EMPLOYEE EXPOSURE TO RADON-222 AND THORON-220 IN THREE FISH CULTURE STATIONS IN PENNSYLVANIA Robert K. Lewis PA Department of Environmental Protection Bureau of Radiation Protection, Radon Division And Naomi H. Harley, PhD New York University School of Medicine Department of Environmental Medicine Abstract Employee exposures to Radon-222 and Thoron-220 were measured in three Commonwealth fish hatcheries using specially designed personal dosimeters to determine whether remediation might be necessary. Employees utilizing the hatch house would wear the dosimeter and keep track of their time in the hatch house. Area detectors were also deployed full time in each hatch house. Exposure measurements were compared to NRC, EPA, and OSHA exposure limits. All measured employee exposures to Radon-222 and Thoron-220 were very low and well below currently established regulatory limits. However, hatch house radon concentrations are significantly elevated above the U.S. Environmental Protection Agency residential guideline of 148 Bq m-3 (37 Bq m-3 = 1 pCi/L). Key words: Radon-222; Thoron-220; Radon/thoron in water; employee exposure Acknowledgements- Partial funding for this work was supplied by the U.S. Environmental Protection Agency State Indoor Radon Grants. We would like to thank Dr. Passaporn Chittaporn of the NYU Department of Environmental Medicine, for the CR-39 film processing and etching, track counting, and QC analysis of all personal dosimeters. We also thank the Pennsylvania Fish and Boat Commission employees for their cooperation throughout this project, particularly Mr. Thomas Cochran, Bureau of Fisheries 1 Introduction The Pennsylvania Fish and Boat Commission operate 14 fish culture stations throughout the Commonwealth. The exclusive purpose of this operation is to provide stocked fish in Commonwealth lakes and streams for licensed anglers. Both warm water and cold water species are raised and stocked. The majority of the fish culture stations uses ground water, either spring or well water, for their daily operations. Additionally, the daily amount of water use can be significant, well into the hundreds of thousands of gallons per day. The measurement of radon-222 in both water and air at these 14 facilities is described in a previous paper by this author (Lewis 2001). Three out of the 14 facilities were found to have elevated radon concentrations of concern in the hatch house, during our previous survey. It is these three facilities that are the focus of this study. For this paper those facilities are coded B, BS, and C. Other investigators (Dwyer and Orr 1992; Kitto 1998; Harris and Craig 1991; and Stillwell 2008) have also found high radon levels in similar facilities in Montana, New York, Missouri, and Maine, respectively. The mechanism of action is that gaseous radon-222 (radon) or radon-220 (thoron) diffuses from the rocks and soil surrounding the well or spring into the facility water supply. Flows can be very variable, but typically ranges from ten’s to several thousand liters per minute. This water is drawn into the hatch house (HH) for the rearing of the young fish. Due to the high Henry’s Constant for radon in water, which at 20 degrees centigrade and 1 atmosphere is 2200, 2 there is a strong tendency for radon to off-gas into the surrounding air of the hatch house. Ground water can often be saturated with nitrogen, depending upon the area of the state, and also have very low oxygen levels. Both of these conditions are deleterious to fish production. Due to these conditions many of the hatch house facilities use various aeration techniques on the influent water supply within the enclosed structure to drive off nitrogen and increase the oxygen; this also favors the release of the waterborne radon, and presumably any thoron into the hatch house air. This mechanism provides for one pathway for both radon and thoron exposure to hatch house personnel. Another mechanism of possible exposure is from the radon/thoron in soil gas. This is the gas contained in the soil directly below the foundation of the building. This is the typical route of gas entry into most residential buildings. The preferable solution to any elevated radon problem is to mitigate the incoming radon. For the case of soil borne radon this almost always entails the use of an active sub-slab depressurization system. For the case of water borne radon, particularly where large quantities of water are being used, aeration is the mitigation method of choice. Although the hatch houses have crude types of aeration systems in place for removal of nitrogen and oxygenation of water they are often located inside the hatch house. These systems are contributing to the radon/thoron exposure within the hatch house. For effective removal of waterborne radon/thoron these systems would need to be outside the enclosed structure. These engineered aeration systems can be quite costly to initially develop and install. Due to limited funding we decided to measure the employee personal exposure, over the work year to determine whether further remedial steps should be taken. 3 We do not know the exact proportion of Radon/Thoron exposure arising from the two potential sources; soil gas, and off-gassing from incoming water. We measured the cumulative exposure arising from potentially both sources in the hatch houses. Materials and Methods The 222 Rn and 220Rn personal or area monitor was designed to measure radon and thoron in duplicate in order to estimate measurement precision. Two chambers have a diffusion barrier and inhibit thoron entry, two chambers permit both radon and thoron entry. Thoron is calculated by signal difference. A conducting foam, directly beneath the entry caps permit only radon or thoron gas passage. The monitor face is four lobed with length 5 cm. and thickness 1 cm. It is molded from electrically neutral ABS (CNi) plastic, i.e., plastic with embedded nickel coated carbon fibers. The nuclear track film used for detection of the alpha particles is laser cut 9x9 mm square solid state nuclear track film (CR-39). The pristine film background is 5 to 15 tracks. Tracks over the entire film area are counted. The unit displays no charge artifacts and the calibration is constant in all situations. Using a video imaging system for track counting**, the efficiency for 222 Rn is ( 0.009 Track per Bq m-3 day) (0.32 Track per pCi/L day) and the lower limit of detection is 220 Bq m-3 day (6 pCi/L day). The efficiency for thoron is 0.013 Track per Bq m-3 day, (0.27 Track per pCi/L day) and the lower limit of detection is 1100 Bq m-3 day (30 pCi/L day). 4 After exposure, the CR-39 alpha track film is etched for 20 hours in 6 N KOH to reveal the alpha particle tracks as shallow pits. Track counting with image analysis and about 10 to 20% of samples are scored visually. Pristine nuclear track film and exposed positive controls are etched with each batch of field samples for Quality Control. The hatchery personal that frequent the hatch house were each equipped with a radon/thoron dosimeter and instructed to wear the dosimeter during their time in the hatch house. They were also instructed to record the number of hours worked in the hatch house, the approximate water flow rate into the hatch house, and the condition of the building (windows open/closed). Dosimeters were worn only during hatch house work. When not in use, dosimeters were stored in the manager’s office, which had previously been determined to be low in radon, with values of 52 Bq m-3 for facility C, 74 Bq m-3 for facility BS, and 37 Bq m-3 for facility B. Radon/thoron dosimeters were also placed in each hatch house (area monitors) during the year in order to determine the annual average radon and thoron concentration. So as not to “saturate” the film of the dosimeters with high track counts they were all changed out on an approximate quarterly basis, this includes workers as well as the hatch house dosimeters. Table 1 shows the deployment schedule and the hatch house radon and thoron concentrations. Previous radon measurements from the hatch houses did vary significantly depending on amount of water use in the hatch house and the amount of ventilation of the building. Facility B hatch house was measured using a continuous radon monitor for one week during May of 2000 and averaged 1000 Bq m-3, with spikes to 2220 Bq m-3, with the hatch house using 757 lpm of 5 water. Facility BS hatch house was measured using a year-long Landauer alpha track detector and recorded an annual average of 640 Bq m-3. Finally, facility C hatch house was measured via two short-term measurements; one in February 2001 with the hatch house using 227 lpm and the radon at 740Bq m-3, and another measurement April 2001 at 1041 lpm showing 1480 Bq m-3. This data does show that hatch house radon values are significantly elevated above U.S. Environmental Protection Agency guideline value of 148 Bq m-3. It should be pointed out that no previous thoron measurements were made in any of the facilities, either in the water or air. Results Table 2 below presents data on the cumulative exposure to all monitored users to the hatch house for each of the three facilities. Both a radon and thoron cumulative exposure is provided, in units of working level months (WLM). Our Radon/Thoron working level calculations assume an equilibrium ratio, Feq of 0.4 for radon and 0.02 for thoron. The radon and thoron decay product concentrations in Table 2 are in units of WL and WLM, so that published dose factors may be applied directly. We had two choices for the calculation of the Radon/Thoron working level values; use the results from the dosimeters worn by each hatch house worker, or use the dosimeters that were left in place in the hatch house (area monitors) to monitor the hatch house concentrations and use employee occupancy time. We decided to use the area monitors because the hatch house reported exposure time from the personal dosimeters was small relative to total time deployed. 6 When not in use the personal dosimeters where stored in the managers office and this did not reflect the hatch house exposure. Average radon concentrations in the three facilities seem consistent with previous measurement data with certain caveats. Facility B hatch house showed a week long average concentration of 1000 Bq m-3 measured in 2000, however, this was during water use and this facility does not use water to the same extent as the other two facilities. Figure 1 shows the details of water use for the three hatcheries over a one year period. Thus, this lack of water use throughout the entire year, which is the major source of the radon, explains why our current annual average is 270 Bq m-3. The BS facility average annual value made in 1999 of 640 Bq m-3 is in reasonable agreement with the current value of 464 Bq m-3. Finally, the 740 Bq m-3 for facility C measured in 2001 is consistent with our current value of 512 Bq m-3. Employee cumulative radon exposure from the three facilities shows a range from 0.186 to 0.001 WLM yr-1. These values are all well below any currently established exposure limits for radon, where both the US EPA and the US Nuclear Regulatory Commission (NRC) have set a limit of 4 WLM yr-1. The fish hatchery facility should technically be regulated by the Occupational Safety and Health Administration (OSHA); however, OSHA still uses 1970 vintage 10 CFR 20 ionizing radiation regulations, which reference 12 WLM yr-1 for radon exposure in the workplace. There is a wealth of radon concentration data for various building types and in various countries. The United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) 2000 report provides valuable data for concentration values. They show a geometric mean concentration for numerous countries around the world at 37 Bq m3 . This value is similar to what the US EPA reports for the United States at 48Bq m-3 (EPA, 7 1992). Pennsylvania however, has much higher concentrations than the US and world-wide averages, with a basement average for the Commonwealth at about 263 Bq m-3 (Lewis, 2007). Our hatchery data is in line with the data for the Commonwealth with a range from 270 to 512 Bq m-3. Employee cumulative thoron exposure ranges from 0.0001 to 0.031 WLM yr-1. The NRC and EPA exposure limit for thoron is 12 WLM yr-1. The hatchery workers are well below this exposure limit for thoron. The world wide data for thoron concentrations in buildings and residences is much sparser. There are also more uncertainties associated with this data due to the various measurement techniques used, less effort devoted to quality assurance, and generally low concentrations are being measured. UNSCEAR 2000 quotes a range of equilibrium equivalent concentrations (EEC) of 0.2 to 12 Bq m-3 . See Appendix A for explanation of EEC. More recent data from 117 Winnipeg houses show a range from 5 to 297 Bq m-3 , with 60 of those homes below the detection limit (Chen, et. al., 2009). Our data above for the hatch house annual average falls within these ranges with values of 29, 53 and 113 Bq m-3. UNSCEAR (2000) adopts central values for the effective dose coefficient of 9 and 40 nSv per Bq m-3 hour for radon (EEC) and thoron (EEC) respectively (6.4 and 1.9 mSv/WLM respectively). The calculated dose is to target cells in bronchial epithelium implicated in bronchogenic carcinoma. Table 3 shows the calculated annual effective dose based on the hatch house area monitor measurements, the reported employee exposure time in Table 2 and the UNSCEAR dose factors. 8 Discussion Employee exposure to both radon and thoron as measured in this study was surprisingly low, as compared to the available exposure limits set by EPA, NRC, and OSHA. These low exposures are likely due to the relatively short work times in the hatch house, combined with work during times of low radon concentration. Previous continuous radon monitor measurements (Lewis, 2001) showed very cyclic daily patterns to radon concentration, with lows around 2-5 PM and highs around 5-7 AM. This is consistent with Porstendorfer et. al. 1994 who found the highest activity during the night and early morning hours when the atmosphere was the most stable. At noon and early afternoon the mixing of the lower atmosphere is strongest and radon concentration is lowest. Finally, the low exposures are also due to building ventilation, when windows and doors are left open or ventilation fans are running. Radon concentration measurements in fish hatcheries are sparse and to our knowledge no thoron measurements are available. Scott (1997) quotes personal dosimetry measurements for radon in two Ontario fish hatcheries ranging from 0.05 to 4.5 mSv yr-1 (10 mSv = 1 rem). This range of values is consistent with our Table 3 calculated doses for radon (0.003 to 1.03 mSv yr1 ). Measurements of radon and thoron in this study show that estimates of decay product exposure and effective dose are well within occupational limits, ranging from 0.001 to 0.186 WLM for radon ( 4 to 1188 uSv per year) and 0.0001 to 0.031 WLM ( 0.2 to 58 uSv per year) for thoron. 9 Conclusions Cumulative exposure to both radon and thoron made over one full year suggest that mitigation measures at this time are not necessary in these hatch house facilities. If future modifications are made to the supply water that significantly increase radon and thoron concentration off-gassing or employee exposure time is increased, well designed aeration systems, outside the building structure should certainly be considered to further reduce radon/thoron concentrations in the supply water with a concomitant reduction in exposure. Repeat measurements during water use should be made at least annually along with employee time records to determine any changes. The maximum effective dose values in the fish hatcheries studied were 1188 and 58 µSv per year for radon and thoron respectively We would urge other facilities to make similar long term baseline radon and possibly thoron measurements, even though our cumulative exposures are low. Other hatch house facilities in Montana, New York, Missouri, and Maine have already recorded indoor concentrations of radon in the hundreds of pCi L-1 with remediation required. 10 **The CR-39 film is etched overnight at NYU for 20 hours in 6N KOH along with pristine blanks and positive controls for QC. The tracks are counted with an Olympus SZ-CTV zoom microscope and tracks scored using Data Translation Global Lab Image software. About 20 percent of all films are scored visually as a QC measure using hard copy printout from a Minolta microfiche reader at 20X magnification. 11 Table 1. Radon and thoron concentrations measured in three Pennsylvania fish hatcheries over a one year interval Facility Time Period (dd-mm-yy) Hatch House Average (Bq m-3) Rn-222 Th-220 C 6-7-07 to 15-8-07 (30days) 6-9-07 to 10-12-07 (95 days) 27-12-07 to 7-5-08 (132 days) 13-5-08 to 6-8-08 (85 days) Average +/- SEM BS 6-7-07 to 15-8-07 (30 days) 6-9-07 to 10-12-07 (95 days) 1-1-08 to 7-5-08 (128 days) 8-5-08 to 8-8-08 (93 days) Average +/- SEM B 6-7-07 to 15-8-07 (30 days) 6-9-07 to 10-12-07 (95 days) 21-12-07 to 7-5-08 (138 days) 8-5-08 to 8-8-08 (98 days) Average +/- SEM SEM= Standard error of the mean= S.D/√n 603 362 543 595 525 +/-22 +/- 22 +/- 52 +/- 41 +/-56 7 +/- 22 15 +/- 7 44 +/- 26 N.D. 22 +/- 11 414 +/- 7 344 +/- 181 492 +/- 85 566+/- 26 455 +/- 48 81 +/- 56 167 +/-2 81 +/- 11 N.D. 111 +/- 28 44 +/- 22 437 +/- 11 204 +/- 118 Missing 237 +/- 114 -85 +/- 0 3 +/- 22 118 +/- 2 Missing 11 +/- 34 12 Table 2. Employee radon and thoron exposures measured over a one year interval in three Pennsylvania fish hatcheries Facility Employee Hrs in HH Rn WL Rn WLM Th WL Th WLM B LD 275 0.0256 0.0413 0.0008 0.0013 TE 12 0.0256 0.0018 0.0008 0.0001 TH 72 0.0256 0.0108 0.0008 0.0003 TW 49 0.0256 0.0074 0.0008 0.0002 BM 4 0.0256 0.0006 0.0008 0.0000 -3 B HH Annual Average Radon Concentration is 237 Bq m Annual Average Thoron Concentration is 11 Bq m-3 BS JV ZF NV DB JB 412 0.0492 0.1192 0.0081 37 0.0492 0.0107 0.0081 629 0.0492 0.1820 0.0081 30 0.0492 0.0086 0.0081 74 0.0492 0.0214 0.0081 Annual Average Radon Concentration is 455 Bq m-3 Annual Average Thoron Concentration is 111 Bq m-3 0.0196 0.0017 0.0300 0.0014 0.0035 CL MH KM 208 0.0568 0.0694 0.0016 108 0.0568 0.0360 0.0016 95 0.0568 0.0317 0.0016 Annual Average Radon Concentration is 525 Bq m-3 Annual Average Thoron Concentration is 22 Bq m-3 0.0019 0.0010 0.0009 BS HH C C HH Radon WL equals (HH Avg. Rn x 0.4)/3700; Thoron WL equals (HH Avg. Th x 0.02)/274 WLM equals (Rn or Th WL x hrs)/170 13 Table 3. Employee annual effective dose in three Pennsylvania fish hatcheries -1 -1 222Rn(µ Sv yr ) 220Rn(µ Sv yr ) Facility Employee B LD TE TH T BM 234 10 62 42 3.4 0.006 0.003 0.017 0.011 0.001 BS JV ZF NV DB JB 675 61 1030 49 121 0.99 0.09 1.5 0.07 0.18 C CL MH KM 393 204 180 0.1 0.051 0.046 1µSv = 0.1 mrem 14 Figure 1 Hatch House Water Use Hatch House Water Use 700 600 500 400 gpm B BS C 300 200 100 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month 15 References Chen, J. Schroth, E., Mackinlay, E., Fife, I., Sorimachi, A, and Tokonami, S. Simultaneous 222Rn and 220Rn Measurements in Winnipeg, Canada. Radiation Protection Dosimetry, April, 2009. Dwyer, W.P., and Orr, W.H. Removal of radon gas Liberated by Aeration Columns in Fish Hatcheries, The Progressive Fish-Culturist, 54:57-58, 1992. Harris, D.B., and Craig, A.B. Control of Radon Releases in Indoor Commercial Water Treatment. USEPA Office of Air and Energy Engineering Research Laboratory, 1991. Kitto, M.E., Kunz, C.O., McNulty, C.A., Covert, S., and Kuhland, M. Radon Measurements and Mitigation at a Fish Hatchery, Health Physics, 74(4):451-455; 1998. Lewis, R. An Investigation of Radon Occurrence in Pennsylvania Fish and Boat Commission Fish Culture Stations. Proceedings of the International Radon Conference, Daytona Beach, Fl. October, 2001. Lewis, R. A Statistical Report of Pennsylvania- Radon-222. Pennsylvania Department of Environmental Protection, Bureau of Radiation Protection, Radon Division. April 2007. Unpublished report. Porstendorfer, J.; Butterweck, G.; and Reineking, A. Daily Variation of the Radon Concentration Indoors and Outdoors and the Influence of Meteorological Parameters. Health Physics 67: 283-287; 1994. Scott A.G. Occupational Doses from Radon. The Health Physics Society’s Newsletter. Volume XXV Number 11, November 1997. Stillwell, R. Fishy Radon Problems or Some Radon Problems Require More than Outreach. Oral Presentation at the 18th International Radon Symposium. September, 2008, Las Vegas, NV. U.S. Environmental Protection Agency. Technical Support Document for the 1992 Citizen’s Guide to Radon, EPA 400-R-92-001, May 1992. 16 Appendix A Equilibrium Equivalent Concentration (EEC) The EEC is another way of expressing the radon (or thoron) progeny concentration in units of activity concentration, either Bq/m3 or pCi/L. This term is more commonly used in Europe than the United States. By definition it is the concentration of radon in air, in equilibrium with its short-lived decay products which would have the same potential alpha energy concentration as the existing non-equilibrium mixture. Expressed mathematically it is: EECrn = 0.104 C(Po-218) + 0.517 C(Pb-214) + 0.379 C(Bi-214) Where the constants are the fraction of the PAEC (WL) C = concentration of the three isotopes in Bq/m3 or pCi/L Also can be expressed as EEC = Rn Gas x ER, or EEC = Th Gas x ER The dose is related to the working level (WL), but the WL is not the measurement of choice in residences and most other places because it requires real time equipment. Therefore, if we want dose but are measuring radon gas with passive detectors the EEC is important. 17 COLD CLIMATE RADON MITIGATIONS A CANADIAN’S PERSPECTIVE Preliminary Report Presented at the 2009 AARST International Radon Conference St. Louis, U.S.A. Robert (Bob) Wood Owner: Mr. Radon (Inc.)-Toronto, Ontario Bachelor of Education B.A. in Adult Education Associates Degree in Industrial Training CRRMT, CRMS, Radon International Speakers List CRPA, NEHA, NRPP, AARST, IAC2 www.mr-radon.ca bob@mr-radon.ca The author has received partial funding from Canada Mortgage and Housing Corporation (CMHC), a Canadian government owned corporation, for the research leading to this publication. 18 Abstract This preliminary report is part of a larger research project that will record how over 50 radon mitigators from North America and Europe effectively mitigate homes in cold climates. This preliminary research paper is an investigation of the methods and reasoning of 29 radon mitigators who work in the cold climate areas of the USA. The author’s hypothesis was that most of the radon mitigators in the USA follow the ASTM E2121 guidelines with regard to fan location and vent termination. In addition, that they would have good “workarounds” to avoid premature fan failures due to icing, freeze up and snow loads that interfere with radon systems. This research study found that all , not most, of the mitigators interviewed were following the standards of ASTM E 2121 in regard to fan location being located outside of the living envelope. Only the two mitigators interviewed from Zones 7 & 8, working in Alaska, stated that occasionally when they had no other warm choice, they would the consider locating the fan inside the living space. The researcher heard from many mitigators that in cold and really cold climates (ASHREA Zones 6, 7 and 8) some elements of home construction design are different than in warmer climates. Most have heated attached garages and separate heated mechanical rooms (outside of the living space). These were the normal construction design practices allowing them to follow ASTM E 2121 by placing the fan in these locations and out of the living space. For the average mitigator, some of the “workarounds” that are identified in the report are simple, easy and make sense. Presenting and sharing these ideas should benefit all mitigators in cold climates. 19 This is a preliminary report that is part of a larger research project that will look at how over 50 radon mitigators from North America and Europe effectively mitigate homes in cold climates. This research paper is an investigation of how and why radon mitigators who, working in the cold climate areas of the USA mitigate residential homes from high levels of radon. This paper looks at whether these US mitigators actually follow the international ASTM E 2121-03 standard or not. The three specific areas of focus for this paper are the location of the fan, the location of the discharge piping and the direction of the termination. This survey will also find ways mitigators actually make these systems work in cold climates and what way these individuals have developed “work arounds” that work within the international ASTM E 2121-03 standard and where they deviate from the standard. These USA working mitigators have the “boots on the ground” experience in getting successful long term radon mitigations for their clients should prove to be an informative resource. Canada is in the process of creating radon mitigation procedures and possibly standards of their own. As a Canadian mitigator I feel certain that this information will be of great value to the professionals who are formulating the protocols or standards for Canada. The information from this survey is tracked in the ASHREA temperature zones 5, 6, and, 7 and 8 to ensure that the conclusions are more useful. Canada’s highest population density is concentrated very close to the US border. This places most of our population in ASHREA zone 5 or zone 6 with limited populations in zone 7 and 8. 20 Canada’s ASHREA zones Canada’s Population Density The author’s hypothesis for this study was that most of the radon mitigators in the USA will be following the ASTM E2121 guidelines with regard to fan location and vent termination and will have come up with good practices as “work arounds” to avoid premature fan failures due to icing and freeze up and snow loads interfering with radon systems. These results are the preliminary results from a larger study. This preliminary paper is the result of tabulating a study of 29 mitigators from the USA. These mitigators were contacted over the phone. The methodology for these interviews was; two radon mitigators having a conversation about radon mitigations and the survey answers were recorded by the author as the conversation happens rather than a list of dry questions. With twenty nine interviews we have seventeen from zone 5, six from zone 6 and six from zone 7 and 8. Zone 6 Zones 7 & 8 Zone 5 1 2 3 Split of interviews 21 Here is what this survey discovered; most radon mitigators are owner operators, 55% are single person businesses while 23% have 2 to 3 employees. The balance, 22%, have more than 4 employees. These numbers seemed to have been consistent across all three ASHREA zones. Most mitigators are some what busy with average mitigations per month being 13.67. Many reported that they are not as busy as in previous years. The numbers stood up well for zone 5 mitigators with an average of 13.5 mitigations but went up to 17 mitigations per month for the zone 6 mitigators and falling to 10 mitigations per month for zone 7 and 8 mitigators. Across all three zones most mitigators, who were busy (over 13 mitigations / month) were servicing wide geographical areas. The average cost to install a radon mitigation system across all three zones averaged $1260.00. The results of the zone 5 data was that mitigations came in at a lower average cost of $1180.00, zone 6 was $1200.00 and zone 7 and 8 $1550.00. It seemed that the cost of the system was driven by amount of available work. If a mitigator is busy (over 13/month) they can charge less because the fixed cost/sale is lowered. These numbers make sense because as we headed into colder climates more mitigations were “inside” jobs hence a higher cost. 93% of the mitigators provide a post mitigation test and the balance recommend one be completed by a third party. Most mitigators (60%) who provide a test use a charcoal test kit, 25% use liquid scintillation and 16% using a CR monitor. We asked where the mitigators typically located the fan across all three zones. 