United States
Environmental Protection
Agency
National Risk Management
Research Laboratory
Research Triangle Park, NC 27711
Research and Development
EPA/600/SR-96/107 December 1996
Project Summary
Contributions of Building
Materials to Indoor Radon
Levels in Florida Buildings
Kirk K. Nielson, Rodger B. Holt, and Vern C. Rogers
The Florida Standard for Radon-Re-
sistant Residential Building Construc-
tion originally contained a provision to
limit the concentration of radium in con-
crete. The provision was designed to
prevent concrete from causing elevated
indoor radon concentrations. It was re-
moved from the October 1994 version
of the standard, however, because con-
crete from commercial sources had not
been shown to be a major radon con-
tributor in Florida. This report docu-
ments subsequent work to character-
ize potential radon sources in concretes
and recommend related changes to the
building materials radium standard.
A mathematical model is presented
to estimate the contributions of build-
ing materials to indoor radon levels.
The model computes radon flux from
concrete surfaces using typical Florida
concrete properties and multiplies the
flux by concrete surface areas to esti-
mate their contribution to indoor ra-
don. The model also accounts for build-
ing ventilation by outdoor air.
Radium distributions in Florida resi-
dential floor slabs had a geometric
mean of 1.3 pCig1 and a geometric
standard deviation (GSD) of 1.62. Ra-
don emanation coefficients for the slabs
averaged 0.10 + 0.04. Radium measure-
ments in concretes with potentially el-
evated radon sources had a similar geo-
metric mean of 1.4 pCi g~1, but a much
greater GSD of 3.0, owing to occasional
elevated-radium samples. Radon ema-
nation coefficients for these samples
were also higher and more variable,
averaging 0.14 + 0.07. Radium and ra-
don emanation in aggregate materials
similarly showed occasionally elevated
radium concentrations.
A concrete and block building in Lake
City, FL, was found to have elevated
concrete radium levels and elevated in-
door radon. Gamma ray surveys sug-
gested elevated radium levels, and sub-
sequent concrete analyses showed 33
pCi g~1 radium in the ceiling slab. In-
door radon concentrations averaged 5.0
+ 0.8 pCi L~1, and radon source calcula-
tions suggested a ventilation rate of
0.43 Ir1 during the elevated radon pe-
riod. The radon source calculations
suggested that approximately 93% of
the radon came from the ceiling slab,
while only 3% came from the floor slab
and block walls. The remaining 4% of
the radon was estimated to have dif-
fused through the floor slab from foun-
dation soils. The calculated radon
source strengths were also consistent
with a gamma ray trend identified from
published data.
A revised building material radium
standard was developed to account for
the areas and radium concentrations
of concretes exposed to building inte-
riors. The standard would limit the in-
door radon increment from building ma-
terials to 2 pCi L~1. It would limit con-
crete radium concentrations to 7 to 9
pCi g~1 if only a single slab or walls
contain elevated radium. However, it
could limit radium to approximately 3
pCi g~1 if floor, ceiling, and walls all
have elevated radium.
This Project Summary was developed
by EPA's National Risk Management
Research Laboratory's Air Pollution
Prevention and Control Division, Re-
search Triangle Park, NC, to announce
key findings of the research project
that is fully documented in a separate
-------
report of the same title (see Project
Report ordering information at back).
Introduction
Radon (222Rn) gas enters buildings pri-
marily from radium (226Ra) in foundation
soils. However, significant radon contribu-
tions can also come from building materi-
als if they contain elevated radium con-
centrations. If the total radon entry rate is
elevated and the building is not well venti-
lated, radon can accumulate to levels that
can significantly increase the occupants'
risks of lung cancer with chronic expo-
sure. The U.S. Environmental Protection
Agency (EPA) attributes 7,000 to 30,000
lung cancer fatalities annually to radon,
and recommends remedial action if indoor
radon levels average 4 picocuries per liter
(pCi L"1) or higher.
