its susceptibility to radon entry, as defined by the slab and crack radon entry velocities.
2.3 THE LUMPED-PARAMETER MODEL
Radon entry equivalent velocities were defined for use in equation (14) by parametric
fitting of empirical functions of the soil and house properties to velocities calculated by the
detailed RAETRAD model (Nie93b). The RAETRAD calculations to support these fits utilized
a reference set of soil and house parameters that were varied only in the parameter of
interest to isolate the parameter's effect. This fitting approach and the reference house
parameters and soil parameters were described in the original lumped-parameter report
(Nie93a). The model calculations were used to define approximate, lumped-parameter
expressions for ua]b and i?crk.
where
^net
h
2-9
-------
The expression developed to represent radon entry through the intact floor slab was
mainly a function of the concrete diffusivity and of the house size. The diffusivity function
was expressed empirically in terms of the concrete water/cement ratio, since diffusion
coefficients for radon in concrete are not commonly available. The house size effect was
incorporated into the intercept of the concrete diffusivity expression, and accounts for the
horizontal diffusive losses of sub-slab radon to the atmosphere and the reduced effect of these
losses from very large slabs. An additional, smaller term also was estimated to account for
the slight increase in radon entry when cracks are localized near the center of a slab, as
opposed to those occupying larger areas near the slab edge. The expression has the form:
t/slb = 2.9xl0"7 e1L4W + 4.6xl0'5/xh + 3.5xl(r5 xcrk/xh (15)
where
W = concrete slab water/cement ratio
xh = house minor dimension (m), assuming an equivalent rectangular footprint.
xcrJ{ = location of the dominant floor openings from the house perimeter (m).
The expression developed to represent radon entry through floor cracks and openings
included both radon diffusion and pressure-driven advective transport of radon. The diffusive
component was a relatively constant function of the crack area and the concentration
gradient, while the advective component depended strongly on the air pressure gradient and
the air permeability of the soil. Using the indoor-outdoor air pressure difference as a
surrogate for the pressure gradient, and the soil moisture as a surrogate for the soil air
permeability, the following expression was developed for ^crk:
Wcrk = 1/70 + AP exp(-3-°.045e6S) (16)
where
1/70 = diffusive component of the effective velocity for radon entry (mm s"1)
fRn = ratio of radon concentration beneath the crack to the average under the
slab.
2-10
-------
Using equations (15) arid (16) in equation (14) gave the following expression for the
lumped-parameter model (Nie93a):
Cn„t = Cin - Cout = (3.6 CsubfssvAX.h) [fc(l/70 + APexp(-3-0.045e6S)
+ 2.9xlO'7e11-4W + 4.6xl0*5/xh + 3.5xl0"5(xcrk/xh)]. (17)
This relation was used for analyzing and comparing the measured and calculated Cnet/Csub
ratios among NHEP houses with capped, passive, and active sub-slab ventilation systems.
2-11
-------
-------
3. HOUSE PARAMETERS AND RADON MEASUREMENTS
Data from the second phase of the FRRP NBEF were compiled in terms of the
parameters used in the lumped-parameter model or their surrogates. The data were taken
from measurements by FSEC (Tys93) and UF (Naj93). The basic house parameters for the
FSEC and UF houses are presented in Tables 1 and 2, respectively. The FSEC set contained
equal numbers of slab-in-stem-wall (SSW) and monolithic-slab houses (seven each), since the
data for house 28 were incomplete and insufficient for model comparisons. The UF set
contained nine SSW houses and five monolithic-slab houses. The house areas, volumes,
widths, and numbers of stories averaged higher for the SSW houses than for the monolithic-
slab houses in both data sets, suggesting the tendency to use SSW designs for larger houses.
The concrete slumps for the floor slabs were higher for the UF houses due to the more
frequent use of super plasticizers, Slab reinforcement included wire mesh, glass libers, and
in the FSEC case, post-tensioning. Sub-slab ventilation systems included both suction pits
and ventilation mat, with well-point pipe being used in some of the FSEC houses, generally
in connection with suction pits.
Measurements of soil and house ventilation parameters are summarized in Tables 3
and 4 for the FSEC and UF houses, respectively. Soil air permeability was generally lower
for the FSEC houses, consistent with their higher average soil moisture contents. House
permeability or leakage, represented by 50-Pa house ventilation rates, was equivalent for
both sets. House pressure exponents, n, computed from blower-door tests, were only reported
for the UF data set. They averaged 0.66 ± 0.05, equivalent to the typical reported value of
2/3 (She87). Therefore, n=2/3 was used to represent the pressure exponents for the FSEC
houses. Observed slab crack areas were much higher for the FSEC houses because physical
measurements of crack area were reported and used. For the UF houses, only effective crack
areas, estimated from air flow measurements, were reported. Soil density was reported only
for the UF houses, averaging 1.6 ± 0.1 g cm"3. Therefore, a constant value of 1.6 g cm"3 was
assumed to also apply to the FSEC houses.
3-1
-------
Table 1. FSEC HOUSE PARAMETERS.
Base Occup, Inside Equiv.
House Area Vol.a Height Wid.fc
ID (m^) (m3) (m) (m) Stories Const
Floor Slab
No. House^
Slump Super-
Detail (in)e plast/
Rein-
fore#
SSY
Syst*
14
113
690
6.1
12.8
2
BL
SSW
4.0
. N
W
WP-SP
19
219
627
2.9
17.1
2
BL-FR
SSW
4.0
Y
F
WP-SP
21
248
738
3.0
19.1
1
BL
SSW
- 4.5
N
W-F
WP
22
211
530
2.5
10.9
1
BL
SSW
4.5
N
w
VM
23
314
1531
4.9
18.1
2
FR-BR
SSW
5.0
Y
F
VM
26
492
1393
2.8
27.8
2
FR
SSW
5.0
N
W
VM
27
278
847
3.0
17.2
2
BL
SSW
4.0
• N
W
WP-SP
28
—
—
—
...
2
—
SSW
5.0
N'
w
VM
Mean
268
908
3.6
17.6
1.7
4.5
±S.D.
±108
±364
±1.2
±5.0
±0.4
±0.4
15
252
769
3,0
16.4
1
BL
Mono
5.0
N
PT
VM
16
174
531
3.0
14.6
1
BL
Mono
6.0
N
W
SP-VM
17
214
602
2.8
16.0
1
BL
Mono
4.5
N
W-F
VM
18
241
626
2.6
19.2
1
BL
Mono
4.0
Y
PT
VM
20
164
500
3.0
15.9
1
BL
Mono
5.0
N
W
VM
24
228
765
3.4
18.0
1
BL
Mono
5.0
N
W
VM
25
174
531
3.0
15.0
1
Mono
7.0
Y
VM
Mean
207
618
3.0
16.4
1
5.2
±S.D.
±33
±103
±0.2
±1.5
±0
±0.9
Overall
Mean
237
763
3.3
17.0
1.4
4.8
±S.D.
±86
o
CO
H
±0.9
±3.7
±0.5
±0.8
"Volume of the occupied space in the house.
6Width of the equivalent rectangular area of the house footprint.
^Construction: block (BL), frame (PR), or brick (BR).
dSIab edge detail; slab poured into stem wall (SSW) or monolithic slab (Mono).
eConcrete slump.
^Super plasticizer used in slab concrete (Yes or No).
gSlab reinforcement; wire mesh (W), glass fiber (F), or post-tensioned (PT).