0% choose to put the fan within the living space, 59% chose fan placement outside at the side of the home, 31% chose the attached garage and 10% in the attic of the home. When we began to review the different climate zones we saw a huge change of attitude. In zone 5, 88% were putting the fan at the side of the home, as a first choice of location and only 12% were installing the fan in attic of the home. When asked why, the overriding reasons were cost or ease of installation of the mitigation system. In zone 6 only 16% reported using ‘outside on the side of the home’ as their first choice and 84 % are using the attic of the home or attached garage as their first choice of location of the radon mitigation system. In zones 7 and 8 the numbers are very similar with only 16% using ‘outside on the side of home’ as their first choice and 84 % were using the attic of the home or attached garage as their first choice of location of the radon mitigation system. The only difference between zone 6 and zone 7 and 8 group data is that in the zone 7 and 8 group we had a higher incidence of using the garage. The reasons for the choice of location followed this trend of similarity between zone 6 and zone 7 and 8. The overriding reason for choice of fan and piping location being ‘warmth’ the mitigators who put the fan in the attic or attached garage and the smaller group choosing to put the fan on the side of the home sighting reduced cost of their outside system as the primary reason for this choice. Even those interviewed from Alaska were finding a location to place the fan in the warmth of the home but have it outside of the envelope of habitable space. 22 When the author enquired about fan failures (for any reason) we were amazed to see how reliable these fan systems are with most mitigators reporting less than a 5% failure rate on fans under 5 years old. In the fans older than five years 26% of the mitigators reported finding no fan failures, 11% reporting manufactures defects (squeaky), 49% from old age, 7% from ice damage and 7% from other causes (squirrels). 100 % of the surveyed mitigators contacted were discharging the soil gases of the radon mitigation system above the line of the eave of the roof including those from Alaska. Those from Alaska are terminating in a horizontal fashion rather than in an upright fashion. They typically try to choose a southern or eastern exposure gable wall if possible to take advantage of sun warming or the lee side away from prevailing winds. The mitigators were asked “do you do anything to deal with the potential of creation of an ice ball, icing and fan blockage in winter”? As a group 66 % did nothing. When reviewing the zone breakdown we had some surprises. In zone 5 we had a 70% to 30% split in favor of do nothing. The proactive group had some interesting work around solutions. One mitigator always installs the last piece of pipe in black ABS pipe to take advantage of sun warming. Another always tries to use southern exposure, another uses all black ABS pipe whenever he can. (He likes the color on brick homes). Condensate bypasses were mentioned quite often as a solution that was used. Surprisingly in zone 6, 100% of the mitigators said that they do nothing to deal with the potential of ice ball icing and fan blockage. In zone 7 & 8, 84 % said that they do make adaptations to be proactive in dealing with potential ice ball, icing and fan blockage in winter. While 16% did nothing. It was recorded that in Alaska mitigators terminate in a horizontal fashion above the eave and in a gable end wall not very far out from siding (typically less than an inch) so that icicles are not formed, or they insulate the pipe like a bathroom vent (wrap with pipe with 1” insulation or install commercial pipe insulation) and go up through roof. Some other solutions that were offered by zone 7 and 8 mitigators included the insulation of piping where exposed to cold which came up 32% of the time. They often talked about this being necessary on horizontal runs. Use of only 4 inch schedule 40 cellular (foam) core pipe came up 66% of the time, many quoted the insulation value of cellular core pipe. One mitigator always spray paints his last piece of pipe through the roof black. (He tries to place fan in the attic.) Another if he has an outside run, pulls armourflex insulation inside 4 inch pipe to create “a neat workmanlike project that just won’t freeze”. Another installs a ¾” valve indoors near the u tube and has the customer adjust the opening based on outside temperature. I spoke to all three major Radon Fan manufactures about placing their fans in cold outdoor situations and any concerns or installation instructions that they would have. 23 RadonAway: “All RadonAway fan models are rated for outdoor use with an operational temperature range of -20F to 120F. (-30C to 50C). However, there are a number of considerations that should be taken into account when designing and operating an Active Soil Depressurization (ASD) system in extreme cold climates. Freeze-up can be a serious problem in such an ASD system and if the system freezes it can totally block off the exhaust pipe, rendering the system completely ineffective for radon reduction. Interior pipe runs are preferred to minimize exposure to the extreme cold. Pipe and fan insulation and auxiliary heat can be included in the system design to minimize freeze-ups. Steps should be taken to prevent ice from falling into the fan that could potentially damage the impellers. Piping supports should be designed for the additional weight of any potential ice load. Exhaust points should be extended to ensure they remain open in normal snow cover. Additional factors may need to be considered to account for local codes and building practices” Fantech: “Fantech fans are rated for outdoors use. Fantech has no concerns about the fan being placed outdoors in cold temperatures provided: 1/ Fan is installed in a vertical position 2/ Do not turn fans off and on in cold temperatures 3/ Installer should ensure that an adequate moisture bypass built into the piping system.” Festa Manufacturing: “Festa fans are rated for outdoor use. They had no concerns about their fans being placed outdoors in cold temperatures provided that “the fan is installed in a vertical position” One of my last questions to these mitigators was what would you change if you could to any part of ASTM E2121? The majority, 62%, stated that they couldn’t think of anything to change. Those that had things that they thought should be changed quite often had several ideas. One was the requirement to have a 4 inch suction header if multiple points, was an issue for many mitigators, their objection was not to this rule when multiple mitigation points were planned, but as a retrofit for those 3” systems that were not able to achieve the desired radon reduction on the first go. These mitigators pointed out that they were concerned about achieving lowered radon levels and that this rule was onerous in a retrofit application. The issue of staying 10 feet away or above a chimney came up almost as often. Most felt that treating a chimney as a potential point of re-entrainment, just did not have good science behind it. Some reported that they had not found research to support this requirement. Several mitigators felt the requirement to provide a fire rated assembly at the garage wall and ceiling was over the top when their system was going in beside other penetrations to the fire wall that were unprotected i.e. central vacuum. More than half the mitigators working in zone 7 felt that minimum stack size outside should be 4” and minimum schedule for pipe should be schedule 40. Two mitigators 24 pointed out that the manufacturers of schedule 20 (SDR 35) do not support its use above ground. As it has no UV inhibitors in it. 10% of the mitigators questioned suggested that putting the fans in the living envelope of the home, could result in a cost savings to the mitigation and perhaps extend the life of the fan. One went so far as to suggest that all the manufacturers had to do was improve their gasketing around fan and electrical connections and this should be practical. The author asked two of the mitigators interviewed who were mechanical engineers about this idea and got similar responses from both. “The basis of sound engineering for extraction of any hazardous or unpleasant gasses is to always use negative pressure inside the building envelope to ensure no cross contamination can occur.” The representative from Fantech that the author interviewed also stated that “their radon fans should not be installed indoors. It is extremely rare, but occasionally gaskets fail. The resulting high radon levels inside a home that the owner felt was protected from radon could detrimental to the occupant’s health” When reviewing the data about the authors hypothesis for this study that most of the radon mitigators in the USA are following the ASTM E 2121 guidelines with regard to fan location and vent termination, our study found that all , not most, of the mitigators interviewed were following the standards of ASTM E 2121 in regard to fan location being located outside of the living envelope. Only the 2 mitigators interviewed from zone 7 & 8 working in Alaska stated that occasionally when they had no other choice and would the consider locating the fan inside the living space but they had to consider their severe climatic conditions. It was interesting to note that they also stated that those severe climatic conditions also led to fact that almost all homes in Alaska were designed with heated attached garages and heated attached mechanical rooms that they could utilize for fan location. The data from the interviews with regard to vent termination show again that all, not most, are terminating the radon vent piping above the eave. When we looked at the data for direction of vent termination only 6% of those surveyed did not terminate in an upward direction. It was again our colleagues from Alaska that stated that they terminate horizontally very close to the siding as not to create a point for an icicle to happen. Should ice form they would prefer it to be on the siding and not as a dangling icicle. It is the conclusion of the author from this data that even in a cold climate radon mitigation systems work and work well. They can be installed in a cold climate situation and meet the requirements of ASTM E2121. Fans and piping can be located outside of building envelope as long as due care for cold climates situations are brought into play. More care may be required as to the location of the fan in colder environments and higher cost choices of attic or attached garage should be utilized to ensure that while exposed to cold the fan is out of the worst of the weather extremes. The mitigator may have to make choices around using piping with insulation values and utilization of 4” pipe instead of a 3 inch pipe that could be used in a warmer climate. In the extreme cold of zones 7 and 8 mitigators may do well to consider insulating piping systems. Vent terminations of above the eave is a workable rule. Vent termination direction may have to be considered in cold zones of 7 and 8. 25 References RADON: A Guide for Canadian Homeowners, CMHC, 2007 ISBN: 0-662-25909-2 Guide for Radon Measurements in Public Buildings, Health Canada, 2008 ISBN: 978-1-100-10183-5 ASTM E 2121 – 03 ASTM International, Feb 2003 Consumer’s Guide to Radon Reduction, U.S. EPA, 402- K-06-0942006 Building Radon Out, U.S EPA, 402-K-01-002 April 2001 RADON MITIGATION Alaska Experiences, Costs, Results. (Richard Seifert) revised The Cooperative Extension Service, March 2007 WHO Working Group, Indoor Air Quality: A Risk Based Approach To Health Criteria For Radon Indoors, WHO Working Group, April 1993 Assessment of Radon Mitigation Methods in Low-Rise Buildings, U.S. Department of Housing and Urban Development, 1996 Model Standards and Techniques for Control of Radon in New Residential Buildings, U.S. EPA 402-R-94-009, March 1994 Effective Interventions to Reduce Indoor Radon Levels, National Collaborating Center for Environmental Health, July 2008 Radon Testing and Remediation What Works? National Collaborating Center for Environmental Health, July 2008 Radon Reduction Technical Installation Drawings, http://www.infiltec.com/inf-drar.htm Radon Legislation and National Guidelines, Swedish Radiation Protection Institute, July1999 ASHREA maps http://www.firestonebpco.com/contractors/Bulletins/mktgBulletins/increasedRvalue/ Canadian Population Map https://travelcanada.wikispaces.com/Population+Map+of+Canada 26 Bill Broadhead, Radon Mitigation Installation Efficiency Techniques AARST International Radon Conference Cherry Hill, NJ. Sept 1998 http://www.wpbradon.com/pdf/Radon%20Mitigation%20Installation%20Techniques%2098.pdf 27 PRE- AND POST-MARKET MEASUREMENTS OF GAMMA RADIATION AND RADON EMANATION FROM A LARGE SAMPLE OF DECORATIVE GRANITES Daniel J. Steck1 Physics Department, St. John’s University Collegeville, MN 56321 ABSTRACT Reports of radiation from granite counter tops generated public interest in the potential exposures to external gamma radiation and internal exposure from radon decay products created by building materials. The gamma radiation, radionuclide content and surface dose rates were measured in 322 slabs of 254 named stone types (area ~50 ft2 per slab) and 14 smaller samples (area ~1 ft2). Average surface gamma dose rates ranged from 1 to 24 µrem h-1 with hot spots up to 160 µrem h-1. Radon emanation, measured for 60 slabs of 24 named stone types, ranged from 1 to 400 pCi ft-2 h-1. Emanation fractions ranged from roughly 5% to 50%. Individual sampling points on the slabs showed emanation rates as high as 1300 pCi ft-2 h-1. Model calculations suggest that some home occupants might receive doses that warrant remedial actions when substantial areas of some types of granite are installed in small, minimally ventilated living spaces. Based on these results, pre-market screening protocols have been developed and are being used to remove potentially problematic slabs from inventory. Post-market screening measurements are being tested in a pilot study of 35 homes. The maximum result radon emanation measured from granites in 11 houses was 90 pCi ft-2 h-1. INTRODUCTION Media reports of radiation from granite counter tops recently generated public interest in this issue. While any building material derived from rock has the potential to create radiation exposures, some natural decorative stone like granite can have elevated concentrations of naturally-occurring radioactive materials (NORM). The 238U family, and to a lesser extent the 232Th family, create two pathways for exposure. In addition to external radiation dose from gamma rays, these families have a member which is a radioactive gas that can escape from the rock matrix into the environment. These gases, often referred to as radon and thoron respectively, decay into chemically active decay products that can deliver internal radiation doses when inhaled. Customers buying decorative granite, and industry that supplies them, want an answer to a deceptively simple question, “Is it safe?” There are a number of underlying issues that help formulate an answer to that question. These finer points may not be of interest to everyone but can help support the answer if the questioner asks for details. For example, while public health officials might measure the “safety” using the average impact of the radiation exposures for the general population, individuals are usually more concerned about their 1 This work was supported, in part, by a grant from the CSB/SJU faculty development program. 28 own specific exposure. A product that might be deemed relatively safe in general, might not be judged so by a specific consumer whose use of the product and radio-sensitivity are out of the ordinary. Thus, a blanket statement of a product’s safety needs to include the judgment of sensitive, yet reasonable consumers. The following guiding philosophy for pre-market screening was developed through discussions with the president of Cold Spring Granite Co.2 (CSG). We wanted to find and implement a procedure to protect public from granite countertops that might generate radiation exposures deemed unsafe or higher than normally encountered in nature. We wanted a reasonable consumer to feel confident that the product was safe when used in a “normal” application. That desire required the development of a cost-effective method to identify products that may (or may be perceived to) cause an excess radiation risk to the public, to test the CSG current countertop inventory using the identification and assessment methods, to apply the identification methods to all incoming stock of exiting stone types, and to apply the assessment method to new types of stones that may be added to the product line. Since there are no comprehensive federal (US) regulations for NORM exposure, the recommendations of national and international radiation protection organizations were adopted as the reference safety standards. Those recommendations are based on the effective annual dose to an individual from controllable sources (HPS, 2009). In this work, the dose was calculated for an individual in a high, but not maximal, exposure scenario. In addition, the dose from an average or typical installation was calculated to estimate of the public health impact of decorative granite. At present, over a thousand named “granite” stones are being used in the US as countertops, desktops, flooring and wall tile. The radiation characteristics of only a few dozen of these stones have been reported in the literature (Kitto, 2005, 2008, 2009; Brodhead, 2008) or been released to the public (EH&E, 2008). Based on those early reports, it is clear that many of the high radiation granites show substantial variation in their radiation characteristics. With the diverse needs of the natural stone industry and consumers in mind, the goals of this project were to substantially expand the number of stones analyzed for radioactive and assessed for radiation exposure. That required the development of measurement methods and predictive models suitable for “in situ” assessment of the external gamma radiation dose and internal radon-related dose for stones in the supply system (pre-market) or in the home (post-market). MATERIALS AND METHODS Two distinct sets of measurements and protocols were needed to fit the needs, environments, and capabilities of the stone industry and consumers. A higher level of technical skill and resources are available in the industry for assessing radiation potential of stones that have been processed but not yet customized and installed. However, they may 2 Cold Spring Granite Company 17482 Granite West Road Cold Spring, MN 56320 29 have to perform the tests more quickly than measurements in the home because of the large number of stones that may need to be tested. Pre-market measurements and protocols Companies that either quarry, process or import decorative stone generally have a central facility where the material can be measured while being processed, either in the polish line or at the receiving dock or warehouse. This environment suggests that an effective assessment might rely on a series of measurements that escalate from quick, simple screens to more complete investigation of radionuclide content or radon flux depending on the results of the screen tests and/or previous experience with the particular type of stone. This is the measurement and assessment approach described in this paper. Granite slabs and samples Decorative granite for countertop installations are usually quarried in large blocks and then cut and processed into slabs that are roughly 5 to 6 ft wide by ~10 ft long. While most slabs are 3 cm thick, some are as thin as 2 cm. Decorative granite for floor or wall tiles are usually 1 to 2 cm thick. One side of the granite is polished after being coated with a liquid polymer which is sucked into cracks and crevices by applying a pressure differential across the slab. Stones that are likely to crack are often covered with a polymer net that is glued to the nonpolished side. At least one slab from every one of the 254 stone types in the 2008-09 CSG inventory was screened for gamma dose rate and radionuclide content during the period October 2008 to May 2009. Smaller samples from fourteen stone types which were selected based on their radiation properties were also analyzed. The polished surface area of these samples was approximately 1 ft2. Two of these samples from the current CSG inventory (CK08 and LD08) were from the same stone types that had been analyzed twenty years earlier (Steck, 1988). These stones had shown the highest radiation potential of the seven analyzed in 1988. One of the samples (CK88) was the actual stone that had been measured in 1988. The non CSG samples, all quarried outside North America, had been sent by concerned individuals because they displayed high radioactivity during screening tests. Gamma dose rate screening Simple gamma ray counters, usually scintillators or G-M detectors, provide the quickest and easiest radiation measurement that can be made. However they do not always give an accurate estimate of the actual effective gamma dose rate because the energy dependant response of the detector is usually different than tissue. Nonetheless, if proper precautions and calibrations are used, they can be useful for pre-screening granite. We used two simple scintillation survey counters; the Ludlum 12S micro R meter3, and the Polimaster PM1703 4 as quick pre-screening devices. These detectors are shown on the right in Figure 1. The Ludlum is in the right rear. These pre-screening detectors may not give an accurate assessment of the effective dose rate from a mixed source of radionuclides like granite. The naturally occurring materials (NORM) in stones emit radiation from three primary families; the uranium-radon (U-Rn) family, the thorium-thoron (Th-Tn) family, and 40K. The energy 3 4 Ludlum Measurements, Inc. 501 Oak Street Sweetwater, Texas 79556 USA POLIMASTER Ltd. 112 M. Bogdanovich str. Minsk, 220040, Republic of Belarus 30 spectrum from granite generated in most scintillation materials is quite different from the spectrum from the source (137Cs) that is normally used to calibrate the detector’s response. One solution to this problem is to use a scintillation material whose absorption is similar to tissue like the Thermo Scientific (TS) Micro Rem Tissue Equivalent Survey Meter5 shown in Figure 1 on the front-left. Another solution is to actually measure the absorption spectrum in the detector and use software and probe characteristics to calculate the effective dose rate like the Canberra Inspector 1000(In1k)6 shown in the left-rear of Figure 1. Fig 1 Scintillation-based gamma detectors used for pre-screening Fig 2 The relative response of three other instruments compared to the TS meter. Figure 2 shows the relative response of these detectors when exposed side-by-side on the surface of the slab of Coral Gold granite shown in Figure 1. The correlation between various detectors is probably good enough for most field pre-screening tests. Some of the variation in response is due to the actual spatial variation of the gamma radiation within the area 5 6 Thermo Fisher Scientific Inc. 81 Wyman Street Waltham, MA 02454 Canberra Industries, Inc. 800 Research Parkway, Meriden, CT 06450, U.S.A. 31 covered by the detector array. However, the slope of the response curves means that the readings from the Ludlum 12S and the PM1703 would need to be converted using their respective regression curves to get reasonable estimates of effective dose rates. In this work, slabs were screened for gamma dose rate by placing either the TS or the IN1k in contact with a spot on the polished surface, waiting for roughly 12 s for the meter to respond before recording the reading. Since the detector on the surface “saw” an effective area of about 1 square foot, readings were taken on each slab at 50 locations in 1 foot intervals. Using this procedure, it took roughly 10 minutes to screen each slab. All dose rates reported in this paper are background corrected. In the slab warehouse the background was approximately 5 µrem h-1(50 nSv h-1). Gamma dose rate spatial variation The In1k was used to measure dose rates on the surface and in the space within 3 feet of two horizontal slabs; one slab that was radon rich and another slab that was thoron rich (shown in Figure 1). The grid points were separated by 1 foot from 0 to 3 feet in both horizontal and vertical directions except for a series of points 0.5ft above the surface and along the edges of the slabs. The grid point data was analyzed by SURFER® using a kriging procedure to give values at all grid points within 3 feet of the horizontal surface. These grid point values were analyzed to determine horizontal and vertical functions that represented the spatial variation. These functions were then used to integrate the dose over the space occupied by an individual whose exposure was being modeled. Gamma radionuclide content Spectroscopic screening measurements were made of the slabs using the In1k with a 2x2 stabilized NaI detector and Genie 2000 analysis software in an attempt to find a quick and inexpensive way to estimate the radon generating potential of granites. The hypothesis was that radon emanation was well enough correlated with the U family content in the material that it would serve as a surrogate for radon flux. Several approaches were used to calibrate the detection efficiency of the probe and geometry for field-grade (+20%) precision and accuracy in the radionuclide content averaged over the slab. The resolution of the probe limited the identification and analysis of many of the NORM peaks usually used in laboratory analysis. The highest energy peaks for each family were used to characterize the family’s radionuclide content. Since these peaks belonged to radon and thoron progeny for the U and Th families, these results are labeled remnant radon, thoron progeny since some radon and thoron generated in the slab can escape before decaying to progeny. Given the short half-life of thoron, it is believed that little thoron escapes from deep in the granite so the thorium content determined from remnant thoron progeny should be a good approximation to the thorium family concentration. That situation may not hold for the radon progeny in all granites depending on the porosity, fracturing and coatings on the granite. The efficiency of probe-detector was calibrated with a multistep procedure. The detection efficiency was calculated for the radon progeny peak using a NIST-traceable radium standard at a variety of locations within a square foot surface location. Then the thorium peak efficiency was slightly adjusted by measuring sample CK88 whose radionuclide concentration had been determined by a NIST-standard calibrated HpGe spectrometer in 1988. 