The Florida Department of Community
Affairs (DCA), under the Florida Radon
Research Program (FRRP), has devel-
oped radon-protective building standards.
These standards are incorporated in pro-
posed rule 9B-52, the Florida Standard
for Radon-Resistant Residential Building
Construction, which is primarily aimed at
controlling radon by blocking its entry from
foundation soils.
An initial criterion was developed under
the FRRP to limit radon sources in build-
ing materials. The criterion was included
in early drafts of the Florida Standard for
Radon-Resistant Residential Building Con-
struction, requiring that materials used in
concrete for habitable structures have less
than 10 pCi g"1 of radium. The criterion
was removed from the October 1994 ver-
sion of the standard after comments from
the Florida Concrete and Products Asso-
ciation indicated that the criterion was un-
necessary because (a) concrete from com-
mercial sources had not been shown to
be a major radon contributor in Florida;
(b) testing and related cost impacts were
not defined; and (c) it appeared that con-
crete was singled out without considering
drywall, lumber, carpets, insulation, and
other materials. Related comments from
FRRP scientists suggested that inclusion
of the radium criterion would encourage
suppliers to use higher-radium materials
because it was allowed, and that the pro-
posed criterion was three to five times
higher than would be expected for a uni-
form material exposed to the indoor envi-
ronment.
The full report presents the findings of a
subsequent task initiated by DCA under
the FRRP to address the first objection to
the concrete radium criterion, that con-
crete from commercial sources had not
been shown to be a major radon contribu-
tor in Florida. The objective of the task
was to identify buildings in Florida whose
source of indoor radon was suspected to
be building materials. The cause of the
problem was also to be examined, and
recommendations were solicited for re-
lated changes to the standard. Further
study of concrete as a radon source was
justified by FRRP scientists, who recog-
nized the potential of concrete to signifi-
cantly contribute to indoor radon, while
the potentials for drywall, lumber, carpets,
insulation, and other materials to contrib-
ute to indoor radon were judged to be ten
to hundreds of times lower, based on lit-
erature surveys. Therefore, this study fo-
cused on concrete and concrete products
(block).
Theoretical Effects of Concrete
Radon Sources
Radon generated by concrete or other
building materials cannot be distinguished
from soil-generated radon once it has en-
tered a structure and mixed with indoor
air. Radon from concrete therefore must
be measured directly as a flux exiting a
slab or wall surface to characterize it sepa-
rately from other sources. Although radon
fluxes from building materials have been
measured in several studies, the proce-
dures are generally difficult and expen-
sive, making alternative approaches such
as modeling preferable whenever possible.
A simple modeling approach was there-
fore used to estimate indoor radon contri-
butions from concrete and other building
material sources.
Indoor radon concentrations reflect a
balance between the rate of radon entry
into a structure and the rate of radon loss
by decay and dilution by ventilating air.
The rate of radon entry is the sum of
radon coming from foundation soils, build-
ing materials, and in unusual cases, wa-
ter supplies, natural gas combustion, and
other potential sources. Radon loss rates
are invariably dominated by the building
ventilation rate, which is commonly ex-
pressed in air changes per hour (ach or
h"1). The simple expression for indoor ra-
don concentration under these conditions
is:
EJ. .A
i i
C =C. -C =•
3,600
(1)
C!n = measured indoor radon
concentration (pCi L1)
Cout = outdoor radon concentration in
ventilating air (pCi L1)
J = radon flux from surface / (pCi
nr2 s-1)
A = area of radon-emitting surface
/(m2)
V = interior volume of the struc-
ture (L)
X = rate of ventilation by outdoor
air(h'1)
3,600 = unit conversion (s h"1)
A, = rate of radon decay (2.1 x 10'6
s-1).
The expression for indoor radon con-
centration can be simplified even further
by neglecting the Cou( and XRn terms. Cou(
seldom approaches the 4-pCi L1 level at
which Cln becomes a concern, so Cout is
often ignored. Similarly, XRn is only 2.1 x
10'6 s-1, which is less than 8% of the loss
rate even for tight buildings (0.1 ach).