ASub-slab ventilation system: well point (WP), suction pit (SP), or ventilation mat (VM).
3-2
-------
Table 2. UF HOUSE PARAMETERS.
House
ID
Base Occup.
Area Vol.°
(m2) (m3)
Inside Equiv,
Height Wid6
(m) (m)
No. House ¦
Stories Const.c
Edged
Detail
Floor Slab
Slump Super-
(inf plast/
Rein-
fore.^
ssv
Syst.A
HPBLl
235
858
3.7
9.9
1
FR
SSW
8.0
Y
F
VM
RDTS1
328
1026
3.1
9.8
2
FR
SSW
7.0
Y
B
VM
OMMJ1
223
655
2.9
14.3
1
FR
SSW
8.0
Y
W
VM
SOGR1
248
665
2.7
15.0
1
FR
SSW
8.0
Y
W
VM
CFSH1
227
623
2.7
7.7
2
FR
SSW
8.0
Y
W
VM
CFSH2
139
401
2.9
6.6
1
FR
SSW
9.0
Y
W-F
VM
GFSH3
164
426
2.6
7.9
1
FR
SSW
8.0
N
—
VM
RDTS2
301
811
2.7
8.7
2
FR-BR
SSW
8.0
N
—
VM
SOGR2
180
494
2.7
12.2
1
FR
SSW
8.0
N
—
VM
Mean
233
683
2.9
10.0
1.4
8.0
±S.D.
±59
±198
±0.3
±2.9
±0.5
±0.5
ASGR1
227
693
3.0
11.3
1
FR
Mono
7.0
Y
W
VM
ASEM1
198
603
3.0
14.6
1
FR
Mono
8.0
Y
W-F
VM
ASEM2
269
912
3.4
12.2
1.5
FR
Mono
6.0
Y
VM
ASGR2
231
668
2.9
14.9
1
FR
Mono
8.0
N
—
VM
RBPE1
164
500
3.0
14.9
1
FR
Mono
8.0
N
VM
Mean
212
645
3.0
13.3
1.1
7.5
±S.D.
±35
±141
±0.2
i+
in
±0.2
±0.8
Overall
Mean 224 667 3.0 11.4 1.2 7.8
+S.D. - ±51 ±177 ±0.3 ±2.9 ±0.4 ±0.7
^Volume of the occupied space in the house.
HVidth of the equivalent rectangular area of the house footprint.
Construction: block (BL), frame (FR), or brick (BR).
dSlab edge detail: slab poured into stem wall (SSW) or monolithic slab (Mono).
'Concrete slump.
'"Super plasticizer used in slab concrete (Yes or No).
^Slab reinforcement: wire mesh (W), glass fiber (F), or post-tensioned (PT).
ASub-slab ventilation system: ventilation mat (VM).
3-3
-------
Table 3. FSEC HOUSE VENTILATION AND SOIL MEASUREMENTS.
Soil Air
Soil"
Filla
Fill
House Press,c Reported**
Slab*
Soil^
House Permeability Moist.
Moist.
Depth Perm.
Expon. Nat. Vent Crk. Area Density
ID
(cm )
(% dry) (% dry)
(cm)
(ach50)
n
(ach)
(cm2)
(g/enr)
14
2.7xl0'7
6
46
5.1
0.67
0.26
9.7
1.60
19
2.7xl0'7
1
7
30
5.2
0.67
0.26
44.
1.60
21
l.lxlO*7
7
4
30
4.8
0.67
0.24
35.
1.60
22
UxlO"7
5
1
15
4.6
0.67
0.23
0.0
1.60
23
3.3xl0'7
3
5
46
3.8
0.67
0.19
69.
1.60
26
4.8x10"®
9
5
30
7.8
0.67
0.39
38.
1.60
27
3.6xl0"7
8
7
18
5.4
0.67
0.27
0.0
1.60
28
2.7xl0'7
19
11
61
...
0.67
0.30
204
1.60
Mean
2.3xl0'7
7.2
5.7
35
5.2
0.67
0.29
50.
1.60
±S.D.
±l.lxl0"7
±5.4
±3.1
±15
±1.2
±0.07
±67.
15
2-lxlO"8
5
5
30
6.5
0.67
0.33
10.
1.60
16
8.3x10"®
13
10
46
7.1
0.67
0.36
0.0
1.60
17
8,8xl0"9
9
4
15
3.9
0.67
0.20
22.
1.60
18
8.8x10-®
5
5
30
6.7
0.67
0.34
0.0
1.60
20
4.6x10-®
—
6
30
5.6
0.67
0.28
41.
1.60
24
1.6xl0'7
...
5
15
5.3
0.67
0.27
540.
1.60
25
3.4xl0"7
11
4
61
—
0.67
0.44
28.
1.60
Mean
l.lxlO"7
8.6
5.6
33
5.8
0.67
0.31
92.
1.60
±S.D.
±1.2xl0'7
±3.6
±2.1
±16
±1.2
±0.08
±200.
Overall
Mean
1.7xl0"7
7.8
5.6
34
5.5
0.67
0.29
69.
1.60
±S.D.
±1.3xl0*7
±4.7
±2.5
±15
±1.2
±0.07
±140.
""Moisture percentage, dry-weight basis.
6Infiltration air changes per hour at 50 Pa pressure, from blower-door test.
cAssumed typical pressure exponent for blower-door test, since none were reported.
^Passive-condition air infiltration rate.
eTotal area of observed slab cracks.
^Assumed typical soil densities, since none were reported.
3-4
-------
Table 4. UF HOUSE VENTILATION AND SOIL MEASUREMENTS
Soil Air
Soil0
Fill
House Press.0 Reported**
Slab®
Soil'
House Permeability Moist,
Depth Perm.6
Expon. Nat. Vent Crk. Area Density
ID
(cm )
(% dry)
(cm)
(ach50)
n
(ach)
(cm )
(g/cm4)
HPBL1
1.3xl0"7
2
0
7.8
0.58
0.50
1.39
RDTS1
2.2X10-8
5
—
4.4
0.62
0.22
—
1.67
OMMJ1
2.4xl0"9
13
—
5.7
0.65
0.28
0.02
1.43
SOGR1
l.SxlO"7
9
0
3.9
0,70
0.19
—
1.66
CFSHl
1.7xl0"7
3
—
3.9
0.67
0.33
0.02
1.49
CFSH2
8.7x10"®
3
—
5.6
0.69
0.21
—
1.64
CFSH3
5,5x10"®
4
—
6.8
0.69
0.32
...
1.74
EDTS2
2,lxl0"6
4
6.6
0.59
0.55
0.01
1.63
SOGR2
2.1x10"®
4
—
5.9
0.72
0.19
—
1.73
Mean
9.1xl0"7
5.2
0
5.6
0.65
0.31
0.015
1.60
±S.D.
±1.9xl0"6
±3.5
±0
±1.3
±0.05
±0.13
±0.005
±0.13
ASGR1
1.3xl0"7
4
5.4
0.71
0.14
0.01
1.79
ASEM1
1.3xl0"7
3
...
6.5
0.72
0.20
—
1.56
ASEM2
7.4x10"®
5
— •
7.3
0,63
0.12
—
1.60
ASGR2
3,9xl0"6
4
—
5.9
0.66
0.18
—
1.58
RBPE1
2.1xl0"7
2
...
4.0
0.59
0.20
0.02
1.64
Mean
9.0xl0"7
3.6
—
5.8
0.66
0.17
0.014
1.63
±S.D.