32 Gamma exposure models In keeping with the radiation protection philosophy described above, the exposure of individuals to external gamma radiation was estimated using parameters that would lead to elevated, but not maximal doses. For example, one model assumed that the exposure to a seated home worker came from geometry where the highest point of radiation on the slab (“hot spot”) was centered at the edge of his desk’s work area. But the worker only spent 40 hours per week for 48 weeks at that desk. Thus, these estimates are only rough approximations and the uncertainty in their value is taken into account when comparing them to dose recommendations. The actual exposure of individuals to external gamma radiation was calculated from a simplified model of the geometry of the person-source exposure conditions and the gamma dose field constructed from measured surface gamma dose rates as described above. Two exposure models were used to estimate the annual dose from countertops: (1) a kitchen worker who spent 4 hours per day within 3 feet of a horizontal countertop and (2) a home worker who spent 40 hours per week sitting near a desk top that had a hot spot at his work area. Two other models were used to assess the annual dose from floor and wall tiles in a room: (1) a worker standing on a granite floor for 8 hours per day, and (2) a sleeper lying 2 feet above the floor and 2 feet from a granite tile wall for 8 hours per day. Radon emanation from slabs and samples After all stone types had been screened for radionuclide content, those with the highest remnant radon progeny were selected for radon emanation measurements. Both the polished and unpolished sides were measured because they are both free to emanate in typical granite installations. A previous study had suggested that there might be different emanation rates from the different surfaces (Brodhead, 2008). Since a nearly vertical orientation was the most convenient way for slabs to be measured in the warehouse, emanation accumulators had to be designed to adhere to polished, netted, and rough surfaces. They also had to be radon leak proof. Figure 3 shows a typical set of emanation measurements being made on a polished and a netted surface. Fig 3: Emanation measurements on slabs. The slab on the left is an example of a netting glued to the non-polished side. At least seven locations were sampled on each slab; four or more on the polished side and three on the other side. At one location on the polished side, a continuous radon monitor7 7 Radon Scout SARAD GmbH Wiesbadener Straße 10 01159 Dresden GERMANY 33 (CRM) was enclosed by a 4.6L accumulator which covered 0.47 ft2. At all the other locations, electret ion chambers 8(EIC) were enclosed by accumulators which were 3L stainless steel bowls which covered 0.43 ft2. The sampled area was only about 5% of the total surface area of the slab. The smaller granite samples were measured with accumulators centered on the sample that covered 20 to 40% of the surface. A new integrated EIC-based radon flux monitor (RFM) was also used to measure the samples. The flux can be calculated by the ingrowth of the radon in an accumulator measured by the CRM using equation 1: C ( Rn) = kT ( FA) (1 ! e ! ) Vk (1) where: T is the accumulation time F is the radon flux A is the area of opening of the accumulator V is the air volume of the accumulator k is the effective loss rate of radon (The loss can come from decay, 7.55x10-3 h-1 , or leaks) C(Rn) is the radon concentration at any accumulation time of T 1250 1250 1000 1000 750 750 500 500 Fig 4 Radon 250ingrowth in an accumulator with a slight leak. 8 Rn pCi/L Rn pCi/L Since accurate flux results require a known effective loss rate, numerous experiments were done to find a system that created a good, reproducible seal between the accumulator and the various surfaces on the slabs. While adhesive clay was acceptable for polished sides, the rough and netted sides required a quick-setting, adhesive along with the clay to hold and seal the accumulator on the surface. Figure 4 shows an ingrowth curve fit with a weighted least squares fit to determine the effective loss rate of radon for a leaky accumulator. 250 RAD ELEC Rad Elec, Inc. 5716-A Industry Lane Frederick, Maryland 21704 0 0 0 20 40 Time Hr 60 80 34 Once the effective loss rate is known, then the radon concentrations during the initial ingrowth can be corrected for loss and fitted to a straight line whose slope can be used to calculate the flux. Figure 5 shows a typical 22 hour ingrowth of a CRM-stainless steel bowl on the polished surface of a granite slab. Figure 5: Typical ingrowth in a 5L accumulator attached to granite with the Radon Scout monitor. Slope used to determine emanation rate Smaller samples can be measured more easily than slabs as the accumulator can enclose the sample and measure the emanation from all surfaces simultaneously. Granite sample emanations were measured in a well sealed 24 L aluminum case. Bowl style accumulators were used to measure the emanation from each side of each sample. For integrating detectors like EICs or ATDs in accumulators, the flux can be calculated from the average radon using equation 2. < C ( Rn) >= ( FA ) & , 1 - e - kT $1 - * Vk % *+ kT )# ''! (" (2) where: T is the accumulation time F is the radon flux A is the area of opening of the accumulator V is the air volume of the accumulator k is the effective loss rate of radon. The loss can come from decay (7.55E-03 h-1 ) or leaks is the average radon concentration during the accumulation time of T Each accumulator system was calibrated using NIST emanation sources (SRM 4974-8 and 4971-3) (Kotrappa 2005 Volkovitski, 2006). Additional details of the performance and calibration of the accumulator systems can be found in another paper from this Symposium (Kotrappa 2009A). 35 Radon Exposure and dose models The high exposure scenario, called the conservative scenario, uses realistic estimates of the parameters like living space area, occupancy factor and ventilation rate. A simple model was used to estimate the radon concentration generated by granite used in a variety of ways indoors. The model assumed complete mixing of the source in the living spaces. The other important model parameters were taken from the EPA exposure factors handbook (EPA, 1997) to create a realistic but conservative estimate of exposures in a small (640 ft2) minimally ventilated (0.35 ach) living space. The annual radon exposure was converted to effective dose using the dose conversion factors from UNSCEAR 2006 and typical high occupancy rates. The conversion was roughly 100 mrem for a year’s exposure to 1 pCi L-1. The annual dose was also calculated for a “more typical” modern home with an area of 3000 ft2 and a ventilation rate of 0.2 ach to assess the public heath impact of granite in more common situations. Post-market home measurements and protocols The instruments and procedures used to assess the radiation impact of granite already installed in homes have to require less technical skill and be more cost effective than those in the pre-market environment. Simpler gamma dosimeters and emanation measurement systems were developed and tested so that “skilled” homeowners could deploy the detectors and make simple measurements. A pilot study of a small number of selected volunteered homes is underway, testing procedures and measurement methods to see if they are practical and useful for assessing post-market radiation exposure from granite. Thirty-five homeowners, mostly from Minnesota, volunteered online to have their homes and granites tested for radon. The selection criteria were that the home was “small” and “tight” and had a substantial installation of “exotic granite”. The measurement protocol included conventional radon in air measurements in the room where the granite was present, a room that was “remote” from any granite, and a basement, if the home had one. AirChek 9 short-term test kits were exposed for 4 days under closed house conditions. Landauer RADTRAK10 detectors were simultaneously deployed for a 90 day (or longer) exposure under normal living conditions. A newly developed radon emanation measurement system was used to measure the radon emanation in situ at two locations on each type of granite in the home. The emanation system consisted of a high sensitivity radon-thoron discriminating track detector inside a radon-retaining tent that covered 1.8 ft2. The track registration detector has been described earlier (Steck, 2006). The radon flux from an integrating detector like this can be obtained from the average radon concentration measured by the detector and equation 2. The locations for emanation measurements on the granite surface were selected after the top surface was slowly scanned for gamma activity using the PM1703 meter. Emanation measurements were made at the location of the highest gamma reading and an “average” gamma location. The emanation systems effective loss rate (k) was measured using the Radon Scout and proved to be reasonably consistent. The track generation rate from radon 9 Air Chek, Inc. 1936 Butler Bridge Rd Mills River, NC 28759 Landauer 2 Science Road Glenwood, Illinois 60425 10 36 was calibrated with a NIST SRM4794 source. The response of the thoron detector to thoron emanation has not yet been calibrated. RESULTS Pre-market summaries of slab measurements Table 1 provides a summary of the pre-market slab measurement protocols. Table 1: Summary of pre-market slab screening measurements and procedures MEASUREMENT INSTRUMENTS Samples and time Gamma Dose Screen (GDS) Gamma Radon Progeny Analysis (GRA) Radon Surface Emanation (RSE) TS microrem meter or 322 slabs of 254 stone types (complete Canberra Inspector 1000 current inventory of stone types) at 50 sample locations separated by ~1 foot for about 12 s per location Canberra Inspector 322 slabs of 254 stone types at 50 1000+ sample locations separated by ~1 foot Genie 2000 software for about 12 s per location; 5 minute post measurement analysis RAD ELEC EIC in 3L 60 slabs of 24 stone types* at7 or more accumulator or locations (4 or more on front, 3 on Radon Scout CRM in back) 3 to 4 square feet sampled for 24 5L accumulator hour *22 of the stone types were suspected to have high radon potential based on screening tests and media reports Table 2 gives the statistical summary of the gamma dose rate, remnant radionuclide concentrations, and radon emanation averaged across the slabs. Table 2 : Statistical parameters for slab-average measurement results MEASUREMENT (N) Median Range Remnant radon progeny (322) 4.3 µrem h-1 (43 nSv h-1) 84 Bq kg-1 (2.3 pCi g-1) 1 to 24 µrem h-1 (10 to 240 nSv h-1) <10 to 2300 Bq kg-1 (0.3 to 62 pCi g-1) Remnant thoron Progeny (322) Radon Surface Emanation (60) 48 Bq kg-1 (1.3 pCi g-1) 62 pCi ft-2h-1 (24 Bq m-2 h-1) <10 to 1300 Bq kg-1 (0.3 to 35 pCi g-1) 3 to 300 pCi ft-2h-1 (1.2 to 120 Bq m-2 h-1) 22% 3 to 59% Gamma dose rate(322) Emanation fraction(60) Slab variation 10% median 0 to 75% range 36% median 0 to 120% range 37 Pre-market gamma radiation detail Gamma Dose Screen (GDS) s Figure 6 shows the gamma radiation dose rates averaged across individual slabs. Individual b sampling points ranged up to a maximum of 80 µrem h-1 (background subtracted). a l s 50 f o 40 0.14 0.12 0.10 r 30 e b m 20 u N 10 0.08 0.06 0.04 0.02 0 5 10 15 20 25 Surface Dose rate microrem/hr 0.00 P r o p o r t i o n p e r Fig 6 Measured gamma radiation dose rate averaged across the slab surface. (The number of observations per bar is shown in the left vertical scale while the fraction of B a the total observations is shown on the left vertical scale.) r Gamma Radon Progeny Analysis (GRA) Figure 7 shows the average progeny concentration that remains in the slab for both radon and thoron. For the slabs as a group, remnant radon progeny averaged 210 Bq kg-1 (5.7 pCi g-1) and ranged up to 2300 Bq kg-1(62 pCi g-1). Remnant thoron progeny averaged 90 Bq kg1 -1 -1 -1 (2.4 pCi s g )and ranged up to 1300 Bq kg (35 pCi sg ). b a l s f o b a l s 50 0.16 40 0.14 0.12 30 r e b m 20 u N 10 0.10 0.08 0.06 0.04 0.02 0 10 100 1000 RemnantprogenyconcentrationBq/kg 0.00 f Po r o r p e o b r m t u i N o n p e r B 60 50 0.2 40 30 0.1 20 10 0 1.0 10.0 100.0 1000.0 RemnantprogenyconcentrationBq/kg Fig 7 Remnant progeny concentrations measured a in slabs: radon (left) and thoron r (right). 0.0 P r o p o r t i o n p e r B a r 38 Figure 8 shows the fraction of the dose created by each NORM family using the coefficients for the contributions to dose by family found in the European Commission report 112 (EC 1999) and the individual concentrations measured for individual slabs. It is worthwhile to note the substantial number of slabs where radon progeny make up the bulk of the dose and the substantial number where the radon progeny contributes little. The average contribution by sfamily are: U/Rn 33 %, Th/Tn 24 %, K40: 43 %. s s b a l s f o r e b m u N b a l s 50 0.16 40 0.14 0.12 30 20 10 0.10 0.08 0.06 0.04 0.02 0 0.00 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 F rac tion of dose 40 b a l s 0.16 0.14 f Po r or pe ob rm tu iN o n p e r 30 0.12 0.10 20 0.08 0.06 10 0.04 0.02 0 0.00 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 F rac tion of dose f Po r or pe ob rm tu i N o n p e r 50 40 0.16 0.14 0.02 P r o p o r t i o n 0 0.00 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 F rac tion of dose p e r 0.12 30 0.10 0.08 20 10 0.06 0.04 a to surface dose from NORM a Fig 8 Relative contributions families 238U, 232Th, and 40a K B B B r r r Pre-market gamma dose rate spatial variation and exposure Figure 9 shows the results of an analysis of the measured spatial variation of the background subtracted gamma dose rate around a slab of Coral Gold. Fig 9 Gamma dose rate contour maps of the in the vicinity of the Coral Gold slab (Figure 1) Preliminary exposure analysis using this distribution as a template suggests that a countertop would have to have an average surface gamma dose rate of 100 µrem h-1 above background to exceed the exposure recommendations for the kitchen worker. In the desk worker exposure scenario, the hot spot would have to exceed 60 µrem h-1. In the case of exposure to floor and wall tiling installations, the safe average surface dose rates are lower; 33 µrem h-1 39 for standing-floor scenario, 29 µrem h-1 for the sleeping – floor scenario, and 13 µrem h-1 for the sleeping-wall plus floor scenario. Pre-market Radon Surface Emanation (RSE) The median value for the slab-average radon emanation from the 60 slabs measured was 24 s Bq m-2 h-1 ( 62 pCi ft-2 hb-1). Fluxes ranged from 1 to 117 Bq m-2 h-1(3 to 300 pCi ft-2h-1). Figure 10 shows the fluxa distribution of the sample set. l s 15 f o r e b m u N 0.2 10 0.1 5 0 1 10 Rn Flux Bq/m^2 h 100 Fig 10 Slab average radon flux distribution for the 60 slabs 0.0 P r o p o r t i o n p e r B a The radon emanation fraction can be calculated from the measured radon flux and the r remnant radon progeny determined by gamma spectroscopy as described in the Methods section. The emanation fraction distribution had a median value of 22% with a range from 3 to 59%. For individual slabs, the flux variation across the polished side showed a median variation of 60% as did the variation across the unpolished side. But the flux often was quite different from the polished to the unpolished side. In particular most slabs that had netting “glued” to the unpolished side usually showed fluxes below the detection limits. But 4 out of 22 such slabs had almost equal flux on both sides. Similarly, while most slabs with untreated (rough) backs had about equal emanation from each side, in 5 out of 38 cases the polished had a higher flux than the rough side and in 3 cases the opposite was true. A small study of the flux dependence using combinations of 1 cm thick tiles suggests that the flux scales better with mass than surface area up to a thickness of 5 cm. The distribution of annual effective radon-related doses from the 60 slab sample is shown for the conservative scenario in Figure 11for three different kinds of installations. The conservative model’s median dose for these types of granites is 17, 30 and 40 mrem for the countertop, floor, and floor plus wall installations while the maximum dose is 83, 150 and 200 mrem respectively. In an exposure scenario that is believed to be more representative of typical modern homes, those same doses are lower as illustrated in Figure 12. The median doses in these cases are 40 6, 11 and 30 mrem for the countertop, floor, and floor plus wall installations while the maxima are 30, 57, and 150 mrem respectively. s b a l S f o r e b m u N s b a l S 15 0.2 10 0.1 5 0 0.0 1. 0 10 .0 Annual Dosemrem P r o p o r t i o n f o r e b m u N 15 0.2 10 0.1 5 0 p e r s b a l S 1 10 Annual Dosemrem 100 0.0 B a r P r o p o r t i o n f o r e b m u N 15 0.2 10 0.1 5 0 1 p e r 10 Annual Dosemrem 100 0.0 B a r Fig 11 Effective annual doses due to radon from countertops, floor, and floor and walls in the conservative exposure case s b a l S f o r e b m u N s b a l S 15 0.2 10 0.1 5 0 1.0 10.0 Annual Dosemrem 0.0 P r o p o r t i o n p e r f o r e b m u N s b a l S 15 0.2 10 0.1 5 0 1.0 10.0 Annual Dosemrem 0.0 P r o p o r t i o n p e r f o r e b m u N P r o p o r t i o n p e r B a r 15 0.2 10 0.1 5 0 1 10 Annual Dosemrem 100 0.0 P r o p o r t i o n p e r Fig 12 Effective annual doses Badue to radon from countertops,Ba floor, and floor and walls in Ba r r a more typical exposure case. r Pre-market small sample results The smaller area granite samples show a wider range of those same radiation characteristics since most of them were selected because they had elevated gamma emissions. Two of the stone types (CK and LD) were selected because they are more representative of the characteristics of the majority of granites. Table 3 summarizes the major radiation characteristics of these 14 samples. The first three samples from the same slab. The next two are samples from the same slab of another stone type. 41 Table 3: radiation properties of granite samples 222 222 Surface dose Rn Rn † rate emanation emanation† -1 -1 -2 -1 -2 -1 Sample§ nSvh (µrem h ) Bq m h (pCi ft h ) % ‡ FS09 1600 (160) 125 (320) 2 ‡ FS08 200 (20) 120 (300) 17 ‡ FSMK 160 (16) 60 (155) 12 ‡ JBMK 1400 (140) 160 (420) 11 ‡ JBSR 600 (60) 170 (440) 10 JDJF 400 (40) 60 (150) 7 NG08 400 (40) 38 (96) 2 CB08 200 (20) 50 (130) 8 ‡ JB08A 160 (16) 35 (90) 8 ‡ JB08 100 (10) 30 (76) 11 SUMK 60 (6) 6 (16) 14 LD08 40(4) 4 (11) 16 CK08 30(3) 6 (16) 25 CK88 30(3) 0.5 (1.4) 3 222 Rn progeny* 220 Rn progeny * Bq kg-1 (pCi/g) Bq kg-1 (pCi/g) 17000 (460) 1900 (51) 1600 (43) 5800 (157) 5000 (135) 2500 (68) 7000 (190) 2000 (54) 1400 (38) 800 (22) 140 (4) 110 (3) 60 (2) 70 (2) 30 (1) 300 (8) 270 (7) 900 (24) 500 (14) 500 (14) 200 (5) 200 (5) 200 (5) 200 (5) 570 (15) 100 (3) 70 (2) 70 (2) § Cut from slabs or floor tiles; area ~ 0.1 m2 (1 ft2) x 2 or 3 cm thick Average of both polished and unpolished sides ‡ Separate samples from the same slab * Remnant concentration in an open, one square foot sample, measured at the center † Table 4 shows the radon flux from some of these samples as measured by different emanation systems. While a more thorough discussion of the implications of these measurements can be found in another paper in these Proceedings, this data illustrates the differences in flux between the sides of the different stones (Kotrappa, 2009A). Both the CB sample and the FS samples (same slab) show atypical emanation in that there is a significant difference in emanation from polished to the rough side of the slab. The JB (Juparana Bordeaux) samples are from the same stone type but different slabs. Even though they are both netted on the back side, the emanation is low from the netted side in JB08 but equivalent to the polished side in the case of JBMK. 42 Table 4 Comparative radon flux results ( in pCi ft-2 h-1) from both sides of selected samples using a variety of measurement systems. 5L Bowl +CRM 3L Bowl +EIC Enclosure Method +CRM Sample/surface CK08 polished 16 11 rough 13 21 * * 15 16 all surfaces 16 CB08 polished 100 90 rough 260 250 all surfaces 155 180 170 FS08 polished 30 30 rough 450 510 all surfaces 310 240 270 FSMK polished 42 46 rough 215 230 all surfaces 160 129 138 JB08 polished 120 105 net 1 10 all surfaces 80 61 58 JBMK polished 495 430 net 365 290 all surfaces 415 430 360 * The cells shaded grey are averages of the two cells above. EIC Radon Flux Monitor 18 21 * 20 70 270 170 25 525 275 52 370 211 160 8 84 829 280 555 Post-market radon and radon emanatation in homes Only short-term radon measurements are currently available for the complete set of 35 upper Midwest homes. To date, the radon emanation from granites in 11 houses have been measured. The average emanation rate was 10 pCi ft-2 h-1 (4 Bq m-2 h-1), the median was 5 pCi ft-2 h-1 (2 Bq m-2 h-1)and the maximum was 90 pCi ft-2 h-1 (35 Bq m-2 h-1). Table 5 Short-term average airborne radon concentrations in homes with decorative granite. Location Rooms with granite Rooms remote from granite Basements Average Radon (pCi/L) 2.7 2.3 4.6 Median Radon (pCi/L) 2.0 1.6 3.8 43 DISCUSSION The central focus of this work was to find practical methods of identifying granite installations that had the potential to generate radiation doses above recommended levels. In light of the significant variation and uncertainties associated with the factors that control NORM dose assessment to individuals, conservative scenarios of the effects of granite installations that produced effective doses less than 25 to 50 mrem yr-1 were deemed acceptably safe (HPS 2009). This choice helped frame acceptable measurable surrogates and models for the dose for the external gamma and internal radon progeny exposure pathways. Thus, the radiation protection goals are converted to finding easily measured surrogates that can be used in verifiable dose construction models. External gamma dose For external gamma radiation, one candidate for a practical surrogate is the radionuclide concentration distribution of the granite installation. The gamma dose field surrounding an installation can be calculated from the measured radionuclide concentration distribution in the installation and a spatial source - transport model that calculates the dose field in the vicinity of the material. The European Commission (EC, 1999) takes this approach by using the average concentration of the NORM families in a material to construct a dose index. The index is actually the predicted annual effective dose (in mSv) for an individual in the center of a room constructed of the NORM material. For thin claddings of the material, an index of 2 (corresponding to a dose of 30 mrem yr-1) is supposed to trigger a detailed dose assessment of the actual geometry and use of the material. Figure 12, which shows the distribution of the index calculated for the material in CSG slabs shows that a significant number exceed 2 and the maximum index is above 9. Of course, most decorative granite installations will have lower doses because they involve less material than complete surface coating, and in the case s of counter tops, have different exposure geometry. An investigation using this approach forb the granites measured in this work is ongoing. a l s f o r e b m u N 120 100 0.3 80 0.2 60 40 0.1 20 0 0 1 2 3 4 5 6 7 8 0.0 P r o p o r t i o n 9 10 p e r Fig 12: Gamma dose index calculated from NORM concentrations measured in 300 slabs. Gamma Dose Index (The dose index corresponds to the annual dose (in mSv) under a hypothetical exposure B scenario.) a r 44 Another candidate for a practical external dose surrogate is the measured gamma dose rate distribution on the slab’s surface. These measurements can be used with a spatial dose field model derived from spatial variation measurements of the dose field near granite slabs. Figure 9 (above) illustrates that the dose falls off rapidly away from the slab vertically and horizontally, especially in the plane of the slab. Two slabs’ dose rate fields have been studied so far. This approach is currently being verified on a third slab. Preliminary results suggest that none of the 300 granite slab materials measured would exceed recommended dose limits when used as a countertop or as flooring. Only two slabs had hot spots that would exceeded the limit when used as a desktop, but 14 granite slabs would have exceeded the limit if a bedroom floor and walls had been clad with that granite. However, since floor and wall tiles are usually one third the thicknesses of slabs, none of those materials in thinner tiles would have exceeded the recommendations. But, had the selected granite samples listed Table 3 been used as surface tiles, two materials would have exceeded the limit for floor installations and 4 would have exceeded the limit for floor plus walls. Internal radon-related dose For the internal dose from radon progeny, the “safe” reference effective dose can be nearer the upper end of the range described above, roughly 40 to 50 mrem y-1. That range corresponds to the dose expected from the lowest practically-achievable indoor radon concentrations, those equivalent to outdoor air concentrations. Of the 60 slabs that were measured for radon emanation, 8 would exceed the reference value for countertop installations, 25 for floor installations, and 34 for floor plus wall installations. Of the smaller granite samples, 2, 5 and 6 of the 14 would exceed the reference levels for the different installations respectively. From an operations perspective, it would be helpful if the external dose surrogate could be used for the internal dose as well. Unfortunately, neither surface dose rate nor radionuclide concentration is well enough correlated with the radon emanation rate to serve as a universal surrogate for internal dose. The radon that escapes from a slab depends on the parent nuclide (radium) content and the fraction of radon generated in the slab that can escape. This latter characteristic depends on the location of the radium bearing minerals in the slab and the physical porosity or fracturing on the slab. These characteristics can lead to large differences in radon emanation fractions from stone type-to-type and in some cases from slab-to-slab within the same type. Figure 13 illustrates one type of spatial variation where the radium bearing minerals are near the surface concentrated in a visible small spot. These “gamma hot spots” often, but not always, corresponded to radon flux “gushers”. 45 Fig 13: Small “hot spots” of mineralization; grey mineral in Juparana Bordeaux on left, and yellow mineral in Niagara Gold on right Gamma dose as a radon flux surrogate Figure 14 shows the surface gamma dose rate and radon flux at the ~450 points sampled on the 60 slabs measured for emanation. Recall that these slabs were pre-selected because they were high in remnant radon progeny, not on total gamma dose rate. The orange line shows the maximum radon flux that would produce internal doses at the recommended level for a counter top application in our conservative scenario (~ 60 Bq m-2 h-1or 150 pCi ft-2 h-1)). Besides the poor correlation, surface dose rates would not make a good diagnostic statistic as plenty of examples of high emanation and low gamma dose as well as low emanation and high gamma dose are evident. For example if you chose a gamma dose rate trigger of 30 µrem h-1 for the hot spots, 8 areas would be classified falsely as “unsafe” and 36 would be classified falsely as “safe” Radon Flux Bq/m^2-h 1000 100 10 1 0 0 30 60 90 Gamma dose rate microrem/h Fig 14 Radon flux and radon emanation at over 400 points on 60 slabs and 14 samples 46 Average measured Rn emanation Bq/m^2-h Other combinations of gamma dose and radon flux, such as the hot spot gamma dose versus slab average emanation shown in Figure 15, suggest that surface gamma dose alone is not an adequate radon flux surrogate. For countertop applications (orange line), a gamma dose rate trigger of 30 µrem h-1, causes 2 false positives and 3 false negatives in the sample of 60. The trigger would have to be lowered to ~ 10 µrem h-1to eliminate false negatives. But that trigger would create almost 25 slabs to be falsely classified as “unsafe”. The situation is exacerbated when more extensive granite applications are analyzed. The yellow and violet lines on Figure 15 show the flux limits that would cause radon-related dose to exceed the recommendation in the case of floor tile, and floor plus wall tile exposure scenarios. 