With these simplifications, Equation 1 can
be rearranged by grouping X with Cn (here-
after called C) to isolate the most variable
building properties from the more con-
stant ones, giving the expression:
,600
(2)
The radon flux for a concrete surface
can be calculated from the radium con-
centration, density, emanation coefficient,
diffusion coefficient, and thickness of the
concrete as:
where:
104
R
P
£
D
x
unit conversion (cm2 rrr2)
concrete radium concentration
(pCi g-1)
concrete bulk dry density
(g cnr3)
concrete radon emanation co-
efficient (dimensionless frac-
tion)
radon diffusion coefficient for
the concrete (cm2 s'1)
concrete thickness (cm).
where:
= net indoor radon from non-
airborne sources (pCi L1)
Using the simplified relationship in Equa-
tion 2, published radon concentrations cal-
culated for building materials in houses
and large buildings were compared with
corresponding calculations of gamma ray
intensity. The CX grouping from Equation
-------
2 was used to obtain a lumped parameter
that is less subject to time and variations
caused by changes in building ventilation
rate. The radon source strengths (Ck) were
plotted versus gamma ray activity to ob-
tain the following relationship by least-
squares linear regression:
"Cl = 0.01277-0.081 <4)
where y = gamma ray activity (\iR rr1).
The empirical correlation of radon source
strength with indoor gamma ray intensity
in Equation 4 could potentially offer a
simple, inexpensive test for radon sources
in building materials. However, actual
gamma ray measurements are subject to
potential biases from natural background
gamma activity, 232Th and 40K gamma ac-
tivity from the building materials, and
source-measurement geometry biases.
The effects of background gamma activity
should be avoidable by simply subtracting
an appropriate background value from the
indoor measurements. Contributions from
232Th activity are often small and predict-
able, since thorium in common Florida
earthen materials seldom exceeds 1 pCi
g~1. Even where exceptions lead to el-
evated gamma ray measurements, the
exceptions would be conservative. Similar
contributions from 40K would be even
smaller and less frequent. Possible bi-
ases from different source-measurement
geometries could generally be made con-
servative by utilizing maximum readings
where the gamma distribution is nonuni-
form. Sampling and laboratory analysis
could then be used only where a confir-
matory measurement is required.
Radium and Radon Emanation
Measurements
A review of radium and radon emana-
tion measurements in Florida concretes
gives insight into their typical radon source
properties. Radium concentrations in con-
crete floor slabs from Florida houses were
directly measured in two previous FRRP
studies, one dealing with new houses and
the other with older houses. Additional
concrete analyses were performed in con-
nection with anomaly investigations for the
statewide mapping study, and in connec-
tion with this study. Together, the con-
crete analyses give an approximate char-
acterization of the range of radium con-
centrations and radon emanation coeffi-
cients in Florida residential concretes. Ad-
ditional data on rock aggregate materials
are also summarized here from separate
FRRP measurements as a possible ex-
planation of the radium distributions ob-
served in Florida concretes.
Floor slabs. In the two previous studies
of Florida residential floor slabs, samples
were obtained from cores drilled from the
floor slabs. The structures were chosen to
represent typical single-family dwellings
without regard to indoor radon levels; in
fact, indoor radon data were not available
for these houses. The data from the first
study showed a geometric mean radium
concentration of 1.4 pCi g"1 and a geomet-
ric standard deviation (GSD) of 1.38, while
the data from the second study showed a
geometric mean radium concentration of
1.3 pCi g-1 and a GSD of 1.76. Although
the variations are larger among the older
homes, the means are not significantly
different, and both sets are represented
here by a single distribution for the 19
slabs with a geometric mean of 1.3 pCi g~1
and a GSD of 1.62. Radon emanation
averaged 0.069 + 0.008 in the first study
and 0.116 + 0.042 in the second study,
with an overall average of 0.101 + 0.041
for all 18 slabs. The measured radium
concentrations are 40% to 80% higher
than typical U.S. or worldwide concrete
radium levels, while the radon emanation
coefficients are slightly lower than previ-
ously reported values.
