Content uploaded by Mahmoud Abo-Elmagd
Author content
All content in this area was uploaded by Mahmoud Abo-Elmagd on May 01, 2015
Content may be subject to copyright.
CALIBRATION OF CR-39 FOR RADON-RELATED PARAMETERS
USING SEALED CUP TECHNIQUE
M. Abo-Elmagd* and Manal M. Daif
National Institute for Standard, Ionizing Radiation Metrology Laboratory, PO Box 136 Giza
Code No. 12211, Egypt
*Corresponding author: abo_elmgd@hotmail.com
Received May 19 2009, revised December 5 2009, accepted December 16 2009
Effective radium content, mass and areal radon exhalation rates of soil and rock samples are important radon-related
parameters and can be used as a better indicator of radon risk. A sealed cup fitted to a CR-39 detector and to the sample
under measurement is an advantageous passive device for the measurement of these parameters. The main factors affecting
the results are the detector calibration factor and the sample weight. The results of an active technique (Lucas cell) and the
CR-39 detector have been found to be correlated resulting in a reliable detector calibration factor. The result illustrates
the dependence of the CR-39 calibration factor with the sample weight which is difficult to use in practice, because each
sample weight has its own calibration factor of CR-39. It is reported to demonstrate the advantage of a back diffusion correc-
tion. After correcting the results for back diffusion effects, one obtains an approximately constant calibration factor for the
sample volumes up to one-third the total sealed cup volume. For this condition the calibration factor is equal to 0.237 track
cm
22
per Bq m
23
d with about 1 % uncertainty.
INTRODUCTION
The amount of radon formed in rocks and soil
depends on their uranium content. If uranium-rich
material lies close to the surface of the earth, there
can be high radon exhalation rate, resulting in high
radon exposure hazards. However, this alone is not
decisive in determining the radon concentration in
air, it is also determined by the extent to which the
formed radon atoms emanate from the mineral
grains and whether radon can leave the pore space
either by diffusion or together with a flow of air or
water. In addition, radon concentration in the soil
air is significantly affected by the occurrence of
moisture or water in the pores
(1)
. The effective
radium content is the fraction of total radium in a
sample, which contributes to radon exhalation
(2)
.
The exhalation rate is the number of radon atoms
leaving the sample per unit time per unit surface
area E
A
(areal exhalation rate in Bq m
23
h
21
) or per
unit mass E
M
(mass exhalation rate in Bq kg
21
h
21
).
Passive measurement of radon exhalation rates is
a widely used method for its simplicity and its
ability for measuring low levels of radon by extend-
ing the exposure time
(3–6)
. This method is based on
using sealed cup equipped with solid-state nuclear
track detector (SSNTD) such as CR-39. The track
density on the detector is used to calculate the exha-
lation rates and other radon-related parameters. The
main factor affecting the results is the detector cali-
bration factor that converts the registered tracks to
the measured parameters. The aim of this work is to
check the detector calibration factor at different
sample volumes. When the radon concentration in
the vessel air starts to approach that of the air in the
sample, radon has significant probability of diffusing
back into the sample, this process is called back dif-
fusion. The effect of back diffusion is studied in the
present work because it is a function of the sample
volume (or thickness) and can alter the emanating
radon from the sample surfaces.
EXPERIMENTAL WORK
Uranium rich-ore (pitchblende) was used as a radon
source, which is grounded, sieved and dried in an
oven at 1108C for 24 h to evaporate the moisture
content
(7)
. Sample weights from 25 to 200 g were
placed in a glass container of 3.75 10
24
m
3
volume and 4.5 10
23
m
2
base area. The emanated
radon gas from the sample was measured by a cali-
brated active technique to be used in calibrating the
passive detectors. A scintillation technique is more
suitable for the experimental setup used. In this tech-
nique, the air sample is allowed inside the cell
through a filter and the concentration is evaluated
using the efficiency factor for the counting system
and from the theoretical factors due to buildup of
radon decay products with respect to sampling and
counting delays. The principle of detection is
achieved through the counting of photons resulting
from the interaction of alpha particles produced by
radon and its progeny with the ZnS (Ag) scintillator.
