Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

MEASURING RADON AND THORON EMANATION FROM CONCRETE
AND GRANITE WITH CONTINUOUS RADON MONITORS AND EPERM’s®
Bill Brodhead
WPB Enterprises, Inc., 2844 Slifer Valley Rd., Riegelsville, PA USA
wmbrodhead@gmail.com www.wpb-radon.com

Abstract
The author investigated the use of commercially available continuous radon monitors (CRM’s)
and S-Chamber E-PERM’s® using short term electrets to measure the radon (222Rn) and thoron
(220Rn) emanation from concrete and granite counter tops. The performance of CRM’s and EPERM’s® placed in 3 to 23 liters metal accumulator chambers sealed to a building material were
compared to the total emanation rate of the building material when the material was placed in a
sealed 122 liter chamber. Concrete slabs were constructed that had radon only versus radon and
thoron and actinon(219Rn) to determine the test equipment response to these isotopes. A thoron
chamber was constructed to test the detectors response to thoron and the reduction in response to
thoron when the detectors were placed in diffusion barriers. Accumulator and sealed chamber
tests on three different granites found significant variation in emanation rates depending on what
side of the granite was tested.

Elevated Indoor Radon Levels due to Building Materials
The author investigated a 200 unit seven story tall condominium unit that had elevated radon
levels in the hallway and individual units on every floor. This building had two levels of
ventilated parking garage under most of the building that precluded ground base soil gas as the
source. The building was constructed with post stressed concrete floors and ceilings and
concrete support columns. All other walls except those adjacent to stairwells were metal stud
and drywall construction. Ventilation measurements indicated the units were typically getting
less than 0.1 air changes per hour (ACH). A simple method was used to measure the concrete
emanation rate by placing single EPERM’s® inside 3 liter stainless mixing bowls that were
sealed against exposed concrete floors, ceilings and walls. The total ingrowth inside the bowl
was determined by doubling the E-PERM® average after first subtracting the back ground radon
when the unit was first sealed. This total ingrowth is then divided by the hours of exposure and
multiplied times the volume of the accumulator and divided by the square feet of exposed
concrete to obtain the emanation rate of the concrete. The exposures were approximately 24
hours long to minimize the effect of ingrowth decay. These emanation measurements made on
every floor of the condominium indicated that the concrete along with the low ventilation rate
was the likely source of the elevated radon. The research presented in this paper was conducted
to determine if an accumulator method using continuous radon monitors (CRM’s) or E-

1

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

PERM’s® with ingrowth correction factors would provide a simple method with reasonable
accuracy to determine the emanation rate of any building material.

Testing Equipment Set Up
The author has two AB5 Pylons® with passive radon detector heads (PRD) and a RAD7® radon
monitor. These units were cross compared with two similar AB5 Pylons with newer CPRD
heads supplied by Pennsylvania DEP Radon Division. In addition the following CRM’s were
generously loaned from the manufacturer/suppliers; two Femto-Tech 510’s®, Sun Nuclear
1029®, Rad Elec Scout®, RadonAway RS500® , RadonAway RS800® and the RTCA On Guard.
Metal test chambers were constructed varying in size from 38 liters to 129 liter size by using
commercially available metal trash cans with removable lids. Each of the cans had all interior
seems sealed with urethane caulking and then covered with 17 mil aluminum tape. A power
cord was installed in each chamber with the cord penetration carefully sealed in a similar
manner. Sampling ports were installed in each of the chambers by mechanically attaching 3/8”
ball valves through the side of each chamber and carefully sealing the penetration. See picture in
Figure 10. The removable lid for each chamber had pliable plumbers putty placed around the
edge. A ball valve would be left open and the lid pressed down on the chamber, compressing the
putty and forming an air tight chamber when the ball valve was closed. The tightness of the
chamber was tested by flowing radon into a chamber with two CRMs and then closing all the
valves to allow the radon to decay. The radon decayed with a normal radon decay rate indicating
that there was no radon leakage out of the chamber. See the decay rate chart below in Figure 1.

Figure 1 Chamber Tightness test

2

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

The author has several radon, thoron and actinon sources that were used to test the performance
of the different CRM’s and E-PERMs®. One of the sources is soil. During an investigation of a
home in need of a radon mitigation system, a suction point was located where a 300 υR/hr
gamma reading was obtained at the slab. The soil excavated from this home produces 0.75
pCi/oz/minute (1.6 Bq/gm/hour). This soil was dried and placed in three 6” by 60” long metal
ducts that were carefully sealed and constructed with sampling ports on either end. Initial sniff
measurements made of the soil indicated it had a low thoron content. Some of this soil was
mixed into a concrete test slab to increase its emanation rate. A more careful grab sample of the
soil source was then made with a Pylon AB5® and scintillation cell with the counter set to 20
second count interval. The tubing length from the soil source to the cell was less than a foot long
and the air flow was set at 4 lpm flow with a 0.8 µm filter. See the graph in Figure 2 below. The
last 20 second cell count while sampling was 661 counts. As soon as the pump was turned off
the next 20 second count fell to 328. The next 20 second count dropped to 117. The counts then
fell off more gradually, dropping to 91 and then a minute latter to 70 and then a minute or two
later to around 55. The sample was aged and counted latter indicating about 140 pCi/l (5,180
Bq/m3) of radon at 4 lpm flow rate. This extreme drop in counts indicates the soil is producing
significantly more actinon which has a 4 second half life than radon. The decay also indicates
there is some thoron in the soil but it is difficult to measure because of the very high level of
actinon in the soil.

