W T RADON EMANATION STANDARD - SRM 4968 THEORETICAL AND PRACTICAL ASPECTS OF IT'S APPLICATIONS Paul Kotrappa Rod Bee,Inc. Frederick, MD ABSTRACT Several years of research in developing a radon emanation standard by the National Institute of Standards and Technology (NIST) has led to NIST Radon Emanation Standard SRM 4968 (Journal of Research of the MST 100:629-639; 1996). This new standard is based on a polyethylene -encapsulated ""Ra solution that has been demonstrated to emanate a well-characterized and known quantity ^"Rn when employed in "accumulation mode*. The encapsulated standard is intended to serve as a more convenient, easier-to-use, alternative to the conventionally employed "^Ra solution standards that have been disseminated by MIST for 5 emanation measurements for the past 40 years. The latter standards were certified only for ^Ra content only, whereas the new standards are certified in termsof two parameters, both the ""Ra content and the "^Rn emanation fractions. Development of this standard is considered as a land mark in the standardization of radon measurement. The ease with which the source can be used lends itself to a wide spread use by any quality conscious radon measurement laboratory. These emanation standards are already in use in the USA, Europe, Canada, and Mexico. Their use is expected to grow. Hie purpose of this presentation is to describe (1) the standard and its characteristics, (2) theoretical and practical aspects, (3) its application for routine calibration of continuous radon monitodintegrating monitors used in indoor radon measurements, and (4) the precautions needed for its use. - Pennisswnfor reproduction of this paper is being obtained from the publisher: Nuclear Tecfmology Publishing. - 1996 International Radon Symposium UP 2.1 TECHNICAL NOTE ~ a d i i~relection ~ i D<Èimu Vol. 55. Na 3. pp. 21 1-218( 1994) Nuclear Technology Publishing APPLICATION OF NIST ^Rn EMANATION STANDARDS FOR CALIBRATING ^Rn MONITORS P. Kotrappa and L. R. Stieff Rad Elec Inc., 5714-C. Industry Lane Frederick, MD 2 1701. USA Received January 21 1994, Amended May 4 1994. Accepted May 10 1994 AbstractÑTh NIST (National Institute of Standards and Technology) has recently made available '"Rn emanation standards. The NIST certified parameters include the '*'Ria strength and the emanation coefficient. When such a source is loaded into a leak tight enclosure of a known volume. '"Rn accumulates over time and it is possible to calculate precisely both the '"Rn concentration after any accumulation time and also the time integrated average '"Rn concentration after any given accumulation time. Equations needed for calculations are given. For example, i f a 25 Bq (676 pCi) NIST source is loaded into a jar with an air volume of 3.72 litres, the time integrated average "'Rn concentration is 973 Bq.m" (26.3 pCi.1"') after an accumulation time of two days. If the "'Rn detector placed in the jar is non-radon-absorbingand a m e integrator, i t must yield the theoretically predicted results. If not. on appropriate calibration correction factor can be calculated The paper describes a study involving 34 randomly chosen commercially available E-PERMtR)(Electret-Passive Environmental Radon Monitor) and 17 NIST sources in 17 different calibration (accumulator) jars. The study indicated that E-PERMt% give results within about 5% of the predicted results. The study also includes a study of continuous radon monitors. The data obtained show that. i n practice, NIST sources can be used to calibrate both passive and active continuous '%I monitors. The availability o f NIST sources with a precisely known ^^Ru emanation characteristic is a major advance in radon metrology. Practical problems in (he successful use of these sources are discussed. INTRODUCTION The NIST (National Institute of Standards and Technology) has recently made available ^Rn emanation standards*'). The NIST certified parameters include the "'Ra strength and the emanation coefficient. When such a source is loaded into a leak tight jar of a known volume 2aRn will accumulate over time. It is possible to calculate precisely the time integrated average radon concentration after any given accumulation time. If the "'Rn detector present in the jar is non-radon-absorbing and a (me integrator, this radon detector must yield the theoretically predicted results. If there is some consistent difference, suitable NIST traceable calibration corrections can be derived. The current study