Radiation Protection Dosimetry Advance Access published April 10, 2014 Radiation Protection Dosimetry (2014), pp. 1–6
doi:10.1093/rpd/ncu080
INDOOR/OUTDOOR RADON DECAY PRODUCTS ASSOCIATED AEROSOL PARTICLE-SIZE DISTRIBUTIONS AND THEIR RELATION TO TOTAL NUMBER CONCENTRATIONS Jun Moriizumi*, Shinya Yamada, Yang Xu, Satoru Matsuki, Shigekazu Hirao and Hiromi Yamazawa Graduate School of Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan *Corresponding author: [email protected]
INTRODUCTION Elevated air concentration of naturally occurring radioactive noble gas nuclide 222Rn (radon) significantly contributes to lung doses and hence increases lung cancer risk of inhabitants living in that environment with a long exposure period(1). The doses are mainly derived from inhalation of not radon gas itself but its short-lived decay products such as 218Po, 214Bi, 214 Pb and 214Po, and a large portion of these decay product nuclides is attached on aerosol particles to form ‘radioactive aerosols’(2). According to aerodynamic characteristics of these radioactive particles, efficiency and amount of their deposition on the inner surface of human respiratory tracts have complicated dependencies on both particle size and region along the tract. Since the difference of radiological sensitivity among regions of the respiratory tract, precise and accurate estimation of effective dose due to inhaled radon decay products requires information on aerodynamic characteristics of the radioactive aerosol, as well as concentrations of the radon decay product, ratio of the aerosol-associated to not associated fraction and breathing conditions(3). The size distribution of aerosol particles generally consists of three modes ( peaks) of nucleation mode (,0.1 mm), accumulation mode (0.1–1 mm) and coarse mode (.1 mm), according to the processes of their formation and growth in size through condensation and coagulation for nucleation and accumulation modes or resuspension for coarse mode only, and of their removal from air into any surface. In case of radioactive aerosol particles associated with shortlived radon decay products, radioactivity size distribution usually has a prominent accumulation mode and a lesser nucleation mode to be frequently approximated
as a unimodal log–normal distribution with an activity median diameter (AMD) and a geometric standard deviation s(2) g . In terms that the size distribution is obtained with an aerodynamic classification such as impactors, the AMD is called as activity median aerodynamic diameter (AMAD). The aerodynamic size distribution of the radioactive aerosol in indoor air, where most of the people stay for long periods, should contribute to variability of lung dose. Some works show that the variation in indoor particle-size distribution results from not only the environmental factors that are inherent in indoor air but also the variation in outdoor aerosol conditions. Harley et al. (4) estimate the contribution of outdoor to indoor aerosol concentration by analysis of temporal variation in long-lived radon decay products 210Pb concentration measurements. Since 210Pb concentration is usually contributed by air-borne particles with diameters in the coarse mode range originated from the resuspension process, indoor radioactive aerosol associated with short-lived radon decay products, which has its sources both in outdoor and indoor air, does not necessarily have the same behaviour as that with 210Pb. Mostafa et al. (5) report on the observations of size distribution of radioactive aerosols associated with short-lived radon decay products in indoor air in Japan, and they show smaller AMADs and decreasing of accumulation mode particles without a significant change in the nucleation mode for the observations after rainfalls at the previous days of their samplings. These results were inconsistent with Porstendo¨rfer et al. (6) that concluded no significant influence of weather conditions on activity of the accumulation particles. Naturally enough, raindrops
# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://rpd.oxfordjournals.org/ at Ondokuz Mayis University on November 6, 2014
The activity size distributions of indoor and outdoor radioactive aerosol associated with short-lived radon decay products were observed at Nagoya, Japan, for some periods from 2010 to 2012, following the indoor observation by Mostafa et al. [Mostafa, A. M. A., Tamaki, K., Moriizumi, J., Yamazawa, H. and Iida, T. The weather dependence of particle size distribution of indoor radioactive aerosol associated with radon decay products. Radiat. Prot. Dosim. 146(1– 3), 19–22 (2011)]. The tendency of smaller indoor activity median aerodynamic diameter (AMAD) after rainfalls showed in the previous study was not consistently obtained, while the consistent tendency of less indoor radioactive particles with diameters in the accumulation mode was observed again after rainfalls. The indoor aerosols showed activity size distributions similar to the outdoor ones. Non-radioactive aerosol particle concentrations measured with a laser particle counter suggested a somewhat liner relationship with AMAD.