Figure 2 Soil Source Checked for Thoron & Actinon

3

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

Once it was determined that there was significant thoron and actinon being produced by the soil
source, a 75 liter decay chamber was constructed that delayed the airflow into the final test
chamber by 15 minutes (75 liters / 5 lpm flow) to ensure that both thoron and actinon were
decayed out. A 47 millimeter 1 µm filter was installed inline before the final test chamber to
collect most decay products produced by the actinon, thoron or radon. 1 to 5 lpm of air was
pushed through the soil sources using a small aquarium pump. Dwyer flow gauges were
installed before and after the test chamber to monitor the flow rate and ensure there was no
leakage out of the chamber. The chamber exhaust air is vented to the outside. A typical flow
through chamber set up is illustrated in Figure 3 below. Note that the test chamber had internal
power outlets and a small mixing fan to create uniform levels inside the chamber.
Each tube produces
125
pC i at 4 lpm flow
WPB Test
C hamber
Exhaust vent to
outside

6 " metal ducts
w ith
48 " of Soil

1- 5 LPM
airflow
Sealable 117 liter
metal chamber

75 Liter chamber
creates at least 15
minute delay

4
3

2

1

Soil

CRM
Ball Valve
sampling
ports for grab
samples
or R AD 7
measurements

CRM

Valved
inlet

CRM
CRM
4

CRM

3
2

1

Low flow mixing
fan inside
chamber

1 um
filter

D w yer
airflow
meter

Gravel
bed

Aquarium
pump
1 -5 LPM flow

Figure 3 Soil Source Test Chamber

Ingrowth Comparison
The CRM’s and E-PERM’s response to ingrowth of radon was tested by sealing the detectors in
a 122 liter chamber with a small radium source placed at the inlet to the interior mixing fan. The
volume of the fans, CRM’s and chargers placed in the chamber was subtracted from the empty
chamber volume. The CRM volume was determined by measuring the components of each unit
rather than the outside dimensions to factor in the free air area inside the CRM’s. Table 1 below

4

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

gives a table of the volume size used for each CRM. Note that the CRM volume was not based
on the external CRM dimensions but the approximate mass volume of all the components of the
CRM. The source of the radon is an antique toy that was manufactured in the 1920’s by the
same company in Pittsburgh that produced the gram of radium that was gifted to Marie Curie.
The source produces no measurable 220 thoron. The radon levels would then ingrow depending
upon the open volume of the chamber and the length of time the chamber was left sealed.
RS 800
0.76 liters

Scout
0.56 liters

EPerms
0.123 liters

SN 1029
0.66 liters

Pylon AB5 PRD
2.5 liters

RS 500
0.70 liters

Femto 510
0.61 liters

RTCA OnGuard
0.81 liters

Table 1 Detector volumes

A comparison of CRM measurements to an ingrowth of radon in a sealed chamber in Figure 5
below indicated the RAD7 which pumps air into its chamber every 5 minutes may have been
responding to the ingrowth of radon more quickly than the other CRMs. All testing was done
with the CRMs set to hourly intervals. A delayed response to an increasing radon concentration
would reduce their calculated ingrowth.

Figure 4 CRM In-growth versus Grab sample measurements

5

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

A second test was performed using a dozen Pylon 300A scintillation cells that were first
measured for background count and then filled with a known radon concentration and then
counted at least four hours later in order to calibrate their individual efficiency. Two Pylon
AB5’s and other CRM’s were then placed in a sealed chamber with 109 liters of free air and a
radium source. The RAD7 was not available for this second test. See the results graphed in
Figure 4 above.
There were five small mixing fans inside the chamber during the 36 hour exposure. Single grab
samples using the calibrated scintillation cells were taken every 2 to 4 hours during the exposure.
The plotted ingrowth and the mathematical calculated ingrowth is shown in Figure 4 above along
with the Pylon response and its mathematical ingrowth line. Note that the grab sample ingrowth
of 510 pCi/hr (18,870 Bq/hr) is 16% greater ingrowth than the Pylon ingrowth of 440 pCi/hr
(16,280 Bq/hr). The varying delayed response of different CRMs to increasing radon levels
could be due to the different radon progeny each CRM counts to determine the radon levels or to
other factors. The table 2 below gives the average amount of additional ingrowth each CRM
would require to match the RAD7 and grab sample results based on this single test. Additional
testing would be needed to confirm this response. The RadonAway RS800 and FemtoTech data
is based only on the initial RAD7 data displayed in Figure 5 because their monitor results during
the grab samples were significantly off.
Unit
Correction %

Pylon
+15%

RTCA
+14%

Scout
+9%

SN1029
+9%

RS500
+6%

RS800
+1%

F-510
+10%

Table 2 Correction factor for CRM in-growth delay

Although Figure 5 is a crowded graph some general CRM performance differences can be seen
that repeated in other similar exposures. In general the Pylon AB5 and the RAD7 produced the
smoothest line that makes determining ingrowth rate more accurate. The RadonAway RS800 had
the next smoothest ingrowth line although the unit tended to under report the radon level increase
during the initial 8 to 10 hours of exposure. It appears radon entry into the chamber of the
RS800 is delayed thus producing the lag and facilitating the smoother line. This may be why the
RS800 also had the least response to thoron. Because the RS800 tends to under respond for the
first six hours, this data cannot be used for the ingrowth calculation. The FemtoTech 510
performed well until the radon levels climbed above 80 pCi/l (3000 Bq/m3) when it would bias
low. In most cases however the ingrowth measurements will not be above 80 pCi/l. Note that
the RTCA On-Guard CRM does not record any results above 100 pCi/l (3700 Bq/m3) and the
RS800 does not record results above 200 pCi/l (7400 Bq/m3). The Sun Nuclear 1029, Scout and
RadonAway RS500 which are less expensive units had greater variability than the other
detectors. Longer exposure period would help minimize the effect of this variability. The small
size of the Scout and SunNuclear 1029 if the handle is removed does allow them to be placed
under a large metal mixing bowl (7.5 liters) which also has one of the lower liters to