Concrete components. Further insight
was sought on radium and radon emana-
tion distributions in Florida concretes from
analyses of dry-mix concrete materials
sampled from four diverse Florida loca-
tions. Portions of these samples were
separated by sieving to isolate the aggre-
gate, sand, and cement fractions so that
each fraction could be analyzed separately.
Additionally, bulk analyses were performed
on concretes prepared from the dry mixes.
The geometric mean radium concentra-
tion for concretes mixed from the four
samples was 0.6 pCi g-1 (GSD=2.3), nearly
identical to the geometric mean of 0.5
pCi g-1 (GSD=2.2) among the mass-
weighted component means. Interestingly,
the geometric mean radium in the cement
components was highest (1.2 pCi g"1,
GSD=1.4), followed by the highly variable
aggregate radium concentrations (0.5
pCi g"1, GSD=4.1) and the uniformly low
sand radium concentrations (0.1 pCi g"1,
GSD=1.4). Although the average dry-mix
radium concentration is only about half
the average for the 19 slabs, both distri-
butions are so variable that this difference
is not statistically significant.
The average radon emanation coeffi-
cient for concretes mixed from the four
samples was 0.19 + 0.14, nearly identical
to the 0.18 + 0.09 average of the mass-
weighted component means that utilized
the moist-paste cement emanation coeffi-
cients. The average emanation for the
moist cement paste (0.31 + 0.06) was
much greater than for the dry cement pow-
der (0.02 + 0.01); however, the average
18% composition of cement in the con-
cretes minimizes the effect of this mois-
ture dependence in the mass-weighted
means. The average emanation of the
sand was lower (0.14 + 0.05), and that for
the aggregate was lower yet (0.07 + 0.07).
The average emanation coefficient for the
dry-mix concretes is nearly 90% higher
than the average for the slab measure-
ments, probably because of higher mois-
ture in the dry-mix samples.
Other concretes. Additional concrete
analyses were performed in connection
with the radon map anomaly investiga-
tions and with this study. The samples for
these analyses were obtained from vari-
ous locations throughout Florida by com-
mercial concrete suppliers, radon
mitigators, and Rogers & Associates En-
gineering Corp. (RAE) personnel. The
samples represented both single-family
dwellings and multistory apartment build-
ings. Although most samples consisted of
cores drilled from floor slabs, some were
also taken from foundation footings, poured
concrete walls, and concrete blocks.
The map-related analyses may be less
representative of all Florida concretes be-
cause the samples were sought from build-
ings with potentially elevated indoor radon
(>4 pCi L"1). However, their radium con-
centrations were only slightly higher (1.4
pCi g~1 compared to 1.3 pCi g-1) even
though they were much more variable than
the previous analyses (GSD of 3.0 com-
pared to 1.6). Their radon emanation co-
efficients were also somewhat higher (0.14
+ 0.07 compared to 0.10 + 0.04). Although
the map-related radon sources (the prod-
uct of radium concentration and radon
emanation coefficient) are expectedly
higher, they are not high enough to sug-
gest a consistent correlation of building
materials with indoor radon. The compari-
sons are more consistent with the usual
trend of indoor radon concentrations that
are dominated by foundation soils rather
than by building materials. However, oc-
casional cases may be dominated or af-
fected by building materials.
Aggregates. A brief survey of con-
crete aggregate materials was conducted
because aggregate is the least-character-
ized major concrete constituent. The sur-
vey of concrete aggregate materials in-
volved collecting and analyzing samples
from sources throughout Florida. The
samples were collected opportunistically
during various field investigations and map
-------
validation studies. They consisted of ag-
gregate materials from active quarries, rock
samples from U.S. Geological Survey in-
vestigations in Dade and Broward Coun-
ties, and road aggregate samples from
various sites.