±1.7x10*®
±1.1
±1.2
±0.05
±0.04
±0.004
±0.09
Overall
Mean
9.1xl0"7
4,6
0
5.7
0.66
0.26
0.014
1.61
+S.D.
±1.7x10"®
±3.0
±0
±1.3
±0.05
±0.13
±0.004
±0.11
"Moisture percentage, dry-weight basis.
^Infiltration air changes per hour at 50 Pa pressure, from blower-door test,
cAir pressure exponent measured in blower-door tests,
^Passive-condition air infiltration rate.
Calculated effective area of observable slab cracks.
^Assumed typical soil densities, since none were reported.
3-5
-------
The air pressure and ventilation characteristics of the FSEC and TJF houses are
summarized in Tables 5 and 6, respectively. The air pressures in both cases are reported
relative to the house interior; therefore, a house pressure of -0.9 Pa indicates the house is
positively pressurized relative to the outdoor atmosphere. Conversely, a house pressure of
0.4 Pa indicates the house is at -0.4 Pa relative to the outdoor atmosphere. The indoor air
pressures are largely scattered about zero, with some being positive and some negative.
Since these data are from short periods of time, they give little information about the long-
term average pressures in these houses that would contribute to long-term average radon
entry rates. The interior pressures with exhaust fans on tend to be more negative relative
to outside pressures than the others, as expected. The interior pressures with the air handler
on and the interior doors closed also tend to be more negative relative to outside pressures
than the measurements with interior doors open.
Air infiltration rates, measured by tracer gas dilution, increased from turning on the
house air handler, and also from closing the interior doors. The lower air infiltration rate
with exhaust fans on, reported for the FSEC data set, is counter-intuitive. Corresponding
infiltration rates were not reported with the UF data set (Naj93). Slab leakage fractions
reported for both sets exhibit considerable scatter, and show higher leakage for both the
monolithic slabs in the FSEC set and the SSW slabs in the UF set.
Radon measurements in the sub-slab and indoor environments are documented in the
original reports by the FSEC (Tys93) and UF (Naj93) research groups. The radon
measurements in soil gas utilized flow-through alpha scintillation cells, through which soil
gas was pumped (typically at 1-2 L min"1 flow rates). With suitable calibrations, these
devices can achieve good accuracy (±<10%) and high precisions due to high count rates and
resulting good counting statistics. Sampling biases, from sub-slab voids or nearly-saturated
moisture conditions, can sometimes limit sub-slab measurement accuracy. Indoor radon
measurements used in this report were generally short-term measurements over several days
or less, and utilized either scintillation cell grab samples or charcoal canister passive
samplers. For both methods, precisions are poorer because radon concentrations are closer
to the detection limits. However, precisions of ±20% are generally attainable in most cases.
Details of the measurement protocols are given in Wil91.
3-6
-------
Table 5. FSEC HOUSE PRESSURE AND VENTILATION MEASUREMENTS
AH Off, DO®
AH On
, DO
AH On,
DC°
AH, EX On, DO®
House
Pres.6
Vent.c
Pres.
Vent.
Pres.
Vent.
Pres.
Vent.
SLF*
ID
(Pa)
(ach)
(Pa)
(ach)
(Pa)
(ach)
(Pa)
(ach)
(%)
14
-0.9
0.13
-0.8
0.44
-0.1
0.39
4.6
0.08
4.3
19
0.4
0.19
0.7
0.26
-2.2
0.44
-3.3
0.13
17.4
21
0.2
0.12
-0.3
0.26
8.1
0.85
2.4
—
—
22
-0.6
0.13
-1.1
0.36
1.7
0.58
-0.8
0.22
9.2
23
-0.4
0.31
-0.4
0.46
1.0
0.54
1.4
0.18
6.6
26
-1.6
0.27
-2.1
0.44
2.1
0.50
-0.1
0.20
14.7
27
OQ
0.0
0.17
0.0
0.41
1.3
0.45
2.7
—
Zo
Mean
-0.41
0.19
-0.57
0.38
1.7
0.54
0.99
0.16
10.4
±S.D,
±0.64
±0.08
±0.82
±0.08
±2.9
±0.14
±2.42
±0.05
±4.9
15
0.0
0.19
-0.3
0,50
0.9
0.70
2.4
0.23
13.3
16
0.2
0.12
-0.2
0.61
0.0
0.57
1.1
0.16
22.0
17
-0.8
0.16
-0.2
0.47
-0.8
0.45
6.0
0.31
34.8
18
0.3
0.06
-2.6
0.18
1.7
0.47
2.7
0.12
10.0
20
0.6
0.20
-0.1
0.51
1.8
0.60
1.5
0.14
18.7
24
-0.3
0.22
-0.8
0.43
0.5
0.35
1.9
0.12
30.2
25
—
0.12
—
0.76
—
0.56
—
0.08
6.9
Mean
0.00
0.15
-0.70
0.50
0.7
0.53
2.60
0.17
19.4
±S.D.
±0.45
±0.06
±0.88
±0.18
±0.9
±0.11
±1.61
±0.07
±9.6
Overall
Mean
-0.22
0.17
-0.63
0.44
1.2
0.53
1.73
0.16
15.7
±S.D.
0.60
±0.07
±0.85
±0.15
±2.3
±0.12
±2.23
±0.06
±9.1
aAir handler (AH), exhaust fans (EX), interior doors open (DO), interior doors closed (DC).
^Outdoor-indoor air pressure difference (pascals).
cHouse ventilation rate from tracer gas measurements (air changes per hour),
rfSlab leakage fraction.
3-7
-------
Table 6. UF HOUSE PRESSURE AND VENTILATION MEASUREMENTS
AH Off, DOa
AH On, DO
AH On, DC
AH,EX On, DOa
SLF1
House
Pres.0
Vent.c
Pres.
Vent.
Pres.
Vent.
Pres.
ID
(Pa)
(ach)
(Pa)
(ach)
(Pa)
(ach)
(Pa)
(%)
HPBL1
0.0
0.50
0.1
0.42
1.0
0.56
0.7
10.1
RDTS1
0.9
0.22
1.3
0.32
9.8
0.63
5.2
66.9
OMMJ1
0.2
0.28
0.0
0.35
2.3
0.74
1.8
23.6
SOGR1
-0.1
0.19
0.5
0.42
6.0
0.69
8.9
28.2
CFSH1
-0.2
0.33
0.0
0.55
0.7
0.73
1.6
15.1
CFSH2
1.3
0.21
0.9
0.30
4.9
0.76
8.1
12.6
CFSH3
0.6
0.32
-1.4
0.93
1.9
0.92
4.3
22,6
RDTS2
2.1
0.55
1.8
0.55
1.9
—
3.3
28.4
SOGR2
-0.6
0.19
-0.1
0.34
1.8
0.49
2.4
21.1
Mean
0.47
0.31
0.34
0.46
3.37
0.69
4.03
25.4
±S.D.
±0.85
±0.13
±0.93
±0.20
±2.98
±0.13
±2.89
±16.9
ASGR1
0.0
0.14
0.0
0.33
6.7
0.63
2.2
16.4
ASEM1
-0.5
0.20
0.1
0.41
0.7
0.44
1.7
26.8
ASEM2
0.0
0.12
-0.6
0.52
2.4
0.79
1.3
6.7
ASGR2
0.0
0.18
-0.2
0.41
2.1
0.76
1.5
14,0
RBPE1
0.2
0.20
1.5
0.40
6.9
0.81
4.5
19.4
Mean
-0.06
0.17
0.16
0.41
3.76
0.69
2.24
16.7
±S.D.