6

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

Figure 5 CRM In-growth Performance

square foot ratio and thus will produce the most radon increase per hour inside the accumulator.
See Table 3 below.
Accumulator

liters

Small mixing bowl

2.95

Large mixing bowl

7.3

2.5 gallon bucket

9.63

Small trash can

8.65

width
8.5” 21.6
cm
12.75” 32.4
cm
10.25”
26 cm
8.375”
21.3 cm

ft2

m2

ft2/liter

m2/liter

0.39

0.036

0.13

0.012

0.89

0.083

0.12

0.011

0.57

0.053

0.06

0.006

0.38

0.035

0.04

0.004

2

2

Table 3 - Accumulator area to liters – larger ft /l or m /l higher the response

CRM Response to Slab with Thoron & Actinon
It was not possible to quantify CRM response to actinon because its half life of 4 seconds is too
short. All of the CRM’s and E-PERMs were however tested to determine their response to

7

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

thoron. In typical indoor air measurements, a detectors thoron response would not be considered
important because it is assumed that thoron’s half life of 55 seconds does not allow enough time
for it to reach the breathing or testing zone if the source is the soil. Thoron sources inside the
dwelling would be more likely to influence CRMs that were sensitive to thoron levels. Flux
measurements made under an accumulator however place a radon detector in very close
proximity to the source which might contain thoron. A detector that is very sensitive to thoron
could cause a false interpretation of the results if thoron is present in the material. In most cases
the diffusion length of the material is long enough to decay out the thoron. Because thoron’s
half life is very short it will reach a maximum concentration inside the accumulator during the 1st
hour of exposure.
Table 4 below is the calculated response that thoron would have if there was equal alpha activity
from thoron and radon and the detector only provided an average such as the Pro-Series 3
monitor or an E-PERM. If CRM’s are used under an accumulator the thoron response can be
eliminated by using the slope of the ingrowth after 4 hours to determine the emanation rate since
the radon will continue to ingrow while the thoron will be steady state.
If only average is used
for this
exposure length
12 hours
24 hours
48 hours

E-PERM bias if equal
Thoron & Radon

Pro-Series 3 bias if
equal Thoron & radon

+ 0.8%
+ 0.4%
+ 0.2%

+ 1.0%

Table 4 Effect of thoron on average of detector results

All of the CRM’s and EPERMs were tested to
determine their response to
thoron. A 114 liter chamber
was constructed with two
computer type fans installed
1/3 up from the bottom of the
chamber and two additional
computer type fans 2/3 of the
way up. The four fans created
a counter clockwise air flow
with a velocity of 1.3 meters
per second or approximately 1
revolution around the
chamber per second. Each
fan had thorium coated
Aladdin lantern mantles
suspended in the fan’s
airflow. See photo below
Figure 6 - 114 liter Thoron Chamber using Aladdin mantles & fans

8

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

in Figure 6. Two sampling ports in the walls of the chamber were used to flow air through a
RAD7 that is capable of measuring thoron concentrations.
The CRM’s and E-PERMs were exposed in the sealed chamber for 18 to 48 hours. The 4 to 12
Aladdin mantles produced enough thoron to maintain the chamber at 200 to 600 pCi/l of thoron
as measured by the RAD7. The thoron concentration was measured by averaging 30 minutes of
sampling data taken during two periods during the exposure length. The RAD7, which was
located outside the chamber, was set up with short tubing and the small desiccant holder to
minimize thoron decay loss.
In each exposure outdoor air was blown into the chamber prior to sealing the chamber to
minimize radon levels. Any activity above background radon levels that the detectors recorded
above the chamber radon background was assumed to be caused by thoron.
Table 5 below demonstrates the dramatically different thoron response of the CRM monitors that
were tested. The RadonAway RS800 had very little response to thoron. When the RS800 was
exposed to 550 pCi/L (20,300 Bq/m3) of thoron it only displayed an average of 3.4 pCi/l (126
Bq/m3). The RadonAway RS500 which has a very similar metal case however responded
dramatically to thoron concentrations. It also had an increasing response which may have been
due to a response to the decay products of thoron inside the chamber. See Figure 7 below. This
increasing response would bias the results if there was significant thoron in the material being
flux tested. The Femto-Tech 510 also responded to thoron but did not have an increasing
concentration over the exposure. The Scout, Sun Nuclear 1029 and Pylon AB5 PRD had some
limited response. The inexpensive Pro-Series 3 radon monitor also responds significantly to
thoron. E-PERM S-Chambers with short term electrets had an average response to thoron of 4%.

RS 800
0.5%

Scout
6.4%

EPerms
4.0%

SN 1029
5.5%

Pylon AB5 PRD
2.8%

RS 500
67%

Femto 510
17%

Pro-Series 3
22%

On-Guard
10.9%

Table 5 - Detector Response to 220 Thoron

9

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

Figure 7 - Detector response to steady state 220 Thoron

Reducing Thoron Response
Any lengthening of the time it takes for thoron to reach the detectors sensors will reduce the
detectors response to thoron. Radon, thoron and actinon will pass through but be slowed down
by thin plastic depending on the plastic density and molecular structure. If one is trying to use
the E-PERM thoron chambers to measure thoron it is important to know the background radon
without having the detector respond to the thoron. Readily available zip lock food storage bags
were tested to determine if they could reduce entry of thoron. The data in table 6 below was
obtained by exposing different configurations of S-Chamber E-PERMs for one to two days in the
thoron chamber at two different concentrations and once to only an ingrowth of radon. In each
thoron test all the EPERMs were suspended in the center of the chamber to allow free circulation
of thoron enriched air around them. Three different diffusion barriers were tested, Tyvek
envelope, Ziploc brand vegetable bag that has pin holes in the plastic every centimeter (3/8”) and
double Hefty One Zip brand bags. The thickness of the plastic was not available. In order to
induce a longer travel path for the thoron an E-PERM was placed inside an open Hefty bag and
both were then placed inside a second open Hefty bag. E-PERMs without any bag covering
were also exposed.