involves 34 randomly chosen E-PERMqR)S*and 17 NIST sources in 17 different calibration jars. The purpose of the study is to show how a NIST source can be used in practice. The availability of NIST sources with precisely known radon emanation characteristics is considered to be a major advance in radon metrology. MATERIALS AND METHODS The recently developed ^"Rn emanation standards that are based on polyethylene-encapsulated ^Ra solutions(u are used in the present study. These sources *E-PERM1') is a registered trade mark of the electret ion chamber system patented and manufactured by Rad Elec Inc., 57144 Industry Lane. Frederick. MD 21701. USA. are certified by NIST in terms of 226Racontent and the ^Rn emanation fraction. These have been demonsinned to emanate a well characterised and known quantity of "'Rn when employed in 'accumulation mode'('*. A total of 17 such sources with "'Ra content ranging from about 4.5 Bq to about 450 Bq with an emanation fraction of 0.890 were made available by NIST for this study. Specially prepared glass jars with a capability of being made leak tight were used as accumulators (see Figure 1). Such glass jars were previously used for the measurement of ^Rn in water"'. The %n detectors used were the standard commercially available EPERM"11 210 ml volume 'S' chambet*'-** configured with either short term (ST) or long term (LT)electrets. Ñ^^^^^^Q Radon leak-tlaht IId Temperature strip €-PE radon monitor Figure 1. Experimental arrangement. Detectors used: two SST or SLT E-PERMS. - 1996 International Radon Symposium IIP 2.2 P. KOTRAPPA and L R. STIEFF given by Equation I since the chamber starts with 0 concentration and builds up to a maximum: THEORETICAL CALCULATIONS OF ^Rn IN AN ACCUMULATOR Equation I gives the ^Rn concentration in the accumulator at any time TA after the start of accumulation. This is the maximum '"Rn concentration seen by the detector in the accumulator. The ^Rn concentration varies from zero at the start of the experiment to the concentration given by Equation 1 at the end of the accumulation lime of TA days: ARn = f AR, ( 1 - exp(-ARnTA)) vA (1) where An, is the '"Ra content of the NIST source. Ann is the ^Rn concentration after an accumulation time of TA days. f is the emanation coefficient of NIST source, ARn is the decay constant of ^=Rn, VA is the air volume of the accumulator. The time integrated concentration of ^Rn is obtained by integrating Equation 1 from time zero to TA. Further, if the time integrated "'Rn is divided by the accumulation time TÈwe arrive at the average concentration over the time TAPCan (Equation 2 gives the result). This is always smaller than the maximum concentration ( - 1 - expi-b.TA)) ÈRnT f C" 'VA (2) The "*Ra concentration is in Bqunits, f is the emanation coefficient and is determined by NIST as 0.890, the volume of the jar should be in units of cubic metres (3720 ml = 0.00372 m3), the time is in units of days and the decay constant of radon is in units of day"' (0.1812). then the "'Rn concentration is in units of Bq.m"'. This is divided by 37 to get the concentration in pCi.1-I. AIR VOLUME OF THE ACCUMULATOR The air volume of the accumulator depends upon the number and the type of E-PERMR*used in the accumulator jar. The different air volumes to be used are given in Table 1. These were determined experimentally by carefully measuring the volume of the empty jar and the volume of the jar when one, two or three 'S' chambers were loaded into the jar. The empty accumulator jars were filled with distilled water and the water needed to Table 1.Air volume of the accumulator when different fill the jar weighed. Measurements were then made by numbers and types of E-PERMS are used in the accumu- loading the accumulator jar with different numbers of 'S' chambers. The difference in the weight of the water lator. between the two measurements gives the air volume of the jars, when the respective number of detectors were Number Air volume E-PERM loaded. Experiments were repeated at least 10 times and of units (ml) type the values reported in the table have an error of less than 3%. Similar measurements were made for one, two, SST or SLT three and four 'L' chambers. The volume of the empty SST or SLT jar is 3996 ml and the air volume occupied by one SST or SLT SST or SLT E-PERM is 123 ml and one LST or LLT LST or LLT LST or LLT E-PERM is 47 ml. LST or LLT Depending upon how many E-PERMS are inside the LST or LIT accumulator, the appropriate air volume (Vi) given by Table 1 should be used. For example, if 2 SST (or SLT) are used in the accumulator, the appropriate value for VA to be used is 3720 ml. See Table 1 for the appropriate value to be used when a different number of detectors are used in the accumulator. Figure 2 is a graphical representation of Equations 1 and 2. The top curve gives the maximum ^ R n concentration seen by the detector (Equation 1) and the lower curve gives the time averaged concentration (Equation 2) seen by the detector. If the detector is a true integrator, the response should be in accordance with the lower curve. If a continuous radon monitor is inside the accumulator, it should track the upper curve. Figure 2. "'Rn concentration in a 3.72 litre accumulator. Detectors used: two SST E-PERMS. Source: NIST source with 25 Bq "'Ra and f of 0.890. ( a ) radon concentration. (+) average radon concentration. EXPERIMENTS AND RESULTS Each source was fixed inside each jar using the clip arrangement. The NIST serial number was marked on each jar. NIST instructions require $at the source - 1996 International Radon Symposium IIP 2.3 M5T STANDARDS FOR CALIBRAT/NG ^Rn MONITORS should be open to the atmosphere for at least 24 h prior to the use of the source in the accumulator mode'". This boundary condition was achieved by keeping the lid of the jars open for at least 24 h. The jars were also left in a low radon area (outside environment) so that the contribution from the residual radon in the jar is minimised before starting the accumulation. A pair of premeasured E-PERMS (initial voltage IV) were loaded into each jar. lids closed, (he collars tightened and the time noted (please see Figure 1). This is the start of the experiment After the desired specific accumulation period. the E-PERMS were removed from (he jar and left in a low radon area in the 'on' position for a period of 3 h. Only after this delay was the final voltage reading taken (FVJ. In the normal use of E-PERMS the final reading is taken immediately after the termination of a measurement. The reasons for taking a delayed final reading in the present case warrant additional discussion. In the accumulator mode the radon concen(ration changes from zero at the beginning to a maximum concentration at the end of (he experiment. The ^"Rn and associated progeny signals at the beginning and at the end are very different. The E-PERMS are usually calibrated in a radon calibration chamber with a steady "'Rn concentration"-*. Boundary conditions are not the same in the accumulator mode and in steady state modes. The end effect in accumulation mode is compensated by leaving the E-PERMS in a low radon area for about 3 h so that the contribution from the deposited decay products is taken into account. By following (his procedure it has been found that the accumulation periods can be as short as 6 h. The "'Rn concentration was calculated by the standard pr~cedure"-~~ using the calibrations based on steady state radon concentration. It is important to note that (he actual accumulation time is used in the calculation and this does not include (he3 h delay time. Results are listed in Tables 2 to 6. In these tables, column 2 gives the source strength and column 3 gives the theoretically calculated average concentration in the Table 2. Experimental results. Detectors used: two SST E-PERMS. Accumulation time 3.07 days. Source CP5 CP5 CP6 CP6 CP7 CP7 CP27 CP27 CP28 CP28 CP30 CP30 CP31 CP3 1 CP32 CP32 CP33 CP33 CP35 CP35 CP36 CP36 CP37 CP37 CP38 CP38 CP4 1 CP4 1 CP42 CP42 CP44 -44 blank blank NS NS (Ra Bq) (I3q.m-') El W Fv: EP EP-B (Bq.m-9 (EP-B)/ Av. NS ratio 0.898 0.897 1.075 0.962 0.984 1.023 1.029 1.152 1.183 0.980 1.096 I .040 1.056 1.014 0.945 0.944 0.88 1 0.995 0.976 0.843 1.038 I .022 0.914 1.001 1.018 1.OX 0.892 0.848 0.881 0.903 0.937 0.997 SK6439 SK47O1 SK4697 SKI 126 S13944 SK3612 SK1301 SK4734 SJ6073 SK4733 SK3574 SK3786 SKI255 SK4722 SK6435 SK6412 SK6446 SKI118 SKI217 SK4696 SK3601 SK3533 SK6405 SK4715 SJ3779 SK3728 SK4699 SK6451 SK3682 95778 S13986 SK3768 S11457 SI14I9 Av. % STD - 1996 International Radon Symposium IIP 2.4 L R. STIEFF accumulator. Column 7 gives the average concentration of 4 to 5 Bq "'Ra. Because the usable range of electret as measured by E-PERMS, Column 8 gives the net E- surface voltage is from 700 to 200 V, electrets of differPERM value after subtracting the blank readingB' and ent surface voltages were used to verify their performcolumn 9 gives the ratio of the concentration measured