Radium measured in five samples from
commercial gravel quarries was distributed
most narrowly, ranging from 1.7 to 5.1 pCi g-1,
and having a geometric mean of 2.7 pCi g~1
and a GSD of 1.7. These samples may over-
estimate the typical radium concentration in
Florida aggregates, since they would lead to
slightly higher concrete radium concentrations
than measured in residential slabs. The ag-
gregate analyses also fall into the upper range
of the radium distribution measured for Florida
soils (geometric mean = 0.6 pCi g"1; GSD =
3.5). Radium in 21 "potential aggregate" rock
samples ranged from <0.2 to 11.3 pCig"1,
and had a lower geometric mean of 1.4
pCi g-1, but a higher GSD of 2.8. Radium in
five road aggregate samples ranged from
0.7 to 57 pCi g"1, with a geometric mean of
13 pCi g-1 and a GSD of 13.2. The overall
geometric mean of the 34 radium measure-
ments in aggregates is 2.1 pCi g-1, and its
GSD is 4. Although the rock materials may
overestimate typical radium concentrations
in Florida concrete aggregates, they show
a potential for elevated radium concentra-
tions in concretes.
Radon emanation coefficients for the
gravels from active quarries averaged 0.05
+ 0.03, significantly less than the 0.35 +
0.23 for the potential aggregate rocks and
the 0.16 + 0.12 for the road aggregate
samples. These differences are probably
dominated by differences in ambient mois-
ture levels, since the emanation measure-
ments were conducted at ambient mois-
ture. Surface samples from gravel piles
were dry, while the "potential aggregate"
rock samples were collected at significant
depths below the soil surface. Road ag-
gregates probably had intermediate mois-
ture, since they were in contact with shal-
low soils, but were mixed with or covered
by asphalt materials. In general, the po-
tential and road aggregate samples sug-
gest emanation coefficients comparable
to the "wet paste" values unless materials
are completely dry.
Association of Concrete
Radium with Indoor Radon
Several of the radium and radon ema-
nation measurements are high enough to
associate with elevated indoor radon con-
centrations using the equations presented
here. However, this study also sought to
determine if actual Florida buildings could
be found in which elevated indoor radon
levels are caused by building materials.
This objective required measurement of
indoor radon in buildings that have el-
evated radium levels in their building ma-
terials. Measurement opportunities were
sought in buildings where elevated con-
crete radium levels had already been mea-
sured. However, access to these build-
ings was limited because the concrete
samples were mostly provided by con-
crete suppliers or construction workers who
could not also provide access for indoor
sampling of the completed buildings.
Therefore, only one building was studied
in sufficient detail to show a link between
its concrete radium level and the indoor
radon concentration.
Empirical Measurements. The study
building was located at 30.179° N latitude
and 82.692° W longitude, in the vicinity of
Lake City, FL, which is entirely within a
green (low radon potential) area of the
Florida radon protection map. The build-
ing was a two-story structure with a con-
crete floor slab, concrete block walls, and
a 20-cm concrete slab separating the first
and second stories. The building was ini-
tially identified by gamma ray surveys,
which showed gamma ray intensities ex-
ceeding 60 |iR h'1 in some locations.
Gamma ray surveys in the vicinity of the
building showed no elevated soil radium
sources, with typical soil gamma intensi-
ties in the 2- to 4-|iR rr1 range. Radon flux
measurements from the bare surfaces of
surrounding soils averaged 0.2 + 0.1
pCi rrr2 s~1, also indicating that the site soils
should not contribute to elevated indoor
radon concentrations.
A detailed gamma ray survey was con-
ducted in the accessible first-floor portion
of the building. The gamma activity near
the floor was consistently lower than cor-
responding measurements at the ceiling
of the first level. The floor measurements
averaged 25.9 + 3.2 jaR rr1, while the ceil-
ing measurements averaged 50.7 + 4.2
|iR h"1. Gamma measurements along the
block walls were intermediate, while
gamma activity at a single accessible lo-
cation on the floor of the second level was
slightly higher than the measurements from
the ceiling of the first level. Because of
the relative uniformity of the gamma ray
distributions over the survey area, it ap-
peared that the concretes were causing
the elevated gamma activity.