10

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

E-PERM
setup
No covering
Inside Tyvek bag
Inside Vegetable bag
Inside 2 zip-loc bags

550 pCi/l
20.3 kBq/m3
thoron
3.6%
N/A
1.7%
0.6%

191 pCi/l
7 kBq/m3
thoron
4.0%
3.7%
2.8%
1.5%

Just
radon
100.0%
100.0%
97.5%
85.0 %

Table 6 – Thoron & Radon Reduction from plastic bags

Building Materials
Two concrete slabs were hand mixed and poured in forms. Sakrete 5000 plus concrete mix was
purchased locally from a building supplier and used for both slabs. This higher strength concrete
was used to closer mimic commercial post stressed concrete. Each slab was carefully mixed
using the water to concrete ratio specified by the manufacturer. The drying time of the slabs was
reduced by keeping the slabs covered and occasional misting them with water for 14 days. The
slabs were allowed to dry for at least 60 days before any testing was done on the slabs. One of
the slabs referred to hereafter as the “mixed slab” had 9 ounces of high radon/thoron/actinon soil
thoroughly mixed in with the cement to raise the radon emanation rates. The mixed slab is 17”
by 17” by 3.5” thick (43x43x8.9 cm) (36.7 kg). The 3.5” edge around the perimeter of the slab
was covered with 17 mil aluminum tape to allow radon emanation from only the two flat
surfaces for a total area of 4 square feet (0.37 m2). See photo of this slab in Figure 19 below.
The second slab referred hereafter as the “cold slab” is 16” round by 4.5” thick (40 cm round x
11.4 cm thick)(31.7 kg). The 4.5” perimeter of this slab is also covered with aluminum tape.
This slab has 2.8 square feet (0.26 m2) of exposed slab.

11

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

WPB building material emanation test
using Flow Through Test Chamber

Exhaust vent to
outside
Sealable metal
chamber

C R M’s
inside
chamber

4
3

2

1

CRM
CRM
Ball Valve
sampling
ports for grab
samples
or R AD 7
measurements
4

Building
Material
In chamber

D esiccant
column

Valved
inlet

3
2

1 um
filter

0 .1 to 1 .0
LPM
airflow

1

Low flow mixing
fan inside
chamber

D w yer
airflow meter

Granular activated C oconut
carbon ( GAC )
Aquarium
pump

Figure 8 – Slab Flow through Measurement Chamber

Testing Cold & Mixed Slab with Ingrowth & Flow through
The radon emanation rate for both slabs was determined by placing them individually inside a
sealed metal chamber and measuring the ingrowth that takes place. To test the ingrowth method
the mixed slab emanation rate was also measured by a flow through method. The cold slab did
not have a high enough emanation rate to allow a flow through test. The flow through method
eliminates the need to know the exact volume inside the chamber and the determination of the
emanation rate is a straight forward calculation but the exact flow rate through the chamber must
be known and the radon levels of the inflowing air must also be know. The flow rate was
determined by using a Dwyer airflow gauge that was cross compared to a lab research bubble
film flow gauge. Note that the authors six flow gauges vary from 8% high to 5% low compared
to the cross calibrated unit.
Any radon in the inflowing air will bias the reading. Even outdoor air can often exceed 1 pCi/l
(37 Bq/m3) at night. In order to eliminate the need to measure the radon levels of the inflowing
air a radon filtering method was tested. A 3” PVC pipe, 10 feet long (3.6 cm X 3 meters) was
filled with 0.5 cubic feet (14 liters) of granular activated coconut carbon (GAC). To test the
effectiveness of the carbon filter, one lpm of desiccant dried air containing 150 pCi/l (5500
Bq/m3) was pushed through the carbon. It took eight days and six hours of continuous steady
flow rate of 1 lpm before radon broke through the carbon. This carbon was then replaced with

12

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

fresh carbon. The GAC filled PVC pipe makes an excellent pre-filter to eliminate any need to
subtract background radon from the measured radon levels produced inside the chambers. The
air entering the carbon tubes should be dried with a desiccant to maintain maximum radon
reduction. Figure 8 above illustrates how a flow through chamber for testing building materials
can be set up. The mixed slab was tested with a modified flow through setup by adding a second
chamber that the CRM’s were placed in to allow thoron to decay out and minimize their
influence on the CRM’s.
The following three charts represent the three different methods of measuring emanation rate of
the slab that had hot soil mixed into it. Figure 9 is the flow through method. Figure 10 is the
total slab in a sealed chamber method. Figure 12 is different CRMs sealed under accumulator
metal buckets. Note that the ingrowth emanation rate is determined by using the formula in
Table 7 below. These methods produced results that varied from a high of 200 pCi/ft2/hr to a low
of 150 pCi/ft2/hr or a total variation of about 25%.

Figure 9 – Slab Flow through Measurement Chamber

The ingrowth of radon into a sealed chamber or accumulator in Figure 10 below is compared to
the mathematical ingrowth using the formula in Table 7. Using this formula allows any
exposure duration to be used and the initial radon in the chamber to be subtracted from the
ingrowth created by the building material. Unless carbon filtered air is used it will be necessary
in most cases to approximate this initial radon concentration based on ambient radon
measurements or make a grab sample measurement.

13

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

The mathematical ingrowth needs to incorporate the free air volume of the accumulator
(accumulator volume minus the volume of the detector(s) & building material), the area of the
slab that is exposed and the initial radon levels when the detector is sealed in the accumulator.