±0.26
±0.04
±0.80
±0.07
±2.85
±0.15
±1.31
±7.4
Overall
Mean
0.28
0.26
0.28
0.45
3.51
0.69
3.39
22.3
±S.D.
±0.73
±0.13
±0.86
±0.16
±2.83
±0.13
±2.54
±14.5
aAir handler (AH), exhaust fans (EX), interior doors open (DO), interior doors closed (DC).
^Outdoor-indoor air pressure difference (pascals).
^ouse ventilation rate from tracer gas measurements (air changes per hour).
j
Slab leakage fraction.
3-8
-------
Sub-slab and indoor radon measurements are summarized in Tables 7 and 8 for the
PSEC and UF data sets, respectively. Although similar levels are apparent for many of the
comparisons between the IT-1 and capped sub-slab measurements, very large variations are
apparent for certain houses in each set. The capped set in these tables is preferred due to
its longer equilibration and measurement times. These variations are as high as a factor of
17 in the FSEC set (house 17), and a factor of 4.9 in the UF set (house CFSH3). Sub-slab
radon concentrations average lower, as expected, with passive sub-slab ventilation (SSV),
despite cases in which higher levels were observed. Passive ventilation reduced FSEC sub-
slab radon levels by 28% for SSW houses and 14% for monolithic slab houses. Passive sub-
slab ventilation in the UF houses similarly reduced sub-slab radon levels by 15% in SSW
houses and by 5% in monolithic slab houses.
Indoor radon levels in the FSEC houses under capped-SSV conditions averaged 2.4±1.5
pCi L"1 and 2.3±1.5 pCi L"1 for SSW and monolithic slab houses, respectively. In the UF
houses, the indoor radon levels under capped-SSV conditions averaged 3.3±3.2 pCi L"1 for the
SSW houses and 1.1±0.4 pCi L"1 for the monolithic slab houses. The SSW mean was
dominated by one house, however, and had a value of 2.3±0.6 pCi L"1 for the other eight
houses. Passive-SSV means in Table 7 show a consistent trend of higher indoor radon than
for capped-SSV for the FSEC houses. The passive-SSV data for UF houses in Table 8 show
equal average indoor radon levels under capped- and passive-SSV conditions. For both data
sets, active SSV systems reduced the indoor radon levels. However the data are sparse for
the active-SSV case because the SSV systems were not activated unless indoor radon levels
approached critical levels associated with the 4 pCi L"1 EPA criterion.
3-9
-------
Table 7. FSEC SUB-SLAB AND INDOOR RADON MEASUREMENTS
IT-16 Capped0 Passive4* Active*' IT-16 Capped0 Passive* Active*
Soil Sub-slab Sub-slab Sub-slab Sub-slab Indoor Indoor Indoor Indoor
House Radon® Radon Radon Radon Radon Radon Radon Radon Radon
ID (pCi L_1)(pCi L^HpCi L'1)(pCi L^HpCi L':)(pCi L^KpCi L^KpCi L^HpCi L1)
14
2330
5580
4530
2.3
1.4
1.5
19
—
—
1160
988
—
3.0
3.0
4.4
1.7
21
6860
3420
4780
4760
—
2.5
5.6
4.8
2.9
22
853
2680
1240
1700
—
0.2
1.1
1.4
—
23
1600
2410
2410
1090
—
0.2
0.9
2.8
—
26
2210
1710
1940
2230
...
1.3
2.5
2.8
—
27
8450
6370
3200
3700
298
1.1
3.5
2.2
LI
28
6280
—
1660
1320
...
.-
0.9
1.0
—
Mean
4080
3320
2750
2540
298
1.5
2.4
2.6
1.9
±S.D.
±3020
±1810
±1650
±1560
—
±1.1
±1.6
±1.4
±0.9
15
3440
1550
2690
2840
...
0.9
1.0
1.9
16
4080
3190
3640
3680
1510
2.8
5.2
4.3
1.8
17
947
424
7180
3650
LI
3.4
3.7
2.8
18
7260
1080
3460
—
...
1.8
2.1
3.4
2.0
20
3000
7220
2670
3040
0.0
2.3
2.1
...
24
418
2740
2820
2660
—
0.5
2.5
1.0
—
25
—
3800
3550
4510
1.2
1.1
0.9
—
Mean
3190
2860
3720
3400
1510
1.2
2.5
2.5
2.2
±S.D.
±2460
±2260
±1580
±690
—
±0.9
±1.4
±1.3
±0.5
Overall
Mean
3670
3050
3200
2910
900
1.3
2.4
2.5
2.0
±S.D.
±2700
±2010
±1640
±1300
±860
±1.0
±1.5
±1.3
±0.7
aSoil gas radon before slab pour, at 0.9 m depth, except 15,16, and 25, which were shallower.
infiltration test 1, one hour duration, air handler off and interior doors open.
cSub-slab ventilation system capped.
dSub-slab ventilation system passively vented.
eSub-slab ventilation system actively vented with suction fan.
3-10
-------
Table 8. UF SUB-SLAB AND INDOOR RADON MEASUREMENTS
IT-16 Cappedc Passive12 Active IT- lb Cappedc Passive** Active8
Soil Sub-slab Sub-slab Sub-slab Sub-slab Indoor Indoor Indoor Indoor
House Radona Radon Radon Radon Radon Radon Radon Radon Radon
ID (pCi L_1)(pCi L_1)(pCi L^XpCi L^XpCi L-1)(pCi L_1)(pCi L_1)(pCi L^MpCi L"1)
HPBL1
5320
6620
7800
7080
2420
5.3
11.6
12.6
0.9
RDTS1
31900
995
1070
1160
...
1,5
2.1
1.7
—
OMMJ1
2680
360
380
363
—
1.0
1.9
1.9
—
SOGR1
10000
11500
3900
3210
—
1.7
3.5
4.2
—
CFSH1
2680
590
1390
1120
—
1.0
1,7
1.2
CFSH2
1900
2810
2220
2430
—
0.7
2.1
2.9
—
CFSH3
4980
600
2940
1830
—
0.7
2.5
1.9
...
RDTS2
2750
2020
2950
3430
...
0.9
1.5
1.2
—
SOGR2
2820
1710
2400
2080
—
0.5
2.7
1.9
...
Mean
7230
3020
2780
2520
2420
1.5
3.3
3.3
0.9
±S.D.
±9580
±3700
±2170
±1980
±1.5
±3.2
±3.6
—
ASGR1
690
1020
829
830
2.3
¦ 1.2
1.3
ASEM1
2070
1710
1020
880
—
0.9
0.6
1.3
—
ASEM2
10800
2850
3880
3810
—
1.2
1.0
0.8
— ^
ASGR2
1870
740
852
740
...
0.6
1.6
1.4
—
RBPE1
1400
338
558
540
—
0.6
0.9
0.7
Mean
3370
1330
1430
1360
- —
1.1
1.1
1.1
...
±S.D.
+4200
±980
±1380
±1380
±0.7
±0.4
±0.3
Overall
Mean
5850
2420
2300
2110
2420
1.3
2.5
2.4
1.1
±S.D.
±8100
±3070
±1980
±1820
±1.2
±2.7
±3.0
±0.3
"Soil gas radon before slab pour, at 0,9 m depth.
^Infiltration test 1, one hour duration, air handler off and interior doors open.
cSub-slab ventilation system capped.