1000 100 10 1 0 10 20 30 40 50 60 70 80 90 Highest spot measured Doserate microrem/hr Fig 15 Slab average radon flux and maximum surface gamma dose for 60 slabs. The orange, yellow, and violet lines show the limits for “safe” countertop, floor, and floor plus wall applications. Figure 8 (results section above) suggests one reason for the failure of surface gamma dose rate as a good surrogate for radon exposure; the remnant radon progeny contribution to the gamma dose has a significant number of low and high fractional contributions to the surface gamma dose rate. Remnant radon progeny concentration as a radon flux surrogate It has been suggested that remnant radon progeny content in the slab would be better correlated with radon emanation. Advanced portable gamma spectrometers like the Canberra IN1k, can determine the radionuclide content of NORM materials in a slab with elevated concentrations within a few minutes. These instruments are affordable and require only modest technical skill once they are calibrated so they are a reasonable alternative for large companies compared to the expense of a facility for measuring slab radon emanation. 47 Average measured Rn emanation Bq/m^2-h Figure 16 shows the relationship between the slab average remnant radon progeny and the radon flux along with flux limits for various dose-exposure scenarios. While the correlation between radon flux and remnant radon progeny is better than surface dose rate, there are still enough variations to require additional measurements for some stone types. For countertop applications, slabs with remnant radon concentrations below about 300 Bq kg-1 would fall consistently in the “safe for counter tops” category. Figure 7 (results) shows that most of the CSG-inventory slabs (75%) meet this condition. The utility of this surrogate can be extended when radon emanation measurements are made on a number of samples of a particular stone which have fluxes near the limits. The trigger limit can be lowered for most stones whose emanation fraction has been repeatedly measured because their emanation fraction is likely to be smaller than the 50% used in setting the a priori limit. 1000 100 10 1 0 500 1000 1500 2000 2500 Average measured remnant RnP Bq/kg Fig 16 Slab average radon flux and average remnant radon progeny for 60 slabs. The orange, yellow, and violet lines show the limits for “safe” countertop, floor, and floor plus wall applications. The radon progeny trigger level for limiting concentrations in floor tile, and floor plus wall tile exposure scenarios would be 150 and 110 Bq kg-1 respectively. Clearly many more stones destined for surface tile applications require radon emanation measurements. Practical pre-market screening protocols The pattern of flux values, remnant radon progeny concentrations and their variations can be combined in a classification scheme to screen existing and new inventory at the processing, distribution or fabrication level. The key elements are multiple measurements of the radionuclide content of a number slabs per stone type repeated over time as the quarried material is taken from different sections of the deposit. An historical database of radon progeny concentrations and radon emanation fractions for potentially troublesome stones would allow for more efficient and effective screening of inventory. Those stones with consistently low remnant radon progeny (exact value would depend on their use in the home 48 and prior emanation measurements) can be sampled less frequently and less intensively. These slabs would have a trigger value on the surface gamma dose which, if exceeded, could indicate that remnant radon progeny was above the trigger level for emanation tests. When the dose trigger value is exceeded, the remnant radon progeny needs to be measured, and if the measured remnant progeny exceeds the emanation trigger level, radon emanation measurements need to be made or the slab kept out of inventory. In this work, which focused primarily on slabs used as counter tops, the radon flux was measured for three slabs of each stone type that had remnant radon progeny greater than 300 Bq kg-1. These measurements helped establish the average value and variation of the emanation fraction for each stone type. Those characteristics were used to calculate a trigger value for the remnant radon progeny of future measurements on that stone type. If the new measurement exceeded the trigger value, then the radon flux had to be measured for that slab or it would be rejected from inclusion in the inventory. This pre-market screening protocol is now being used by Cold Spring Granite to insure the safety of their counter top inventory. CONCLUSIONS Most decorative granites create acceptably low radiation exposures in most home installations. However, some stones should not be used in large scale installations in small living spaces with low ventilation because they generate enough radon to create doses in excess of those recommended by radiation protection organizations for controllable radiation sources. Those stone types that may create excessive internal radiation dose cannot be identified based on gamma measurements alone. In particular, the surface gamma dose rate which is easily measured with survey instruments is ineffective in screening out all potentially troublesome stones and falsely identifies others as troublesome. However, a system of increasingly sophisticated screening measurements and trigger values combined with more extensive analysis of the radon generating potential and radionuclide content of stone types can effectively insure the safety of decorative granite made available to the public. ACKNOWLEDGMENTS I wish to thank John Mattke, president of Cold Spring Granite Company, for his generous cooperation with this project and to Jim Fuchs, and Jerry Middlestadt from the Engineering and Quarry Equipment Department for their assistance and hard work. Thanks to Dr. Paul Kotrappa, RAD ELEC Inc., for helpful conversations and radon emanation equipment and supplies. Some post-market measurement equipment and supplies were generously donated by B.V. Alvarez, AirChek Inc. Thanks to Al Gerhart, The Carpenter Shop, Linda Kincaid, Industrial Hygiene Services, Jeff Burg and Mike Spaniol of Granite Services for granite samples. Dan Franta and David Harrison assisted in sample emanation measurements and gamma radiation modeling. 49 REFERENCES Brodhead WB. Measuring radon and thoron emanation from concrete and granite with continuous radon monitors and EPERM’s®. Proceedings of the American Association of Radon Scientists and Technologists 2008 International Symposium Las Vegas NV, September 14-17, 2008. Environmental Health and Engineering Inc. Assessing exposure to radiation and radon from granite countertops. 2008 http://www.marbleinstitute.com/industryresources/assessingexposurereport.pdf accessed 8/7/2009 EPA National Center for Environmental Assessment Exposure Factors Handbook (1997 Final Report) available at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=12464; accessed 8/7/2009. European Commission, Directorate-General, Environment, Nuclear Safety and Civil Protection Radiological protection principles concerning the natural radioactivity of building materials. Radiation Protection 112, p 8 (1999) Health Physics Society. IONIZING RADIATION-SAFETY STANDARDS FOR THE GENERAL PUBLIC, http://www.hps.org/documents/publicdose_ps005-3.pdf accessed 5 August 2009 Kitto ME, Green J Emanation from granite countertops. Proceedings of the American Association of Radon Scientists and Technologists 2005 International Symposium, San Diego CA ( 2005) Kitto M.E, Haines D.K, Aruzo H.D. Emanation of radon from household granite. Health Physics: 6:477-482 (2009) Kotrappa P, Dempsey J.C, Ramsey R.W and Stieff L.R. A practical E-PERM® System for indoor radon measurement. Health Physics 461-467 (1990) Kotrappa P, Stieff LR, Volkovitsky P. Radon monitor calibration using NIST radon emanation standards: steady flow method. Radiation Protection Dosimetry 113:70-74:2005 Kotrappa P and F Stieff “Radon exhalation rates from building materials using electret ion chamber in accumulators” Health Physics 97:163-166 (2009) Kotrappa P Stieff F Steck DJ. Radon flux monitor for in situ measurement of granite and concrete surfaces. Proceedings of the American Association of Radon Scientists and Technologists 2009 International Symposium St. Louis MO. (2009A) Steck DJ Unpublished report to Cold Spring Granite Company, 1989. archived online at http://www.solidsurfacealliance.org/files/Radon_Results.htm accessed 8/7/2009 50 Steck DJ. A preliminary survey of thoron in the Upper Midwest. Sixteenth International Radon Symposium, Kansas City Mo. September 2006 Available at http://www.aarst.org/radon_research_papers.shtml Volkovitski P. NIST 222 Radon emission standards. Applied Radiation and Isotopes. 64:1249-1252; 2006 51 WHOLE BUILDING VENTILATION OF HIGH RISE CONDOMINIUM WITH ELEVATED RADON FROM CONCRETE Bill Brodhead WPB Enterprises, Inc., 2844 Slifer Valley Rd., Riegelsville, PA USA bill@wpb-radon.com www.wpb-radon.com Tom Hatton Clean Vapor, LLC, 3 Maple Street, Andover, NJ 07821 thatton@cleanvapor.com www.cleanvapor.com ABSTRACT The authors designed and installed a whole building ventilation system for a 100 unit five story high rise condominium building that had elevated radon levels on every floor. The building had two levels of ventilated garages under part of the building. Each floor was constructed with post stressed concrete floors and ceilings. Previous diagnostics revealed the low ventilation rate and high radon emanation rate from the concrete was the primary reason for the elevated radon. The radon reduction in the bedrooms of one unit was pre-tested with the bedroom doors closed, the HVAC air handler off and outdoor air introduced into only the main living area of the unit to determine if single point ventilation would treat an entire condominium. The installed whole building ventilation system routed conditioned air from the roof to a single location in each unit of the building. Post mitigation radon measurements indicated that the system is successfully reducing the radon levels below the guideline. 52 1.0 GENERAL BUILDING INFORMATION The building described in this paper is five stories tall with 100 condominium units. Most of the building has ground contact. A portion of the building has a two story garage that is connected below grade to an adjoining eight story 110 condominium unit building that is not part of this project. The building was constructed between 2001 and 2002. In a previously published paper in 2008 Brodhead described elevated radon levels in different units on every floor of both buildings that was caused by the low ventilation rate of the condominium units and radon emanation coming from the concrete. The building was initially having condensation problems on the windows caused by low ventilation rates. The units were all constructed with concrete floors and ceilings. All of the units have exterior and interior metal stud framed walls that are covered with drywall. There are drywall covered suspending ceilings in most of the units. Units that are adjacent to stairwells and utility rooms have additional concrete exposure from the concrete walls that surround these rooms. There are multiple concrete beams throughout the building supporting the concrete slabs. The flooring material is a combination of carpet, wood, vinyl and tile floors. Eleven different units were measured with closed house conditions in 2007. The radon levels varied from 5.9 pCi/L to 13.2 pCi/l (218 to 481 Bq/m3). In the Fall of 2007 WPB made a diagnostic study of the radon emanation rate coming from the concrete and the air change per hour (ACH) ventilation rate.1 The ventilation rates were determined to be as low as 0.05 air changes per hour (ACH). The radon emanation from the concrete of both buildings was measured using a direct flux test of the concrete by sealing E-Perms in an accumulator. This method determined the radon emanation rate to range from about 30 to 50 pCi/ft2/hour (200 to 332 Bq/m2/min). This emanation and ventilation rate was determined to be the cause of the elevated radon levels. The building was not designed with any common mechanical exhaust or ventilation fans since there are no common bathrooms or laundry rooms that would require this type of ventilation. Each individual condo unit has bathroom exhaust fans and a dryer exhaust. The range hoods are re-circulating. All units have operable windows and most units have one or more glass doors leading to an outside patio. Heating and air conditioning in each unit is provided by a heat pump with the compressor on the roof. The air handling components of these units are located totally inside the condo unit. No condo unit shares an HVAC system. The hallways are also conditioned by their own heat pumps with no introduction of outside air. 2.0 VENTILATION DISTRIBUTION TEST The HVAC operation in the condo units is predominately for cooling and humidity control. The location of the building in a mid-Atlantic state, air tight construction and size of the condo units 53 minimizes the need for supplemental heat. There are significant periods during the heating season when the air handler does not need to operate. The Condo Association and engineering company hired by the association questioned whether introduction of outdoor air in a single location in the building when the air handler was not operating would provide adequate radon reduction throughout the condo unit since the source was in each room. WPB was hired by the condo association to perform a ventilation test in order to determine both if single location outdoor ventilation would reduce the radon levels in the adjoining bedrooms when the bedroom doors are closed and the air handler is off and to confirm the amount of ventilation needed. Note that when the air handler is operating there will likely be no differentiation in radon levels when the bedroom doors are closed. The ventilation test was performed in a single unit on the fifth floor. See floor plan below in Figure 1. In 2007 this unit had measured 7.5 pCi/l (278 Bq/m3). This unit has about 1763 square feet of floor space with 10 foot high ceilings. The ceilings, walls and floors expose 3500 square feet of concrete to the inside of the unit. The unit is next to a stairwell that has concrete walls. The stairwell walls and concrete support columns add about 650 additional square feet of concrete exposure for a total of 4150 square feet. The unit has two exterior balconies. One balcony is off the master bedroom and the other one is off the main living area. The only windows in the unit are in the two bedrooms. There are three French doors that open onto the outdoor balconies. The existing outside dryer vent was used to provide the inlet for the fan ventilation test by propping open the outside dryer vent louvers located on the outside patio and hooking a 4” flexible duct into the laundry room dryer vent piping. The flexible duct was then routed to a radon ventilation fan. Four inch duct was then routed from the fan to a flow grid to measure the airflow and then the duct outlet was positioned so that the discharge was into the main entrance foyer. A digital micro-monometer was hooked up to the flow grid that had previously been calibrated to measure the airflow. See floor plan drawing below. Three continuous radon monitors (Radalink AirCats )were used during the test. A monitor was placed in each of the two bedrooms. A third monitor was placed in the kitchen great room. The air handler was turned off and all windows and interior doors were closed. The radon monitors were operated during the entire test with the following sequence. 1) No outdoor air was introduced for about 12 hours with bedroom doors closed. 2) 47 CFM (22 l/s) of outdoor air with bedroom doors closed for 24 hours 3) 47 CFM (22 l/s) of outdoor air with bedroom doors open for 24 hours 4) 47 CFM (22 l/s) of outdoor air with bedroom doors closed for 24 hours 5) No outdoor air with bedroom doors closed for 48 hours Note that during the test there was unusually cold weather with temperatures below freezing and strong winds. This increased the natural ventilation to the unit and caused the radon levels with the outdoor air off to be around 4.5 pCi/l (167 Bq/m3) compared 54 with a previous measurement of 7.5 pCi/l (278 Bq/m3). See the average radon results listed in Table 1 and the graphs of the actual radon changes in Figures 2, 3, 4, 5, and 6 below. Set up Fan off doors closed Fan on doors closed Fan on doors open Fan on doors closed Fan off doors closed Kitchen pCi/L Kitchen reduction 3.4 Master pCi/L Master reduction 5.1 Guest pCi/L 4.8 1.0 - 75% 2.7 - 45% 3.1 1.8 - 54% 1.9 - 61% 2.1 1.0 - 75% 2.4 - 51% 3.0 4.5 Guest reduction 4.7 4.7 Table 1 Radon levels during each testing phase Note that the radon levels in the bedrooms and the kitchen are almost the same when the bedroom doors are left open. This indicates the ventilation system discharge into the main living area does not need any additional forced circulation when these doors are open. When the bedroom doors are closed the kitchen radon levels decrease from 1.8 pCi/L to 1.0 pCi/L (67 to 37 Bq/m3) while the bedroom radon levels increase from an average of 2.0 pCi/L to an average of 2.8 pCi/L(74 to 104 Bq/m3). The radon levels in the bedroom are still below the EPA guideline at this ventilation rate into the building (0.26 ACH). The master bedroom door had a larger gap under the door than the guest bedroom. This is the likely cause of the guest bedroom having a higher radon level when the doors were closed than the master bedroom. If the induced mechanical ventilation rate is 0.3 ACH and the natural ventilation rate is only 0.03 ACH during mild weather conditions, the total ventilation will be 97 CFM (46 l/s) which will likely reduce the average radon in the bedrooms down to 2.2 pCi/L (81 Bq/m3) when the bedroom doors are closed and 1.5 pCi/L (56 Bq/m3) when the bedroom doors are open. See Figure 2 through 6 below for actual recorded data. 55 Bedroom 10' 0" 3/16" = 1' 0" CRM Connect to Dryer vent Prop open Dryer exhaust WH 4" flexible duct Study Damper 4" flexible duct Radon Fan Seal Hallway fan opening 2' - 4" PVC 2 ft & 4 ft of 4" PVC Flow grid Micro-monometer between PVC pipes Hallway Outside balcony Place nylon stocking on exhaust to catch any lint Bedroom CRM Closet Kitchen CRM Bathroom Note: Adjoining concrete walls Figure 1 Unit 504 Floor Plan & Ventilation Test 56 Figure 2 Kitchen results 57 Figure 3 Guest Bedroom results 58 Figure 4 Master Bedroom results 59 Figure 5 Guest Bedroom & Kitchen results 60 Figure 6 Master Bedroom & Kitchen results 61 4.0 DETERMINING THE EFFECT OF INTRODUCING OUTDOOR AIR If one assumes the radon emanation rate is consistent and uniform throughout each condo unit then the variation in radon levels will be directly related to the amount of outdoor ventilation taking place. The amount of ventilation added by the test fan can be carefully measured. The radon reduction in effect is the inverse of the ventilation increase. I-Rn N-Rn I-Vn F-Vn N-Vn = = = = = Initial radon New radon level with added ventilation Initial natural ventilation in CFM Fan Forced Ventilation New ventilation rate (Initial + Fan forced ventilation) N-Rn = I-Rn X (I-Vn ÷ N-Vn) Formula 1 Calculating New Radon level based on ventilation change The one variable that is not measured in the above calculation is the initial natural ventilation. Previous radon measurements made in main living area of the test unit measured 7.5 pCi/l (278 Bq/m3). The initial radon measurements made during this ventilation test measured 3.5 pCi/l (130 Bq/m3). This means the natural ventilation rate was about 2 times higher during the ventilation test. This is likely due to the cold windy weather happening during the ventilation test. See table 5 below that lists the outdoor high and low temperature swing during the ventilation test. The reduction in radon that can be achieved with outdoor air requires determining the natural ventilation rate by modifying the above formula. I-Vn = F-Vn X N-Rn ÷ ( I-Rn - N-Rn ) Formula 2 Calculating Initial Natural Ventilation Rate In the graph of the kitchen radon variation in Figure 2 above, the average of the final radon levels was used because of a longer measurement period. The radon levels averaged about 4.5 pCi/l (167 Bq/m3) without the fan operating. When the fan was supplying 47 CFM (22 l/s) of outdoor air and all the bedroom doors were open the radon levels dropped to 1.8 pCi/l (67 Bq/m3). Using the above Formula 2 the following initial natural ventilation rate is determined. 31.3 CFM = 47 X 1.8 ÷ (4.5 – 1.8 ) Once we have determined the ventilation rate necessary to produce a radon level in the building we can use the following formula to determine natural ventilation rate that was occurring when a previously higher or lower radon concentration was happening in the 62 unit. It is important to determine this ventilation rate because it indicates the minimum natural ventilation rate of the unit and this ventilation rate is added to the fan forced introduction of outdoor air to predict the final radon levels within the dwelling during mild or windy cold weather. I-Vn = N-Vn X (N-Rn ÷ I-Rn) Formula 3 Determining the ventilation rate when other radon levels were measured Formula 3 can be used to determine that the ventilation rate when the radon levels were at 7.5 pCi/l (278 Bq/m3) was about 18.8 CFM. 18.8 CFM = 78.3 X ( 1.8 ÷ 7.5 ) Or the ventilation rate without the fan running can also be used to determine the ventilation rate when the radon levels were at 7.5 pCi/L (278 Bq/m3). 18.8 CFM = 31.33 X ( 4.5 ÷ 7.5 ) Note to obtain a more representative radon measurement of the effects of induced ventilation into a single room for this type of building it is important to have the air handler operating or to leave all the bedroom doors open. See the radon level variation that happens when the bedroom doors are closed in the figures and graphs above. The bedroom doors do not have to be left open when no additional outdoor air is being introduced. 5.0 CALCULATING THE CONCRETE RADON EMANATION RATE If the ventilation rate of the unit is known and the radon is coming from the concrete the emanation rate of the concrete can be determined if the square footage of concrete is known. Unit 504 has 1763 square feet of floor area. Since there is concrete floors and ceilings this equals 3,526 ft2. The unit however is adjacent to the stair well. The concrete columns and stairwell concrete walls is assumed to equal about 600 ft2 for a total exposure of 4126 ft2. If we use the 78.3 CFM air exchange rate that was maintaining 1.8 pCi/l (67 Bq/m3) we can determine the emanation rate from the concrete is 58.0 pCi/ft2/hour by the following formula. pCi/ft2/hr = (N-Vn X 28.3 X 60 X N-Rn) ÷ ft2 of concrete 58.0 pCi/ft2/hr = ( 78.3 X 28.3 X 60 X 1.8 ) ÷ 4126 Formula 4 Calculating the Concrete Emanation Rate 63 Previous radon flux measurements of bare concrete on the 3rd, 4th and 5th floor of the building in the Fall of 2007 found a range of 32 pCi/ft2/hr to 49 pCi/ft2/hr (212 to 325 Bq/m2/minute). This calculated radon emanation rate of 58 pCi/ft2/hour (385 Bq/m2/minute) is 30% to 50% higher. In a paper presented by the author (see second reference at end of paper) the author found the E-PERM accumulator method against concrete used by the author had a 33% lower emanation rate compared to exposing a CRM under an accumulator. The concrete emanation rate for three other units in the building and the adjoining building also had emanation rates around 60 pCi/ft2/hour (398 Bq/m2/minute) using the ventilation measurements to determine the emanation. Unit 408 had emanation rates that were a third less at 39 pCi/ft2/hour (258 Bq/m2/minute). See the results in Table 3 below. 6.0 DETERMINING THE NECESSARY VENTILATION RATE Necessary ventilation is typically determined using units of air changes per hour (ACH). Some of the condos in the building had additional exposure to concrete walls that adjoined the stairwells. There were also a number of condo units on the first floor that had living space on the two floors. The following formulas were used for determining the necessary ventilation for each unit. ACH = Ventilation rate in CFM X 60 ÷ Volume Formula 5 Determining the ACH rate based on CFM ventilation Therefore when the radon levels were 7.5 pCi/l (278 Bq/m3) in the kitchen and the air handler was likely operating, the ACH was around 0.06 ACH based on the calculated outdoor air ventilation of 18.8 CFM (8.5 l/s) . 0.06 ACH = 18.8 X 60 ÷ 17630 When the radon levels were 4.5 pCi/l (167 Bq/m3) in the kitchen and bedrooms, the natural ACH was around 0.1 ACH based on the calculated natural outdoor ventilation of 31.3 CFM (15 l/s). 0.1 ACH = 31.3 X 60 ÷ 17,630 When the ventilation fan was adding 47 CFM (22 l/s) and the bedroom doors were open, the radon levels in the kitchen dropped to around 1.8 pCi/l (67 Bq/m3) and the bedrooms dropped to about 2.1 pCi/l (163 Bq/m3). The ACH rate with the ventilation fan was 0.26 ACH.. 0.26 ACH = 78.3 X 60 ÷ 17630 If the new additional outdoor air ventilation is sized for 0.3 ACH the fan forced outdoor 64 air ventilation rate would be 88 CFM (42 l/s). 0.3 X 17630 ÷ 60 = 88 CFM In addition to the fan forced ventilation one can assume a natural ventilation rate during mild weather of at least 0.05 ACH or 15 CFM (7 l/s). This will give a new total ventilation rate in unit 504 of 103 CFM (49 l/s) during mild weather. Using the Formula 1 above, the radon levels in Unit 504 would be expected to be 1.4 pCi/L (52 Bq/m3) with the bedroom doors open or the air handler operating. The radon levels would be lower during any weather patterns or occupant activity that increased this minimal natural ventilation rate. 1.4 pCi/L = 4.5 X (31.3 ÷ 103) If units of CFM per ft2 of concrete is used the 88 CFM (42 l/s) in the above calculation is divided by the square footage of concrete that is adjacent to the unit ( 4126 ft2 ) 88 ÷ 4126 = 0.021 CFM per ft2 of concrete In the previous study done on this building it was recommended to provide at least 0.015 CFM per ft2 of concrete to obtain a radon level concentration of 2.0 pCi/l (74 Bq/m3) or less depending upon the weather. Using 0.3 ACH ventilation rate raises the original ventilation rate to 0.02 CFM per ft2 of concrete. 