Radiation Protection Dosimetry Advance Access published May 15, 2015 Radiation Protection Dosimetry (2015), pp. 1–4

doi:10.1093/rpd/ncv316

CALIBRATION SYSTEM FOR RADON EEC MEASUREMENTS Y. A. M. Mostafa1, M. Vasyanovich1,2, M. Zhukovsky1,2,* and N. Zaitceva1 1 Ural Federal University, Ekaterinburg, Russia 2 Institute of Industrial Ecology UB RAS, Ekaterinburg, Russia *Corresponding author: [email protected]

INTRODUCTION Radon is the most important source of natural radiation exposure. The inhalation of short-lived radon progeny in ambient atmosphere is responsible on internal exposure and can lead to lung cancer(1 – 3). Therefore, the measurement of radon and radon progeny concentrations is very important for dose evaluation. Due to these facts, measurements of the activity concentration of radon and its progenies are performed worldwide either at workplace (e.g. mines) or in dwellings. So, it is necessary to have the calibration facilities in which the activity concentration of radon and its progenies can be measured under welldefined conditions. Improving of radon metrology also is necessary for the traceability of secondary measurements of radon or radon progeny concentration in air. Nowadays, a large number of commercial radon progeny monitors were developed and widely applied in environmental survey. For quality assurance on measurement, a large number of reference radon chambers were built for establishing radon and radon progeny standard in addition to assessing the quality of those devices and instruments as well(4). Unfortunately, the possibility of checking equipment is not available permanently. Therefore, there is a necessity to develop metrology approaches for measuring the concentration of radon progeny using available equipment and simple procedures to interpret the data. Each measuring method requires a suitable calibration procedures and standards for accurate measurements. Radon progeny concentration usually can be expressed either by their individual concentration or by their equivalent equilibrium concentration (EECRn). In this paper, without exception, the radon progeny concentration means the equivalent equilibrium concentration of radon progeny, EECRn.

Usually, the most preferable standard measuring methods for the measurements of Rn progeny are alpha radiometry and alpha spectrometry. Nevertheless, the need of correction of the results of the measurements on the self-absorption in filter and the effectiveness of registration of alpha particles with different energies decreases the accuracy of the measurements. Therefore, creation of standard calibration system requires another independent method for the measuring of Rn progeny. To the authors’ opinion, such independent method like gamma spectrometry can be used for the determination of the precision values of EECRn. MATERIALS AND METHODS Investigated atmosphere from equilibrium radon box with volume 2 m3 containing high activity of radon (EECRn from 2000 to 6000 Bq m23) was pumped through three aerosol filters by a sampling pump during 10 min. Three filters were used to collect all aerosol activity, and total deposition of aerosols on filters was 99.95 %. Standard AFA-RMP-20 perchlorovinyl filters with surface density of 3.5 mg cm22 were used. The sampling rate y was in the range from 13 to 17 l per minute and was monitored in each experiment separately. Each filter was separately measured on alpha radiometer with semiconductor surface barrier detector. After that, all three filters were simultaneously measured on high-purity germanium (HPGe) gamma spectrometer. Measurements of alpha activity were carried out to confirm the results obtained by gamma spectrometer. The purpose of all measurements was not to identify individual activity values of individual decay products of radon (214Po, 214Pb and 214 Bi), but to measure EECRn. Alpha activities on the filters were consequently measured by alpha radiometer after 30 min from the

# The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]

Downloaded from http://rpd.oxfordjournals.org/ at University of Waterloo on October 16, 2015

The measurement of radon equivalent equilibrium concentration (EECRn) is very simple and quick technique for the estimation of radon progeny level in dwellings or working places. The most typical methods of EECRn measurements are alpha radiometry or alpha spectrometry. In such technique, the influence of alpha particle absorption in filters and filter effectiveness should be taken into account. In the authors’ work, it is demonstrated that more precise and less complicated calibration of EECRn-measuring equipment can be conducted by the use of the gamma spectrometer as a reference measuring device. It was demonstrated that for this calibration technique systematic error does not exceed 3 %. The random error of 214Bi activity measurements is in the range 3– 6 %. In general, both these errors can be decreased. The measurements of EECRn by gamma spectrometry and improved alpha radiometry are in good agreement, but the systematic shift between average values can be observed.