^Sub-slab ventilation system passively vented.
eSub-slab ventilation system actively vented with suction fan.
3-11
-------
-------
4. COMPARISONS WITH THE LUMPED-PARAMETER MODEL
The comparisons of measured radon concentrations with predictions from the lumped-
parameter model were made using Cnet/Cgub ratios to normalize the different radon source
strengths for each house to a common basis. Parameters for determining the measured
Cnet/Csub ratios were estimated from measured sub-slab radon concentrations. Parameters
for use in the lumped-parameter model were defined directly from measured values or were
calculated from surrogate measurements in certain cases. The resulting calculated values
of Cnet/Csub were then compared with measured Cnet/C85lb ratios defmed directly from the
indoor and sub-slab radon measurements.
4.1 DEFINITION OF MEASURED Cnrt/Cml: RATIOS
Outdoor radon concentrations for converting the measured Cin values to Cnet [as in
equation (14)] were not available, and were therefore estimated as described in the original
lumped-parameter model report (Nie93a) as:
C0ut = Csub (9xl0"5 - 8.7xl0"5 S). ¦ (18)
The calculated outdoor concentrations are listed in Tables 9 and 10 for the FSEC and UF
data, respectively. They are generally small compared to the measured indoor levels, and
therefore have relatively little impact on the measured Cnet/Csub ratios, except for cases
where the indoor radon levels were very low.
Sub-slab radon concentrations for estimating the measured Cnet/Csub ratio [as in
equation (14)] were defined from the capped-SSV measurements in Tables 7 and 8. This
approach provided the best estimate of the effective radon source for each house, independent
of the operation of the SSV system.
4-1
-------
Table 9. LUMPED-PARAMETER VALUES CALCULATED FOR FSEC HOUSES
Average0
Average0
Soild
Concrete6
fW
Outdoor0
Air
House Air
Moisture
Slab
Crack
House
Radon
Infiltration
Pressure
Saturation
W ater/Cement
Location
ID
(pCi L*1)
(ach)
(Pa)
(fraction)
(ratio)
(m)
14
0.4
0.30
0.73
0.24
0.60
0
19
0.1
0.31
0.73
0.04
0.57
0
21
0.3
0.23
0.52
0.27
0.61
0
22
0.1
0.22
0.52
0.20
0.61
0
23
0.2
0.23
0.73
0.12
0.58
0
26
0.1
0.46
0.73
0.35
0.62
0
27
0.2
0.32
0.73
0.31
0.60
0
28
0.0
0.33
0.73
0.75
0.62
0
Mean
0.2
0.30
0.68
0.28
0.60
0
±S.D.
±0.1
±0.08
±0.10
±0.21
±0.02
±0
15
0.2
0.31
0.52
0.20
0.62
7
16
0.2
0.34
0.52
0.51
0.63
—
17
0.4
0.19
0.52
0.35
0.61
3
18
0.2
0.32
0.52
0.20
0.57
—
20
0.2
0.27
0.52
0.05
0.62
3
24
0.2
0.25
0.52
0.05
0.62
7
25
0.2
0.29
0.52
0.43
0.61
4
Mean
0.2
0.28
0.52
0.26
0.61
4.8
±S.D.
±0.1
±0.05
±0.00
±0.18
±0.02
±2.0
Overall
Mean
0.2
0.29
0.60
0.27
0.61
—
±S.D.
±0.1
±0.07
±0.11
±0.19
±0.02
"Calculated from sub-slab radon as in Nie93a.
^Calculated using equation (19).
Calculated using equation (20).
' j
Calculated using equation (6),
'Calculated using equation (21).
^Estimated from floor crack diagrams in Tys93.
4-2
-------
Table 10. LUMPED-PARAMETER VALUES CALCULATED FOR UF HOUSES
House
ID
Outdoor0
Radon
(pCi L"1)
Average6
Air
Infiltration
(ach)
Average^
House Air
Pressure
(Pa)
Soild
Moisture
Satur ation
(fraction)
Concrete*
Slab
Water/Cement
(ratio)
Floor'
Crack
Location
(m)
HPBL1
0.7
0.42
0.32
0.06
0.63
0
RDTS1
0.1
0.28
0.59
0.22
0.61
0
OMMJ1
0.0
0.28
0.47
0.39
0.63
0
SOGE1
0.2
0.18
0.59
0.39
0.63
0
CFSH1
0.1
0.23
0.72
0.10
0.63
0
CFSH2
0.2
0.26
0.57
0.13
0.64
0
CFSH3
0.2
0.31
0.58
0.20
0.66
0
RDTS2
0.2
0.44
0.49
0.17
0.66
0
SOGR2
0.2
0.26
0.64
0.19
0.66
0
Mean
0.2
0.30
0.55
0.20
0.64
o
±S.D.
±0.2
±0.08
±0.11
±0.12
±0.02
±0
ASGR1
0.1
0.24
0.64
0.21
0.61
2
ASEM1
0.1
0.29
0.66
0.11
0.63
—
ASEM2
0.3
•. 0.40
0.52
0.20
0.60
—
ASGR2
0.1
0.29
0.49
0.15
0.66
...
RBPE1
0.1
0.21
0.35
0.08
0.66
3
Mean
0.1
0.29
0.53
0.15
0.63
2.5
±S.D.
±0.1
±0.07
±0.13
±0.06
±0.03
±0.7
Overall
Mean
0.2
0.29
0.54
0.18
0.64
...
±S.D.
±0.2
±0.08
±0.11
±0.10
±0.02
Calculated from sub-slab radon as in Nie93a.
^Calculated using equation (19).
Calculated using equation (20).
^Calculated using equation (6).
Calculated using equation (21).
^Estimated from floor crack diagrams in Naj93.
4-3
-------
4.2 DEFINITION OF LUMPED PARAMETERS
Several model parameters were calculated from the measured data to compare the
measurements with the lumped-parameter model. These include the house air infiltration
rate (Xh), the indoor air pressure (AP), the soil water saturation fraction (S), and the concrete
slab water/cement ratio (W). Other parameters, including fgsv, h, fc, xh, and xc were defined
more directly from house measurements.
Several measurements of the house air infiltration rate were considered for use in the
model analyses, including those listed in Tables 5 and 6. However, the variations of these
rates and their short measurement durations weakened their basis for representing long-term
average house ventilation rates. An alternative approach therefore was chosen, based on the
measured 50-Pa infiltration rates (ACHg0) and pressure exponents (n) from blower door tests
(from Tables 3 and 4). These data utilize the more reliable leakage estimates from -50-Pa
depressurization, thereby avoiding biases from wind turbulence and thermal stack effects,
which can become significant compared to ventilation at pressures of a few Pa. The
relationship between and ACH50 was defined as by Sherman (She87):
Xh = ACH50 / (N0 fx f2 f3) (19)
where
ach50
= house ventilation rate at -50 Pa depressurization (air changes per hour)
N0
= leakage-infiltration ratio (dimensionless)
h
= height correction (= 1 for 1-story; 0.9 for 1.5-story; 0.8 for 2-story buildings)
= shielding correction (=1.2 for well-shielded; 1.0 for normal; and 0.9 for
exposed buildings)
h
= leakiness correction (=1 for normal; 1.4 for tight; and 0.7 for leaky
buildings).