7.0 VENTILATION RECOMMENDATIONS Table 2 below gives the projected radon levels based on a set emanation rate from the concrete with different fan forced ACH rates and weather patterns. The radon emanation rate from the concrete was based on the reduction in radon from introduced ventilation and the calculated ft2 of concrete exposure. The radon levels are based on this emanation rate and the total ventilation from the fan forced air and the natural ventilation rate. The last column radon levels are based on the bedroom doors being closed and no air handler operation is taking place. Note that both unit 318 and unit 504 are adjoining stairwells which causes these units to have less radon reduction if raises the final radon levels The bold radon levels in Table 2 below are the projected radon concentrations based on using 0.3 ACH. All of the units are below 2.0 pCi/L (74 Bq/m3) during mild weather. The bedroom radon levels even when the doors are closed and the air handler is not operating are below 3.0 pCi/L (2.2 pCi/L to 2.7 pCi/L). Reducing the ventilation rate from 0.3 ACH to 0.2 ACH would likely reduce the energy costs by 25% and still maintain radon levels below 2.7 pCi/L (100 Bq/m3) and the bedroom levels below 4.0 pCi/L (150 Bq/m3) even when the bedroom doors are closed. 65 It was decided to introduce the equivalent of 0.3 ACH to each of the units. Most of the first floor units were actually two story units. Although the stairwells were open to each floor and the location of the air inlets was attempted to be located near the stairwells it was not possible in most cases to introduce the outdoor air on both floors. In order to assure adequate radon reduction on both floors the outdoor air was increased to 0.4 ACH in all two story units. Note that the distribution of outdoor air throughout the unit is only a concern when the unit’s air handler is not operating. Unit # pCi/ft2/hr concrete Fan Induced ACH 0.03 ACH Mild Weather pCi/L L-504 L-504 L-504 58.0 0.20 0.25 0.30 2.1 1.7 1.5 1.6 1.4 1.2 Mild weather Bedroom closed doors No HVAC 3.1 2.6 2.2 E-619 E-619 E-619 59.0 0.20 0.25 0.30 2.4 2.0 1.7 1.8 1.6 1.4 3.6 3.0 2.5 60.0 0.20 0.25 0.30 2.6 2.1 1.8 2.0 1.6 1.5 3.8 3.0 2.7 L-408 L-408 L-408 39.0 0.20 0.25 0.30 1.4 1.2 1.0 1.1 0.9 0.8 2.1 1.7 1.5 L-318 L-318 L-318 61.0 0.20 0.25 0.30 2.6 2.1 1.8 2.0 1.7 1.5 3.8 3.1 2.7 E-511 E-511 E-511 0.10 ACH Cold Windy pCi/L Table 2 Radon levels depending on ACH 66 8.0 WHOLE BUILDING VENTILATION VERSUS INDIVIDUAL ERV’S A number of options were explored to deliver conditioned air to all 100 resident units and the hallways. First consideration was installing individual Energy Recovery Ventilator (ERV) units. This option had been chosen on an adjacent similar building. The ERV’s installed in the adjacent building were effective in reducing radon concentrations to below 4.0 pCi/l, however, there were several negative issues including increased interior humidity and odor transfer from outdoors. The Condo association had concerns with individual ERV’s about protecting the exterior styrofoam/stucco layer from water damage from the 200 necessary exterior penetrations necessary for ERV installation. The pros and cons of installing individual ERV’s versus a central conditioned ventilation system (CCVS) are listed below. After considering the pros and cons of both approaches the building condo association decided to have a central conditioned ventilation system (CCVS) installed. This required installing one or more air handlers on the roof that would condition 100% outdoor air. Two deliver systems were considered. The first option was to deliver the necessary air directly to each unit using a ducted system. A less costly second option was to supply all of the necessary conditioned air to the hallways. This air would be distributed to the hallways in multiple locations on each floor at timed intervals. Bathroom blowers in each unit would simultaneously be activated in the exhaust mode. Passive transfer grills would be installed between each unit and the hallway to allow the hallway conditioned air to pass into each unit from the depressurization created by the bathroom exhaust fans. This concept was rejected for several reasons. 1) There was concern that there would be large variation in the hallway ventilation to the units because of greater or lesser negative pressure in the units. 2) Maintaining consistent exhaust airflow from bathroom fans would be difficult 3) Bathroom fan noise would be too great and occupants would request turning them off 4) There would be a large ongoing expense to maintain the bathroom fan motors 5) Noise transfer from the hallways through the passive vents would not be acceptable. After considering the pros and cons of both approaches the building condo association decided to have a central conditioned ventilation system (CCVS) installed. The cost to have a central conditioned system with ducted air to each unit was similar to the cost of having individual HRV/ERV’s installed. Concerns and Advantages for individual HRV/ERV versus whole building CCVS 1) 0.3 ACH for the units in this building requires airflow from 50 CFM to 110 CFM depending on the unit size. CCVS allows easier individual adjustment of the necessary ventilation. Obtaining the necessary airflow to achieve 0.3 ACH would require some units to have oversized HRV/ERV’s that would then be damped down to obtain the necessary airflow. This increases the electrical cost, system noise and system intrusion in the unit. 67 2) HRV/ERV’s do not typically transfer humidity. HRV/ERV’s will contribute humidity to the unit while a CCVS will reduce indoor humidity. This became a factor in the decision process since the building had experienced interior condensation problems. 3) HRV/ERV’s require penetrations of the exterior shell of the building at inaccessible locations that could create a potential for water leakage and building warrantee issues. 4) Maintaining adequate distance between intakes and exhaust vents for HRV/ERV’s would be difficult. 5) Each of the units has an outdoor patio deck that is often used for smoking. These fumes could be brought back into an adjoining unit whose intake is too close to the adjoining patio. 6) HRV/ERV’s located on lower floors could draw in dumpster orders, vehicle exhaust, and street level restaurant exhaust. 7) HRV/ERV’s recover heat better than they recover cooling. In this climate and with this type of air tight construction there is mostly a cooling/ dehumidification requirement and very little heating requirement 8) Individual HRV/ERV units must be installed inside the unit in a location where it can be serviced. This may take up closet space within the unit. The HRV/ERV requires ductwork to the outside that is drywalled and finished to match the interior décor. CCVS only requires a grill adjacent to the hallway to be installed. 9) Each HRV/ERV installation requires an electrical inspection. 10) Higher maintenance cost to change HRV/ERV filters twice a year in each unit which requires visiting all 100 units versus servicing two roof top CCVS units. 11) The occupants of the building could interfere with ERV/HRV operation. 12) CCVS installation would be quieter operation for the occupants than the ERV/HRV. 13) If individual HRV/ERV dual motors fail it requires 100 replacements which may be difficult to access and replace any finish covering. 9.0 IS SUPPLEMENTAL HEAT REQUIRED? During the study the temperature in the kitchen of unit 504 did not drop below 71 degrees (21.7 c) while 47 CFM (22 l/s) of outdoor air was being introduced when the outdoor temperatures were below freezing each night setting record lows for the area. See Table 3 and Figure 7. Note that there was no supplemental heat or occupancy induced heat or lighting that would add additional source of heat during the test. Note that the total CFM of outdoor air during the test was 47 CFM (22 l/s) induced and 31 CFM (15 l/s) natural for a total of 78 CFM (37 l/s) of outdoor air. The proposed airflow of 88 CFM (22 l/s) plus a natural ventilation of 31CFM (15 l/s) would give a total of 119 CFM (56 l/s) which is about 50% greater. Date Tues Wed Thurs Fri Sat Sun Mon 68 High Low 43° 24° 45° 18° 61° 26° 44° 21° 45° 13° 53° 29° 47° 30° Table 3 Outdoor temperatures in F during the ventilation test Note that the indoor kitchen temperature on Wednesday only dropped to 71 degrees (21.6 c) even though the outdoor temperatures had dropped to a low of 18 degrees (-7.7 c). Note that there was no supplemental heat, or lights left on or occupancy during this period to contribute additional heat from occupancy or from cooking or cleaning. The rise in temperature each day was from the solar gain into the unit. The guest bedroom climbed to 80 degrees during 5 of the 6 afternoons from solar gain. The master bedroom climbed to 91 degrees (32.7 c) during 2 of the afternoons and was above 80 degrees (26.7 c) every day. Figure 7 Kitchen temperature & relative humidity 69 Providing outdoor air that is not conditioned when the outdoor temperature is below 68 degrees would actually lower the occupancy cooling cost assuming the outdoor air did not significantly raise the humidity. It was recommended to install supplemental heat in the air handler that is set to activate only when the outdoor temperatures reaches a pre-determined level and to have the capacity to adjust this activation temperature and the temperature of the air that is provided during cold weather. It may in fact not be necessary to add any supplemental heat. 10.0 BUILDING VENTILATION STANDARDS ASHRAE, the American Society of Heating, Refrigerating and Air-Conditioning Engineers has been updating ventilation guidelines for commercial buildings for 100 years. The society however has not directed much attention to residential dwellings. Residential ventilation was traditionally not a major concern because it was felt that between operable windows and envelope leakage, occupants were getting adequate air. The latest standard developed by ASHRAE for ventilation of residential buildings is Standard 62.2-2003. This standard makes the target ventilation rate comprise a sum of the ventilation necessary to dilute background sources and sources attributable to occupancy. To find the total amount of outside air needed add 3 cfm/100 ft2 (15 l/s/100 sq.m.) to 7.5 cfm/person (3.5 l/s/person). Thus the air change rate requirement varies by the size of the house and the occupancy. If one assumes two persons for a single bedroom unit and four persons for a two bedroom unit then the total ventilation rate might be the following listed in Table 4 below. 1000 ft2 1500 ft2 2000 ft2 Single bedroom 45 CFM / 0.3 ACH Two bedroom 60 CFM / 0.4 ACH 75 CFM / 0.33 ACH 90 CFM / 0.3 ACH Table 4 ASHRAE current total ventilation recommendations 12.0 CONSTRUCTING A DESIGN AND INSTALLATION TEAM A team was establishing early in the process that consisted of the following groups. 1) Building Association and point of contact person 2) Radon and ventilation consultant 3) HVAC Engineering company and point of contact person 4) General Contractor consultant 5) HVAC sub-contractor consultant 6) Electrical sub-contractor consultant 70 7) Building Association attorney There were competing objectives among this group. On one side of the project is the group responsible for putting the project together. On the other side is the occupants who will be enduring the installation and performance of the system. Along with all of this are the issues of installation cost, and ongoing system energy and maintenance costs. As-built building plans were not available for this project and caused extra design work. Legal issues prevented obtaining plans from the normal sources: builder, architect, sub-contractor, or office where the permits were filed. A Request for Proposal (RFP) was distributed to HVAC engineering companies that had experience designing systems for similar buildings located in the same climate. The HVAC company was requested to consider all the issues of temperature, humidity, noise, installation issues, maintenance expense, and time table objectives of the project. The initial engineering company selected for this project had to withdraw because their insurance company would not cover the liability of the project. This project required extensive on-site review of duct sizing and routing because of the limited availability of duct runs down through the building and in particular the allowable duct runs above the drop ceiling down the hallways. The hallway duct runs were difficult because the building association requested that the drop ceiling not be lowered in the hallways. There were biweekly meetings throughout the design and installation process to discuss concerns and share preliminary information. The building owner selected Clean Vapor LLC as the commercial mitigation contractor based on cost, their previous experience with similar commercial building ventilation installations, their availability to complete the project on time, their references, and their willingness to work towards all the installation objectives. A project management schedule listing all the functions and task times through completion of the project was used. The site manager was onsite every day work took place in the building. Extra effort was provided by the selected mitigation contractor to minimize system installation dust and debris and minimize installation impact on the building occupants. The Building Association POC informed residents by email weekly of all project time tables and potential impacts. Residents were given advanced notifications of X-rays, utility shut off’s and items that require residents to evacuate their units. A project Health and Safety Plan (HASP) was developed and implemented by the Site Safety Officer. 13.0 INSTALLATION OF SUPPLEMENTAL VENTILATION SYSTEM The central conditioned ventilation system (CCVS) for the building was designed by a professional engineering company. On each floor of the building there were two storage rooms near the center of the building and a storage room on each of the outer wings of the building that 71 lined up with each other from floor to floor. In addition there was a drop ceiling on each floor above the hallway with about 8 inches of space. Two roof top ventilating units were installed on the roof. Two 20 Ton AAON package units, rated at 3500 cfm were installed on the roof and lined up over the outer wing storage rooms below. Each unit weighed 2900 pounds (1311 Kg) and provided 100% conditioned outside air. During the cooling season the unit is delivering 65 degree (18 c) air at 50% relative humidity (rh). During the heating season the unit will deliver 70 to 75 degree (21 to 24 c) air at approximately 20% rh. On humid days during the cooling season each unit is producing condensation at a rate of ½ gallon per minute. An additional drain to the storm sewer was installed to accommodate the condensate discharge. A fully ducted system was installed. The two main vertical ducts extended from the roof top units down to the first floor. The main trunks, that runs down through the building, required cutting through the concrete floors that had post tension stressed steel reinforcing cables. In order to avoid cutting any of these cables an X-ray scan of each slab penetration was performed. On each floor horizontal duct runs were routed from the main trunk above the drop ceiling along the entire hallway. In most cases an individual duct run from the hallway trunk was routed through the common wall of each resident’s unit, to a supply louvered grill installed in the foyer of the unit. Fire dampers were installed in the main trunk at each floor and in the duct that passed through the resident’s common wall to the hallway as per fire code requirements. An airflow damper was also installed in each duct to the unit to allow manual system balancing. Each of the floors had four equally spaced ceiling mounted supply diffusers installed in the hallways and an additional diffuser installed in the central elevator area to maintain necessary hallway ventilation. The original hallway air conditioning and heating units were removed and donated to Habitat for Humanity. As part of start up, an independent test and balance company conducted the testing and balancing to meet the design airflow specifications. 14.0 RADON REDUCTION FROM VENTILATION SYSTEM INSTALLATION After the central ventilation system was installed and balanced all of the units were tested with individual radon charcoal detectors. It appeared that the necessary closed house conditions were maintained by almost all of the occupants. Units Tested 1st Floor Quad units 1st Floor units 1st Floor without Quad units Average Post Mitigation Radon Level 2.1 1.2 0.8 72 3rd Floor units 4th Floor units 5th Floor units Hallways 0.8 0.8 0.7 0.6 Table 5 Post mitigation average per area unit 118 145 Quad 312 318 322 410 428 500 504 Hallway 112 Hallway 313 Hallway 413 Hallway 506 Pre-Mitg 9.0 5.9 16.3 13.2 10.6 10.0 9.5 5.1 7.5 9.3 5.5 4.9 6.6 Post Mitg 0.8 2.7 1.3 0.8 0.5 0.3 0.8 1.3 0.8 0.8 0.3 0.5 0.7 Table 6 Post mitigation versus Pre-Mitigation levels After the CCV was completed 91of the 100 units tested below1.5 pCi/l (56 Bq/m3). The average radon levels for each floor of the building are displayed in table 5 above. There are 12 two story units that are grouped into four units that are referred to as “Quad” units. See the first floor building layout in Figure 7 below. These units had the greatest distance from the roof top central ventilation air handlers. Four of the twelve Quad units had radon reductions between 2.0 pCi/l. and 3.0 pCi/l (74 to 111 Bq/m3). The average of the Quad units and the average of the first floor without the Quad units is displayed in Table 5 above. Note that the average radon levels for all the floors including the hallway measurements and excluding the Quad units is between 0.6 and 0.8 pCi/l (22 to 30 Bq/m3). Table 6 above compares the radon measurements that were obtained before the ventilation system was installed with the post mitigation radon results for these same units. 73 Overall the radon levels were dramatically reduced below the guideline, the indoor air quality was significantly improved, window condensation problems were eliminated, disruption to the occupants during construction and with operation of the system was minimized. 74 6.3 1st Quad Unit 101 103 1.9 1.1 105 107 1.1 0.7 109 1.4 0.3 110 102 1.8 1.6 0.3 0.8 100 9.3 112 2007 Pre-Mitigation Results 1st & 2nd Floor Final Test Results 0.8 111 1.4 113 1.1 115 114 0.8 1.4 June – July 2009 Post-Mitigation Results 117 116 1.6 1.0 9.0 118 0.8 122 120 1.9 1.5 124 126 2.0 1.8 2.3 1.7 nd 2 Quad Units 119 0.3 127 128 0.3 1.0 130 129 0.3 0.6 132 131 0.5 0.3 133 0.8 134 1.2 144 3rd Quad Units 135 142 136 0.3 1.2 2.8 2.0 2.6 1.6 145 143 5.9 1.8 141 139 1.8 1.1 0.9 2.7 137 0.5 2.7 Figure 8 First Floor Final Radon Measurements 75 References 1) Brodhead B. Elevated radon levels in high rise condominium from concrete emanation, Proceedings of AARST International Radon Conference; 2008 2) Brodhead, Measuring radon and thoron emanation from concrete and granite with continuous radon monitors and E-PERMs. Proceedings of AARST International Radon Conference; 2008 3) Burkhart, J.F. and Carnley, R.E. Using a model to estimate the ffects of ventilation and exhaust appliances on the accuracy of a 48 hour radon test. Proceedings of AARST International Radon Conference; 2000 4) Lane-Smith, D.R,, Residential Radon levels in Hong Kong. Proceedings of AARST International Radon Conference; 1997 5) Leonard, B.E. Ventilation rate and source emanation rates by measurement of induced radon time-dependent behavior. Proceedings of AARST International Radon Conference; 1994 76 RADON FLUX MONITOR FOR IN SITU MEASUREMENT OF GRANITE AND CONCRETE SURFACES Payasada (Paul) Kotrappa1 and Fredrick Stieff1 Rad Elec Inc. 5716-A Industry Lane Frederick, MD 21704 Daniel J. Steck2 Physics Department, St. John’s University Collegeville, MN 56321 ABSTRACT Recent interest in radon (222Rn) emanation from building materials like granite and concrete has sparked the development of a measurement device that is suitable for field or home measurements. Based on test with discrete component flux monitors, a large volume (960 ml) hemispherical electret ion chamber (EIC) was modified to integrate the accumulator and detector, into a single device. The device entrance is covered by a Tyvek sheet to allow radon from granite surface to diffuse into the monitor and further into the upper EIC chamber. The diffusional entrance of radon into EIC is through small area filters. This process reduces thoron sensitivity of EIC. This device is calibrated with a NIST radon emanation standard whose radon emanation rate is precisely known. Side-byside measurements with other emanation techniques on various granite surfaces in lab and field environments produce comparable emanation results. For low emitting building materials like concrete, a flux of 110 Bq m-2 d-1 (11 pCi ft-2 h-1) can be measured with 10% precision using short term electret in 8 hours. Sensitivities, ranges and applicable errors are discussed. INTRODUCTION There has been an increased interest in the radon emanation from granite used for countertops and tiles in homes. Radon originates from naturally occurring uranium present in granites and other building materials made from stone. The radon emanation rate of granite, along with the home’s volume and ventilation rate determines its contribution to room radon. Radon emanation rate is also called the radon flux and is defined as radon activity released per unit area per unit time. (While the appropriate scientific flux unit would be Bq m-2 s-1, units like Bq m-2 d-1 and pCi ft-2 h-1 are easier to use in practice. Results are provided in these units with some conversion factors for other units in the text.) 1 These authors are the developers of, and have a commercial interest in, the radon flux monitor described in this paper. 2 This author does not have a commercial interest in the devices described in this paper. 1 77 There are few detailed published studies of granites marketed in the US, but recent reports suggest that most granites have low fluxes with most granites showing a radon flux below 250 Bq m-2 d-1( 26 pCi ft-2 h-1) with few as high as 3000 Bq m-2 d-1 ( 320 pCi ft-2 h-1) (Kitto, 2009). Radon flux can be used in models to estimate the contribution of building materials to indoor airborne radon concentrations under various exposure scenarios. Low indoor radon concentrations (<0.4 pCi/L) are predicted by the models. However for small living spaces with low ventilation and large installations of tile or countertops, much higher radon emanation is predicted. (Steck, 2009) Radon flux measurements can be difficult in the field because large emanation variations occur in some material surfaces and it is often difficult to sample large areas of material. The best radon generating potential of a material would come from a complete enclosuretype accumulator (Kotrappa, 2009). Since that is usually impractical and measuring samples in the laboratory often mischaracterizes the true average emanation, the goal of this work was to develop a practical, light weight, simple radon flux measuring device that could quickly make measurements at multiple locations on the suspected building materials. After experiments with discrete flux monitors using separate components for the accumulator and radon detector, a new radon flux monitor, was developed, calibrated, and field tested. MATERIALS AND METHODS RADON FLUX MONITOR (RFM) In this paper the integrated radon flux monitor is referred to as RFM. A schematic of the RFM is shown in Figure 1. Radon emanating from the granite surface diffuses through the buffer volume and though four filtered holes (one quarter inch diameter) into an electret ion chamber portion of the flux monitor. The electret ion chamber (EIC) measures radon concentration accumulated during the measured period and is related to radon flux. Most building materials generate both radon (222Rn) and thoron (220Rn) gases. But the difference in decays of these two gases makes the dose generated indoors by thoron a small fraction of the dose generated by radon. However, most field-grade “radon” detectors will respond to thoron at about rate as radon unless steps are taken to limit the thoron that gets into the active volume. This monitor has a unique feature that minimizes the response to thoron. Thoron emanating from granite surface while diffusing through the buffer volume and though four filtered holes effectively decays because of short half life of thoron. The use of small area filtered openings to minimize the response of electret ion chamber for thoron, is a standard method, used in all E-PERM®, EIC radon monitors (Kotrappa, 1990). This RFM has an accumulation volume of 960 ml and a Tyvek-covered window area of 189 cm2. This monitor uses a well known “accumulation theory” of determining the radon flux from the objects (Kotrappa, 2009). The entire volume of the chamber serves as an accumulator and the electret ion chamber (EIC) part serves as a radon monitor. Equation (1) relates the average radon measured by the EIC and the radon flux F when the accumulator has no radon losses other than radioactive decay. 2 78 ( F . A ) & , 1 - e - 0.1814T )# '! $1 - * C ( Rn) Av = V . 