Figure 10 – Slab In-Growth Measurement Chamber

The following formula, which can be entered into a spreadsheet program, is repeated each hour
to obtain the mathematical radon concentration at each hour during the exposure:
SR
HR
AR

= Starting Radon under the accumulator
= Exposure hour
= Area accumulator covers in square feet
(this can be m2 if source strength is changed to Bq/m2/hr)
VOL = Free air volume inside accumulator in liters
X
= Multiplication addition & subtraction symbols
CAV = Constant Adjustment Value
SS
= Source Strength in pCi/sq ft/hr
(this can be Bq/m2/hr if area is changed to m2 and Bq units are used)
(SR X (exp(-0.1813 X (HR / 24))))
+
((((SS X AR) X 24) / (VOL X 0.1814)) X (1 – (exp(-0.1814 X (HR / 24))))) – CAV
Table 7 – CRM Accumulator Source Strength (SS) Formula

14

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

Note that the first part of the formula is used to subtract out the diminishing effect of radon
trapped under the accumulator at the start of the exposure. The second formula includes a
Constant Value Adjustment (CAV) which is used to adjust the mathematical ingrowth line up or
down so that it lines up with the CRM data plotted on the chart. The CAV must be a constant
value throughout the exposure so that it does not affect the slope of the mathematical ingrowth
only it’s placement on the chart. The need to adjust the mathematical ingrowth is due to CRM
response delay, thoron, or different detector starting times versus chamber sealing. The CRM
data and mathematical ingrowth value from the accumulator formula are both plotted in a
spreadsheet. The Source Strength value and Constant Value Adjustment (CAV) of the
accumulator formula are varied until the slope of the mathematical ingrowth matches the slope of
the CRM data.

Flux Testing the Slabs with Accumlator Chambers
The emanation rate of both slabs was tested by sealing a metal chamber (accumulator) on top of
the slabs with a CRM installed inside. In each case the slab was elevated off the floor to allow
open air circulation. The accumulator, which should be made of metal or glass to avoid any
diffusion of radon out of the chamber, can be a metal mixing bowl or metal bucket. All seams in
the accumulator must be caulked or foil taped diffusion tight. The accumulator should be just
large enough for the CRM to fit inside to maximize the radon ingrowth. The volume of the
accumulator in liters needs to be obtained by carefully measuring the interior dimensions or by
filling the accumulator up with a known quantity of water. The material volume of the CRM or
E-PERM needs to be known and subtracted out of the accumulator volume The volume of each
CRM was measured and the approximate values are given in Table 1 above.
Flexible plumbers putty was used to seal the accumulator with the CRM inside to the slab. The
area of exposure needs to include one half of the area the putty covers. Most putty’s have some
oil content and will leave a stain if the surface is porous. See photo in Figure 11 below of the
RS800 and a 2.5 gallon (9.6 liter) metal bucket. All seams inside the bucket were sealed.

Figure 11 – RS800 sealed under 9.6 Liter Accumulator for Mixed Slab

15

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

The CRM’s need to be left under the accumulator for 12 to 48 hours. The data is input into a
spreadsheet graph and the source strength value of the mathematical ingrowth formula and the
Constant Adjustment Value (CRV) are adjusted until the mathematical ingrowth matches the
actual CRM ingrowth. The source strength value is the emanation rate of the material in
pCi/ft2/hr or Bq/m2/hr.

Figure 12 – CRM’s sealed under Accumulator over Mixed Slab

Figure 12 shows the difference in CRM performance and the emanation rate based on using the
mathematical ingrowth formula as previously discussed. RS 800, which is the least sensitive to
thoron lags behind and then over responds by 25% compared to the SN1029 and Femto-Tech
510. The Femto-Tech 510 starts to fall off the ingrowth slope at 85 pCi/l but it responds well up
to that point. The RS 500 has a 10% higher ingrowth than the Scout and SN 1029. No EPERMs were exposed under the accumulator with the mixed slab because of a high gamma
reading.

Emanation from Unaltered Retail Concrete
A second slab without any additional soil added was made with “Sakrete 5000 plus” concrete
mix obtained from a local building supplier. The emanation rate of radon from this mix is too
low to use the flow through method to measure the entire slab (16” round by 4.5” thick - 40 cm
round by 11.4 cm thick). Instead the entire slab was sealed inside a chamber with two AB5
Pylons. See the results in Figure 13 below.

16

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

Figure 13 – Pylon ingrowth data compared to mathematical ingrowth

The mathematical ingrowth is determined by using the formula in Figure 17 above and adjusting
the source strength and CAV until the CRM data and the mathematical ingrowth match. A
thoron sniff measurement using the RAD7 during the ingrowth did not reveal any significant
thoron concentrations coming from the cold slab.
The ability to obtain similar results using CRM’s and E-PERMS under smaller accumulators is
displayed in the graph in Figure 14. The CRM’s were placed under either a 9.6 liter or 7.4 liter
metal bucket that was sealed on top of the slab. Note the variation in measurements when using
a less precise Scout, SN1029 or RS500. These monitors need to be exposed for longer periods to
improve accuracy. The full slab test in Figure 13 indicated an emanation rate of 8.2 pCi/ft2/hr
(28.2 Bq/m2/hr). The SN1029 and RS 500 are within 10% of the total slab measurement while
the RS 800 was 27% lower and the FemtoTech 510 is 39% lower.