J. MORIIZUMI ET AL.
MATERIALS AND METHODS Locations and periods The observations of activity particle-size distributions of the radioactive aerosol associated with short-lived radon decay product nuclides in indoor air and accompanying items were made in a laboratory, the same one as those by Mostafa et al.(5), with a volume of 205 m3, three doors and one large wall-to-wall glass window in the fifth floor of a concrete building of Nagoya University, Nagoya, Japan (35.168N, 136.978E). During all the measurements, no air conditioner, circulation fan and air cleaner were operated in the room, and entering into the room was refrained as well as possible. The outdoor air observations were carried out just the outside of the room, in which the indoor observation was performed, with 15 m height above the ground and surrounded with trees. The periods of the indoor observations performed were from 14 May 2010 to 10 September 2010 (mentioned as Period I), 19 September 2011 to 25 October 2011 (Period II), 31 October 2011 to 7 November 2011 (Period III), 5 January 2012 to 12 January 2012 (Period IV) and 26 April 2012 to 17 November 2012 (Period V). The observation for outdoor air was also made in Period V, but not simultaneously with each measurement for indoor air. The intervals between each measurement of the size distribution were not constant. Several intensive measurements with twice or three times air samplings a day were implemented during the observation periods. Throughout Period V, the measurements of the total aerosol particle concentrations with an LPC were simultaneously
performed during both the indoor and the outdoor air measurements. Radon and size distribution of radioactive aerosol The methodology and instrumentation of this study were similar to those of Mostafa et al. (5) and Rahman et al. (8). Here, they are briefly stated with slight differences from them. For aerodynamic classification of aerosol particles with a diameter of ,1 mm, the observation adopted an Andersen-type low-pressure cascade impactor (LP-20RPS47; Tokyo Dylec Co., Ltd, Tokyo, Japan), which had 13 classification stages (12 stainless steel plates coated with a thin layer of grease and one backup filter sheet) corresponding to theoretical 50 % cut-off diameters of 12.0, 8.1, 5.5, 3.5, 1.9, 1.0, 0.73, 0.48, 0.28, 0.19, 0.11 and 0.05 mm(9). At the intake of the impactor, a set of wire screen mesh #635 for the observation Periods I and III or a diffusion battery with five meshes of #60, #100, #200, #400 and #635 for Periods II, IV and V was installed to remove unattached fraction of radon decay products from the sampled air stream beforehand. For each observation, 1 h sampling of air at a flow rate of 25.7 l min21 was followed by 20 min cooling time and subsequent 1 h exposure of an imaging plate (IP) sheet (BAS-SR, FUJIFILM Corporation, Tokyo, Japan) to alpha particle emitted from radon decay products collected on the 13 stages. Alpha emissions for each stage were ‘counted’ from the spotty image of alpha particle incidents on IP sheet read by a BAS-5000 image reader (FUJIFILM Corporation), and then an activity size distribution was obtained. Indoor air radon concentration was measured with AlphaGUARD (SAPHYMO GmbH, Frankfurt a.M., Germany) and outdoor radon concentration with an electrostatic radon monitor(10, 11). Ventilation rate of the observing room was evaluated from a decreasing rate of CO2 concentration elevation due to instantaneous CO2 gas release into the room relative to outdoor CO2 concentration. Number concentration of total aerosol particles A hand-held laser particle counter Model 8330 (Kanomax, Osaka, Japan) was adopted in this study for the measurements of number concentration of total aerosol particles with five range classification of particle diameters of .0.3, .0.5, .1.0, .3.0 and .5.0 mm. The classification was not aerodynamically but optically achieved. As this LPC was designed for usage in clean rooms, typical number concentration in ordinary air in Nagoya was comparable with the upper limit of its application. Thus, some devices to dilute sample air with aerosol-free air must be jointly used; otherwise, a part of counts of the particles would be missed to underestimate the number concentration. A method to correct these lost counts was
Page 2 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Ondokuz Mayis University on November 6, 2014
do not enter into the indoor laboratory directly or indirectly. No significant correlation is, however, reported between AMAD and meteorological factors such as humidity in indoor air. These infer that variations in the formation, growth and removal processes of indoor aerosol particles could not critically control the variation in the size distribution, and the variation in outdoor aerosol size distribution would significantly contribute to that in indoor’s via ventilation. While few works directly related variations in size distributions of outdoor and indoor radioactive aerosols associated with short-lived radon decay products(7). In this paper, first, the results of the indoor observation extended after that by Mostafa et al. (5) and newly started outdoor observation at the same location are reported. Secondly, variation in concentration of general, that is, almost non-radioactive aerosol particles with diameters corresponding to the accumulation mode is related with that in AMAD to discuss whether a laser particle counter (LPC) is feasible or not for a proxy measurement, which may enable continuous and/or high-frequency indirect measurement of the radioactive aerosol size distribution.
INDOOR/OUTDOOR RADON DECAY PRODUCT SIZE
applied to perform measurement without any dilution, as discussed later. RESULTS AND DISCUSSIONS Variation in activity size distribution and its correlation to rainfall
Table 1. The variations in AMAD and sg in relation to rainfall. Periods
Indoor/outdoor
Rainfall
na
AMAD/nm a
Mostafa et al. (5) 22 July 2009 to 24 November 2009 Indoor
No-rainfallb Rainfallb
This study Period I (14 May 2010 to 10 September 2010) Indoor No-rainfallc Rainfallc Period II (19 September 2011 to 25 October 2011) Indoor No-rainfallc Rainfallc Period III (31 October 2011 to 7 November 2011) Indoor No-rainfallc Rainfallc Period VI (5 January 2012 to 12 January 2012) Indoor No-rainfallc Rainfallc Period V (26 April 2012 to 17 November 2012) Indoor No-rainfallc Rainfallc Outdoor No-rainfallc Rainfallc Total Indoor No-rainfallc Rainfallc
sg/– a
Mean
Minimum– maximum
Mean
Minimum– maximum
— —
190 156
107–287 83–283
2.6 3.1
2.1– 3.4 2.3– 5.7
5 6
189 152
144–228 106–248
2.4 2.7
2.2– 2.6 2.2– 3.0
7 8
188 205
133–245 159–262
2.5 2.4
2.2– 2.7 2.0– 2.6
3 3
169 209
145–204 145–289
2.4 2.6
2.3– 2.5 2.5– 2.8
1 4
200 179
131–219
2.7 2.7
2.2– 2.9
3 12 16 23
183 209 194 206
161–198 105–298 116–312 119–311
2.5 2.7 2.6 2.8
2.4– 2.5 2.2– 4.0 2.1– 4.2 2.3– 3.5
19 33
185 194
133–245 105–298
2.4 2.6
2.2– 2.7 2.0– 4.0
a
When more than one sampling was made in a day, their daily average value was adopted for calculation of the mean AMAD for each observation period. b At the previous day of sampling. c In ‘Rainfall’ cases, a rainfall of .0.0 mm h21 for any 1 h in 24 h prior to the air sampling was detected at Nagoya Meteorological Observatory of JMA.