0.1814 $ *+ 0.1814T '(! % " --------------- (1) Notation: F is the radon flux in Bq m-2 d-1 A is the area of the granite measured in m2 A is also the area of the radon flux monitor window in m2 (F × A) is the exhalation rate in Bq d-1 0.1814 is the decay constant of radon in d-1 C(Rn) Av measured by EIC, is the integrated average radon concentration in Bq m-3 T is the accumulation time in days V is the air volume of the accumulator in m3 The radon flux F is calculated using equation (1) leads to a flux in units of Bq m-2 d-1. Flux in other units can be obtained by multiplying the result by; 0.0116 to convert to mBq m-2 s-1, 0.105 to convert to pCi ft-2 h-1 , and 1.125 to convert to pCi m-2 h-1. Fig 1a Schematic details of Radon Flux Monitor 3 79 Figure 1a illustrates that the radon flux monitor and Fig 1b illustrates the deployment of radon flux monitor. Radon flux monitor should be positioned on a granite sample which is allowed to emanate freely from both surfaces. It is important to make a good seal with the surface using adhesive clay because the analysis equation assumes that there are no leakage losses from the surface around or through the RFM. A premeasured electret is screwed tight into the radon flux monitor. The edge of this electret should also be sealed with adhesive clay. At the end of desired period (typically 8 hours) the electret is screwed out and measured. The method of calculating radon concentration is similar to that for any EIC. Fig 1b Deployment of the Radon Flux Monitor Calibration factors are obtained for this EIC using standard procedures (Kotrappa, 1990). Average radon concentration in the accumulator is calculated using equation (2) Rn concentration = (I-F)/ (CF x T) ---- (2) Where CF = 4.4757 + 0.002634 [(I+F)/2] T is the accumulation time in day units Rn is the radon concentration in pCi/L I and F are the initial and final voltages of ST electret Multiply the measured radon concentration by 37 to convert the radon concentration to units of Bq m-3 before using it in equation (1). BACKGROUND CORRECTIONS FOR RFM Electret ion chambers respond to background gamma radiation as well as radon. Therefore, both the ambient gamma radiation and room radon can affect the RFM result. These effects will be most important for low emanation materials and in high radon rooms. To correct for those effects, a background measuring RFM is employed. The background measuring radon flux monitor is identical to the standard RFM, but the window is covered with 1.5 mm thick PVC disk. This stops the radon from the surface, 4 80 but not the gamma radiation or the ambient radon in the room. The background RFM is positioned on the granite, and the background equivalent radon flux calculated by the same procedure. This background flux is subtracted from the measured radon flux to calculate the net radon flux. We have observed that the signal from gamma radiation is very small (2 to 5%) compared to the signal from radon. Background corrections are needed when very accurate measurements are needed. CALIBRATING THE RFM WITH NIST RADON EMANATION STANDARDS Radon emanation standards (SRM 4971 to 4974) are available from NIST (National Institute of Standards and Technology). Characteristics of radon emanation standards are provided in terms of 226Ra strength and emanation fraction (Volkovitsky, 2006). The following equation is used to calculate the emanation rate from the characteristics of the standard provided by NIST. Radon emanation rate, R (222Rn) is calculated using equation (3) (Kotrappa, 2006). R (222Rn) = f . A (226Ra) . λ ------ (3) Where, f is the emanation fraction A (226Ra ) is the radium content in Bq λ is the decay constant of radon in day-1 The primary standard used in this work has NIST certified characteristics as: A is 5604 Bq, f is 0.844, λ is 0.1814 day-1.The emanation rate is calculated using Equation (3) to be 858 Bq/day with a precision of better than 2 %. (Kotrappa et al 2005). To calibrate the RFM, peel back the Tyvek and use a little adhesive clay to attach the source to the chamber separator. Seal the edges of RFM to the plastic disk with adhesive clay. Screw in a premeasured long term electret (LT) into the RFM. Put a seal of clay around the screw parts of electret to ensure leak tightness. After 8 hours, unscrew the electret and take a final reading. Follow the standard procedure to calculate the radon emanation rate. Compare this with the theoretically calculated radon emanation rate. The ratio between the expected to the calculated is the correction factor that should be used for multiplying the measured emanation rate for an 8 hour measurement. Table-1 gives the results of the calibration experiments of author PK. 5 81 Table-1 Results of calibration of RFM with a NIST Source Expected emanation rate: 856 Bq/day Expt # 1 2 3 Radon Conc. Bq/m3 127473 133617 132253 Calculted Emnation Rate F x A Bq/day 764.7 785.2 777.2 Calculated/Measured Acc time hours 7.833 8 8 Average 775.7 1.104 An independent calibration by author DJS using a different NIST source (4974-8) whose emanation was 752 Bq d-1 had a Calculated to Measured ratio of 1.20. The RFM with short-term electrets was calibrated by DJS using a lower activity NIST standard (4971-3) whose emanation was 0.821 Bq d-1. The Calculated to Measured ratio was 1.00. RADON FLUX USING LARGE ACCUMULATORSAND EICs In early experiments, a radon flux measurement system that used discrete components was tested. A 3L hemispherical metal bowl with an opening of 0.04 m2 (0.43 ft2) was attached to the surface and sealed with adhesive clay around the edge. On some rough and dirty vertical surfaces, it was necessary to add a quick fixing adhesive to the rim in addition to the clay seal to insure mechanical stability and low radon leakage. A standard EIC was enclosed in the bowl. At the end of desired period, typically 8 to 24 hours, the bowl is removed from the surface and the EIC electret measured similar to the standard procedure with any electret ion chamber. Then equation (1) is used to calculate the flux. Long-term electrets were used when the radon emanation was expected to be high. Figure 2 shows two of these systems and an RFM on a vertical granite surface. Fig 2 Two discrete component flux monitors and one RFM deployed on a vertical granite surface. 6 82 RADON FLUX USING LARGE ACCUMULATORS AND CRMs Another discrete component radon flux system consisted of an inverted 4.6 L metallic bowl with a small continuous radon monitor (CRM) inside. Edges were sealed with adhesive clay. At the end of the accumulation period, the bowl removed and the hourly data down loaded. Equation (4) (Kotrappa, 2009) is used for calculating the radon flux from the data obtained from hourly readings of a continuous radon monitor C ( Rn) = k !T ( FXA) (1 " e " ) VXk ---- (4) Notation: F is the radon flux in Bq m-2 d-1 A is the area of opening of the accumulator in m2 k is the effective decay constant of radon. This is 0.1814 d-1 if there is no loss from leaks through seals or the granite itself. C(Rn) is the radon concentration at any accumulation time of T in Bq m-3 T is the accumulation time in days V is the air volume of the accumulator in m3 The radon flux F is calculated using equation (1) leads to a flux in units of Bq m-2 d-1. Flux in other units can be obtained by multiplying the result by; 0.0116 to convert to mBq m-2 s-1, 0.105 to convert to pCi ft-2 h-1 , and 1.125 to convert to pCi m-2 h-1. Alternatively, the in growth equation (4) can be fit to determine F. In fact, if there are few leaks, a linear approximation can easily be fit to the in growth of the early hours (say 3 to 20 hours) to give a good estimate for F. RESULTS RADON FLUX MONITOR(RFM) The RFM was used to measure the flux from a granite sample repeatedly with exposure periods ranging from 4 to 24 hours. Table 2 and Figure 3 show that the flux calculated using equation (1) is quite reproducible with a variation of ~14%. Table- 2 Radon Flux Monitor on F granite for different accumulation time Rn pCi/L Hours Volume M3 A m2 23.2 4 0.00095 0.01887 49.3 8 0.00095 0.01887 55.0 12.5 0.00095 0.01887 73.7 16 0.00095 0.01887 114.8 20 0.00095 0.01887 101.3 24 0.00095 0.01887 7 FxA Bq/day 9.8647 10.6158 7.6557 8.0869 10.1819 8.9775 Flux 523 563 406 429 540 476 83 Flux in Bq m2/day Flux Bq/m2/day 600 500 400 300 200 100 0 0 10 20 30 Accumulation hours Fig 3 Calculated Radon Flux at stated accumulation time RADON FLUX USING LARGE ACCUMULATORS AND CRMs Figure 4 shows the results of calculating the radon flux using equation (4) at each hour after the exposure started. While the variation of the calculated flux during each hour during the entire experiment was 11%, it would only be 6% if the average flux result during the last 10 hours was used. Granite F 600 Bq/m2/day 500 400 300 200 100 0 0 5 10 15 20 25 Time, hours Fig 4 Calculated Radon Flux at stated accumulation time 8 84 COMPARATIVE PERFORMANCE OF RADON FLUX SYSTEMS Six different granite samples were repeatedly measured on both sides with a variety of radon flux measurement systems. In addition to the 3 flux measuring systems described above, an enclosure-style system was used to get the radon emanation from all surfaces of a granite sample. That system consisted of a 24 L metal enclosure in which a granite sample and a continuous radon monitor could be sealed. Figure 5 shows this system. Fig 5 Enclosure accumulator with CRM and granite sample In this CRM based systems, the flux was determined by fitting the in growth curve using an appropriate function (equation 3) reflecting the actual radon loss rate (k). Figure 6 shows an experiment with bowl accumulators applied to the “non-polished sides” of the samples. Five of the granite samples had quite high radon emanation that varied significantly across the surface. Since the RFM “ footprint” would cover only 20% of a top or bottom surface, while the bowl accumulators covered roughly 40%, it is not surprising that there is some variation in flux measured by the different systems. Table 2 shows the comparative performance of the different systems on the six granite samples. 9 85 Fig 6 Accumulators applied to the “non-polished” sides of six granite samples Table 3 Radon fluxes in pCi ft-2 h-1 for different granite samples measured with 4 different systems. Method Sample/surface CK08 polished rough all surfaces CB08 polished rough all surfaces FS08 polished rough all surfaces FSMK polished rough all surfaces JB08 polished net all surfaces JBMK polished net all surfaces Enclosure +CRM 5L Bowl +CRM 3L Bowl +EIC RFM 16 16 13 15 11 21 16 18 21 20 155 100 260 180 90 250 170 70 270 170 310 30 450 240 30 510 270 25 525 275 160 42 215 129 46 230 138 52 370 211 80 120 1 61 105 10 58 160 8 84 415 495 365 430 430 290 360 829 280 555 1 0 86 The grayed numbers are averages of the two results above for comparison to the flux in the first column. Sample CK08 has flux typical of many granites and is quite spatially homogeneous. The other samples show variation in gamma activity across their surfaces which suggests that the emanation too may be inhomogeneous across their surfaces. While the 3L and 5L bowls cover about 0.45 ft2, the RFM only covers about 0.2 ft2 . Samples FS08 and FSMK are from the same slab. Samples JB08 and JBMK are from granite with the same name but are from different slabs. JBMK has a small “hot spot” at the center of the polished side that covers about half of the RFM opening and about 20% of the opening for the bowl accumulator systems. SENSITIVITY AND RANGE The standard procedure of calculating sensitivity and range, applicable to other electret ion chambers is also applicable to the EIC in this radon flux monitor (Kotrappa, 1990). A 20 volt drop is considered as the detectable limit for 10% accuracy. A 500 volt drop is considered as the range. It is found that a 50 volt drop in 8 hours leads to 267 Bq m-2 d-1. Using this as the basis, the table of sensitivity and range is calculated. This means about 107 Bq m-2 d-1 can be measured with 10% accuracy in an 8 hour. The ST electret in the EIC goes out of range for more than 2670 Bq m-2 d-1 in an 8 hour measurement. In such cases a LT electret (Kotrappa, 1990) with a lower sensitivity by a factor of 12.5, is used. Following Table gives the sensitivity and range for different electrets for 8 hour measurement. Table-4 Sensitivity and Range Electret Type ST (short term) LT (Long term) Sensitivity (Bq m-2 d-1) 120 1504 Range (Bq m-2 d-1) 3003 37547 DISCUSSION OF RESULTS Table-1 gives the calibration data with a NIST radon emanation standard. The three sets of data show that the results are highly reproducible. The ratio of theoretical results is about 10% higher than the measured results. This may be taken as the calibration correction. This difference may be due to uncertainties in EIC calibration and in the measured dimensions of radon flux monitor. Additional calibration work done by DJS with NIST sources showed good agreement for the ST system and reasonable agreement for the LT system as discussed above. Table-2 and Fig 3 gives results on measurement done on Granite sample F. This granite sample is obtained from a local granite manufacturer. The dimensions of the sample are 1 1 87 12 inch long, 12 inch wide and a thickness of 1.25 inches. All measurements are done on the polished side. The unpolished side has a resin with fiber glass mesh. This is one of the typical types sold. The radon flux is in the range of 400 to 550 Bq m-2 d-1. The results do not appear to vary systematically with the accumulation time. An 8 hour measurement is recommended for routine measurement. Figure 4 give results of another independent measurement done on the same slab with large inverted bowl and a continuous radon monitor. There are wide fluctuations in results in the range of 1 to 8 hours. This is largely due to low sensitivity of the CRM during these periods. However after 8 hours, the calculated flux appears to stay stable. The results appear to agree fairly well with the results obtained with radon flux monitor (Table-2). Agreements prove that both methods are acceptable for field use. Table-4 is generated by the author (DJS) on large number of samples using different accumulators and different types of radon monitors. The radon flux monitor results, shown in the far right column generally agree with the other devices. Some differences in the results can be due to the non uniformity of radon flux on surfaces along with the differences in the accumulators’ foot prints. CONCLUSIONS The light weight radon flux monitor is simple to use and provides results similar to the other advanced accumulator based flux methods. The measurements for this radon flux monitor, and others, were validated using NIST radon emanations standards. 1 2 88 REFERENCES Kitto ME, Green J Emanation from granite countertops. Proceedings of the American Association of Radon Scientists and Technologists 2005 International Symposium, San Diego CA ( 2005) Kitto M.E, Haines D.K, Aruzo H.D. Emanation of radon from household granite, Health Physics: 6:477-482 (2009) Kotrappa P, Dempsey J.C, Ramsey R.W and Stieff L.R. A practical E-PERM® System for indoor radon measurement, Health Physics 461-467 (1990) Kotrappa P, Stieff LR, Volkovitsky P. Radon monitor calibration using NIST radon emanation standards: steady flow method. Radiation Protection Dosimetry 113:7074:2005 Kotrappa P and F Stieff. . Radon exhalation rates from building materials using electret ion chamber in accumulators. Health Physics 97:163-166 (2009) Steck DJ. Pre- and post-market measurements of gamma radiation and radon emanation from a large sample of decorative granites, Proceedings of the American Association of Radon Scientists and Technologists 2009 International Symposium St. Louis MO.(2009) Volkovitski P. NIST 222 Radon emission standards, Applied Radiation and Isotopes. 64:1249-1252; 2006 1 3 89 2009 International Radon Symposium, St. Louis, MO LABORATORY INTERCOMPARISON OF RADON-IN-WATER STANDARDS Michael E. Kitto1,2, Abdul Bari1, Douglas K. Haines1, Traci A. Menia1, and Eileen M. Fielman1 1 Wadsworth Center, New York State Department of Health, P.O. Box 509, Albany, NY 12201 2 School of Public Health, State University of New York, Rensselaer, NY 12144 Abstract The standardization of instruments used for radon-in-water measurements typically involves handling and disposal of 226Ra in solution. To avoid contact with the Class A carcinogen, radium-free solutions were prepared and tested for use as radon-in-water standards. Filters containing known amounts of 226Ra were sealed in polyethylene and placed in vials filled with distilled water for >30 days to allow the decay products to establish secular equilibrium. Voluntary intercomparisons of the radon-in-water standards were conducted to investigate the accuracy of analyses by commercial, government, and private companies. Various analytical methods were utilized by the participants. Result show that, at radon concentrations of 18,700 and 3850 pCi/L (693 and 143 Bq/L, respectively), most participants reported concentrations within 25% of the known amounts. Introduction Naturally occurring radionuclides are frequently measured in groundwaters. Of these radionuclides, radon (222Rn) has sufficient properties, solubility, and half-live (3.8 d) that during household use of groundwater (showering, cooking, or washing dishes and clothes) the dissolved radon can contribute substantially to indoor-air levels (Fitzgerald, 1997). The radon released from the water into the home's air will contribute to the radon concentration in the indoor air. Liquid scintillation (LS) counting is a recommended method for the measurement of radon in water (Whittaker, 1989). The LS method is inherently easy, rapid, and in common usage. During the efficiency standardization of an LS spectrometer, a solution containing radium (226Ra), a Class A carcinogen, is typically used. Use of 226Ra means that loss of radon 90 during transfer of the water sample is undetectable, since equilibrium of 222Rn and its decay products is re-establishment in the cocktail. Also, as radioactive equilibrium is maintained in the cocktail during counting, the decay of radon, which occurs in real samples, cannot be monitored. The inability to detect transfer losses and radioactive decay, in standardization cocktails that contain radon in equilibrium with 226Ra, limits the reliability of radon results obtained from groundwater samples that do not contain supporting 226Ra. An alternative method involves creation of a 226Ra-free solution in which radon emanates from a sealed 226Ra source into an encapsulated aliquot of water. This rarely-used method includes sample-transfer proficiency and decay-correction calculations in determination of the method efficiency. Unfortunately 226Rafree standards for radon-in-water analyses are not readily available. The objective of the present study was to examine results from the analysis of 226Ra-free standards by laboratories that standardize their methods using solutions containing 226Ra. Methodology Sealed sources were prepared by pipetting a known volume of a 226Ra-standard solution onto filter papers of 1.27 cm diameter. After drying, each filter paper was sealed inside 0.05-mm (2-mil) thick polyethylene sheeting and placed in a 42-ml glass vial, which was filled with distilled water and capped with a Teflon-lined septum. Concurrently, the 226Ra-standard solution was pipetted into 10 additional vials, which were filled with distilled water, capped, and used as reference solutions. Solutions containing a sealed 226Ra source of a known activity (A; pCi/L) can be measured to determine the emanating fraction (ƒ) of the source according to the equation: ƒ= (G - B) (2.22*V*ε*A*D* I) (1) where G and B are the respective gross and background count rates (cpm), 2.22 is the conversion factor between dpm and pCi, V is the sample volume, and ε is the detector efficiency. The time span for the ingrowth factor (I) extends from initial encapsulation to sample extraction, while the time for the decay factor (D) extends from sample extraction to the midpoint of the measurement. Prior to sending the vials to participants of the intercomparison, the solutions were allowed to attain full ingrowth (>40 d) of radon and measured to identify any problematical 91 standards. For each vial, 10 ml of radon-laden water was extracted and injected beneath 10 ml of high-efficiency mineral oil. The LS cocktail was shaken to expedite transfer of the radon into the scintillation fluid. Counting was delayed a minimum of 3 h, to allow radioactive equilibrium of the short-lived radon decay products to be established in the cocktail. Radon activities were determined using 50-min counts on a LS spectrometer, with an absolute efficiency of 64% for radon and for each of its four short-lived (α and β) decay products (ε = 3.2 cpm/dpm). Each sample was counted three times within a week and was decay-corrected to the transfer time. Repetitive counts of the four method blanks resulted in a mean of 0.61 ± 0.07 dpm. While negligible compared to the standards, the blank count rate was subtracted from all LS measurements. To examine the utility of the 226Ra-free radon-in-water standard, and to investigate the accuracy of radon-in-water measurements, we conducted a voluntary intercomparison. The participants included commercial laboratories, individuals, and state and federal agencies. Each of the sealed 226Ra sources described above was placed in a 42-ml vial filled with distilled water and provided to the participants after full radioactive ingrowth. Results and Discussion The 10 vials that contained a known quantity of 226Ra in solution provided a reference value for the LS measurements. There is no decay associated with the activity in the cocktails, and radioactive equilibrium is maintained in the solutions. Further, the radon is dissolved in solution and does not have to penetrate polyethylene, as it does with the sealed sources. Activities measured in the 10-ml vials averaged 99.6% of the value expected based on the NIST certificate that accompanied the standard. Count rates for five determinations of each of the 10 vials (n=50) were within 7% of the known value. For the radium standards, there was no further ingrowth evident during the consecutive counts, implying that the solution pipetted onto the filters is of a known activity and the transfer of the radon-laden water to the LS vials was quantitative. The efficacy of the Teflon-lined septum in retaining radon in the vials is demonstrated by the precision and accuracy of the activity values for the 226Ra-dissolved solutions. Cocktails were created from water in the vials containing the sealed filters and were measured three times by LS counting. Results (Fig. 1) indicate that after full ingrowth the radon 92 Fig. 1. Decay-corrected activities in the cocktails indicate that secular equilibrium is established within 40 d. concentration in the solutions averaged 86% of that determined in the 226Ra standards, and was identical to the emanating factor reported (Volkovitsky, 2006) for a polyethylene-encapsulated radium source. Decay during diffusion through the polyethylene and retention of radon in the polyethylene may be responsible for diffusivity below 100%. Errors due to counting statistics were <1% and cannot account for the 14% reduction in activity. The analytical devices employed by the intercomparison participants included liquidscintillation counters, alpha-scintillation (“Lucas”) cells, electrets, gamma-ray detectors, a continuous radon monitor (CRM), and etching of polycarbonate (Pressyanov, 2008). For those utilizing the liquid-scintillation method, sample volumes were 8, 10 or 15 ml, count times varied from 9 to 100 min, and methods of extraction included needle and large-bore syringes. All participants who utilized a LS counter had standardized the instrument using 226Ra. For the 93 Fig. 2. Results of a radon-in-water intercomparison conducted in 2008 show that 18 of 21 participants reported concentrations within 25% of the known value (18,700 pCi/L). methods based on electret, gamma-ray, and CRM measurements, the entire volume of the vial was measured for >24h, 2 h, and 20 min. respectively. As shown in Figure 2, reported waterborne radon concentrations for the first intercomparison varied from 12,400 to 22,300 pCi/L. Results for most (86%) of the participants were within 25% of the known (18,700 pCi/L); the three outliers reported low concentrations. One laboratory initially reported 5670 pCi/L using a needle-syringe extraction, and 18,300 pCi/L for a repeat sample using a syringe containing a large bore for sample extraction. For the second intercomparison, reported waterborne radon concentrations varied from 2760 to 6910 pCi/L (Fig. 3). With the exception of two results, the participants reported concentrations within 25% of the known (3850 pCi/L). Losses occur during sample transfer when the water is exposed to air in 94 Fig. 3. Results of a 2009 radon-in-water intercomparison conducted in 2009 show that 21 of 23 participants reported concentrations within 25% of the known value (3850 pCi/L). the syringe. In contrast, reported values above the expected concentration indicate error in the detector efficiency. The inter-method comparison demonstrated that laboratories often provide accurate results; however, results are often skewed toward lower values. Conclusions Standards containing radon in water were created. Results of measurements of 10 liquid scintillation cocktails containing 226Ra in solution were nearly identical to the known value. Water in vials that held encapsulated 226Ra sources were found to contain only 86% of the known activity, after full ingrowth had been attained. The uniformity of radon emanation through polyethylene demonstrates that 226Ra-free radon standards can be reliably produced and used repeatedly. The 226Ra-free standards are a functional method to generate samples for proficiency testing of laboratories that conduct radon analyses of water. For both radon-in-water 95 intercomparisons, a large majority of the participants reported results within 25% of the known activities using an assortment of analytical methods. References Fitzgerald, B.; Hopke, P.K.; Datye, V.; Raunemaa, T.; Kuuspalo, K., 1997. Experimental assessment of the short- and long-term effects of 222Rn from domestic shower water on the dose burden incurred in normally occupied homes. Environ. Sci. Technol. 31, 1822-1829. Pressyanov, D.; Dimitrova, I.; Georgiev, S.; Mitev, K., 2008. Measurement of 222Rn by absorption in polycarbonates - reseach and practice. International Radon Symp., Las Vegas, NV. Volkovitsky, P., 2006. NIST 222Rn emission standards. Appl. Radiat. Isotopes 64, 1249-1252. Whittaker, E.L.; Akridge, J.D.; Giovano, J., 1989. Two test procedures for radon in drinking water: interlaboratory collaborative study. Las Vegas, NV: U.S. EPA Environmental Monitoring Systems Laboratory; EPA 600/2-87/082. 96 A LIVING RADON REFERENCE MANUAL Robert K. Lewis Pennsylvania Department of Environmental Protection Bureau of Radiation Protection, Radon Division and Paul N. Houle, PhD University Educational Services, Inc. Abstract This “living” manual is a compilation of facts, figures, tables and other information pertinent and useful to the radon practitioner, some of which can be otherwise difficult to find. It is envisioned as a useful addition to one’s desk and radon library. This reference manual is also intended to be a “living” document, where its users may supply additional information to the editors for incorporation in revisions as well as updates to this document on-line. Topics contained within the current version include radon chemistry and physics, radon units, radon fans, epidemiology, ambient radon, diagnostics, dosimetry, history, lung cancer, radon in workplace and radon statistics. In some cases motivations and explanations to the information are given. References are included. Introduction This reference manual is a compilation of facts, figures, tables and information on various aspects of radon science. It is hoped that this manual may prove useful to federal and state employees, groups such as AARST and CRCPD, and industry. There are numerous other reference manuals that have been produced on the various aspects of radon science; however, we hope that this manual will have a more “applied” use to all of the various radon practitioners who may use it. Many of the snippets on the various pages are highlights from referenced sources. The snippet will obviously only provide one with the briefest of information. To learn more about that item go to the reference and read the whole paper. It is our intent to continually update this manual with the help of all those mentioned above. Especially helpful would be remarks, comments and additions to the manual from anyone inclined to help. Please forward any contributions to rolewis@state.pa.us. The authors who will act as editors reserve all editorial rights. September 14, 2009 97 Table of Contents Introduction............................................................................................................... 1-1 Chemistry.................................................................................................................. 2-1 Physics...................................................................................................................... 3-1 Units ......................................................................................................................... 4-1 Derivation of Rn/Tn working level............................................................................ 5-1 Radon-222 Dose Calculation..................................................................................... 6-1 Dosimetry ................................................................................................................. 