Y. A. M. MOSTAFA ET AL.

end of sampling. For selected intervals of the measurements, the measured alpha activity corresponds to the decay of 214Po in equilibrium with 214Bi. The measuring time was 100 s for each side of filters (front and back) to estimate self-absorption in the filter(5). The alpha activity on the filter was calculated as follows: A¼

Ifront þ Iback 1  ; 1 þ ðI0;filt =I0 Þ 1

ð1Þ

† † †

radon progeny concentrations remain constant during sampling period; sampling rate, collection efficiency of the filter and counting efficiency remain the same during measurements; environmental parameters have no influence on measuring process.

RESULTS AND DISCUSSIONS Estimation of systematic and random errors

EECRn ¼

Nðts ; T1 ; T2 Þ 1  h  y  Kkuz

ð2Þ

where N is the number of counts accumulated in the photo peak during gamma measurement (net counts minus background), 1 is the detector efficiency, h is the transition probability of measured gamma line of 214Bi (for 609.3 keV, h ¼ 0.46), y is the flow rate of pumping and Kkuz is the Kuznets coefficient depending on time of sampling tS and time of the beginning and finishing of the measuring T1 and T2, respectively, counted from the time of the finishing of the sampling. It was assumed that:

For any radiation standard measuring systems, both systematic and random errors should be considered. Random error is generally associated with the estimation of the count rate of measured radon progenies (214Po and 214Po for alpha measurements or 214Pb and 214Bi for gamma measurements), but systematic errors are associated with the errors introduced by the instrumentation and measurement technique. In the authors’ case, the systematic errors include: † †

† †

error due to variations in equilibrium shift in radon box influencing the Kuznets coefficient calculation; error in determining the activity of nuclides using polynomial approximation of registration efficiency of the HPGe detector especially in the range of radon progeny emission lines; error in determining the detection efficiency of alpha radiation radiometer; error in flow rate measurement by airflow meter.

The first stage of calculations to determine the EECRn systematic errors of measurement is the determination of radioactive equilibrium in the measurement reference box. To determine this shift, special measurements by the method of Thomas(7) were performed and the values of the concentrations of 218Po, 214 Pb and 214Bi calculated. It was obtained that the equilibrium shift during all set of experiments was 218Po:214Pb:214Bi ¼ 1:(0.26+0.08):(0.10+0.06). From these data, it can be concluded that the error caused by the influence of the equilibrium shift on the value of Kuznets coefficient is in the range Deq sys ¼ +2  3 %: This error can be reduced by controlling the equilibrium shift during each calibration experiment by simultaneous separated collecting of radioactive aerosols on the filters. The polynomial dependence of efficiency HPGe detector was determined using standard calibration sources 241Am, 133Ba and 152Eu with precisely known activity and degree of activity uncertainty. The obtained energy-dependent polynomial efficiency curve of absolute full-energy peak of HPGe gamma spectrometer is shown in Figure 1. The absolute detection efficiency for 214Bi line (609.3 keV) was found to be 8.65`  1023 +3 % for

Page 2 of 4

Downloaded from http://rpd.oxfordjournals.org/ at University of Waterloo on October 16, 2015

where Ifront and Iback represent count rate from the front and back sides of aerosol filter, respectively; 1 denotes the detector efficiency; I0 is count rate from calibration source with the energy of alpha particles corresponded to the measured nuclides and I0,filt is count rate from calibration source covered by aerosol filter. Such measuring technique allowed taking into account the self-absorption of alpha particles in the filter without any assumption about the activity distribution in the bulk of filter. The calibration source was prepared by exposure of polished aluminium disc in atmosphere with high radon concentration during 24 h. In this case, the significant surface activity of deposited radon progeny was observed on the disc surface. After waiting for 30 –40 min, the surface activity of 218Po decays and the alpha activity of calibration source formed only 214Po in equilibrium with 214 Bi. After the measurements, equilibrium equivalent concentration of radon for each filter on the moment of finishing of the sampling was calculated by the modified Kuznets method(6) with the correction on the self-absorption in filters. Gamma activity on the filters was consequently measured after 50 –70 min from the end of sampling during 30 min. The measurements of gamma activity count rate of 214Bi (609.3 keV) in the three aerosol filters using HPGe detector were conducted. Equilibrium equivalent concentration of radon, taking into account the time shift between the beginning of gamma activity measurement and the end of sampling activity on filter, was calculated as follows:

CALIBRATION SYSTEM FOR RADON EEC MEASUREMENTS Table 1. Systematic and random errors values of gamma spectrometry and alpha radiometry. Type of measurement Systematic errors

Alpha radiometry

Gamma spectrometry

Deq sys ¼ +2 %

Deq sys ¼ +2 %

Dfrsys ¼ +1 %

Dfrsys ¼ +1 %

D1sys1 ¼ +4:3 %

D1sys ¼ +0:4 %

D1sys2 —unknown

Dsa sys ¼ 0 %

Dsa sys2 —unknown Figure 1. Polynomial fitting of efficiency of HPGe spectroscopy using standard sources (241Am, 133Ba and 152 Eu).

the 95 % confidence interval. The uncertainty in this detection efficiency was obtained by the least-squares method using the experimental data of the energy dependence detection efficiency in Spectra Line Ultimate Gamma Lab program. The quality of fitting this polynomial efficiency was directly tested by the measurement of the activity of 137Cs standard source, Eg ¼ 661 keV (close to the energy line of 214Bi, 609.3 keV). The difference between the measured value and passport value corrected on radioactive decay was 0.4 %. This value was considered as systematic uncertainty D1sys of gamma measurements in the range of energies from 600 to 700 keV. For the comparison purposes, there is the need to know the systematic errors of detection efficiency of alpha-radiation radiometer. Alpha radiometer calibration was performed using a set of standard alpha sources 239Pu with different values of activities. The error in determining the detection efficiency was calculated as follows:

d1sys ¼

Pn

i¼0 ð1i  1av Þ ; ðn  1Þ  1av

ð3Þ

where 1i is the detection efficiency for the ith sample, 1av is the average value of the detection efficiency (1av ¼ 0.30) and n ¼ 5 is the number of measurements (sources). The obtained uncertainty for alpha radiometry efficiency is D1sys1 ¼ 1:96: d1sys ¼ 4:3 % at 95 % confidence level. It should be noted that additional systematic error can exist due to difference between the energy of standard source 239Pu (Ea ¼ 5.15 MeV) and the energy of 214Po (Ea ¼ 7.69 MeV). Moreover, the detection efficiency can be different for the alpha particles with different energies due to loss of energy in the filter.

Random errors

Dar ¼ +2:4 %

Dgr ¼ +6:3 % (comparison experiments) Dgr ¼ +3:3 % (working value)

Also, the systematic error of the coefficient used for taking into account the self-absorption of alpha particles in aerosol filter should be considered. By the authors’ estimates, the systematic error of correction coefficient 1/(1 þ I0,filt/I0) (Equation 1) obtained by the measurements of absorption of alpha particles in the filter is Dsa sys1 ¼ 2:5%: It should be noted that Equation 1 for the estimation alpha activity of the filter taking into account self-absorption was obtained assuming linearity of alpha particle trajectories in filter material(5). Therefore, the systematic error of alpha activity measurements Dsa sys2 caused by this assumption can exist, and one cannot calculate or measure the value this error. The last source of systematic error from the measurements of air sampling rate according to air flow meter passport not exceeds Dfrsys ¼ 1 %: In gamma measurements, uncertainty level of measured net count (random error) is nearly 4–8 %. These values of uncertainty depend on the time after the end of sampling and beginning of activity measurement. The average value of the random error for the time interval between the end of sampling and start measuring 50–70 min is +6.3 %. Net count uncertainty (random error) value can be reduced to Dgr ¼ +3:3 %; if the measurement of gamma spectrometric will be carried out not later than 30 min after the end of sampling. Due to higher effectiveness of the registration of alpha particles and lesser time between the end of the sampling and the beginning of the measurements, the mean value of random measurement error of the alpha radiometer is Dar ¼ +2:4 %: The results obtained by the two different techniques were used to calculate and compare their total uncertainties. These results are given in Table 1.