17

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

Figure 14 – CRM compared to mathematical ingrowth

E-PERM’s were also exposed a number of times under accumulators. For E-PERM
measurements it is important to know the initial
radon levels at the start of the measurement and
the gamma emanation which might be elevated in
some cases above background from the building
material. Gamma measurements can be made
with a properly calibrated gamma survey
instrument or more accurate measurements can be
obtained by using 2 mR gamma dosimeters
obtainable from Rad Elec Inc that are exposed
over a one to two day period and then re-charged
with a portable charger. See photo in Figure 15.
The ambient radon in air concentration
trapped inside the accumulator that is
decaying during the exposure period needs
to be factored out of the ingrowth measurement. A less precise method is to approximate the
initial radon measurement based on average radon in air measurements made in the same
location and then use the first part of the formula in Table 8 to determine the Starting Radon
Influence (SRI). A more precise method if the initial radon concentration is not known is to seal
an E-PERM in a glass jar at the beginning of the measurement. The average radon concentration
of the E-PERM in the jar is the SRI value. The SRI value is subtracted from the radon
measurement obtained under the accumulator (RUA). The formula for obtaining this
Figure 15 – 2 mR Gamma Dosimeter & re-charger

18

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

measurement is given in Table 8. Note that the lower the emanation rate the more critical it is to
measure the starting radon concentration.
If the E-PERM is immediately closed up at the end of the accumulator exposure the E-PERM’s
response delay to the in-growing radon will bias the average reading low. This effect is more
pronounce for an in-growth exposure of increasing radon concentration than a steady state
exposure because the highest radon concentration happens at the end of the exposure. One
method of compensating for the final out-gassing of radon in the chamber after the exposure and
the final decay of the radon short lived decay products left in the E-PERM chamber is to leave
the E-PERM open an additional 3 hours in a low radon environment. This would be in-practical
in most cases because of the availability of a low radon environment and the time constraint of
waiting three hours. To test the amount of bias at the end of an ingrowth exposure, 12 E-PERMs
were exposed to a radon ingrowth inside a sealed chamber. Six of the E-PERMs were read
immediately and 6 were left open in a lower radon environment (outside mid-afternoon) and read
three hours later. The difference equaled about 10% higher emanation rate which is added to the
formula in Table 8. This is similar to the CRM bias.
If the influence of the starting radon concentration is not measured using the E-PERM sealed in a
jar method then the starting radon influence (SRI) is determined by the first formula given below
which can be entered into a spreadsheet. The SRI is then included in the second formula to
determine the emanation rate.
RUA = E-PERM average radon under accumulator measured by an E-PERM
ARL = Approximate ambient radon level when E-PERM is sealed
SRI = Starting Radon influence
(use either E-PERM average in sealed jar or 1st part of formula in Figure 24)
EXD = Exposure Days
AR
= Area accumulator covers in square feet
(this can be m2 if source strength is changed to Bq/m2/hr)
VOL = Free air volume inside accumulator in liters
X
= Multiplication symbol
SS

= Source strength in pCi/sq ft/hr
SRI = (ARL X (1- exp(-(0.1813 X EXD))) / (0.1814 X EXD))

SS = ((((RUA-SRI) X VOL X 0.1814) / AR) / (1-((1- exp(-0.1814 X EXD)) / (0.1814 X EXD))) / 24) X 1.1
Table 8 – E-PERM Accumulator Source Strength (SS) Formula

Note that the first calculation (SRI) determines the diminishing effect of radon trapped under the
accumulator at the start of the exposure. Single S-Chamber E-PERM’s were exposed under an
accumulator sealed on top of the cold slab.

19

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

Pylon average
8.2 pCi/ft2/hr
Ingrowth correction
1.15
9.4 pCi/ft2/hr
3.7 Bq/m2/hr

E-PERM smooth side
5.5 pCi/ft2/hr
Final decay correction
1.1
6.0 pCi/ft2/hr
3.5 Bq/m2/hr

E-PERM rough side
6.0 pCi/ft2/hr
Final decay correction
1.1
6.6 pCi/ft2/hr
3.6 Bq/m2/hr

Table 9 – E-PERM under accumulator versus Pylons with cold slab ingrowth

The E-PERM’s were placed under 3 liter accumulator bowls and they calculated emanation rate
was 33% lower than the Pylon. It is unclear why they responded so much less. The Pylon
exposure with the total concrete slab was not repeated to determine if the total emanation rate out
of the concrete was reduced because the summer months had higher outdoor humidity levels that
the slab was exposed and there may have been a decreased emanation rate because of humidity
being greater than 80%. See paper on radon emanation and moisture content of concrete in
references below.

Measuring Granite Tiles & Countertops
Granite typically contains 238 uranium and 226 radium which decays into 222radon which will
escape into the air. There is concern that granite with unusually high levels of radium could
significant increase the radon levels if it was installed in an air tight homes or areas of the home
that had limited air exchange. Several pieces of granite were obtained that had higher than
average emanation rates of radon in order to test the ability to measure the emanation rate using
the accumulator method. These pieces of granite were first measured by placing them in a sealed
chamber with one or two AB5 Pylons. The graph of 1.25” (3 cm) thick granite in Figure 18 is
the highest emanating granite slab that was tested. Granite emanation rate in this study uses
units of square feet or square meters of the polished top side but the granite is actually emanating
from both sides of the material although not at equal rates. Note that the emanation rate across
the granite surface is also likely to vary significantly.

20

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

Figure 16 – Granite Emanation Rate in a sealed Chamber

The accumulator method was used to measure the emanation rate of the granite samples by
sealing a metal trash can (7 to 8 liter size) with a CRM’s inside to either side of the granite. The
volume of the CRM is subtracted from the volume of the accumulator to determine the actual
free air. It was determined that the granite area the accumulator covered needed to include half
the width of the putty placed around the accumulator as additional area emanating into the
chamber.

Figure 17 – Granite Emanation Rate Variation

21

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

Table 10 and 11 below depicts the significant difference between emanation rates of the polished
side versus the un-polished side. The JB granite had a plastic fiber re-enforced coating that is
apparently stopping 98% of the radon emanation out of the un-polished side. The CB granite
was the reverse with almost 8 times more radon emanation from the un-polished side versus the
polished side. The NG granite had 40% more emanation from the un-polished side versus the
polished side. The difference between emanation rates is due to the increased surface area of the
un-polished side and the different sealing methods used on the polished side. Polished granite
typically has fillers installed to fill the small indentations in the granite before it is final polished.
These results indicate the need to measure both sides of a granite counter top to avoid significant
errors. It may be possible to test the underside of a granite kitchen slab by removing a cabinet
drawer to gain access for the accumulator.
The last three columns in Table 10 gives the radon emanation rate determined by measuring the
entire piece in a sealed chamber. The sum of the CRM accumulator measurements versus the
total emanation matches within a few percentage points for two of granites. The NG granite has
a total emanation rate that is almost 20% less than the sum of measurements of the two sides.
Variation between measuring the total granite piece and individual sides could be due to
variations in emanation across the surface of the granite.