7-1 Effective Whole Body Dose Equivalent (EDE) ......................................................... 8-1 Risks From Indoor Radon-US EPA ........................................................................... 9-1 Radon-222 and Lung Cancer Risk Estimates ........................................................... 10-1 Epidemiology.......................................................................................................... 11-1 Lung Cancer............................................................................................................ 12-1 Lung Cancer Rate by State ...................................................................................... 13-1 NCRP Pie Chart ...................................................................................................... 14-1 History of Radon-222 Occupational Limits ............................................................. 15-1 Radon Geology ...................................................................................................... 16-1 Radon Decay Products ........................................................................................... 17-1 Radon Variability.................................................................................................... 18-1 Radon in the Workplace .......................................................................................... 19-1 Diagnostics ............................................................................................................. 20-1 Fan Selection .......................................................................................................... 21-1 Fan Comparison ...................................................................................................... 22-1 Post-mitigation Radon Data..................................................................................... 23-1 Ambient Radon ....................................................................................................... 24-1 Radon Statistics....................................................................................................... 25-1 State Radon Data..................................................................................................... 26-1 State Radon Rankings ............................................................................................. 27-1 EPA State Rankings ................................................................................................ 28-1 PA Radon Analyzer Data vs. UK Data .................................................................... 29-1 International Radon Section .................................................................................... 30-1 10 CFR 20 Table..................................................................................................... 31-1 Other Radon Reference Manuals ............................................................................. 32-1 Radon in Water .............................................................................................In Progress Radon and Granite/Building Materials ..........................................................In Progress Radon Testing Equipment/Devices................................................................In Progress Mitigation Equipment ...................................................................................In Progress Thoron ..........................................................................................................In Progress Pertinent Web Sites .......................................................................................In Progress September 14, 2009 98 Chemistry of Radon Atomic Number: 86 Symbol: Rn Atomic Weight: 222.0176 Discovery: Fredrich Ernst Dorn 1898 or 1900 (Germany) discovered the element and called it radium emanation. Ramsay and Gray isolated the element in 1908 and named it niton. Word Origin: from radium. Radon was once called niton, from the Latin word nitens, which means “shining.” Isotopes: At least 20 isotopes of radon are known. Radon-220 is commonly called “thoron” and emanates naturally from radium-224, one of the decay products of thorium-232. Thoron is an alpha-emitter with a half-life of 55.6 sec. Radon-219 is commonly called “actinon” and emanates from radium-223, a decay product of actinium-227. It is an alpha-emitter with a half-life of 3.96 sec. Properties: Radon has a melting point of -71° C, boiling point of -61.8 °C, gas density of 9.73 g/l, specific gravity of the liquid state of 4.4 at -62°C, specific gravity of the solid state of 4 and usually a valence of 0 (it does form some compounds, however, such as radon fluoride). Radon is a colorless gas at normal temperatures. It is also the heaviest of the noble gases. When it is cooled below its freezing point it displays brilliant phosphorescence. The phosphorescence is yellow as the temperature is lowered, becoming orange-red at the temperature of liquid air. Inhalation of radon decay products presents a health risk. Radon build-up is a health consideration when working with radium, thorium or actinium. It is also an issue in uranium mines. September 14, 2009 99 Sources: It is estimated that each square mile of soil to a depth of 6 inches contains approximately 1 g of radium, which releases radon to the atmosphere. The average concentration of radon is about one in sextillion parts of air (1 radon atom in 1021 air atoms.). Radon naturally occurs in some spring waters. Element Classification: Inert Gas Density (g/cc): 4.4 (@ -62°C) Melting Point (°K): 202 Boiling Point (°K): 211.4 Appearance: heavy radioactive gas Specific Heat (@20°C J/g mol): 0.094 Evaporation Heat (kJ/mol): 18.1 First Ionizing Energy (kJ/mol): 1036.5 Lattice Structure: Face-Centered Cubic References: Los Alamos National Laboratory (2001), Crescent Chemical Company (2001), Lange's Handbook of Chemistry (1952), CRC Handbook of Chemistry & Physics (18th Ed.) and http://chemistry.about.com/od/elementfacts/a/radon.htm September 14, 2009 100 Physics of Radon Atomic and Nuclear Structure Let’s first recognize that atoms consist primarily of electrons, protons and neutrons. Interestingly, the protons and neutrons exist in a very small central region called the nucleus, while the electrons orbit outside this region at quite a distance comparatively as seen in this drawing. The number of protons in the nucleus uniquely defines the element of which the atom is a part. For example, if there is only one proton in the nucleus of an atom, that atom is a hydrogen atom, no matter how many neutrons or electrons are also part of that atom. If there are 86 protons in the nucleus of an atom, that atom is a radon atom, again regardless of how many protons or neutrons are in that atom. Radioactive decay is defined to be the spontaneous breakup of an atom. Not all atoms are radioactive and those that are not are called “stable.” If we have a group of radioactive atoms, such as radon, or uranium or plutonium, if we just wait a while, some of them will break up or explode spontaneously. Moreover, it is interesting to note that this “break up” or decay is relatively easy to model. An analysis of this decay points out that every radioactive element has a unique half-life. Half-life being defined as the time it takes for half of a sample’s atoms to decay to the next nuclide in the decay chain. For example, U-238 is radioactive with a half-life of 4.47 billion years. This means if you have 1000 U-238 atoms in your hand today and wait 4.47 billion years you will have only 500 U-238 atoms in your hand, and in another 4.47 billion years, you will then have only 250 U-238 atoms in your hand. All right, so there’s a slight problem here in having you live that long, but you get the point. If we take Radon as an example, its half-life is 3.82 days. Therefore, if you have 1000 radon atoms in your hand today, in 3.82 days you will have just 500 radon atoms and in September 21, 2009 101 another 3.82 days, or a total of 7.64 days, you will have 250 atoms. The graph below depicts this decay rate. # of radon atoms # of radon atoms vs. time 1000 800 600 400 200 0 0 1 2 3 4 5 6 7 8 9 10 time (days) This is also sometimes called exponential decay, simply because the number of radon atoms N(t) existing at the time t is given by the equation: N (t ) = N i # exp " !t Where Ni is the number of radon atoms existing initially at t = 0, λ is the decay constant, or the probability of decay per unit of time (e.g. 2 per second, or in this case 0.181 per day) and t is the time as measured from when the initial number Ni exist. Notice from the graph and the equation above that the number of radon atoms is predicted to go to zero only after a very long time. Actually, after the number of atoms decreases to a small number, the statistical assumptions leading to the concept of half-life fail. Fortunately, one very rarely comes upon situations where the number of atoms is that small. Distribution of the Heavy Elements Uranium is a common element found almost everywhere within both the earth’s crust and seawater in varying concentrations. In nature, uranium atoms exist as uranium-238 (99.284%), uranium-235 (0.711%) and a very small amount of uranium-234 (0.0058%). Uranium decays slowly by emitting an alpha particle. The half-life of uranium-238 is about 4.47 billion years and that of uranium-235 is 704 million years, making them useful in dating the age of the Earth. There are four decay chains of importance when considering naturally occurring radioactive materials, each defined by its heaviest natural (not man-made) element. These are the U-235 series (also called the Actinium Series), the thorium-232 series (the Thorium Series), the U-238 series (the Uranium Series) and the Np-237 series (the September 21, 2009 102 Neptunium Series); graphical depictions of each are shown below. The one of most interest to us is the decay chain that includes radon-222, namely the U-238 series. 1µs = 10-6 s, 1 ms = 10-3 s,1 My = 106 y, 1 Gy = 109 y September 21, 2009 103 1µs = 10-6 s, 1 ms = 10-3 s,1 My = 106 y, 1 Gy = 109 y September 21, 2009 104 1µs = 10-6 s, 1 ms = 10-3 s,1 My = 106 y, 1 Gy = 109 y September 21, 2009 105 1µs = 10-6 s, 1 ms = 10-3 s,1 My = 106 y, 1 Gy = 109 y The members of this series are not presently found in nature because the halflife of the longest-lived radionuclide in the series is short compared to the age of the earth. Further, this chain does not include an isotope of radon. Reference: http://hyperphysics.phy-astr.gsu.edu/HBASE/hframe.html September 21, 2009 106 The U-238 series is the one that produces radon 222 and if you’ll note from that series, radon is the only element which is gaseous at STP. This is the crux of the problem for as the uranium-238 decays into its decay products, all the solids remain within the earth (or seawater), and the radon, being gaseous, has the mobility to percolate up through the earth and into the atmosphere, or into a house which may be above the percolating radon. If you will consider for a moment the radioactive decay series above, which begins with U-238, we see that one of the radionuclides in that decay chain is radon 222. And when radon 222 decays it does so by giving off an alpha particle, leaving behind a polonium218 atom, which then decays further to At-218 or Pb-214. In this case, it has two routes by which it can decay; the one is by alpha decay, the other by beta decay. The route which is called “alpha decay” is called that simply because the Po-218, or for that matter the Rn-222, emits an alpha particle, which consists of two protons and two neutrons held tightly together. (This alpha particle is also the nucleus of the helium atom.) The route called beta decay is called that simply because the Po-218 can decay when one of its neutrons, within the nucleus, breaks up into a proton and an electron, emitting the electron immediately. So the net result in this case is an electron is shot out from the nucleus. Now this is not one of the electrons that have been orbiting the nucleus, but a new one made up from the neutron that decayed. So to prevent us from confusing that electron from the orbiting ones, that electron is called a “beta” particle. It is identical to all electrons in all regards, but is called a beta particle to remind us of its origin, the nucleus. As a further aside, it is not possible for an electron to exist inside a nucleus—the result of an interesting quantum mechanical phenomenon, so that when it’s created, it must immediately exit the nucleus. The neutron that decayed cannot decay simply into an electron for a host of reasons, so it decays into a proton and that emitted electron. The proton, however, is free to remain in the nucleus, which then increases the number of protons and changes the element itself. Radioactive Decay Laws and Equilibrium Activity: Consider a number of radioactive nuclei of the same isotope, e.g. Rn222 which has a half-life of 3.82 days. Let N(t) represent the number of those nuclei that exist at the time “t.” The equation which describes the number that exist at the time “t” is given by N (t ) = N i exp ( " !t ) Where Ni represents the number of that nuclei that exist initially, at t = 0, and λ is the decay constant, which represents the decay probability for that nuclei, e.g. for Rn-222 its value is 0.181/day. That is, the probability of a given radon atom decaying in one day is 0.181. This decay constant is related to the half-life t1/2 by the equation September 21, 2009 107 != .693 t1 2 The “activity” of a sample of N(t) radioactive atoms can be calculated from Activity = λ N(t). This represents the number of radioactive atoms that decay per second or per day. For example, the activity of 1,000,000 radon-222 atoms is "N = .18 ! 1,000,000 = 180,000atoms / day This can easily be changed to decays per second by converting days to seconds. In this case, the activity becomes about 2 atoms/second. That is, λN , the activity or quantity of a radionuclide, is the number of atoms that are decaying per second at a given time. Now we can consider equilibrium. If we continue with the radon-222 atom as the example, we can recognize that it decays from radon-222 into Po-218, which has a halflife of 3.1 minutes, or a decay constant, λPo, of .0037 atoms/second. So it decays rather quickly, at least in comparison to radon-222 which has a decay constant, λRn, of 0.000002 atoms/second. So what happens if we start with 1,000,000 radon-222 atoms initially and nothing else in our container? As the radon decays, it becomes Po-218 which then decays further. So initially, the number of radon-222 atoms decreases and the number of Po-218 atoms increases. After a while though, because the Po-218 half-life is short compared to that of Rn-222, the number of Po-218 atoms reaches a maximum and then eventually decreases at the same rate that the Rn-222 decays. This does not mean the half-life of Po-218 has changed, it has not. But the number of new atoms of Po-218 being created by the decay of the Rn-222 and the number of Po-218 atoms decaying yield a net effect that the rate at which the number of Po-218 atoms changes is the same rate with which the number of radon-222 atoms decreases. This is called “secular equilibrium” and occurs when the half-life of the decay product is short compared to that of the parent radionuclide. This result, can be written as ! Po N Po " ! Rn N Rn . It takes about four hours for all of the short-lived decay products of radon-222 to come into complete secular equilibrium with an initially pure amount of radon. September 21, 2009 108 Radiation Type Energy content Range in medium Radio Waves 1 millionth eV Visible Light 1 to 3 eV Ultraviolet Light 3 to 10 eV X-rays 10 eV to 120 keV Gamma rays Few keV to 10 MeV Hundreds of miles in air Beta particles Few keV to 1 MeV Dozens of feet in air Alpha particles 4 to 8 MeV A few inches in air Some radiation types with typical energy values and in some cases their range in a given medium. Plate Out Radon atoms are typically electrically neutral, that is they have as many electrons surrounding the nucleus as there are protons within the nucleus. Being neutral they are not attracted to materials in their surroundings, such as furniture, walls, carpeting, etc. But this is not the case with the radon progeny such as Po-218 and Pb-214, for example. These progeny may be electrically charged upon their creation during the radioactive decay process and may be attracted to surfaces found within the rooms where the radon decayed into these radionuclides. Therefore, while the radon floats around in the room quite easily, many of the progeny become attached to furniture, etc. and therefore are removed to a large degree from becoming a health hazard. If there is no plate out then it is generally taken that 100 pCi/l of radon yields 1 working level (WL) of radon progeny in the air. However, with a typical amount of plate out of approximately 50%, then of course one can easily see that it would take 200 pCi/l of radon to produce a concentration of 1 WL of radon progeny in the air. Plate out can be affected by a. Attachment rates b. Ventilation c. Deposition of unattached progeny One can easily picture that if there is a modest amount of air movement within a given environment that the progeny will more frequently find themselves in contact with other surfaces in the room, therefore yielding a greater amount of plate out than if there were little air movement within the room. Thoron Thoron, the common name for Rn-220 is also a gas and is found in the thorium-232 decay series (see above). Its half-life is 55 seconds and it decays via alpha emission producing an array of radon progeny somewhat like Rn-222. The health risk due to Rn220, like that of Rn-222, is due mostly to the alpha emission of its progeny. September 21, 2009 109 Because of its short half-life, it has not been considered to be a significant radiobiological hazard. However, there is some research to indicate occasions where that is not the case. Further some “grab-sample” measurements may be made in such a way that thoron has an impact on the concentrations of radon-222 that are reported. September 21, 2009 110 Radon Units Energy BTU is defined as the amount of energy needed to raise one pound of water held at one atmosphere pressure by one degree Fahrenheit, from 60 °F to 61 °F. Calorie is defined as the amount of energy needed to raise one gram of water from 14.5 °C to 15.5 °C at one standard atmosphere of pressure. Electron volt is defined as the amount of energy by which an electron’s energy increases as it passes through one volt of potential difference. MeV is defined as one million electron volts. Erg is defined as the amount of work done by a force of one dyne exerted for a distance of one centimeter. Joule is defined as the amount of energy expended by a force of 1 newton over a distance of one meter. Disintegration Each time a radioactive atom decays, that is termed one “disintegration.” As an example, radon-222 decays by emitting an alpha particle and becoming polonium-218. Half-life If you begin with a number of radon-222 atoms, e.g. 1,000,000, then after 3.82 days you would have only 500,000 radon-222 atoms left and as you watched the number of radon atoms decay, even though their half life is 3.82 days, some would decay every second you are watching them. Hence, we would have a number of disintegrations per second. The number of disintegrations per second is proportional to the number of radon atoms we have at that given second and hence is a measure of the health risk of the radon atoms. So the unit of dis/sec becomes important. The number of disintegrations/second is also called the activity of the sample. Some other units which are related to dis/sec are: 1 becquerel (Bq) 1 curie (Ci) 1 picocurie (pCi) 1 picocurie/liter (pCi/l) 1 pCi/l September 14, 2009 = 1 dis/sec = 3.7 x 1010 dis/sec = 0.037 dis/sec = 0.037 dis/sec/liter = 37 Bq/m3 111 Exposure 1 roentgen is the amount of photon energy (either x-rays or gamma rays) required to produce 1.610 x 1012 ion pairs in one cubic centimeter of dry air at 0°C and 760 mm Hg. The main advantage of this unit is that it is easy to measure directly with a survey meter, as survey meters function on measuring the number of ion pairs produced as x-rays or gamma rays pass through the detector. The main limitation is that it is valid only for deposition of energy in air. Obviously, it is more important to recognize the amount of energy absorbed by tissue than by air, so the concept of absorbed dose is developed. 1 rad is defined as 100 ergs/gram and this unit does not depend on the time of radiation depositing that energy nor the material into which that energy is absorbed. For example, if 100 ergs of energy is deposited by some alphas and betas into one gram of tissue that tissue is said to have absorbed 1 rad of radiation. “It can be shown that one gram of air will absorb 87 ergs of energy and that one gram of soft tissue will absorb 96 ergs of energy when exposed to a radiation field which produces an exposure of one roentgen. This is true to within two percent for gamma energies from 0.1 MeV to 3 MeV. Thus, for many practical health physics problems, over the range of energies usually encountered, the rad and roentgen are often used interchangeably”1 Another unit of absorbed dose is the gray: 1 gray (Gy) is defined as one joule/kilogram, and 1 Gy = 100 rad. The rem is a unit designed to take into account the effect that different types of radiation have on tissue. For radiation protection purposes it is useful to define a quantity, the dose equivalent, which describes the effect of radiation on tissue. Equal absorbed doses of radiation may not always give rise to equal risks of a given biological effect, since the biological effectiveness may be affected by differences in the type of radiation or irradiation conditions. Thus, the dose equivalent is defined to be the product of the absorbed dose and a modifying factor or factors: Dose Equivalent (rem) = Absorbed Dose (rad) x Quality Factor. Another unit of dose equivalent is the sievert: September 14, 2009 112 1 sievert (Sv) = 100 rem. Dose Equivalent (sievert) = Absorbed Dose (gray) x Quality Factor where the quality factor, the most common modifying factor, takes into account the relative effectiveness of the radiation in producing a biological effect. The special unit of dose equivalent is the rem. Quality Factor The values for quality factor given in the table below are those recommended by the International Commission on Radiological Protection in ICRP Publication 26: Types of Radiation Quality Factor (QF) x or gamma rays 1 beta particles 1 *neutrons and protons of unknown energy 10 singly charged particles of unknown energy with rest mass 10 greater then 1 amu alpha particles 20 particles of multiple or unknown charge of unknown energy 20 The value of the quality factor for each type of radiation depends on the distribution of the absorbed energy in a mass of tissue. For example, the increased effectiveness of neutrons relative to gamma rays is believed to be related to the higher specific ionization of the recoil protons liberated by neutron bombardment as compared to the specific ionization of the secondary electrons arising from gamma-ray irradiation. The values of quality factor are known to vary with the biological effect being observed, and in some cases are still a matter of controversy for the same biological effect. (See http://web.princeton.edu/sites/ehs/radsafeguide/rsg_app_e.htm) Working Level Radon decay product concentrations are typically measured in the unit of working level (WL). 1 working level (WL) is defined as any combination of short-lived radon decay products in one liter of air that will result in the ultimate emission of 1.3 x 105 MeV of potential alpha energy. The historical more definition of the working level is the potential alpha energy of radon decay products in equilibrium with 100 pCi/L of radon. This value is 128,000 MeV. An excellent reference for this September 14, 2009 113 definition is Evans, R.D. Engineers’ guide to the elementary behavior of radon daughters. Health Physics 17(2):229-252, August 1969. The number 130,000 MeV was chosen for its simplicity and because it is approximately the alpha energy released from the decay products in equilibrium with 100 pCi of Radon-222 per liter of air assuming an equilibrium ratio of 1.0 Working Level Month (WLM): An exposure to a concentration of radon progeny of 1 WL for one working month. Working Month: For purposes of calculating working level months, a month is 170 working hours. For example, a worker exposed to a radon progeny concentration of 0.5 WL for a period of 170 hours would have been exposed to 0.5 WL x 1 month = 0.5 working level months. The calculation is simply # wlms =# workinglevels!# months Worker exposures are measured in working level months (WLM). (See http://www.radon.com/pubs/homprot7.html) Number of Digits Radon concentrations are typically reported in pCi/l to one digit to the right of the decimal. For example: 3.7 pCi/l, 0.3pCi/l. Radon progeny concentrations are typically reported in WL to three digits to the right of the decimal. For example: .002 WL, 1.222 WL. Equilibrium Ratio The equilibrium ratio is simply defined as the ratio of the collective concentration of all the short-lived radon progeny to the concentration of the parent radon gas and is defined analytically by the following equation pCi / l wl radongasactivity ( pCi / l ) Daughteractivity ( wl ) ! 100 ER = For example: Given the radon progeny concentration to be 0.020 WL and the parent activity to be 25 pCi/l, the ER becomes ER = 0.020 ! 100 25 ER = 0.08 September 14, 2009 114 Generally it is not known what the equilibrium ratio is in a given environment. Current (2009) recommendations suggest using an ER of 0.4 to convert from WL to pCi/l as follows: WL = ER ! radonconc( pCi / l ) pCi / l 100 wl For example, if the equilibrium ratio is 0.4 and the radon concentration is 20 pCi/l, the radon progeny concentration is calculated as: WL = 0.4 ! 20 wl 100 WL = 0.080 wl Clearly, the range of values of the Equilibrium Ratio falls between zero and 1.0. The minimum value of zero would indicate there are no decay products in the environment at that time relative to the parent gas. An Equilibrium Ratio value of 1.0 indicates that the concentration of each of the radon progeny is equal to the concentration of the parent radon-222. If the radon concentration is 100 pCi/l, then the concentration of each of the decay products is also 100 pCi/l, and the collective concentration of the radon progeny is 1 WL. Another equivalent view of the ER is sometimes found as: Equilibrium Ratio, radioactive: The total concentration of radon decay products (RDPs) present divided by the concentration that would exist if the RDPs were in radioactive equilibrium with the radon gas concentration that is present. (This is not the common use of the term within the radon industry.) At equilibrium (i.e., at an equilibrium ratio of 1.0), 1 WL of RDPs is present when the radon concentration is 100 pCi/l. Although the equilibrium ratio can temporarily be greater than 1.0 as the radon progeny concentrations lag behind a decrease radon concentration, on the average the ratio is never 1.0 in a house. Due to ventilation and plate out, the RDPs never reach equilibrium in a house environment if the radon concentration is stable. A commonly assumed equilibrium ratio is 0.4 (i.e., the progeny are 40% toward equilibrium), in which case 1 WL corresponds to 250 pCi/l. However, equilibrium ratios vary with time and location, and ratios of 0.3 to 0.7 are commonly observed. Large buildings, including schools, often have equilibrium ratios less than 0.5. Some factors affecting the equilibrium ratio in a given environment are: a. Attachment rate b. Deposition of unattached progeny c. Ventilation September 14, 2009 115 Clearly the greater the amount of “plate out,” the fewer progeny will be in the air and therefore the lower the equilibrium ratio. Equilibrium ratios for outdoors have been reported in the vicinity of 0.6. Equilibrium Factor • Equilibrium factor (F): The ratio of the radon progeny concentration in WL actually present to the radon progeny concentration which would be present if the short-lived progeny were in equilibrium with the radon that is present. This quantity is the same as the equilibrium ratio. Equilibrium Equivalent Concentration Equilibrium equivalent concentration (EEC): The concentration of radon that would be present if the radon progeny that are present were in complete equilibrium with the radon that is present (i.e., if the ER were 1.0). The ratio of the EEC to the actual radon concentration is equal to the equilibrium ratio. Analytically the EEC may be calculated by either EEC = WLactual ! 