A photomultiplier tube assembly counts the photon
events.
#The Author 2010. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
Radiation Protection Dosimetry (2010), Vol. 139, No. 4, pp. 546–550 doi:10.1093/rpd/ncp300
Advance Access publication 13 January 2010
by guest on July 29, 2010rpd.oxfordjournals.orgDownloaded from
Calibration of radon concentration
A calibrated Lucas cell coupled with an electronic
apparatus AB-5 (Pylon Electronics, Canada) was
connected with the sealed cup by two plastic pipes
and valves. Such a system enables the air-containing
radon from the cup to be removed only to the Lucas
cell and vice versa. Each sample was sealed for a
known time followed by grab sampling for 10 min
using the AB-5 pump at 0.5 l m
21
flow rate
(8)
. After
waiting 4 h from the end of sampling, the radon
content in the Lucas cell was counted. Each result
consisted of six measurements, each of them lasted
5 min and then averaged. Then the net count per min
was calculated by subtracting the background count.
Using the following equation, the radon concen-
tration in the cell, Rn (pCi l
21
), can be calculated
(9)
:
Rn ¼NCPM C
32:22 EVcA;ð1Þ
where Cand Aare the correction factors for decay
during the counting time and time interval between
the end of sampling and the start of counting, respect-
ively. The digit 3 accounts for the three alpha emitters
in equilibrium, 2.22 is used to convert disintegrations
per min (DPM) to pCi, V
c
is the Lucas cell volume
(0.151 l) and Eis the efficiency of the Lucas cell for
radon measurements ( 0.74 +0.003 CPM/DPM) as
measured in the National Institute for Standard,
Ionizing Radiation Metrology Laboratory, Egypt.
The radon activity Q(pCi) of the used sample
weight is given by:
Q¼Rn VT
1e
l
t
ðÞ
;ð2Þ
where, V
T
is the total volume in litre (0.755 l ), which
is the sum of cell volume (0.151 l), sealed cup
volume (0.375 l), pump volume (0.154 l) and the
total pipe and connector volumes (0.075 l), t(h) is
the exposure time that elapsed from sealing the cup
to the end of pumping and
l
(7.55 10
23
h
21
)is
the radon decay constant.
From the sample activity Q(after converting it to
Bq), the effective radium content Ra
Eff.
(Bq kg
21
),
areal E
A
(Bq m
23
h
21
) and mass E
M
(Bq kg
21
h
21
)
exhalation rates can be determined:
RaEff ¼Q
M;ð3Þ
EA¼Q
l
S;ð4Þ
EM¼Q
l
M;ð5Þ
where, M(kg) is the sample weight and S(m
2
) is its
surface area.
Calibration of CR-39 detector
The same prepared sample weights (from 25 to
200 g) were measured by Lucas cell was used for
calibrating the CR-39 detectors. Each cup was fitted
to a CR-39 detector placed in the top inside of the
cup. The sensitive part of the detector was facing the
emanated radon gas from the sample so that it
could record the alpha particles resulting from the
decay of radon and its daughters in the whole air
volume of the cup
(10)
. Such assembly was left for a
known exposure time. At the end of the exposure
period, CR-39 detectors are collected and etched
together under their optimum conditions of 6.25 N
NaOH at 708C for 6 h and counted under optical
microscope of 400magnification power.
RESULTS AND DISCUSSION
The active measurements of radon-related par-
ameters such as the activity Q(in Bq), the areal
exhalation rate E
A
(Bq m
23
h
21
) and the mass exha-
lation rate E
M
(Bq kg
21
h
21
) as a function of
sample weights are shown in Figures 1–3. The
results are presented for three cases A, B and C in
the following:
Case A: no correction of the results;
Case B: correct the results for air volume variation
with sample weights M; and
Case C: correct the results for back diffusion effect.