Page 3 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Ondokuz Mayis University on November 6, 2014
The observations of the activity size distribution are summarised in Table 1 in terms of the mean values and fluctuations of AMAD and sg with their correlations to rainfall. In summarising, when more than one air sampling with the low-pressure cascade impactor was made in a day, their daily average value is adopted for calculation of the arithmetic means of AMAD and sg for each observation period to avoid partiality and hold analytical consistency with Mostafa et al. (5) having air sampling intervals longer than 1 d. If a rainfall of .0.0 mm h21 for any 1 h in 24 h prior to the air sampling was detected at Nagoya Meteorological Observatory of Japan Meteorological
Agency (JMA), 1.4 km distant from the laboratory, the measurements are classified as ‘Rainfall’ in Table 1. All the measured AMAD and sg for the indoor air samples in this study range from 105 to 298 and from 2.0 to 4.0, respectively, being consistent with those by Mostafa et al. (5). The same tendency as Mostafa et al. (5) that mean AMAD for the indoor air was smaller in case of rainfall than without rainfall was obtained for Period I. For the other observation periods, the mean AMADs for Rainfall were, however, larger than for No-rainfall. The ranges of fluctuations of both AMAD and sg were not significantly different between the Rainfall and No-rainfall classes for each observation period and for the total of the observation in this study. A consistent correlation of the variation in indoor air AMAD to existence of rainfall in previous 24 h was not obtained through all the measurements in this study. The means and fluctuation ranges of AMAD and sg for the outdoor air in Period V were comparable with those for the indoor air observed in Period V.
J. MORIIZUMI ET AL.
Observation of total aerosol particle concentration with LPC The LPC Model 8330 can measure particle concentration with a diameters of ,1 mm. Since these diameter ranges seem to correspond to those of the accumulation mode of the radioactivity size distribution, fluctuation in AMAD due to variation in a number of radioactivity attached particles belonging to the accumulation mode could be detected as change in total number concentrations of these ranges. The measurements of uncorrected (i.e. 1 ¼ 1) number concentrations of total aerosol particles with the LPC Model 8330 were ‘saturated’ by presumable missing of the particle counts at high concentrations more than several 107 particles m23 as shown in Figure 2, where c, c0 h and 1 are a number concentration diluted with the diluting device, a not-diluted number concentration of the identical ambient air sample, the dilution ratio defined by c/c0 and a factor for correcting the loss of particle counted. If ideally no loss of counting occurs, (c/h) has to be theoretically equal to c0 by the definition of h, and 1 ¼ 1, such as the case of the .0.5 mm range of the LPC where h was experimentally decided to be 0.101. Assuming the value of h was common between the .0.3 and .0.5 mm ranges, the correction factor 1 as a function of c0 (108 m23) was obtained as 1(c0) ¼ 0.939 exp(2.00c0) for the .0.3 mm range, from regression curve for the measurements of 1 ¼ (c/h)/c0 and c0. Figure 2 shows that the practical counting-loss-corrected number
Figure 1. Mean alpha emission counts normalised to radon concentration of 1 Bq m23 with their standard deviations (error bars), for the nucleation mode (n ¼ 23) and the accumulation mode (n ¼ 30) of the indoor air through Periods I– V and those of the outdoor air (n ¼ 12 and 15, respectively) for Period V.
Page 4 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Ondokuz Mayis University on November 6, 2014
The unexpected tendency of larger mean AMAD for Rainfall than for No-rainfall for this period was common between the indoor and the outdoor. To discuss more details of the variation in the size distribution, averages of the alpha emission counts normalised to radon concentration of 1 Bq m23 for the nucleation mode stages (cut-off diameter ,0.11 mm) and the accumulation mode stages (0.11 – 1.9 mm) of the cascade impactor are compared in Figure 1. For the indoor measurements, the same tendency as Mostafa et al. (5) that show smaller alpha counts of the accumulation mode for Rainfall than for No-rainfall and no significant difference between counts of the nucleation mode regardless of the existence of rainfall was obtained again through all the periods in this study. This tendency also significantly holds even if the period for averaging is limited to Period V. While, for the outdoor observations, limited in Period V, significant differences of the normalised alpha counts between Rainfall and No-rainfall were not shown for the accumulation mode and the nucleation mode. The ventilation rates observed through this study almost ranged from 0.7 to 1.1 h21 except several extremely low values of ,0.4 h21; therefore, it seems to be reasonable to consider that a significant fraction of indoor air aerosol particles was attributed from entering outdoor air. However, the result of the analysis with the alpha counts could not support so simple understanding of contribution of the outdoor radioactive aerosol to the indoor one via ventilation and mixing.