Using Sherman's climate-geographic definition of No=20 to 22 for central and
peninsular Florida, an average value of N0=21 was chosen for use in equation (19), The
4-4
-------
number of stories for each house in Tables 1 and 2 were used to define fls and a normal-
shielding value of unity was used to define f2. The value of f3 was defined as f3=4n 2/3, based
on the reference pressure and exponent used by Sherman (She87). The resulting house
ventilation rates calculated by equation (19) are listed in Tables 9 and 10 for the FSEC and
UF houses, respectively. They are generally comparable to the ventilation rates in Tables
5 and 6 for the open-door, no-air-handler cases, but the present values exhibit considerably
less scatter among different houses.
Indoor air pressures similarly were evaluated using the long-term average relationship
for house ventilation, as given by equation (19), Measured values from Tables 5 and 6
initially suggested considerable scatter and dominance by short-term fluctuations from wind
and thermal effects. Since the values are all small, and include both negative and positive
pressures, a longer-term average that represents the net depressurization of the house was
desired. This was calculated from the infiltration-pressure relationships at -50 Pa and at
passive conditions [as in equation (19)] to obtain:
APp«ss - "50 /
-------
The water/cement ratios of the concrete slabs were estimated from the concrete slump
and the use of super-plasticizers as reported in the FSEC and UF reports. The relationship
between the water/cement ratio and the concrete slump was estimated by fitting slump
versus water content data from American Concrete Institute (ACI) specifications for concrete
with 1.3-cm aggregate, without air entrainment (ACI89). The resulting fit was normalized
to a 5-inch slump at 0.55 water/cement ratio to yield the relationship:
W = (0.0144 Sc + 0.48) V (21)
where
W = water/cement ratio (dimensionless)
Sc = slump of poured concrete (inches)
\|f = 0.9 if superplastieizer was used, or 1.0 otherwise.
The reduction in water/cement ratio using super plastieizer was defined from the ACI
estimate of water reduction in concretes with water-reducing admixtures (ACI89). The
resulting estimates of water/cement ratio for the FSEC and UF houses averaged 0.61±0.02
and 0.64±0.02, as listed in Tables 9 and 10, respectively.
As noted previously (Nie93a,c), the effective size of floor crack openings is difficult to
characterize and is generally not well known. For the present analyses, the floor crack
openings for SSW houses were defined from the sum of the observed crack areas (in Tables
3 and 4) plus the previous estimate (Nie93a) for openings associated with hollow-block stem
walls, fc=0.0029. The stem wall portion of the crack areas completely dominated the
resulting estimates of fc. For monolithic-slab houses, only the observed crack areas were used
(divided by total slab areas to obtain fc). The fc values for the monolithic-slab houses were
much lower than those used in the previous study (Nie93a).
The locations of the dominant floor cracks were defined to be at the perimeter (xcrk=0)
for the SSW houses, since their crack sizes were dominated by routes associated with the
hollow block stem walls. For the monolithic-slab houses, xcrk was defined graphically from
4-6
-------
the floor crack diagrams in Tys93 and Naj93. The values estimated for xcr({ are listed in
Tables 9 and 10, The widths of the houses, xh, were estimated from the house slab areas and
the aspect ratios of equivalent rectangles, which were also estimated graphically from the
floor crack diagrams in Tys93 and Naj93. The resulting estimates of xh are presented in
Tables 1 and 2, along with measured values for the height of the house, h.
The sub-slab ventilation factor was defined as fssv=l for capped-SSV systems, and as
fssv=0.936 for passive (uncapped) SSV systems, as in the previous lumped-parameter study
(Nie93a). However, fssv was defined from the average ratio (active-ventilation and capped-
SSV) of the measured sub-slab radon concentrations from Tables 7 and 8 for the active-SSV
condition. This gave a value of fssv=0.27 for the active-SSV condition. The passive-SSV value
of fssv was not calculated in the same manner because of the large uncertainty in the
corresponding ratio calculated for passive-SSV conditions, and its proximity to the value
obtained in the previous study.
4.3 COMPARISON OF MEASURED AND CALCULATED RADON RATIOS
The measured and calculated Cnet/Csub radon ratios estimated for each NHEP house
are presented in Table 11 for the SSW houses and in Table 12 for the monolithic-slab houses.
The geometric means of each data set are also plotted in Figure 2 for graphical comparisons.
The means and 68% (1-standard deviation) confidence limits for the lumped-parameter ratios
(calculated) are consistently within the corresponding 68% confidence limits of the measured
radon ratios for both capped and passive-SSV systems in both SSW and monolithic-slab
houses. The average and standard deviation of the ratios of the calculated/measured
geometric means for the capped-SSV and passive-SSV houses (both SSW and monolithic) is
1.01 ± 0.16. This suggests only about a 1% overall net bias in the lumped-parameter model
compared to the measured ratios, and a precision of about ±16% in its average agreement
with the measured radon ratios. For active-SSV houses, where data are more sparse and
scattered, the agreement is not as good. Calculated ratios for active-SSV houses are less than
half of the measured ratios for the six FSEC houses, but are about five times higher than the
measured ratio for the UF house.
4-7
-------
Table 11. COMPARISON OF MEASURED AND CALCULATED Cnet/Csub RATIOS
FOR SLAB-IN-STEM-WALL HOUSES
House Capped-SSV System Passive-SSV System Active-SSV System
ID
Measured"
Calculated^
Measured®
Calculated
Measured0
Calculated''
' 14
l.SlxlO"4
5.1bdO"4
1.99xl0"4
4.78xl0"4
19
2.50x1c3
8.69xl0'4
3.71X10"3
8.13xl0"4
1.38xl0"3
2.37xl0"4
21
l.llxlO"3
1.30x1c3
9.38xl0"4
1.21xl0'3
5.36xl0"4
3.54xl0"4
22
8.14xl0"4
1.65X10"3
1.06x1c3
1.54x1c3
—
—
23
2.93xl0"4
7.24xlC4
1.08xl0'3
6.77xl0"4
...
...
26
1.23X10"3
7.36xl0'4
1.38xl0"3
6.89xl0"4
—
—
27
1.03x10"®
9.28xl0"4
6.25xl0"4
8.68xl0'4
9 oi„in-4
MiMiSk JL \j
2.53X10"4
28
5.17xl0"4
7.23xl0"4
5.77xl0"4
6.77xl0"4
...
—
Geom. Mean
7.27xl0"4
8.73xl0"4
8.90xl0"4
8.17x1c4
5.93xl0'4
2.77xl0"4
Geom. S.D.
2.33
1.45
2.30
1.45
2.23
1.24
HPBL1
1.40xl0"3
5.11xl0'4
1.53xl0"3
4.78xl0"4
2.78xlC5
1.40x1c4
RDTS1
1.85xl0"3
9.48xlC4
1.55x1c3
8.87xlC4
—
—
OMMJ1
5.00xl0"3
9.43xl0"4
4.92x1c3
8.82xl0"4
—
—
SOGR1
8.44xl0"4
1.71X10"3
1.03x10"®
1.60x1c3
—
—
CFSH1
1.15xlC3
1.53xl0"3
7.60xl0"4
1.43X10"3
—
—
CFSH2
8.80xl0*4
1.31xl0"3
1.21x1c3
1.22xl0"3
—
—
CFSH3
7.83X10"4
1.64x1c3
5.76xl0"4
1.54X10"3
—
—
RDTS2
4.23xl0"4
LlOxlO"3
3.45xl0"4
1.03X10"3
...
SOGR2
1.04X10"3
1.87X10"3
7.40xl0"4
1.75x1c3
—
—
Geom. Mean
1.16X10-3
1.20xl0"3
1.06xl0"3
1.13X10"3
2.78xlC5
1.40xl0"4
Geom. S.D.