Granite type
NG granite
JB granite
CB granite

Polished
emanation
pCi/ft2/hr
240
120
1.0

Unpolished
emanation
pCi/ft2/hr
345
2
7.8

Total
emanation
pCi/ft2/hr
490
125
8.6

Total
emanation
pCi/m2/hr
5274
1345
92.6

Total
emanation
Bq/m2/hr
195
50
3.4

Table 10 – Granite Emanation Rate calculated with CRM’s

Granite type
NG granite
JB granite
CB granite

Polished
emanation
pCi/ft2/hr
199
107
1.6

Unpolished
emanation
pCi/ft2/hr
376
1.4
11.9

Table 11 – Granite Emanation Rate calculated with E-PERMs

The likely variation across the granite surface and the difference in emanation rate between the
polished and un-polished side can easily produce a significant bias in a single accumulator test of
only one side of a granite slab.

22

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

The gamma rates of the four granite pieces were measured with a Bicron Micro Rem gamma
meter that had been recently calibrated as well as with the Rad Elec 2 mR/hr gamma dosimeters.
There was only a 10 to 15% difference in measurement results between the two types of gamma
measurements. See picture of gamma dosimeters in Figure 15 above. The gamma
measurements are compared to the measured radon emanation rate in Table 12 below. In each
case the average of the background gamma was subtracted from three gamma dosimeters placed
on top of the granite pieces. This small sample of four granite pieces indicates the ratio between
the gamma emanation rate and the radon emanation rate varies by a factor of 8. The variation in
the ratio between gamma measurements and radon emanation rate will likely indicate which
granite pieces are unlikely to increase radon levels but are not likely to be able to indicate how
much radon emanation is coming off granite based on gamma measurements.

Granite type
NG granite
FS granite
JB granite
CB granite

Gamma µR/hr
above
background
99.3
25.0
12.7
3.4

Total
emanation
pCi/ft2/hr
490
508
125
8.6

pCi/ft2/hr
per µR/hr
above background
4.9
20.3
9.8
2.5

Table 12 – Gamma emanation rate versus radon emanation

Calculating Radon Increase from Building Material Emanation Rates
Determining the increase in radon levels from a building material is difficult even if the building
material has a uniform emanation rate. Radon emanation from concrete may be reduced by the
materials placed over the concrete such as vinyl flooring or ceramic tile although drywall, paint,
texture coatings or carpeting may provide very little reduction in emanation rate.
This total emanation rate per hour from the material is divided by the liters of outdoor air moving
into the structure or room every hour to obtain the radon level increase. The amount of outdoor
air entering a building can obviously change hour by hour depending upon wind load,
temperature difference inside to outside, exhaust fan operation and window and door position.
Any change to this ventilation rate will have a linear effect on the radon levels since the
emanation rate from building materials is likely to be fairly consistent. The introduction of
outdoor air into the dwelling will likely be well mixed if the unit has an air handling system that
is operating. If there is no air handling system or it is not operating then the increased radon in
air from the building material will vary from room to room depending on the room’s volume
versus exposure to the building material and the natural mixing taking place from room to room.
If some assumptions are made, one can calculate the contribution of increased radon in a small
home that is very air tight. An air tight home would be most influenced by building material
emanation rate. The condominiums the author worked on had air change per hour (ACH) rates
less than 0.1 ACH. If we use 0.1 ACH with a 1250 ft2 (116 m2) condominium the ventilation
rate would be 28,316 liters per hour. If 40 ft2 (3.72 m2) of granite was installed in this size
dwelling assuming even mixing of the air by an operating air handling system the radon levels
would be increased by the amounts given in Table 13 below. If the condominium floors and

23

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

ceilings were constructed of concrete, as they typically are, there would be 2500 ft2 of concrete
exposure. If the emanation rate of the cold slab (9.4 pCi/ft2/hr) is used, the radon increase will
be around 0.8 pCi/l. The cold slab was however only 3.5 inches thick (8.9 cm) while a typical
condominium slab is 7 to 8 inches thick (18 to 20 cm). The “diffusion length” of concrete (point
where only 37% of the element is escaping) has been measured by other researchers to be around
10 cm (4”). The double thickness of the actual slab versus the tested cold slab will increase the
surface radon emanation but it would likely not be linear. Note however that the concrete even
using the cold slab emanation rate increases the radon levels a greater amount than the granite
having an unusually high emanation rate.
CB Granite
0.1 pCi/l
4 Bq/m3

JB Granite
0.2 pCi/l
7 Bq/m3

NG Granite
0.7 pCi/l
26 Bq/m3

Concrete
0.8 pCi/l
30 Bq/m3

2

Table 13 – Radon increase in 1250 ft (116 m2) dwelling with 0.1 ACH
2
2)
2
2
2
from 40 ft (3.7m granite or 2500 ft (232m )concrete at 9.4 pCi/ft /hr emanation

Summary
In most cases it will not be possible to take a sample of a building material and place it inside a
sealed chamber with a radon monitor to measure the emanation rate. This study has
demonstrated that placing a continuous radon monitor or E-PERM inside a metal or glass
accumulator that is sealed to the emanating material surface is a reasonably reliable method for
determining the emanation rate assuming the emanation rate is consistent across the surface of
the material. It appears from the small number of granite samples tested that granite can have
significant variation in emanation rates between surfaces . Concrete slabs however are likely to
have significantly more uniform emanation rates assuming the material came from the same
source and if there have been no coatings applied to one side of the concrete. To obtain
emanation rates, it is necessary to know the exact volume of the accumulator, the amount of free
space taken up by the detector and the area the accumulator is covering. The detector should be
in place from 24 to 48 hours. Materials with low emanation rates should have 48 hour
exposures. The emanation rate in pCi/sq ft/hr or Bq/m2/hr can be determined by using the
formulas given in this paper. The CRM ingrowth rate will need to be matched to a mathematical
ingrowth rate obtained from the formulas and adjusted until it matches the ingrowth of the CRM
data to determine the emanation rate. This emanation rate times the area of the material exposed
inside the dwelling divided by the ventilation rate will give the expected radon increase provided
by the material. Changes in the radon concentration will therefore be directly related to the
ventilation rate of the dwelling.