INDOOR/OUTDOOR RADON DECAY PRODUCT SIZE
concentration measured without the dilution, (1.c0), was nearly equal to the theoretical original concentration estimated from the measured diluted concentration, (c/h). Thus, without any dilution device, the total number concentration can be obtained as (1.c0) by only measuring c0 and correcting with 1(c0). The observations of .0.3 mm range total number concentrations (without the dilution) and AMAD for both the indoor and the outdoor air shown in Figure 3 seem to have linear relationships, and the ranges of values scattered in both panes are similar. Although their correlations are not so much high, the difference of the relationships between Rainfall and No-rainfall classes is not significant for the outdoor measurements. It is not clear at present whether these quasi-liner relationships for AMAD from 100 to 400 nm also hold for AMADs ,100 and .400 nm. Size distributions with these extreme AMAD values would be contributed by substantial nucleation mode or course mode, respectively, comparing with the accumulation mode. Thus, in extreme cases, relating the .0.3 mm range concentration to AMAD would fundamentally be less reasonable and inappropriate. The larger scattering around the quasi-liner relationship suggests that the application of the LPC as a proxy for size distribution measurement with the impactor would be insufficiently feasible based on the present methodology. However, long-term, continuous and highly frequent concentration measurement enabled by adopting the LPC substantially increases
Figure 3. Correlations between number concentrations of total (non-radioactive) aerosol (.0.3 mm range) and AMAD for the indoor (a) and the outdoor air (b). The criteria for Rainfall and No-rainfall are common to Table 1.
the size of observation dataset. To clarify complicated processes of outdoor radioactive aerosol, which is hardly controlled in contrast to laboratory experiments, modelling approaches with theories of aerosol attachment, meteorology and atmospheric transport phenomena should be essential. These analyses need large volume of field observation, and the LPC method potentially supplies a measure for it. The long-term highly frequent measurement would also make this method feasible for surveying a representative condition for activity size distribution measurement in terms of lung dose assessment and screening to decide to start a precise measurement with the impactor on suitable timing for some purpose, such as catching variation in activity size distribution in outdoor air.
Page 5 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Ondokuz Mayis University on November 6, 2014
Figure 2. Comparison of measured non-diluted number concentrations of total aerosol particles (c0) multiplied with the correction factor (1) to theoretically estimated one (c/h) from diluted concentration (c) and the dilution ratio (h). If 1 ¼ 1, the values of (1.c0) are without the correction.
J. MORIIZUMI ET AL.
REFERENCES
Through the observation of short-lived radon decay products associated radioactive aerosol in the indoor air for 2010–12 in Nagoya, Japan, consistent correlation of the variation in AMAD to existence of rainfall in previous 24 h was not obtained. The means and fluctuation ranges of AMAD and sg for the outdoor air were comparable with those for the indoor air observed in the same period. The alpha emission counts of the nucleation and the accumulation modes in the indoor air showed the same tendency in terms of the existence of rainfall as Mostafa et al. (5), while those in the outdoor have insignificant difference between Rainfall and No-rainfall classes. The method of total aerosol particle number concentration measurement by the LPC without any dilution was developed. The positive quasi-linear relationships were observed between the AMAD and the total number concentration with weaker correlations. Although exact estimation of AMAD by the LPC as proxy would be insufficiently feasible at present, long-term, continuous and highly frequent measurement with the LPC substantially enlarges the volume of observation and would be applicable to survey and screening to perform precise size distribution measurement with the impactor at suitable timing.
1. Marcinowski, F. M. Nationwide survey of residential radon levels in the US. Radiat. Prot. Dosim. 45, 419–424 (1992). 2. Papastefanou, C. Radioactive Aerosols. Radioactivity in the Environment, Vol. 12. Elsevier (2008). ISBN:978 0 08 044075 0. 3. ICRP. Human respiratory tract model for radiological protection. ICRP Publication 66. Ann. ICRP 24(1– 3), (1994). 4. Harley, N. H., Chittaporn, P., Fissene, I. M. and Perry, P. 222Rn decay products as tracers of indoor and outdoor aerosol particle size. J. Environ. Radioact. 51, 27– 35 (2000). 5. Mostafa, A. M. A., Tamaki, K., Moriizumi, J., Yamazawa, H. and Iida, T. The weather dependence of particle size distribution of indoor radioactive aerosol associated with radon decay products. Radiat. Prot. Dosim. 146(1– 3), 19–22 (2011). 6. Porstendo¨rfer, J., Zock, Ch. and Reineking, A. Aerosol size distribution of radon progeny in outdoor air. J. Environ. Radioact. 51, 37– 48 (2000). 7. Reineking, A., Becker, K. H. and Porstendo¨rfer, J. Measurements of activity size distributions of the shortlived radon daughters in the indoor and outdoor environment. Radiat. Prot. Dosim. 24(1), 245–250 (1988). 8. Rahman, N. M., Iida, T., Yamazawa, H., Yokoyama, S., Moriizumi, J., Saito, F. and Ito, T. The study of activity median aerodynamic diameter using imaging plate technique for assessment of effective dose from radon and its decay products. Radiat. Prot. Dosim. 124(4), 360–371 (2007). 9. Rahman, N. M., Iida, T., Saito, F., Koarashi, J., Yamasaki, K., Yamazawa, H. and Moriizumi, J. Evaluation of aerosol sizing characteristic of an impactor using imaging plate technique. Radiat. Prot. Dosim. 123(2), 171– 181 (2007). 10. Iida, T., Ikebe, Y., Suzuki, K., Ueno, K., Wang, Z. and Jin, Y. Continuous measurements of outdoor radon concentrations at various locations in East Asia. Environ. Int. 22(Suppl. 1), S139– S147 (1996). 11. Moriizumi, J. et al. Continuous observation of atmospheric 222Rn concentrations for analytic basis of atmospheric transport in East Asia. J. Nucl. Sci. Tech. Suppl. 6, 173– 179 (2008).
FUNDING This study was partly supported in the project ‘Construction of natural radiation exposure study network’ in the Programme of Promotion of International Joint Research under the Special Coordination Funds for the promotion of Science and Technology operated by MEXT, the Japanese government. This work was partly supported by JSPS KAKENHI (grant no. 24110002).