1.98
1.50
2.12
1.50
—
—
Overall
Geom. Mean
9.31xl0"4
1.04xl0"3
9.74xl0"4
9.68X1C4
2.76xl0"4
2.33xl0"4
Geom. S.D.
2.18
1.51
2.16
1.51
5.28
1.47
"Calculated using equation (14),
^Calculated using equation (17).
4-8
-------
Table 12. COMPARISON OF MEASURED AND CALCULATED Cnet/C8ub RATIOS
FOR MONOLITHIC-SLAB HOUSES
House Capped-SSV System Passive-SSV System Active-SSV System
ID
Measured0
Calculated
Measured®
Calculated
Measured"
Calculated0
15
2.99xl0"4
6.54xl0"4
6.33xl0"4
6.13xl0"4
mmrn
16
1.38X10"3
6.43xl0"4
l.MxlO"3
6.02X10"4
4.49xl0*4
1.76xl0'4
17
4.14xl0*4
1.04xl0"3
4.56xl0"4
9.74X10"4
3,31xl0'4
2.84xl0"4
18
5.34xl0"4
3.19xl0"4
9.10xl0"4
2.98xl0"4
B.OSxlCT4
8.70xl0'5
20
7.75xlQ"4
7.27xl0"4
7.00xl0"4
6.80X10-4
—
—
24
8-OOxlO*4
7.62xl0'4
2.69xl0*4
7.13x10*4
—
—
25
2.57xl0'4
S.OOxlO"4
2,QlxlCT4
4.68X10"4
—
...
Geom. Mean
5.47xl0"4
6.28xl0"4
5.27X10"4
5.88xl0"4
4.22xl0"4
1.63xl0'4
Geom. S.D.
1.82
1.45
1.88
1.45
1.24
1.81
ASGR1
1.38XKT3
5.84xl0"4
1.51X10"3
5.46xl0*4
ASEM1
4.69xl0"4
5.37xl0"4
1.17X10"3
5.03xl0"4
—
—
ASEM2
1.77xl0"4
2.60xl0"4
1.33X10"4
2.43xl0"4
—
—
ASGR2
l.SlxlO"3
l.llxlO"3
1.52xl0"3
1.04xl0"3
—
...
EBPE1
1.58x10-®
1.46xl0"3
l.OSxlG"3
1.37X10"3
—
—
Geom. Mean
8.00xl0"4
6.67xl0"4
8.27xl0"4
6.24xl0"4
—
—
Geom. S.D.
2.71
1.97
2.80
1.97
—
—
Overall
Geom. Mean
6.40xl0'4
6.44xl0"4
6.36xl0"4
6.03xl0'4
4.22xl0"4
1.63X10"4
Geom. S.D.
2.16
1.64
2.25
1.64
1.24
1.81
Calculated using equation (14).
^Calculated using equation (17).
4-9
-------
10
-2
Capped
SSV
FSEC UF
£}
3
W
O
o
c
Q
10-3rj„ 5"lj" {"
Passive
SSV
"FSFC IJ F
10~4
Active
SSV
FSEC
iLE
° Calculated
• Measured
10-5
Slab Poured into Stem Wall
Capped • Passive j Active
SSV { SSV j SSV
FSFC : FSFC U F ;
II
il
\ FSFC
I J
Monolithic Slab & Footing
6
a.
Figure 2. Comparison of calculated and measured Cnet/Csub ratios for capped,
passive, and active SSV systems in SSW and monolithic-slab houses.
Error bars represent one standard deviation (68% confidence interval).
The measured ratios for both the capped-SSV and passive-SSV systems were higher
(by factors of 1.60 and 1.19, respectively) for the UF houses than for the FSEC houses. This
indicates slightly better radon resistance for the FSEC houses. The lumped-parameter model
calculations showed similar trends, with ratios for both capped-SSV and passive-SSV UF
houses averaging approximately 1.38 higher than those for FSEC houses. The single UF
house with an active-SSV system had about 20 times lower radon entry than the other SSW
houses, only a factor of two of which was explained by the lumped-parameter model.
4-10
-------
The comparison of passive-SSV systems to capped-SSV systems involves large
uncertainties for both the FSEC and UF data sets, but suggests slightly higher radon entry
rates for the passive systems. The measured passive/capped ratio was 1.05 for SSW houses
(1.22 for FSEC and 0.91 for UF), and 0.99 for monolithic-slab houses (0.96 for FSEC and 1.03
for UF). Although the data do not indicate a significant benefit from capping the SSV vents,
they also show no benefit from leaving them uncapped, contrary to earlier hypotheses of a
stack-driven passive SSV system that would have intermediate performance between that of
a capped system and that of an active-SSV system. Although the geometric standard
deviations in the measured capped- and passive-SSV data sets in Figure 2 average about 2.24
(1.59 for the calculated sets), the variations are much smaller than those in the initial
lumped-parameter comparisons (Nie93a). The present data sets thus give a more definitive
estimate of the performance of capped-SSV and passive-SSV systems, but the measurement
variations are too large to conclude that the average measured differences are significant.
The comparison of monolithic-slab construction to SSW construction also involves large
uncertainties, but shows a significantly greater difference. The monolithic/SSW ratio for
average capped-SSV system performance (from Tables 11 and 12) is 0.69 (0.75 for FSEC
houses and 0.69 for UF houses). The corresponding monolithic/SSW ratio for passive-SSV
systems is 0.65 {0.59 for FSEC houses and 0.78 for UF houses). These ratios suggest that
monolithic-slab construction may give about 33% better radon resistance (lower radon by a
factor of 0.67) than SSW construction.
Similar conclusions about active-SSV systems are complicated by the large amount
of scatter in the data. Much of this scatter probably results from the variations in sub-slab
air flow and pressure-communication under different houses, and the varying ability to
ventilate the sub-slab region. Using the overall means of the measured ratios from Table 11
(for SSW houses), the active-SSV systems reduced radon to about 30% of the levels observed
with capped-SSV systems. Corresponding reductions in monolithic-slab houses (from data
in Table 12) were to about 66% of the capped-SSV levels. These data indicate that active-
SSV systems are much more effective in reducing indoor radon levels, and that they have
nominally equivalent radon resistance to that estimated in the initial lumped-parameter
study (Cnet/Csub=4xl0"4, Nie93a).
4-11
-------
5. CONCLUSIONS AND RECOMMENDATIONS
The analyses of NHEP radon data in this study have provided a more precise definition of the
effectiveness of radon-resistant construction features proposed under the draft Florida radon-resistant
building code (DCA91) than was provided by the initial lumped-parameter study (Nie93a), The analyses
also have indicated that the lumped-parameter model may represent very closely the actual radon-
resistance of houses built according to the proposed radon standard. Several important estimates of the
effectiveness of different construction features are provided by the analyses in this report. These
estimates are presented in terms of the indoor/sub-slab radon concentration ratio, Cm/Csub. This ratio
was proposed in the initial lumped-parameter study to normalize the different radon source strengths for
different houses to a common basis for comparison.
Analyses of the NHEP measurements made during FY-92 (Naj93, Tys93) lead to the following
conclusions about radon-resistance effectiveness:
• SSW construction, complying with the proposed radon-resistant construction code,
reduces indoor radon to about 9X10"1 of the sub-slab concentration (with an
uncertainty of a factor of 2.2).