Sampling within the building consisted
of triplicate radon flux measurements from
the floor slab, single concrete samples
from the floor slab and the ceiling slab,
and indoor radon measurements in the
first level of the building. The radon flux
measurements utilized the small charcoal
canister method described and used pre-
viously for the statewide radon flux sam-
pling. The concrete samples were obtained
by drilling several 1.6-cm-diameter, 5-cm-
deep holes in the slabs and collecting the
drill cuttings on plastic sheets for analysis.
The concrete cuttings were analyzed by
the same gamma assay procedure used
previously for soil samples. Indoor radon
measurements utilized a continuous ra-
don monitor that circulated approximately
2 L min"1 of room air through its scintilla-
tion cell while continually recording alpha
activity over 20 min intervals. Radon con-
centrations were computed from the con-
tinuously measured alpha counts using
the calibration method and equations of
Thomas and Countess.
The radon flux measurements from the
floor slab averaged 0.083 + 0.049
pCi rrr2s-1, typical of the range expected
from ordinary diffusion of radon through a
slab from underlying soils. The concrete
radium concentrations were more surpris-
ing, however, indicating 0.6 + 0.4 pCi g"1
of radium in the floor slab and 32.8 + 1.7
pCi g-1 in the ceiling slab. Based on these
assays, most of the gamma activity at the
floor surface was hypothesized to come
from the ceiling. The intermediate values
along the walls are consistent with this
gamma shine interpretation, suggesting
that any radium activity in the concrete
block walls is too low to significantly affect
the gamma measurements.
The indoor radon concentrations in-
creased at an initial rate of approximately
0.24 pCil_-1h-1 during the first 10 h of
measurements. They reached the 3 to 4-
pCi L1 range, and then decreased during
a period when outdoor gusty winds were
observed. The outside door was briefly
opened four times during the measure-
ment period for entry or exit of personnel.
The increased ventilation from door open-
ings may also have contributed to de-
clines observed during the 10- to 16-h
and 22- to 26-h periods.
Radon concentrations increased at a
higher rate of about 1.2 pCi L1 Ir1 during
the period from 18 to 22 h. They reached
the 4 to 6-pCi L1 range and then de-
creased to levels that were mostly below
4 pCi L1. The measurements demonstrate
that the building had sufficient radon po-
tential to exceed 4 pCi L1 for sustained
periods of several hours when perturbing
effects such as winds or mechanical open-
ings were not increasing its natural venti-
lation rate. For calculation purposes, the
indoor radon concentration was estimated
from an average of 13 points during the
19 to 23-h period to be 5.0 + 0.8 pCi L1.
Calculated Effects. The contributions of
various building materials in the study
-------
building were calculated using Equations
1-4. Table 1 shows the results of these
calculations. Radon fluxes from the ceil-
ing slab were calculated from its
32.8-pCi g"1 radium concentration using
Equation 3, assuming typical density, ema-
nation, and diffusion properties for con-
cretes as measured in the previous stud-
ies. The indoor radon source resulting from
this flux was computed from Equation 2
using the 25.4-m2 slab area and 61.9-m3
volume of the study room. Contributions
from the block walls were estimated simi-
larly, assuming a radium concentration
equal to that of the floor slab, 0.6 pCi g-1.
The wall area used to calculate CX was
estimated to be 40.9 m2. The radon flux
and resulting source from radium in the
floor slab were calculated from the mea-
sured slab radium concentration in the
same way as the corresponding values
were calculated for the ceiling.
The flux of radon diffusing through the
floor slab from foundation soils was esti-
mated from the difference between the to-
tal measured floor flux and the portion that
was explained by radium in the slab. The
measured floor flux of 0.083 pCi rrr2 s~1 was
strongly dominated by underlying soils when
compared to the flux of 0.025 pCi rrr2 s"1
calculated to result from radium in the con-
crete. The soil contribution to the total ra-
don source strength was also estimated
using Equation 2. The last column in Table
1 shows the relative contributions of each
of the four components to the total indoor
radon concentration.