Page 6 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Ondokuz Mayis University on November 6, 2014
CONCLUSIONS
doi:10.1093/rpd/ncu080
INDOOR/OUTDOOR RADON DECAY PRODUCTS ASSOCIATED AEROSOL PARTICLE-SIZE DISTRIBUTIONS AND THEIR RELATION TO TOTAL NUMBER CONCENTRATIONS Jun Moriizumi*, Shinya Yamada, Yang Xu, Satoru Matsuki, Shigekazu Hirao and Hiromi Yamazawa Graduate School of Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan *Corresponding author: [email protected]
INTRODUCTION Elevated air concentration of naturally occurring radioactive noble gas nuclide 222Rn (radon) significantly contributes to lung doses and hence increases lung cancer risk of inhabitants living in that environment with a long exposure period(1). The doses are mainly derived from inhalation of not radon gas itself but its short-lived decay products such as 218Po, 214Bi, 214 Pb and 214Po, and a large portion of these decay product nuclides is attached on aerosol particles to form ‘radioactive aerosols’(2). According to aerodynamic characteristics of these radioactive particles, efficiency and amount of their deposition on the inner surface of human respiratory tracts have complicated dependencies on both particle size and region along the tract. Since the difference of radiological sensitivity among regions of the respiratory tract, precise and accurate estimation of effective dose due to inhaled radon decay products requires information on aerodynamic characteristics of the radioactive aerosol, as well as concentrations of the radon decay product, ratio of the aerosol-associated to not associated fraction and breathing conditions(3). The size distribution of aerosol particles generally consists of three modes ( peaks) of nucleation mode (,0.1 mm), accumulation mode (0.1–1 mm) and coarse mode (.1 mm), according to the processes of their formation and growth in size through condensation and coagulation for nucleation and accumulation modes or resuspension for coarse mode only, and of their removal from air into any surface. In case of radioactive aerosol particles associated with shortlived radon decay products, radioactivity size distribution usually has a prominent accumulation mode and a lesser nucleation mode to be frequently approximated
as a unimodal log–normal distribution with an activity median diameter (AMD) and a geometric standard deviation s(2) g . In terms that the size distribution is obtained with an aerodynamic classification such as impactors, the AMD is called as activity median aerodynamic diameter (AMAD). The aerodynamic size distribution of the radioactive aerosol in indoor air, where most of the people stay for long periods, should contribute to variability of lung dose. Some works show that the variation in indoor particle-size distribution results from not only the environmental factors that are inherent in indoor air but also the variation in outdoor aerosol conditions. Harley et al. (4) estimate the contribution of outdoor to indoor aerosol concentration by analysis of temporal variation in long-lived radon decay products 210Pb concentration measurements. Since 210Pb concentration is usually contributed by air-borne particles with diameters in the coarse mode range originated from the resuspension process, indoor radioactive aerosol associated with short-lived radon decay products, which has its sources both in outdoor and indoor air, does not necessarily have the same behaviour as that with 210Pb. Mostafa et al. (5) report on the observations of size distribution of radioactive aerosols associated with short-lived radon decay products in indoor air in Japan, and they show smaller AMADs and decreasing of accumulation mode particles without a significant change in the nucleation mode for the observations after rainfalls at the previous days of their samplings. These results were inconsistent with Porstendo¨rfer et al. (6) that concluded no significant influence of weather conditions on activity of the accumulation particles. Naturally enough, raindrops
# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://rpd.oxfordjournals.org/ at Ondokuz Mayis University on November 6, 2014
The activity size distributions of indoor and outdoor radioactive aerosol associated with short-lived radon decay products were observed at Nagoya, Japan, for some periods from 2010 to 2012, following the indoor observation by Mostafa et al. [Mostafa, A. M. A., Tamaki, K., Moriizumi, J., Yamazawa, H. and Iida, T. The weather dependence of particle size distribution of indoor radioactive aerosol associated with radon decay products. Radiat. Prot. Dosim. 146(1– 3), 19–22 (2011)]. The tendency of smaller indoor activity median aerodynamic diameter (AMAD) after rainfalls showed in the previous study was not consistently obtained, while the consistent tendency of less indoor radioactive particles with diameters in the accumulation mode was observed again after rainfalls. The indoor aerosols showed activity size distributions similar to the outdoor ones. Non-radioactive aerosol particle concentrations measured with a laser particle counter suggested a somewhat liner relationship with AMAD.