Figure 1. Radon activity at different sample weights.
CALIBRATION OF CR-39 FOR RADON USING SEALED CUP TECHNIQUE
547
by guest on July 29, 2010rpd.oxfordjournals.orgDownloaded from
In case A, the total volume of the sealed cup is
used, therefore V
T
in equation (2) is equal to 7.55
10
24
m
3
, in case B the air volume above the sample
is used as the cup volume, which decreases as the
sample volume increases.
The back diffusion effect (case C) can be easily
taken into account if the used decay constant of radon
l
in above equations is replaced by
l
*(¼
l
þ
l
b
),
where
l
b
is the decay constant correcting for the first
order of removal of radon by back diffusion
(11)
.From
the work of Jang et al.
(12)
,
l
b
can take the following
form:
l
b¼
a
S
Va
¼
l
VS
Va
ð6Þ
where
a
is the correction term for back diffusion effect
and it is equal to lx,wherexis the sample thickness
(m), Sis the sample surface area (m
2
), V
a
is the
volume of the air space of the sealed cup (m
3
)andV
S
is the sample volume (m
3
). This formula is valid only
if the thickness of the sample, x, is less than the diffu-
sion length of radon
(13)
.
The relation between sample weights and their
activities as measured by Lucas cell is shown in
Figure 1. For uncorrected results (case A) the
relation is linear with 0.998 correlation coefficient,
the slope of this line is the specific activity of the
used sample, which is equal to 14.2+0.4 kBq kg
21
and it corresponds to the effective radium content
Ra
Eff
.
After correcting the results for the volume effect
(case B), the relation deviates from linearity. This
deviation increases as the sample weight increases.
In this case, the calculated value of the mean effec-
tive radium content is reduced to be equal to 12.4 +
1.1 kBq kg
21
.
The effect of back diffusion is obvious in case C
where the measured activity Qis increasing with M
up to 150 g followed by saturation, i.e. sample
weights from 150 to 200 g have the same activity.
This means that more and more radon atoms are
diffused back to the sample as well as some radon
atoms cannot escape from the sample to the sur-
rounding air due to its large thickness. For this
reason, the net result is the reduction of Ra
Eff
with
high uncertainty. In this case Ra
Eff
is equal to 9.7 +
2.4 kBq kg
21
. Therefore, correcting the results leads
to reducing the measured Ra
Eff
with increased sys-
tematic uncertainty.
Figure 2shows the relationship between the areal
exhalation rate E
A
and the sample weight. From
equation (4), E
A
is directly proportional to Q, there-
fore cases A and B have the same behaviour as
obtained in Figure 1for Q(where
l
is not cor-
rected). Different behaviour in case C is obtained
when the back diffusion effect takes place where no
plateau of E
A
is obtained and its value is always
higher than in case B. This is due to the decay con-
stant
l
, which is replaced by
l
* in the back diffusion
correction. From equation (6),
l
* is increasing with
increasing the sample volume (or weight) and there-
fore it is responsible for increased E
A
with Mwith
no saturation behaviour at higher weight (unlike the
case of Q(Figure 1)).
For mass exhalation rate E
M
as a function of
sample weight as shown in Figure 3, the mean value
of E
M
is equal to 106.8 +3.0 (2.8 %) Bq kg
21
h
21
for uncorrected results (case A), 93.6 +8.1 (8.6 %)
Bq kg
21
h
21
for volume correction (case B) and
96.3 +6.7 (7 %) Bq kg
21
h
21
for back diffusion cor-
rection (case C). From these values the significance
of the correction used can be estimated. Ideally, if
there is no correction, E
M
and consequently Ra
Eff
Figure 2. Areal exhalation rate at different sample weights.