ance over the entire usable voltage range. Table 3 is a by E-PERM and the theoretically expected concen- repeat of the experiment, using the same detectors and tration. Column 10 gives the average ratio of the two same sources. Table 4 gives the results of experiments detectors in the same accumulator. The grand average with stronger sources (33 to 460 Bq). Table 5 and Table ratio and the associated standard deviation is given at 6 give results using SLT E-PERMS. Results indicate that, on average, SST devices give the end of column 10. To obtain the blank reading, two E-PERMS were introduced into an empty jar similar to results nearly 3% lower than (he NIST predicted values. the one used for experiments. These E-PERMS were Similarly, on average, SLT devices give results nearly exposed for the same time that was used for the experi- 6% lower than the NIST predicted values. This is within the accuracy expected from E-PERMS~'.~'.However, menial ones. individual E-PERMS can vary by wider margins. For very accurate measurements, individually calibrated DISCUSSION OF RESULTS E-PERMS can be used (please see later section). The Table 2 gives the results for a 3.07 day accumulation. procedure developed in this study is demonstrated to Two SST E-PERMSwere used. NIST sources used were be usable for different accumulation times, for different P. KOTRAPPA and Table 3. Experimental results. Detectors used: two SST E-PERMS. Accumulation t h e 3.00 days. Source CP5 CP5 CP6 CP6 CP7 CP7 CP27 CP27 CP28 CP28 CP30 CP30 CP31 CP31 CP32 CP32 CP33 CP33 CP35 CP35 CP36 CP36 CP37 CP37 CP38 CP38 CP40 CP40 CP4 1 CP4 1 CP42 CP42 CP44 CP44 blank blank NS NS (RaBq) (Bq.m") El IV FV, EP (Bq.m-I) EP-B (EP-B)/ NS 0.934 0.886 1.079 0.98 1 0.965 I .008 1.014 1.103 1.219 1.000 1.139 1.O8O 1.063 0.994 0.947 0.988 0.799 0.968 0.975 0.832 1.O87 1.024 0.910 0.98 1 1.038 1.O37 0.90 1 1.100 0.842 0.898 0.870 0.893 0.960 1.a24 SK6439 SK47O1 SK4697 SKI 126 $13944 SK3612 SKI301 SK4734 SJ6073 SK4733 SK3574 SK3786 SK I255 SK4722 SK6435 SK6412 SK6446 SKI118 SKI217 SK4696 SK3601 SK3533 SK6405 SK4715 SJ3779 SK3728 SK6434 SK6415 SK4692 SK6422 SKI253 SK6448 SI3986 SK3768 SI1457 SI1419 GRD A % STD 214 - 1996 International Radon Symposium IIP 2.5 Av. ratio N1ST STANDARDS FOR CALIBRATING '"Rn E-PERM configurations and for sources o f different strengihs. E-PERMS were found to function as truly integrating "Qn measuring devices, integrating radon concentrations from near zero to the highest (17.500 Bq.me3) encountered in the study. Before using a certain configuration of E-PERMwith a source of certain strength, one should do a theoretical analysis so as not to discharge the electret beyond the operating voltages of electrets. I n order to use NIST sources i n accumulation mode: (1) It is important to know the exact air volume of the accumulator, when test devices are inside the MONITORS accumulator. This may be more complicated when complex instruments such as continuous radon monitors are located inside the accumulator. ( 2 ) The accumulator should be leak tight so that accumulated radon does not escape. (3) The test devices themselves should not absorb or affect the radon concentration i n the accumulator. (4) Instruction given by NIST on the use of these sources must be followed'". One imponant requirement is to leave the source in infinite volume for at least 24 h before using the source in the accumulator. amends the sources be stored in lms R,ative ,,umidiv, when not in uw. Table 4. Experimental results. Detectors used. two SST E-PERMs. Accumulation time: 1 day for CP4S to CPS9 and 2 days for CP17 to CP26. Source NS NS (Ra Bq) (mm-') . El IV Fv2 EP (6q.m-') EP-0 (EP-BY NS 0.951 0.959 I .043 0.992 0.971 0.949 0.975 0.96 1 0.960 1.009 0.986 0.960 0.954 1.035 0.9 17 0.893 1 .on 0.9 19 0.941 0.946 1.058 I .O53 1.038 1 .