J. MORIIZUMI ET AL.
MATERIALS AND METHODS Locations and periods The observations of activity particle-size distributions of the radioactive aerosol associated with short-lived radon decay product nuclides in indoor air and accompanying items were made in a laboratory, the same one as those by Mostafa et al.(5), with a volume of 205 m3, three doors and one large wall-to-wall glass window in the fifth floor of a concrete building of Nagoya University, Nagoya, Japan (35.168N, 136.978E). During all the measurements, no air conditioner, circulation fan and air cleaner were operated in the room, and entering into the room was refrained as well as possible. The outdoor air observations were carried out just the outside of the room, in which the indoor observation was performed, with 15 m height above the ground and surrounded with trees. The periods of the indoor observations performed were from 14 May 2010 to 10 September 2010 (mentioned as Period I), 19 September 2011 to 25 October 2011 (Period II), 31 October 2011 to 7 November 2011 (Period III), 5 January 2012 to 12 January 2012 (Period IV) and 26 April 2012 to 17 November 2012 (Period V). The observation for outdoor air was also made in Period V, but not simultaneously with each measurement for indoor air. The intervals between each measurement of the size distribution were not constant. Several intensive measurements with twice or three times air samplings a day were implemented during the observation periods. Throughout Period V, the measurements of the total aerosol particle concentrations with an LPC were simultaneously
performed during both the indoor and the outdoor air measurements. Radon and size distribution of radioactive aerosol The methodology and instrumentation of this study were similar to those of Mostafa et al. (5) and Rahman et al. (8). Here, they are briefly stated with slight differences from them. For aerodynamic classification of aerosol particles with a diameter of ,1 mm, the observation adopted an Andersen-type low-pressure cascade impactor (LP-20RPS47; Tokyo Dylec Co., Ltd, Tokyo, Japan), which had 13 classification stages (12 stainless steel plates coated with a thin layer of grease and one backup filter sheet) corresponding to theoretical 50 % cut-off diameters of 12.0, 8.1, 5.5, 3.5, 1.9, 1.0, 0.73, 0.48, 0.28, 0.19, 0.11 and 0.05 mm(9). At the intake of the impactor, a set of wire screen mesh #635 for the observation Periods I and III or a diffusion battery with five meshes of #60, #100, #200, #400 and #635 for Periods II, IV and V was installed to remove unattached fraction of radon decay products from the sampled air stream beforehand. For each observation, 1 h sampling of air at a flow rate of 25.7 l min21 was followed by 20 min cooling time and subsequent 1 h exposure of an imaging plate (IP) sheet (BAS-SR, FUJIFILM Corporation, Tokyo, Japan) to alpha particle emitted from radon decay products collected on the 13 stages. Alpha emissions for each stage were ‘counted’ from the spotty image of alpha particle incidents on IP sheet read by a BAS-5000 image reader (FUJIFILM Corporation), and then an activity size distribution was obtained. Indoor air radon concentration was measured with AlphaGUARD (SAPHYMO GmbH, Frankfurt a.M., Germany) and outdoor radon concentration with an electrostatic radon monitor(10, 11). Ventilation rate of the observing room was evaluated from a decreasing rate of CO2 concentration elevation due to instantaneous CO2 gas release into the room relative to outdoor CO2 concentration. Number concentration of total aerosol particles A hand-held laser particle counter Model 8330 (Kanomax, Osaka, Japan) was adopted in this study for the measurements of number concentration of total aerosol particles with five range classification of particle diameters of .0.3, .0.5, .1.0, .3.0 and .5.0 mm. The classification was not aerodynamically but optically achieved. As this LPC was designed for usage in clean rooms, typical number concentration in ordinary air in Nagoya was comparable with the upper limit of its application. Thus, some devices to dilute sample air with aerosol-free air must be jointly used; otherwise, a part of counts of the particles would be missed to underestimate the number concentration. A method to correct these lost counts was
Page 2 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Ondokuz Mayis University on November 6, 2014
do not enter into the indoor laboratory directly or indirectly. No significant correlation is, however, reported between AMAD and meteorological factors such as humidity in indoor air. These infer that variations in the formation, growth and removal processes of indoor aerosol particles could not critically control the variation in the size distribution, and the variation in outdoor aerosol size distribution would significantly contribute to that in indoor’s via ventilation. While few works directly related variations in size distributions of outdoor and indoor radioactive aerosols associated with short-lived radon decay products(7). In this paper, first, the results of the indoor observation extended after that by Mostafa et al. (5) and newly started outdoor observation at the same location are reported. Secondly, variation in concentration of general, that is, almost non-radioactive aerosol particles with diameters corresponding to the accumulation mode is related with that in AMAD to discuss whether a laser particle counter (LPC) is feasible or not for a proxy measurement, which may enable continuous and/or high-frequency indirect measurement of the radioactive aerosol size distribution.
INDOOR/OUTDOOR RADON DECAY PRODUCT SIZE
applied to perform measurement without any dilution, as discussed later. RESULTS AND DISCUSSIONS Variation in activity size distribution and its correlation to rainfall
Table 1. The variations in AMAD and sg in relation to rainfall. Periods
Indoor/outdoor
Rainfall
na
AMAD/nm a
Mostafa et al. (5) 22 July 2009 to 24 November 2009 Indoor
No-rainfallb Rainfallb
This study Period I (14 May 2010 to 10 September 2010) Indoor No-rainfallc Rainfallc Period II (19 September 2011 to 25 October 2011) Indoor No-rainfallc Rainfallc Period III (31 October 2011 to 7 November 2011) Indoor No-rainfallc Rainfallc Period VI (5 January 2012 to 12 January 2012) Indoor No-rainfallc Rainfallc Period V (26 April 2012 to 17 November 2012) Indoor No-rainfallc Rainfallc Outdoor No-rainfallc Rainfallc Total Indoor No-rainfallc Rainfallc
sg/– a
Mean
Minimum– maximum
Mean
Minimum– maximum
— —
190 156
107–287 83–283
2.6 3.1
2.1– 3.4 2.3– 5.7
5 6
189 152
144–228 106–248
2.4 2.7
2.2– 2.6 2.2– 3.0
7 8
188 205
133–245 159–262
2.5 2.4
2.2– 2.7 2.0– 2.6
3 3
169 209
145–204 145–289
2.4 2.6
2.3– 2.5 2.5– 2.8
1 4
200 179
131–219
2.7 2.7
2.2– 2.9
3 12 16 23
183 209 194 206
161–198 105–298 116–312 119–311
2.5 2.7 2.6 2.8
2.4– 2.5 2.2– 4.0 2.1– 4.2 2.3– 3.5
19 33
185 194
133–245 105–298
2.4 2.6
2.2– 2.7 2.0– 4.0
a
When more than one sampling was made in a day, their daily average value was adopted for calculation of the mean AMAD for each observation period. b At the previous day of sampling. c In ‘Rainfall’ cases, a rainfall of .0.0 mm h21 for any 1 h in 24 h prior to the air sampling was detected at Nagoya Meteorological Observatory of JMA.