Figure 3. Mass exhalation rate at different sample weights.
M. ABO-ELMAGD AND M. M. DAIF
548
by guest on July 29, 2010rpd.oxfordjournals.orgDownloaded from
should be the same for all experimental arrange-
ments. The higher uncertainty in E
M
after correcting
the result for back diffusion is from the cases when
the sample weight is between 150 and 200 g. This is
because the back diffusion correction fails; it is only
valid if the thickness of the sample is less than the
diffusion length
(13)
.
The above-mentioned active measurements of
radon-related parameters were used to calibrate the
CR-39 detector using the sealed cup configuration.
To check the correlation between the active and
passive technique, Figure 4shows the relationship
between the rate of track density registration on the
CR-39 detector and the sample weight. The behav-
iour is approximately similar to that obtained in
Figure 1for the measured activity. The difference is
that the volume correction (case B) and the back dif-
fusion correction (case C) have approximately the
same values, for this reason plotting the curve of the
back diffusion correction is sufficient.
This figure assures the correlation between active
(Lucas cell) and passive (CR-39) measurements, and
consequently, one can get a reliable calibration of a
CR-39 detector.
The calibration factor of a CR-39 detector K
(track cm
22
per Bq m
23
d) can be calculated from
the following equation:
K¼DV
Qt1e
l
t
l
ð7Þ
D(track cm
22
) is the track density on CR-39,
V(m
3
) is the sealed cup volume (case A) or the
volume of the air space of the sealed cup (cases B
and C), Q(Bq) is the sample activity as calculated
from equation (2) and t(d) is the exposure time of
CR-39, which starts from sealing the detector with
sample to its removal. During this exposure period,
the detector accumulates the tracks from the buildup
radon concentration.
The variation of the calibration factor (K) with
the sample weights Mis shown in Figure 5for
uncorrected and corrected (for back diffusion) cases.
For uncorrected results (case A), Kincreases with
Mup to 150 g followed by a drop with the average
value of Kequal to 0.27 +0.02 (6.5 %) track cm
22
per Bq m
23
d. Up to 150 g, the relationship is linear
and therefore each sample weight has its own CR-39
calibration factor, which is difficult to use practically.
For back diffusion correction (case C) Kis fluctu-
ated around 0.237 track cm
22
per Bq m
23
d fol-
lowed by hard breakdown for Mlarger than 150 g.
This breakdown introduces high uncertainties in K
values, for this reason the data should be limited to
the sample weight up to 150 g in the used sealed cup
to reduce the uncertainty.
Up to a 150 g sample weight, the average value of
Kis equal to 0.237 +0.002 (1 %) track cm
22
per Bq
m
23
d. This value of calibration factor can be used
when back diffusion effect is applied.
Table 1lists the Kvalues for uncorrected and cor-
rected results at different sample volume to the total
sealed cup volume (V
s
/V) for sample weight up to
150 g. From this table, it is recommended to use V
s
/V
up to 0.333, i.e. sample volume is equal to one-third
Figure 4. The rate of registered tracks as a function of
sample weights.
Figure 5. CR-39 calibration factor as calculated for
different sample weight for uncorrected and corrected
results, the dashed line represents the limit of the sample
weight for the used sealed cup, which is equivalent to
V
s
/V¼0.333.
CALIBRATION OF CR-39 FOR RADON USING SEALED CUP TECHNIQUE
549
by guest on July 29, 2010rpd.oxfordjournals.orgDownloaded from
the total sealed cup volume. In this limit, Kfor back
diffusion correction is realistic and has low
uncertainty.
Using the calibration factor corrected for back dif-
fusion, the sample activity can be determined using
equation (7) and then radon-related parameters can
easily be determined using equations from (3) to(5)
after replacing
l
by
l
*.
CONCLUSION
Active and passive measurements of radon-related
parameters are well correlated and so the calibration
of a CR-39 detector across a Lucas cell is reliable.