Page 3 of 4

Downloaded from http://rpd.oxfordjournals.org/ at University of Waterloo on October 16, 2015

Dsa sys1 ¼ +2:5 %

Y. A. M. MOSTAFA ET AL. Table 2. The results of parallel EECRn measurements, Bq m23. Alpha radiometry

Ratio of alpha/gamma measurements

2350 3160 3250 2390 1940 2560 1960 3050 2410 2020 2290 5080 6150 5470 4010

0.82 0.94 0.89 0.82 0.89 0.89 0.79 0.88 0.75 0.85 0.91 0.85 0.82 0.80 0.83

CONCLUSIONS

Uncertainties associated with the uncertainties of equilibrium shift in radon box and air flow rate sensor error effect on results both gamma and alpha measurements. Calibration measurements of EECRn are a direct single measurements and systematic, random and total errors should be assessed properly. The total systematic error can be defined as follows: qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 1 2 fr 2 sa 2 Dsys ¼ +K ðDeq sys Þ þ ðDsys Þ þ ðDsys Þ þ ðDsys Þ ; ð4Þ where K ¼ 1.1—correction factor determined for the confidence level of p ¼ 0.95. For known and assessed sources of systematic errors, the overall errors are Dgsys ¼ 2:5 %; Dasys ¼ 6:0 %: It can be observed that the systematic errors for alpha measurements are bigger than those for gamma measurements in spite of the fact that the authors did not consider the unknown errors D1sys2 and Dsa sys2 : Comparison of the results obtained by gamma spectrometry and alpha radiometry The set of 15 parallel EECRn measurements by alpha radiometry and gamma spectrometry was conducted. The results are presented in Table 2. A good correlation between the two types of measurements was observed, but the systematic shift between the results also presented. The average ratio between the measurements conducted by alpha radiometry and gamma spectrometry for 95 % confidence level was as follows:   EECaRn ¼ 0:84 + 0:10: ð5Þ EECgRn

The different kinds of activity measurements can be used for the control of EECRn value in calibration system. In all calibration systems working as a primary standard, the activity control should be conducted by absolute measurement techniques. Traditionally, for the measurement of radon progeny concentration in the air, the alpha radiometry or alpha spectrometry are used, but the need to take into account the self-absorption of alpha particles in aerosol filter decreases their applicability at absolute measurements. To avoid the need of the calculation of self-absorption, the prototype of EECRn calibration system is created on the base of HPGe gamma spectrometer. The results of determining EECRn by gamma spectrometry are consistent with the improved method of alpha radiometry but have fewer sources of error. The observed systematic error between the alpha and gamma activity measurements can be explained by the influence of self-absorption effects of alpha particles in the aerosol filter. Developed prototype allows obtaining the levels of systematic error of +2.5 % and the random error of +3.3 %, which correspond to the required accuracy of the national standards of EECRn. The developed system can be recommended as a national or regional standard of EECRn. The intercomparisons of developed system with Russian national and international centres of standardisation are planned.

REFERENCES 1. World Health Organization. WHO Handbook on indoor radon: a public health perspective. WHO, Geneva (2009). 2. UNSCEAR. Effects of ionizing radiation. Volume 2, Annex E. Sources-to-effects assessment for radon in homes and workplace. UNSCEAR, Report. UN (2006). 3. UNSCESR. Sources and effects of ionizing radiation UNSCEAR. Report. UN (2000). 4. Kadir, A., Zhang, L., Guo, Q. and Liang, J. Efficiency analysis and comparison of different radon progeny measurement methods. Scientific World J. 2013 (2013), Article ID 269168, 6 pages. 5. Budyka, A. K. and Borisov, N. B. Fiber Filters to Control Air Pollution. IzdAt (2008) (in Russian). 6. Nazaroff, W. W. An improved technique for measuring working level of radon daughters in residences. Health Phys. 45, 509–523 (1980). 7. Thomas, J. W. Measurement of radon daughters in air. Health Phys. 23(6), 783–789 (1972).

Page 4 of 4

Downloaded from http://rpd.oxfordjournals.org/ at University of Waterloo on October 16, 2015

1940 2970 2880 1950 1720 2280 1540 2700 1800 1710 2080 4340 5050 4390 3340

Gamma spectrometry

The observed difference cannot be described by known and estimated systematic errors Dgsys and Dasys : In the authors’ opinion, such systematic bias can be explained by the additional influence of the systematic errors D1sys2 and Dsa sys2 with unknown values.

Calibration system for radon EEC measurements.

The measurement of radon equivalent equilibrium concentration (EECRn) is very simple and quick technique for the estimation of radon progeny level in ...
134KB Sizes 0 Downloads 11 Views