Acknowledgements
This study would not have been possible without the very generous donation of equipment from
Rad-Elec, Femto-Tech, RadonAway, Sun Nuclear, RTCA and Pennsylvania DEP Radon
Department. In addition the author is very indebted to the following individuals for much
needed advice and help: Mike Kitto, Paul Kotrappa, Dan Steck, Phil Jenkins and Al Gerhart.

24

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

References
Ackers J.G., Den Boer J.F., DeLong P., Wolschrijin R.A., Radioactivity and radon exhalation
rates of building materials in the Netherlands. Sci.Total Environ. 45: 151-156; 1985
Chao C., Tung T., Chan D., Burnett J.. Determination of radon emanation and back diffusion
characteristics of building materials in small chamber tests. Building and Environ. 32:355-362;
1997
Chyi L.L.. Radon testing of various countertop materials final report. University of Akron,
May 2008
Cozmuta I., Van der Graaf E.R.. Methods for measuring diffusion coeffients of radon in
building materials. Sci. Total Environ. 272: 323-335; 2001
Cozmuta I., Van der Graaf E.R., Meijer R.J.. Moisture dependence of radon transport in
concrete: measurements and modeling. Health Phys. 85: 438-456; 2003
Folkerts K.H., Keller G., Muth H.. An experimental study on diffusion and exhalation of 222Rn
and 220Rn from building materials. Rad. Prot. Dosim. 9: 27-34; 1984
Gadd M.S., Borak T.B.. In-situ determination of the diffusion coefficient of 222Rn in concrete.
Health Phys. 68: 817-822; 1995
Janga M., Kanga C.S., Moonb J.H.. Estimation of 222Rn release from the phosphogypsum
board used in housing panels. J. Envirn. Radioact. 80: 153-160; 2005
Keller G., Hoffman B., Feigenspan T.. Radon permeability and radon exhalation of building
materials. Sci.Total Environ. 272:85-89; 2001
Kitto, M., Green J.. Emanation from granite countertops. Proceedings of AARST International
Radon Conference; 2005
Kotrappa P., Stieff L.R.. NIST traceable radon calibration system for calibrating true integrating
radon monitors – E-PERM . Proceedings of AARST International Radon Conference; 1993
Kotrappa P., Stieff L.R. Volkovitsky P.. Radon monitor calibration using NIST radon emanation
standards: steady flow method. Journ. Rad. Protection Dosimetry; 2005
Kovler K. Perevalov A., Steiner V., Rabkin E.. Determination of the radon diffusion length in
building materials using electrets and activated carbon. Health Phys. 66: 505-516; 2004
Kovler K.. Radon exhalation of hardening concrete: monitoring cement hydration and prediction
of radon concentration in construction site. Journ. Environ. Radioactivity 86: 354-366; 2006

25

Proceedings of the American Association of Radon Scientists and Technologists 2008 International
Symposium Las Vegas NV, September 14-17, 2008. AARST © 2008

Llope W.J.. Radiation and radon from natural stone. Rice University, 5/7/2008
Misdaq, M.A., Amghar A.. Radon and thoron emanation from various marble materials: impact
on the workers. Radiat. Meas. 39: 421-430; 2005
Pavilidou S., Koroneos A., Papastefanou C., Christofides G., Stoulos S., Vavelides M.. Natural
radioactivity of granites used as building materials. Journ. Environ. Radioactivity 89: 48-60;
2006
Righi S., Bruzzi L.. Natural radioactivity and radon exhalation in building materials used in
Italian dwellings. Journ. Environ. Radioactivity 88: 158-170; 2006
Roelofs L.M.M., Scholten L.C.. The effect of aging, humidity, and fly ash additive on the radon
exhalation from concrete. Health Phys. 67: 225-230; 1991a
Rogers V.C., Nielson K.K., Holt R.B.. Radon diffusion coefficients for residential concretes.
Health Phys. 67:261-265; 1994
Rogers V.C., Nielson K.K., Holt R.B.. Radon diffusion coefficients for aged residential
concretes. Health Phys. 68:832-834; 1995
Sahoo B.K., Nathwani D., Eappen K.P., Ramachandran T.V., Gaware J.J., Mayya Y.S..
Estimation of radon emanation in Indian building materials. ScienceDirect 42:1422-1425; 2007
Stoulos S., Manolopoulou M., Papastefanou C.. Assessment of natural radiation exposure and
radon exhalation from building materials in Greece. J. Environ. Radioact. 69:225-240; 2003
Stoulos S., Manolopoulou M., Papastefanou C.. Measurement of radon emanation factor from
granular samples: effects exposure and radon exhalation from building materials in Greece. J.
Environ. Radioact. 69:225-240; 2004
Sun H., Furbish D.J.. Moisture effect on radon emanation in porous media. J. Contam Hydrol
18:239-255; 1995
Yu K.N., Chan T.F., Young E.C.M.. The variation of radon exhalation rates from building
materials of different ages. Health Phys. 68:716-718; 1995

26