Correcting the results for the back diffusion effect
will reduce the measured effective radium content,
mass and areal exhalation rates. Any increment in
sample weight more than 150 g has no effect in the
measured parameters and so it is unsuitable to use
the sealed cup technique in this configuration. The
obtained calibration factor for CR-39 detectors has
been found to be reliable up to a 150 g sample
weight. The calibration factor has an uncertainty
around 1 %. To generalise the results for other con-
figurations, the condition for using the obtained cali-
bration factor in terms of sample and cup volumes
should be V
s
/V0.333.
REFERENCES
1. Munazza Faheem and Matiullah. Radon exhalation
and its dependence on moisture content from samples of
soil and building materials. Radiat. Meas. 43,
1458–1462 (2008).
2. Bossew, P. The radon emanation power of building
materials, soils and rocks. Appl. Radiat. Isot. 59,
389–392 (2003).
3. Singh, A. K., Sengupta, D. and Prasad, R. Radon
exhalation rate and uranium estimation in rock samples
from Bihar uranium and copper mines using the SSNTD
technique. Appl. Radiat. Isot. 51, 107–113 (1999).
4. Al-Jarallah, M. I., Abu-Jarad, F. and Fazal-ur-
Rehman. Determination of radon exhalation rates from
tiles using active and passive techniques. Radiat. Meas.
34, 491–495 (2001).
5. El-Bahi, S. M. Assessment of radioactivity and radon
exhalation rate in egyptian cement. Health Phys. 86(5),
517–522 (2004).
6. Kumar, R., Mahur, A. K., Sengupta, D. and Rajendra
Prasad. Radon activity and exhalation rates measure-
ments in fly ash from a thermal power plant. Radiat.
Meas. 40, 638– 641 (2005).
7. Shafi-ur-Rehman, Nadia Imtiaz, Munazza Faheem,
Shakeel-ur-Rehman, Matiullah. Determination of
238
U
contents in ore samples using CR-39-based radon dosi-
meter—disequilibrium case. Radiat. Meas. 41, 471– 476
(2006).
8. Chau, N. D., Chrusciel, D. and Prokolski, L. Factor
controlling measurements of radon mass exhalation rate.
J. Environ. Radioact. 82, 363– 369 (2005).
9. Pylon Instruction. Instruction manual for using Pylon
Model 110A and 300A Lucas cells with the Pylon AB-
5. Pylon Electronic Company, Canada (1989).
10. Singh, H., Singh, J., Singh, S. and Bajwa, B. S. Radon
exhalation rate and uranium estimation study of some
soil and rock samples from Tusham ring complex, India
using SSNTD technique. Radiat. Meas. 43, S459– S462
(2008).
11. Kovler, K. Radon exhalation of hardening concrete:
monitoring cement hydration and prediction of radon
concentration in construction site. Environ. Radioact.
86, 354–366 (2006).
12. Jang, M., Kang, C.-S., Hyun and Moon, J. Estimation
of
222
Rn release from the phosphogypsum board used in
housing panels. Environ. Radioact. 80, 153– 160 (2005).
13. Samuelsson, C. The closed-can exhalation method for
measuring radon. J. Res. Natl. Inst. Stand. Technol. 95,
167–169 (1990).
Table 1. Uncorrected and corrected (for back diffusion)
calibration factor for different sample volume to the total
cup volume.
V
s
/VK(track cm
22
per Bq m
23
d)
Uncorrected Corrected for back diffusion
0.056 0.2470 0.2404
0.111 0.2544 0.2397
0.167 0.2600 0.2367
0.222 0.2682 0.2351
0.278 0.2791 0.2353
0.333 0.2945 0.2348
Average 0.267 0.237
Uncertainty 6.5 % 1 %
M. ABO-ELMAGD AND M. M. DAIF
550
by guest on July 29, 2010rpd.oxfordjournals.orgDownloaded from