Page 3 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Ondokuz Mayis University on November 6, 2014
The observations of the activity size distribution are summarised in Table 1 in terms of the mean values and fluctuations of AMAD and sg with their correlations to rainfall. In summarising, when more than one air sampling with the low-pressure cascade impactor was made in a day, their daily average value is adopted for calculation of the arithmetic means of AMAD and sg for each observation period to avoid partiality and hold analytical consistency with Mostafa et al. (5) having air sampling intervals longer than 1 d. If a rainfall of .0.0 mm h21 for any 1 h in 24 h prior to the air sampling was detected at Nagoya Meteorological Observatory of Japan Meteorological
Agency (JMA), 1.4 km distant from the laboratory, the measurements are classified as ‘Rainfall’ in Table 1. All the measured AMAD and sg for the indoor air samples in this study range from 105 to 298 and from 2.0 to 4.0, respectively, being consistent with those by Mostafa et al. (5). The same tendency as Mostafa et al. (5) that mean AMAD for the indoor air was smaller in case of rainfall than without rainfall was obtained for Period I. For the other observation periods, the mean AMADs for Rainfall were, however, larger than for No-rainfall. The ranges of fluctuations of both AMAD and sg were not significantly different between the Rainfall and No-rainfall classes for each observation period and for the total of the observation in this study. A consistent correlation of the variation in indoor air AMAD to existence of rainfall in previous 24 h was not obtained through all the measurements in this study. The means and fluctuation ranges of AMAD and sg for the outdoor air in Period V were comparable with those for the indoor air observed in Period V.
J. MORIIZUMI ET AL.
Observation of total aerosol particle concentration with LPC The LPC Model 8330 can measure particle concentration with a diameters of ,1 mm. Since these diameter ranges seem to correspond to those of the accumulation mode of the radioactivity size distribution, fluctuation in AMAD due to variation in a number of radioactivity attached particles belonging to the accumulation mode could be detected as change in total number concentrations of these ranges. The measurements of uncorrected (i.e. 1 ¼ 1) number concentrations of total aerosol particles with the LPC Model 8330 were ‘saturated’ by presumable missing of the particle counts at high concentrations more than several 107 particles m23 as shown in Figure 2, where c, c0 h and 1 are a number concentration diluted with the diluting device, a not-diluted number concentration of the identical ambient air sample, the dilution ratio defined by c/c0 and a factor for correcting the loss of particle counted. If ideally no loss of counting occurs, (c/h) has to be theoretically equal to c0 by the definition of h, and 1 ¼ 1, such as the case of the .0.5 mm range of the LPC where h was experimentally decided to be 0.101. Assuming the value of h was common between the .0.3 and .0.5 mm ranges, the correction factor 1 as a function of c0 (108 m23) was obtained as 1(c0) ¼ 0.939 exp(2.00c0) for the .0.3 mm range, from regression curve for the measurements of 1 ¼ (c/h)/c0 and c0. Figure 2 shows that the practical counting-loss-corrected number
Figure 1. Mean alpha emission counts normalised to radon concentration of 1 Bq m23 with their standard deviations (error bars), for the nucleation mode (n ¼ 23) and the accumulation mode (n ¼ 30) of the indoor air through Periods I– V and those of the outdoor air (n ¼ 12 and 15, respectively) for Period V.
Page 4 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Ondokuz Mayis University on November 6, 2014
The unexpected tendency of larger mean AMAD for Rainfall than for No-rainfall for this period was common between the indoor and the outdoor. To discuss more details of the variation in the size distribution, averages of the alpha emission counts normalised to radon concentration of 1 Bq m23 for the nucleation mode stages (cut-off diameter ,0.11 mm) and the accumulation mode stages (0.11 – 1.9 mm) of the cascade impactor are compared in Figure 1. For the indoor measurements, the same tendency as Mostafa et al. (5) that show smaller alpha counts of the accumulation mode for Rainfall than for No-rainfall and no significant difference between counts of the nucleation mode regardless of the existence of rainfall was obtained again through all the periods in this study. This tendency also significantly holds even if the period for averaging is limited to Period V. While, for the outdoor observations, limited in Period V, significant differences of the normalised alpha counts between Rainfall and No-rainfall were not shown for the accumulation mode and the nucleation mode. The ventilation rates observed through this study almost ranged from 0.7 to 1.1 h21 except several extremely low values of ,0.4 h21; therefore, it seems to be reasonable to consider that a significant fraction of indoor air aerosol particles was attributed from entering outdoor air. However, the result of the analysis with the alpha counts could not support so simple understanding of contribution of the outdoor radioactive aerosol to the indoor one via ventilation and mixing.