&to 0.978 I.024 1.030 1.018 1.062 1.064 1.030 1.010 1.057 0.990 0.994 0.924 0.990 1.004 Av. % STD - 1996 International Radon Symposium IIP 2.6 Av. ratio P. KOTRAPPA andL R. STIEFF Table 5. Experimental results. Detectors used: two SLT E-PERM& AccumulaUon time: 1 day. -- Source (Ra Bq) NS El IV W, (Bq.m-') EP (Bq.rn-') EP-B (EP -By AV. NS ratio 0.901 0.881 0.91 1 1.O53 0.897 0.9 18 0.838 0.903 0.973 0.902 0.900 1.015 0.935 0.948 0.980 0.913 0.912 0.9 16 0.932 0.932 Av. %STD Table 6. Experimental results. Detectors used: two SLT E-PERMS. Accumulation Ume: 2 days. Source NS NS (Ra Bq) (Bq.m-') El IV Wa Ep (Bq.m-') EP-B (Bq.m-') (EP-BY Av. NS ratio 0.950 0.923 0.947 1.016 0.927 0.99 1 0.890 0.945 0.965 0.923 0.910 1.014 0.923 0.9 15 0.972 0.953 0.922 0.938 0.976 0.964 Av. %ST - 1996 International Radon Symposium IIP 2.7 0.936 0.982 0.959 0.918 0.944 0.962 0.9 19 0.962 0.930 0.970 0.948 2.204 NIST STANDARDS FOR CALIBRATING '"Rn MONITORS CALIBRATING CONTINUOUS RADON MONITORS Table 7 gives the comparison between NIST estimated values with the values given by a CRM (Femtotech model 210, manufactured by Femto-Tech Inc.. Carisle, OH, USA). The results indicate that the instniment is well calibrated. The NIST sources have been previously used for determining the pressure correction factors, also called elevation correction factors by maintaining different pressures simulating different elevations inside the accumulators(". The use of NIST emanation standards also gives an elegant method of calibrating a continuous radon monitor (CRM) in a single experiment over a wide range of radon concentrations. This application-is based on the fact that Equation I gives the radon concentration at any accumulation time and that these theoretical estimates should match the reading given by CRM. For example, if one uses a 127 litre accumulator with a source of 880 Bq, the concentration varies from 46.4 Bq.m"' after I h to I980 Bq.m-' after 48 h. If such an instrument also gives a cumulative average, then this can be compared to the NIST calculated value. Larger accumulation times will yield higher concentrations. CONCEPT OF INDIVIDUALLY CALIBRATED E-PERM or CRM If an E-PERM shows 5% lower than what is predicted Table 7. Experimental results. NIST source strength: 879.9 Bq. Detector used: Femto-tech 210 continuous radon monitor. Accumulation time: 17 to 48 hours. R l Is the ratio of radon measured by CRM to radon calculated. R2 is the ratio of average radon measured by CRM to the average radon calculated. Accumulator volume 126.6 litres. (hours) NIST Rn (Barn-') NIST Av. Rn (Bq.m-') CRM CRM Rn Av. Rn (Bq.rn-') (f3q.m-') 444.00 466.20 484.70 510.60 536.50 555.00 573.50 595.70 617.90 640.10 654.90 669.70 691.90 717.80 736.30 754.80 773.30 791.80 810.30 828.80 847.30 865.80 884.30 906.50 928.70 947.20 965.70 987.90 1006.40 1024.90 1039.70 1058.20 Av. STD 217 1996 International Radon Symposium IIP - 2.8 RI R2 P. KOTRAPPA and L R. STIEFF by the NIST source, the results of the measurements made by using this unit in a later measurement should be divided by 0.95 to get the correct results. Using a portion (say from 700 to 650 V) it is possible to calibrate an E-PERM to NIST traceability. This still has a large range of usable voltage. An individually calibrated E-PERM can be used as reference E-PERM or as transfer standard for calibrating continuous "^Rn monitors used in '"Rn Calibration Test Chambers. The same logic applies to a CRM calibrated using NIST source. CONCLUSIONS For the first time. it is possible to calibrate both passive and continuous radon monitors to NIST traceability. The procedures developed in this work are usable with simple equipment. Typical passive devices (E-PERMS) were found to give predicted measurement with an accuracy of about 5%. Similarly, commercially available continuous radon monitors (Femto Tech) also gave satisfactory performances. With the availability of this technology, ^Rn measurement instruments can all be made NIST traceable. This is indeed a great step forward in radon metrology. ACKNOWLEDGEMENTS The authors are grateful to Dr Ron Colle of NIST for providing the sources for this study. They also wish to thank Mr Shiva Veerabhadraiah for technical help and for data processing. REFERENCES 1. Colli, R. and Hutchinson, J. M. R. Technical Notes on the Use of Standard Reference Material. Radon-222 Emnation Standard. (RadioactivityGroup. National Institute of Standards and Technology. MD.USA) Technical Report No 4968 (1993). 2. Kotrappa, P. and Jester, W. A. Measurement of Dissolved '"Rn in water using Electret Ion Chamber "'Rn Monitors. Health Phys. 64. 397-405 (1993). 3. Komppa, P.. Dcmpscy. J. C.. Hickey. L. R and Stieff. L. R. An Electret Passive Environineml Rn-222 Monitor Based on Ioniwion Measurement. Health Phys. 5427-56 (1988). 4. Kotrappa. P.. Dempscy, J. C.. Ramsey. R W. and Stieff. L. R. A Practical E-PEKM"' (Passive Environmental '"Rn Monitor) for Indoor lZzRnMeasurement. Health Phys. 58.461-167 (1990). 5. Kouappa. P. and Stieff, L. R. Elevation Correction Factors for E-PERM Radon Monitors. Health Phys. 62. 82-86 (1992). - 1996 International Radon Symposium IIP 2.9