The indoor radon concentration ex-
pected from the calculations in this sec-
tion is equal to the total value of CK =
2.15 pCi L1 h'1 from Table 1 divided by
the ventilation rate of the room. Although
the ventilation rate was not directly mea-
sured, previous estimates of ventilation in
Florida residential structures have usually
been in the 0.25-lr1 to 0.50-lr1 range. This
range of ventilation rates corresponds to
a radon concentration range of 4.3 to 8.6
pCi L1 for the calculated radon source po-
tential. The measured concentration of
5.0 + 0.8 pCi L1 is within this range, and
corresponds to a ventilation rate of X =
0.43 h1. This ventilation rate is higher
than values estimated for many Florida
buildings, suggesting that the measured
radon source could potentially cause
higher indoor radon levels in a more tightly
sealed building. Ventilation rates as low
as 0.1 h"1 have been measured in Florida,
and rates as low as 0.04 h"1 have been
reported for unoccupied buildings when
ventilation systems were not operating.
The indoor radon source strength was
also estimated independently, using the em-
pirical relationship in Equation 4. The aver-
age gamma ray intensity of 50.7 |iR h"1
measured near the ceiling gives a radon
source estimate of 0.56 pCi L1 rr1, which is
within the measurement uncertainty of the
0.52-pCi L1 h'1 value estimated in Table 1.
The study building satisfies the objec-
tive of identifying a Florida building whose
source of indoor radon is suspected to be
from building materials. Based on the build-
ing material contributions demonstrated in
Table 1, the indoor radon is clearly domi-
nated by radium in the ceiling slab. The
long-term average radon concentration in
the study building remains unclear be-
cause of the short duration of the radon
measurements and the lack of information
on its average ventilation rate. However,
the short-term radon measurements and
ventilation estimates for Florida buildings
(X » 0.25-0.50 h'1) both suggest the po-
tential for long-term radon concentrations
exceeding 4 pCi L1. The consistency of
the calculated radon potential with that
estimated from the gamma ray correlation
in Equation 4 suggests a potential for
screening buildings for building-material
radon sources using gamma ray surveys.
Building Materials Radium
Standard
The present empirical measurements
and model analyses show that building
materials can and do contribute signifi-
cantly to indoor radon concentrations in
some instances. To protect the public
against unknowingly incorporating harm-
ful radon sources into building materials,
a standard is proposed for limiting radium
concentrations in the building materials.
The standard is based on the typical con-
crete properties used in the analyses in
Table 1, from which Equation 3 gives the
following relationship between concrete ra-
dium concentration (R in pCi g"1) and ra-
don flux (J in pCi irr2 s'1) for a 20-cm con-
crete wall:
= 0.041^.
(5)
Substituting Equation 5 into Equation 2
then gives a relationship that expresses
indoor radon concentration as a function
of concrete radium concentration, concrete
area, ventilation rate, and occupied vol-
ume. Assuming a ventilation rate of
X = 0.25h"1, as in previous modeling of
Florida residences, the resulting equation
can be simplified as:
where:
C
R.
A
V
indoor radon concentration
caused by concrete materials
(pCi L1)
concrete radium concentration
in slab /(pCi g"1)
area of interior concrete sur-
face / (m2)
interior occupied volume (L).
C =
600 ^
V
(6)
Equation 6 can be used to predict in-
door radon contributions from concrete
building materials under various construc-
tion scenarios. For example, a 140-m2
(1,500-ft2) residence could have 140 m2 of
floor slab area plus another 140 m2 of
ceiling slab area if it were part of a multi-
story building separated by concrete slabs.