INDOOR/OUTDOOR RADON DECAY PRODUCT SIZE
concentration measured without the dilution, (1.c0), was nearly equal to the theoretical original concentration estimated from the measured diluted concentration, (c/h). Thus, without any dilution device, the total number concentration can be obtained as (1.c0) by only measuring c0 and correcting with 1(c0). The observations of .0.3 mm range total number concentrations (without the dilution) and AMAD for both the indoor and the outdoor air shown in Figure 3 seem to have linear relationships, and the ranges of values scattered in both panes are similar. Although their correlations are not so much high, the difference of the relationships between Rainfall and No-rainfall classes is not significant for the outdoor measurements. It is not clear at present whether these quasi-liner relationships for AMAD from 100 to 400 nm also hold for AMADs ,100 and .400 nm. Size distributions with these extreme AMAD values would be contributed by substantial nucleation mode or course mode, respectively, comparing with the accumulation mode. Thus, in extreme cases, relating the .0.3 mm range concentration to AMAD would fundamentally be less reasonable and inappropriate. The larger scattering around the quasi-liner relationship suggests that the application of the LPC as a proxy for size distribution measurement with the impactor would be insufficiently feasible based on the present methodology. However, long-term, continuous and highly frequent concentration measurement enabled by adopting the LPC substantially increases
Figure 3. Correlations between number concentrations of total (non-radioactive) aerosol (.0.3 mm range) and AMAD for the indoor (a) and the outdoor air (b). The criteria for Rainfall and No-rainfall are common to Table 1.
the size of observation dataset. To clarify complicated processes of outdoor radioactive aerosol, which is hardly controlled in contrast to laboratory experiments, modelling approaches with theories of aerosol attachment, meteorology and atmospheric transport phenomena should be essential. These analyses need large volume of field observation, and the LPC method potentially supplies a measure for it. The long-term highly frequent measurement would also make this method feasible for surveying a representative condition for activity size distribution measurement in terms of lung dose assessment and screening to decide to start a precise measurement with the impactor on suitable timing for some purpose, such as catching variation in activity size distribution in outdoor air.
Page 5 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Ondokuz Mayis University on November 6, 2014
Figure 2. Comparison of measured non-diluted number concentrations of total aerosol particles (c0) multiplied with the correction factor (1) to theoretically estimated one (c/h) from diluted concentration (c) and the dilution ratio (h). If 1 ¼ 1, the values of (1.c0) are without the correction.
J. MORIIZUMI ET AL.
REFERENCES
Through the observation of short-lived radon decay products associated radioactive aerosol in the indoor air for 2010–12 in Nagoya, Japan, consistent correlation of the variation in AMAD to existence of rainfall in previous 24 h was not obtained. The means and fluctuation ranges of AMAD and sg for the outdoor air were comparable with those for the indoor air observed in the same period. The alpha emission counts of the nucleation and the accumulation modes in the indoor air showed the same tendency in terms of the existence of rainfall as Mostafa et al. (5), while those in the outdoor have insignificant difference between Rainfall and No-rainfall classes. The method of total aerosol particle number concentration measurement by the LPC without any dilution was developed. The positive quasi-linear relationships were observed between the AMAD and the total number concentration with weaker correlations. Although exact estimation of AMAD by the LPC as proxy would be insufficiently feasible at present, long-term, continuous and highly frequent measurement with the LPC substantially enlarges the volume of observation and would be applicable to survey and screening to perform precise size distribution measurement with the impactor at suitable timing.
1. Marcinowski, F. M. Nationwide survey of residential radon levels in the US. Radiat. Prot. Dosim. 45, 419–424 (1992). 2. Papastefanou, C. Radioactive Aerosols. Radioactivity in the Environment, Vol. 12. Elsevier (2008). ISBN:978 0 08 044075 0. 3. ICRP. Human respiratory tract model for radiological protection. ICRP Publication 66. Ann. ICRP 24(1– 3), (1994). 4. Harley, N. H., Chittaporn, P., Fissene, I. M. and Perry, P. 222Rn decay products as tracers of indoor and outdoor aerosol particle size. J. Environ. Radioact. 51, 27– 35 (2000). 5. Mostafa, A. M. A., Tamaki, K., Moriizumi, J., Yamazawa, H. and Iida, T. The weather dependence of particle size distribution of indoor radioactive aerosol associated with radon decay products. Radiat. Prot. Dosim. 146(1– 3), 19–22 (2011). 6. Porstendo¨rfer, J., Zock, Ch. and Reineking, A. Aerosol size distribution of radon progeny in outdoor air. J. Environ. Radioact. 51, 37– 48 (2000). 7. Reineking, A., Becker, K. H. and Porstendo¨rfer, J. Measurements of activity size distributions of the shortlived radon daughters in the indoor and outdoor environment. Radiat. Prot. Dosim. 24(1), 245–250 (1988). 8. Rahman, N. M., Iida, T., Yamazawa, H., Yokoyama, S., Moriizumi, J., Saito, F. and Ito, T. The study of activity median aerodynamic diameter using imaging plate technique for assessment of effective dose from radon and its decay products. Radiat. Prot. Dosim. 124(4), 360–371 (2007). 9. Rahman, N. M., Iida, T., Saito, F., Koarashi, J., Yamasaki, K., Yamazawa, H. and Moriizumi, J. Evaluation of aerosol sizing characteristic of an impactor using imaging plate technique. Radiat. Prot. Dosim. 123(2), 171– 181 (2007). 10. Iida, T., Ikebe, Y., Suzuki, K., Ueno, K., Wang, Z. and Jin, Y. Continuous measurements of outdoor radon concentrations at various locations in East Asia. Environ. Int. 22(Suppl. 1), S139– S147 (1996). 11. Moriizumi, J. et al. Continuous observation of atmospheric 222Rn concentrations for analytic basis of atmospheric transport in East Asia. J. Nucl. Sci. Tech. Suppl. 6, 173– 179 (2008).
FUNDING This study was partly supported in the project ‘Construction of natural radiation exposure study network’ in the Programme of Promotion of International Joint Research under the Special Coordination Funds for the promotion of Science and Technology operated by MEXT, the Japanese government. This work was partly supported by JSPS KAKENHI (grant no. 24110002).
Page 6 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Ondokuz Mayis University on November 6, 2014
CONCLUSIONS