• Capping the SSV system does not significantly alter its radon-resistance effectiveness
compared to leaving it in a passive mode.
• Monolithic slab and stem wall construction may improve radon resistance by
approximately 33% (reducing indoor radon levels by a factor of 0.67) compared to
SSW construction.
• Activation of SSV systems with exhaust fans may improve radon resistance by
approximately 70% (reducing indoor radon levels by a factor of 0.3).
Comparisons of the present CMl/Cwb ratios (as in Figure 2) with those from the initial lumped-
parameter study (Nie93a) suggest that the radon resistance of the present capped-SSV and passive-SSV
houses (both SSW and monolithic) is considerably better than was estimated earlier. This may result
from better refinement and understanding of the proposed radon-resistant construction features, i
5-1
-------
It also may result from closer monitoring and inspection to implement the features correctly. In any
event, the construction methods used for the present set of NHEP houses provide an inherent radon
barrier that reduces indoor radon levels to approximately iff3 of the levels that occur in the sub-slab
region.
Although the present analyses give a relatively precise indication of radon resistance (compared
to the uncertainties in the radon measurements), further analyses of other houses are needed to identify
the confidence that can be placed in the present conclusions. The NHEP data being collected in FY-93
will improve the basis for these conclusions. These data should be analyzed with the methods employed
here to examine consistency and variability of other data sets. Furthermore, several of the houses
reported in this study have subsequently had their SSV systems activated by suction fans. Further
analyses of the subsequent data on the same houses will better identify the effectiveness of active-SSV
systems. The present data without these systems activated will serve as a good control to estimate the
relative effectiveness of the active-SSV systems.
5-2
-------
6. LITERATURE REFERENCES
Aci89 American Concrete Institute, Standard Practice for Selecting Properties for Normal,
Heavyweight, and Mass Concrete, Detroit, MI: American Concrete Institute standard
AC! 211.1-89, 1989,
Cra75 Crank, J., The Mathematics of Diffusion, 2nd edition. New York; Oxford University
Press, 1975,
Dca91 Department of Community Affairs, Florida Standard for Radon-Resistant Building
Construction, Tallahassee, FL; State of Florida, Department of Community Affairs,
draft standard, October 1991.
EPA92a Environmental Protection Agency, National Residential Radon Survey Summary
Report, Washington D.C.: EPA-402/R-92-011 (NTIS unassigned), October 1992.
EPA92b Environmental Protection Agency, Technical Support Document for the 1992 Citizen's
Guide to Radon. Washington D.C.: U.S. Environmental Protection Agency, Office of
Radiation Programs, report EPA-400/R-92-011(NTIS PB92-218395), May 1992.
Fic55 Fick, A., Ueber Diffusion. Annalen Physik 170: 59-86, 1855 (in German).
Naj93 Najafi, F.T., Lalwani, L, Peng, C., Shehata, H., Shanker, A., Meeske, M., Roessler,
C.E., Noble, J.W., and Hintcnlang, D.E., New House Evaluation of Potential Building
Design and Construction for the Control of Radon in Marion and Alachua Counties,
Florida, Gainesville, FL: University of Florida Final report for Contract 92RD-66-13-
00-22-008, June, 1993.
Ner88 Nero, A.V., Radon and Its Decay Products in Indoor Air: An Overview. In: Radon and
Its Decay Products in Indoor Air, Nazaroff, W.W. and Nero, A.V., New York: Wiley
& Sons, p. 1-53, 1988.
Nie82 Nielson, K.K., Rich, D.C., and Rogers, V.C., Comparison of Radon Diffusion
Coefficients Measured by Transient-Diffusion and Steady-State Laboratory Methods,
Washington, D.C.: U.S. Nuclear Regulatory Commission report NUREG/CR-2875,
1982.
Nie84 Nielson, K.K., Rogers, V.C., and Gee, G.W., Diffusion of Radon Through Soils: A
Pore Distribution Model, Soil Science Society of America Journal 48: 482-487, 1984.
6-1
-------
Nie89 Niclson, K.K., Bollcnbachcr, M.K., Rogers, V.C., and Woodruff, G., Users Guide
for the MK-II Radon/Permeability Sampler, Salt Lake City, UT: Rogers & Associates
Engineering Corp, draft report RAE-8829/8-1, 1989.
Nie91 Nielson, K.K., Rogers, V.C., and Rogers, V,, Modeling Radon Generation, Transport,
and Indoor Entry for Building Construction Standards, Salt Lake City, UT: Rogers &
Associates Engineering Corp, report RAE-9127/3-2, 1991.
Nie93a Nielson, K.K., Rogers, V.C., and Holt, R.B., Development of a Lumped-Parameter
Model of Indoor Radon Concentrations ,EPA-600/R-94-201 (NTIS PB95-
142048), November 1994.
Nic93b Nielson, K.K., Rogers, V.C., Rogers, V., and Holt, R.B., The RAETRAD Model of
Radon Gas Generation, Transport, and Indoor Entry, EPA-600/R-94-198 (NTIS PB95-
142030), November 1994.
Nie93c Nielson, K.K. and Rogers, V.C,, Feasibility of Characterizing Concealed Openings in
the House-Soil Interface for Modeling Radon Gas Entry, Salt Lake City, UT: Rogers
& Associates Engineering Corp. report RAE-9226/1-8, August 1993.
Nie94 Nielson, K.K., Rogers, V.C., Rogers, V., and Holt, R.B., The RAETRAD Model of
Radon Generation and Transport from Soils into Slab-on-grade Houses, Health Physics,
67, 363-377, 1994.
Rog89 Rogers, V.C., Nielson, K.K., and Merrell, G.B., Radon Generation, Adsorption,
Absorption, and Transport in Porous Media, U.S. Department of Energy report
DOE/ER/60664-1, 1989.
Rog91a Rogers, V.C. and Nielson, K.K., Multiphase Radon Generation and Transport in
Porous Materials, Health Physics 60: 807-815, 1991.
Rog91b Rogers, V.C. and Nielson, K.K., Correlations for Predicting Air Permeabilities and
222Rn Diffusion Coefficients of Soils, Health Physics61 *.225-230, 1991,
Rog94 Rogers, V.C., Nielson, K.K., Lehto, M.A., and Holt, R.B., Radon Generation and
Transport Through Concrete Foundations, EPA-600/R-94-175 (NTIS PB95-101218),
September 1994.
6-2
-------
San90 Sanchez, D.C., Dixon,R., and Williamson, A.D., The Florida Radon Research
Program: Systematic Development of a Basis for Statewide Standards, In:
Proceedings: The 1990 International Symposium on Radon and Radon
Reduction Technology, Vol. 3, EPA-600/9-91 -026c (NTIS PB9I-234468),
July 1991.
She87 Sherman, M.H., Estimation of Infiltration from Leakage and Climate Indicators,
Energy and Buildings 10, 81-86, 1987.
Tys93 Tyson, J.L. and Withers, C.R., Demonstration of Radon Resistant Construction
Techniques, Phase II, Cape Canaveral, FL: Florida Solar Energy Center report
FSE C- CR-608-93, 1993.
Wil91 Williamson, A.D. and Finkel, J.M., Standard Measurement Protocols, Florida
Radon Research Program, EPA-600/8-91-212 (NTIS PB92-115294),November,
1991.
Yua81 Yuan, Y.C. and Roberts, C.J., Numerical Investigation of Radon Transport
Through a Porous Medium, Transactions of the American Nuclear Society
38:108-110, 1981.
6-3
------- |