In addition, concrete or block perimeter
walls could comprise an additional 115m2
of concrete area exposed to the occupied
space. If all of the concrete contained
background radium at the 0.5-pCi g~1 level,
the concrete would contribute a total of
only 0.35 pCi L1 to the indoor radon con-
centration. However, if the concrete con-
tained elevated radium concentrations, it
would cause higher radon levels, as shown
by the limiting radium concentrations in
Table 2. These concentrations are the cal-
culated limits for the total concrete to con-
tribute no more than 2 pCi L1 to the in-
door radon levels.
The standard proposed for limiting ra-
dium concentrations in building materials
is designed to permit no more than 2 pCi
L1 of indoor radon to be caused by the
building materials. The 2-pCi L1 limit is
purposely defined lower than the 4-pCi L1
standard to accommodate radon contribu-
tions from other sources, such as soil gas
from foundation soils. The proposed stan-
dard gives specific guidance for concrete
products, since concrete presently appears
to be the dominant building material con-
tributing to indoor radon. The standard is
also formulated to give credit for different
occupied volumes, for different concrete
surface areas, and for different radium
concentrations. The standard is based on
Equation 6, which is restated for clarity.
Radium concentrations specified by the
standard and by Equation 6 are intended
to be measured by protocols accepted by
the FRRP. The following standard is there-
fore proposed for avoiding elevated in-
door radon concentrations caused by ra-
dium in building materials:
Building materials used in the con-
struction of habitable structures shall
not contain quantities of radium that
increase the indoor radon concen-
tration by more than 2 pCi L1. The
-------
Table 1. Calculated Contributions of Building Materials to Radon in the Study Building
Radon
Source Material
Ceiling slab
Wall blocks
Floor slab
Foundation soil
Total
Radon Flux
(pd m-2sr1)
1.353 *
0.013"
0.025 *
0.058 c
CA
Radon Source
(pd L-1lr1
1.996
0.031
0.037
0.086
2.15
Contribution to
Indoor Radon
(%)
92.9
1.4
1.7
4.0
100.0
'Calculated from measured radium concentration, 10% emanation, 2.1 g cm3 density, and
0.001 cm2 s~' radon diffusion coefficient.
"Same as * but assuming 0.6 pC; g-1 radium.
"Difference between measured flux and floor flux calculated from measured radium.
Table 2. Limiting Concrete Radium Concentrations for Contributing 2 pCi L' of Radon to a 140-m2
Residence Using Equation 6
Concrete Structures with a
Background Radium
Concentration of 0.5 pCi g~1
Concrete Structures
with Elevated Radium
Concentrations
Limiting Elevated Radium
Concentration
2 Slabs
Walls + 1 slab'
Walls
None
Walls
1 Slab*
2 Slabs
2 Slabs + walls
8.6
7.2
3.8
2.9
'Either floor or ceiling slab.
contribution of concrete materials to-
ward the 2-pCi L1 limit shall be de-
fined as:
C=
600
v
where:
C = radon concentration from
concrete materials (pCi L1)
V = volume of the habitable space
(L)
Rf = radium concentration in the
floor slab(s) (pCi g-1)
Af = area of the concrete floor
slab(s) (m2)
Rc = radium concentration in the
ceiling slab(s) (pCi g~1)
Ac = area of the concrete ceiling
slab(s) (m2)
Rw = radium concentration in the
concrete walls (pCi g~1)
Aw = area of concrete walls facing
the interior volume (m2).
Radium concentrations used to com-
pute radon contributions shall be
measured in accordance with "Stan-
dard Measurement Protocols, Florida
Radon Research Program," or other
procedures accepted by the Depart-
ment.
-------
KirkK. Me/son, Rodger B. Holt, and Vern C. Rogers are with Rogers and Associates
Engineering Corp., Salt Lake City, UT 84110-0330.
David C. Sanchez is the EPA Project Officer (see below).
The complete report, entitled "Contributions of Building Materials to Indoor Radon
Levels in Florida Buildings,"(OrderNo. PB97-104681; Cost: $21.50, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air Pollution Prevention and Control Division
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268
Official Business
Penalty for Private Use $300
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT NO. G-35
EPA/600/SR-96/107
------- |