100 or EEC = Rnactual ( pCi / l ) ! ER Typically, we don’t know the ER, so we might take it as 0.4 which yields EEC = Rnactual ( pCi / l ) ! 0.4 For example, assume we know the radon progeny concentration in a room is 1.0 WL. What radon concentration would produce that concentration of radon progeny, assuming no plate out? The answer is of course 100 pCi/l and we can find that using the equation above: EEC = WLactual ! 100 pCi / l wl EEC = 1.000 ! 100 = 100 pCi / l As a second example: Assume we know that the actual radon concentration is 100 pCi/l, and the equilibrium ratio is 0.4. From our equation for equilibrium ratio above September 14, 2009 116 WL = ER ! radonconc( pCi / l ) pCi / l 100 wl we find that the radon progeny concentration is WL = 0.4 ! 100 = 0.4 wl 100 Therefore, we know without plate out, we would need 40 pCi/l, since without plate out 100 pCi/l yields a radon progeny concentration of 1 working level. We can calculate that radon concentration by using the equation above which reads EEC = Rnactual ( pCi / l ) ! ER EEC = 100 ! 0.4 = 40 pCi / l If one knows the concentrations of the individual radon-222 decay products, then the EEC can be calculated from the following expression: EEC = 0.105 x C(218Po) +0.516 x C(214Pb) + 0.379 x C(214Bi) where EEC is the equilibrium equivalent radon concentration in pCi/l or Bq/m3 (depending on the unit used for the decay product concentrations) and C(218Po), C(214Pb) and C(214Bi) are the concentrations of 218Po, 214Pb and 214Bi in pCi/l or Bq/ m3. Note that the concentration of 214Po is not included in this equation, as it does not contribute significantly to the expression. PAE Potential Alpha Energy. This quantity is the total alpha energy of an atom of one of the short-lived decay products of radon-222 as it decays to the long-lived decay product 210Pb. For example, if we have one atom of 218Po, as it decays to 210Pb, the total alpha energy emitted is 13.68 MeV (6.0 MeV from the decay of 218Po + 7.68 MeV from the subsequent decay of 214Po). If we also have at that moment one atom of 214Bi, as it decays to 210Pb, the total alpha energy emitted is 7.68 MeV (from the subsequent decay of 214Po). So, if we have in a given room at a given moment one 218Po atom and one 214Bi atom, the total potential alpha energy emitted as these two atoms decay to 210Pb is the sum of 13.68 MeV and 7.68 MeV, or 21.36 MeV. This total alpha energy is useful in determining health risks. (See Introduction to Radiation Protection dosimetry, Sabol, Weng, Weng.) September 14, 2009 117 PAEC Potential Alpha Energy Concentration, used as a measure of the decay-product concentration. Units used are Working Level, or joules/m3. This is the sum of all the PAEs of the radon progeny: 218Po, 214Pb, 214Bi and 214Po per unit volume of air. (See Introduction to Radiation Protection dosimetry, Sabol, Weng, Weng.) If one knows the concentrations of the individual radon-222 decay products, then the PAEC can be calculated from the following expressions: PAEC (µJ/m3) = 0.000578 x C(218Po) +0.00285 x C(214Pb) + 0.000210 x C(214Bi) where C(218Po), C(214Pb) and C(214Bi) are the concentrations of 218Po, 214Pb and 214 Bi in Bq/ m3, PAEC (WL) = 0.00103 x C(218Po) +0.00507 x C(214Pb) + 0.00373 x C(214Bi) where C(218Po), C(214Pb) and C(214Bi) are the concentrations of 218Po, 214Pb and 214 Bi in pCi/l. September 14, 2009 118 Derivation of the Relationship of the Concentration of Radon Needed for Its Decay Products to Produce 1 WL Before we begin this derivation an issue of “units” arises. One often finds in the literature the statement that 100 pCi/l = 1 WL. If we take a careful look at the left-hand side of this equation, pCi is 2.22 disintegrations/min which has units of min-1 since the term disintegration is in fact not a “unit,” just like atom or nucleus are also not “units,” rather just a handle by which it is easier to talk about. For example, if one asks how many atoms there are in a jar, the answer might be 1000 atoms, but the term atom here is not a unit, the answer to that question is most correctly simply “the number is 1000.” But, adding the word atom makes it a bit more clear. So returning to the equation 100 pCi/l = 1 WL, we have on the left side of the equation the unit min-1 l-1, i.e. 1/(min l). On the right we have the “WL” which has units of MeV/l, or energy/l. So, the right side of the equation does not have the same units as the left side leading us to recognize that this is not an equation at all. In particular, if we reduce this equation just a bit we have 100 ! 2.22 1 1.3 ! 105 MeV = 1! min• l l which can be reduced to: 222 = 1.3 ! 105 MeV min This makes very little sense. It would be better to write 100pCi/l →1WL, but this has not become the common use, which is unfortunate. We shall use the “→” as it is a more correct notation. For the case where radon decay products are in equilibrium with radon, it is typically taken that 100 pCi ! 1WL l Its derivation proceeds as follows: Recognize that the definition of 1 working level is any combination of radon or thoron decay products which will potentially produce 1.30 x 105 MeV (more accurately 128,400 MeV) of alpha particles in one liter of air. For the case of radon decay products this is September 14, 2009 119 1.30 x 105 MeV of alphas emitted by radon decay products when they’re in equilibrium with approximately 100 pCi/l of radon which we show below. The decay scheme of radon 222 is as follows: 222 Rn T1/2=5506 m 218 Po T1/2=3.05 m α 6.00MeV 0.02% 214 218 Pb T1/2=26.8 m At T1/2=.022 m α 6.7 MeV 214 Bi 19.7 m α 5.5MeV; .021% 214 210 Po 2.7x10-6 m Tl α 7.68MeV 210 Pb T1/2 = 2.3 y The pertinent alpha emissions and their approximate alpha energies are noted in the drawing and in some cases along with the percentage of that decay branch. September 14, 2009 120 Let’s consider a volume of 1 liter and an activity of 100 pCi for radon with its decay products in full equilibrium for the purpose of this derivation. Therefore, at equilibrium the “activity” of each decay product in this 1 liter of air is equal to that of the radon itself, 100 pCi. In this case, we can calculate the number of atoms of each decay product in that 1-liter volume at that moment of equilibrium as follows: That is, the number of 218Po atoms is ! N Po N Po = Po ! Po Which seems trivially true and where N is the number of atoms and λ is the decay .693 constant ( ! = ) and where t1/2 is the half-life. t1 2 We now recognize that the activity, ! Po N Po is 100 pCi or 220 dis/min. So N Po = 220dis / min = 978atoms ! 1 .227 min That is we have initially 978 polonium-218 atoms. Similarly, we can easily show that we also have 8.49 x103 214Pb atoms, and 6.25 x 103 214 Bi atoms. We won’t take into account here the 218At radionuclide since there are so few of them at 100 pCi (6.9 atoms). Also not entirely surprising the number of 214Po atoms at 100 pCi is less than 1, so we’ll take it as zero. Further, we don’t need to look down into the decay chain beyond the 210Pb since its half-life is comparatively long. These omissions will have only the slightest impact on the accuracy of our results and the additional detail serves only to detract from the clarity of this derivation. Next let’s consider these 978 218Po atoms which we have initially in this one-liter volume. Each one of these 218Po atoms will produce one alpha particle of 6.00 MeV as it decays into 214Pb, and then later another alpha of 7.68 MeV as a resulting 214Po atom decays. So each 218Po atom is responsible for two alpha particles, totaling 13.68 MeV, approximately. Therefore, the entire 978 218Po atoms yield a total alpha energy of September 14, 2009 121 978 ! 12.68MeV = 1.34 ! 104 MeV Now we can perform the same calculation for the 214Pb atoms as follows: The total number of 214Pb atoms is found as before from: ! N Pb N Pb = Pb ! Pb N Pb = 220dis / min = 8.49 !103 atoms " 1 .0259 min Each one of these 214Pb atoms is responsible for one alpha particle of 7.68 MeV when the eventual 214Po atom decays. Hence the total alpha energy released by these 8.49 x 103 atoms is 8.49 ! 10 3 ! 7.68MeV = 6.52 ! 10 4 MeV Performing a similar calculation for 214Bi yields 6.25 x 103 atoms yielding 4.80 x 104 MeV. So the total alpha energy eventually released by these atoms in equilibrium with the 100 pCi of radon is 1.34 x 104 MeV + 6.52 x 104 MeV + 4.80 x 104 MeV = 1.27 x 105 MeV ≈ 1.30 x 105 MeV. If we compare this calculated value to that of the more correct value of 1.28 x 105 MeV value of 1 WL, then the result presented here differs by less than 1% The discrepancy between the above calculated value and the more commonly reported value is attributed to the approximations used in this derivation. September 14, 2009 122 Derivation of the Relationship of the Concentration of Thoron Needed for Its Decay Products to Produce 1 WL Before we begin this derivation an issue of “units” arises. One often finds in the literature the statement that 100 pCi/l = 1 WL. If we take a careful look at the left-hand side of this equation, pCi is 2.22 disintegrations/min which has units of min-1 since the term disintegration is in fact not a “unit,” just like atom or nucleus are also not “units,” rather just a handle by which it is easier to talk about. For example, if one asks how many atoms there are in a jar, the answer might be 1000 atoms, but the term atom here is not a unit, the answer to that question is most correctly simply “the number is 1000.” But, adding the word atom makes it a bit more clear. So returning to the equation 100 pCi/l = 1 WL, we have on the left side of the equation the unit min-1 l-1, i.e. 1/(min l). On the right, we have the “WL” which has units of MeV/l, or energy/l. So, the right side of the equation does not have the same units as the left side leading us to recognize that this is not an equation at all. In particular, if we reduce this equation just a bit we have 100 ! 2.22 1 1.3 ! 105 MeV = 1! min• l l which can be reduced to: 222 = 1.3 ! 105 MeV min This makes very little sense. It would be better to write 100 pCi/l →1WL, but this has not become the common use, which is unfortunate. (See the definition of WL elsewhere in this manual.) We shall use the ‘→’ as it is the correct notation. It is interesting to calculate how many pCi/l of thoron (220Rn) are needed for its decay products to produce 130,000 MeV of alpha energy. Recognize that the definition of 1 working level is any combination of radon or thoron decay products which will potentially produce 1.30 x 105 MeV (more accurately 128,400 MeV) of alpha particles in one liter of air. For the case of thoron decay products this is 1.30 x 105 MeV of alphas emitted by thoron decay products when they’re in equilibrium with thoron. (Editor’s Note: Thoron progeny cannot come into equilibrium with the parent thoron, because the half-life of thoron is short compared to that of its decay products, namely that of Pb-212 with a half-life of 10.64 h. However, the discussion here is correct if one considers that the thoron and its progeny come into equilibrium with the source of the thoron, Ra-224. Further, because of the short half-lives of thoron and its immediate decay product, Po216, the concentrations of these two radionuclides are typically significantly reduced at September 14, 2009 123 some sampling point at a practical distance from the source of the thoron. This still does not affect the validity of the discussion here, if one recognizes that the contribution of Po216 to the PAEC is insignificant.) The concentration of thoron needed for its decay products to produce 1 WL may be calculated as follows: The decay scheme of radon-220 is as follows: September 14, 2009 124 220 Rn T1/2 = 55.6 s 216 Po T1/2 = 0.150 s α 6.78MeV 212 Pb T1/2 = 638 m 212 Bi T1/2 = 60.7 m α 6.05 MeV 35.9% 64.1% 212 208 Po T1/2 = 300 ns Tl T1/2 = 3.07 m α 8.78 MeV 208 Pb Stable The pertinent alpha emissions and their approximate alpha energies are noted in the drawing and in some cases along with the percentage of that decay branch. September 14, 2009 125 We will approach this case slightly differently than the radon case. Here we will calculate the total alpha energy emitted by thoron decay products each with a concentration of 1 pCi in one liter of air. This will provide us with a number of MeV. Then we can easily calculate the number of pCi of thoron decay products needed to produce 1.30 x 105 MeV in that one liter of air. It is important to recognize that the decay products of thoron will be in equilibrium with the thoron in this discussion. Therefore 1 pCi of thoron means we have 1 pCi of each of the thoron decay products. So we will calculate the number of atoms of each thoron decay product assuming there is 1 pCi of each of them. Again, we need to remember we are doing this for one liter of air. Beginning with 216Po, we will use the equation: N Po = !Po N Po !Po Which is trivially obvious and where N is the number of atoms and λ is the decay .693 constant ( ! = ) and where t1/2 is the half-life. t1 2 Recognizing now that the numerator on the right is the activity of the 216Po which is 1 pCi (2.22 dis/min), and that the decay constant of 216Po is 2.77 m-1, we find N Po = 2.22dis / min = 0.00801 atoms !1 277m We proceed this way to calculate the number of atoms (or nuclei) of each of the other decay products, using their decay constants of 0.00109 m-1 for 212Pb, 0.0114 m-1 for 212Bi, and 1.39 x 108 m-1 for 212Po and find N212Pb= 2040 atoms N216Po= 0.00801 atoms N212Bi= 195 atoms N212Po= 1.60 x 10-8 atoms By looking at the decay chain above, we see for each 216Po atom that exists initially, there will be produced one alpha of 6.78 MeV, 0.359 alphas of 6.05 MeV as the 212Bi decays 35.9% of the time to 208Tl and .641 alphas of 8.78 MeV. (Please note that 64.1% of the September 14, 2009 126 time the 212Bi will decay into 212Po) as the 212Po decays into stable lead. So the total alpha energy emitted in this case by the 0.00801 atoms of 216Po atoms will become 8.01x 10-3(6.78+.641 x 8.78 +.359 x 6.05) MeV = .117 MeV For each atom of 212Pb that exists initially, there will be produced .641 alphas of 8.78 MeV, and .359 alphas of 6.05 MeV. This yields for the 2040 212Pb atoms a total alpha energy of 2040 x (.359 x 6.05 + .641 x 8.78) MeV = 1.59 x 104 MeV. For each atom of 212Bi that exists initially, there will be produced 0.359 alphas of 6.05 MeV and .641 alphas of 8.78MeV. This yields for the 212195 Bi atoms a total alpha energy of 195 x (.359 x 6.05 + .641 x 8.78) MeV = 1.52 x 103 MeV. For each atom of 212Po we will find one alpha of 8.78 MeV but this will still yield zero MeV since there are effectively no 212Po atoms in this case. So, for the entire decay chain for thoron, we see that for 1 pCi of each decay product we will get a total alpha energy of Total alpha energy = .117 MeV + 1.59 x 104 MeV +1.52 x 103 MeV = 1.74 x 104 MeV. We now have that 1 pCi of each of the thoron decay products causes 1.74 x 104 MeV of alphas to be emitted. To accumulate 130,000 MeV of alpha energy (1 WL) we would then need 130,000 Mev = 7.47 pCi 1.74 x104 Mev / pCi If we use the more correct value of 1WL, 128,400 MeV, then we would find 7.38 pCi ! 1WL l It is quite common to see the figure quoted as 7.43 pCi/l and the discrepancy between the above calculated value and the more commonly reported value is attributed to the approximations used in this derivation. If we compare this calculated value with the more commonly quoted value of 7.43 pCi/l, we find the difference to be less than 1%. September 14, 2009 127 It’s of some interest to note that it takes approximately 100 pCi/l of radon to produce 1 WL of decay products but only approximately 7.43 pCi/l of thoron to produce 1 WL of its decay products or about 13 times more radon activity than thoron activity. This is easily explained by comparing the number of atoms of each decay chain for the case of 1 pCi/l of radon and thoron as shown in this table: Total number of atoms which emit alpha particles for 1 pCi/l of radon or thoron for the case where the decay products are in secular equilibrium. Radon Po218------------10 Pb214------------85 Bi214------------62 Thoron Pb212-------------2040 Po216--------------0 Bi212--------------195 Po212--------------0 Total------------2235 Total------------157 There are more atoms of the decay products for thoron than radon by a factor of almost 14 so 1 pCi/l of thoron decay products yield about 14 times more alpha energy than the decay products of 1 pCi/l of radon. So approximately 14 times less thoron activity is needed to provide the same amount of alpha energy. Further this is due to the fact that the equation for the number of atoms existing for a given activity “λN” which has been shown above to be N = !N ! Is easily rearranged by recognizing that != .693 t1 2 which yields this expression for N: "N ! t 1 N = September 14, 2009 2 .693 128 And taking as has been done for a given activity of 1 pCi/l, N becomes 1! t 1 N= 2 .693 which shows that the number of atoms of a given decay product is proportional to its half-life for a given activity which makes sense in that the shorter the half-life the more quickly the atoms decay and hence the more quickly it produces an alpha particle with its attending energy. The difference in the half-lives of the decay products between the two decay chains is what gives rise to the difference in the number of atoms, and therefore in the amount of alpha energy given off by these decay products. September 14, 2009 129 Radon-222 Dose Calculation Initial Assumptions: Equilibrium Factor (EF): 0.4 BEIR VI Dose Conversion Coefficient 0.54 rad/WLM UNSCEAR 2006 Occupancy Factor for home: 0.7, or 6136 hr/yr, BEIR VI Radiation Weighting Factor (Wr) for Alphas: 20 2005 ICRP Recommendations Tissue Weighting Factor (Wf), Lung: 0.12 Bronchial Tree: 0.08 2005 ICRP Recommendations (NCRP/ICRP) UNSCEAR 2006 Equations: WL = (Rn x EF)/ 100 WLM = (WL x Exposure in hrs)/ 170 working hours per month Now for a problem. 1. 4 pCi/L in home environment (4 pCi/L x 0.4)/ 100 = 0.016 WL; WLM = (0.016 WL x 6136 hr/yr) / 170 hrs/month = 0.58 WLM/yr Now convert to dose equivalent: (WLM/y) (DCC) (Wr) (Wf) (0.58 WLM/yr) (0.54 rad/WLM) (20) (0.08) = 0.5 rem/yr, effective whole body dose equivalent References: United Nations Scientific Committee on the Effects of Atomic Radiation, Volume I: Sources, 2006. National Research Council. Health Effects of Exposure to Radon, BEIR VI. National Academy Press, 1999. September 14, 2009 130 Dosimetry Section Dosimetry is the calculation of the absorbed dose in matter and tissue resulting from the exposure to indirectly and directly ionizing radiation. The dose to matter is measured in gray (Gy) and the dose to biological tissue in sieverts (Sv), where 1 Gy or 1 Sv is equal to 1 joule per kilogram. The United States does not use the SI units for dose and still uses rad and rem, where 1 Gy = 100 rad, and 1 Sv = 100 rem. It should be pointed out that we do not measure the dose to the lung tissue, it must be calculated. Additionally, the vast majority of the dose (energy) deposited in the lung tissue is due to the radon progeny and not the radon gas. The majority of the lung tumors are found in the tracheobronchial region of the lung. The cells at risk in this region of the lung are the basal and secretory cells. We are exposed to radon gas and to radon progeny, and these are measured in pCi/L and working levels; however, it is the decay products that result in the majority of the dose being delivered to the lung tissue. In calculations of dose, we therefore need a dose conversion factor to convert a cumulative exposure in working level months to rads of absorbed dose in the lung. This dose conversion factor as found in UNSCEAR 2006 is 0.54 rad WLM-1 or 9 nGy (Bq h m-3)-1. An example dose calculation follows: Radon-222 Dose Calculation Initial Assumptions: Equilibrium Factor (EF): 0.4 UNSCEAR 2006 Dose Conversion Coefficient 0.54 rad/WLM UNSCEAR 2006 Occupancy Factor for home: 0.7, or 6136 hr/yr, BEIR VI Radiation Weighting Factor (Wr) for Alphas: 20 2005 ICRP Recommendations Tissue Weighting Factor (Wf), Lung: 0.12 Bronchial Tree: 0.08 2005 ICRP Recommendations UNSCEAR 2006 September 14, 2009 131 Equations: WL = (Rn x EF)/ 100 WLM = (WL x Exposure in hrs)/ 170 working hours per month An example problem shows: 2. 4 pCi/L in home environment (4 pCi/L x 0.4)/ 100 = 0.016 WL; WLM = (0.016 WL x 6136 hr/yr) / 170 hrs/month = 0.58 WLM/yr Now convert to dose equivalent: (WLM/y) (DCC) (Wr) (Wf) (0.58 WLM/yr) (0.54 rad/WLM) (20) (0.08) = 0.5 rem/yr, effective whole body dose equivalent References: United Nations Scientific Committee on the Effects of Atomic Radiation, Volume I: Sources, 2006. National Research Council. Health Effects of Exposure to Radon, BEIR VI. National Academy Press, 1999. September 14, 2009 132 Effective Whole Body Dose Equivalent From Exposure to Radon and Radon Progeny in Air (July 2007) pCi/L WLM/year Rem/year Lungs Rem/year Annual Whole Body Whole Body Dose Comparison (EDE) 0.15 0.02 0.23 0.03 Single view Mammogram (10); Cosmic Radioactivity At Earth’s Surface (8). 2 0.29 3.12 0.37 Near twice avg. dose for Nuclear Power Plant Worker (6). 4 0.58 6.24 0.75 About 900 Medical X-Rays (7). 8 1.16 12.47 1.50 Between one to two times the dose from a whole body CT Scan. (8). 10 1.44 15.59 1.87 20 2.89 31.19 3.74 50 7.22 77.96 9.36 100 14.44 155.93 18.71 200 28.88 311.85 37.42 Near Limit of Annual Occ. Exp.– 5.0 Rem (9). WL = ((pCi/l)(ER))/100. WLM = ((WL)(hrs))/170hrs/month. WLM/yr = (WLM)(6136hr/yr). EDE = (WLM/y)(DCC)(Wr)(Wf) September 14, 2009 133 Assumptions/references: 1. Occupancy: 70% of time (16.8 hours/day). Ref. BEIR VI Report (1999). 2. Equilibrium Ratio (ER) :0.4, Ref. EPA 402-R-03-003, Assessment of Risks from Radon in Homes (6/2003). 3. Dose Conversion Coefficient (DCC) 0.54 rad/WLM. Ref. UNSCEAR 2000 Report Vol. 1. 4. Weighting factor (Wf) Lung Dose to Whole Body (EDE) = 0.12 5. Radiation Weighting Factor (Wr) for alpha particles = 20 rem/rad. 6. Occupational Radiation Exposure at Commercial Nuclear Power Reactors and Other Facilities 2005, NUREG 0713, Vol. 27 section 4.4. 7. CRCPD Pub. E-05-2 Nationwide Evaluation of X-ray Trends, Tabular and Graphical Summary of 2001, Survey of Adult Chest Radiography. 8. Radiological Society of North America, Inc. 2007. 9. 10 CFR Part 20 Radiation Protection, 1201. 10. CRCPD Pub. E-03-2 Patient Exposure and Dose Guide (2003). September 14, 2009 134 Risks from Indoor Radon-US EPA The EPA’s revised risk assessment (EPA, 2003) is based on the National Academy of Sciences, BEIR VI report, 1999. This short presentation provides some of the most relevant facts from that assessment. • • • • The risk estimates are still based solely on the miner data, 11 cohort studies. The estimates in BEIR VI are based on an average annual exposure of 0.181 WLM/yr, and this is based on the EPA National Residential Radon Survey value of 1.25 pCi/L for the U.S. housing stock, assuming 70% occupancy and 40% Equilibrium ratio (ER). Example Calc.: ((1.25 pCi/L) (0.4))/ 100 = 0.005 WL; ((0.005 WL) (6136 hrs/yr))/ 170 working hrs/ mo. = 0.181 WLM/yr ER Occupancy Time Avg. Progeny Exp. Risk Estimate Mortality Data Model used Miner Data K-factor Rn Induced LC deaths 1992 Assessment 2003 Assessment 0.5 75% 0.242 WLM/yr 2.24E-4/WLM 1980 data BEIR IV relative risk model 4 cohorts 0.7 13,600 0.4 70% 0.181 WLM/yr 5.38E-4/WLM 1990 data BEIR VI Age Conc. Relative risk model 11 cohorts 1.0 21,100 Note: Changing the exposure from 0.242 to 0.181 has little effect on risk per WLM. References: EPA Assessment of Risks from Radon in Homes. EPA 402-R-03-003, June 2003. United States Environmental Protection Agency. Pawel, D.J. and Puskin, J.S. The U.S. Environmental Protection Agency’s Assessment of Risks from Indoor Radon. Health Physics, July 2004, Volume 87, No. 1, pp. 68-74. September 14, 2009 135 Radon-222 and Lung Cancer (LC) Risk Estimates Below is a compilation of various sources for risk estimates and other useful information for radon-induced LC. UNSCEAR 2006 Report, Vol. II. Effect of Ionizing Radiation. United Nations Scientific Committee on the Effects of Atomic Radiation, Annex E- Sources-to-effects Assessment for radon in Homes and Workplaces. • • • • • • • Though dated 2006 this report was not officially released until July 2009. The report is a review and compilation of all of the most significant work on sources of radon exposure, dosimetry, experimental studies, epidemiological studies of miners, epidemiological studies of residential exposures, effects of radon on organ and tissues other than the lung, and implications of risk assessment. The excess relative risk (ERR) per unit exposure from the combined (9) miners studies is 0.59 per 100 WLM, with 95% CI of 0.35 to 1.0. At this time adopted the ERR from the pooled European study as an appropriate, though possibly conservative, estimate of lifetime risk at 0.16 per 100 Bq/m-3, with a 95% CI of 0.05 to 0.31. In other words the risk is about 16% for smokers and nonsmokers. It is the baseline risk that is significantly higher for smokers. The Committee now believes that risks from residential radon exposure can be directly estimated from the pooled residential case-control studies. Found that the excess relative risk of lung cancer from residential radon exposure is about the same for smokers and nonsmokers. This is consistent with the European and North American pooled studies. The European, North American, and Chinese pooled case-control residential studies, overall provide for a clear association between the risk of lung cancer and residential radon exposure. Radon in homes and risk of lung cancer: collaborative analysis of individual data from 13 European case-control studies. S Darby et al., December 21, 2004, British Medical Journal. • • • • Absolute risk of LC by age 75 at “usual” radon concentration of 0, 2.7, 10.8 and 21.6 pCi/L would be about 0.4%, 0.5%, 0.7%, and 0.93% respectively, for lifelong nonsmokers, and 10%, 12%, 16%, and 21.6% for smokers. Relevant exposure for risk of LC was 30 years ending 5 years prior to diagnosis of LC. Risk of LC increased by 8.4% per 2.7 pCi/L. This corresponds to an increase of 16% per 2.7 pCi/L in “usual” radon, or an excess relative risk of 1.16. Radon poses a much greater absolute hazard to cigarette smokers, and to recent ex-smokers, than to lifelong nonsmokers. Note: “Usual” radon is radon corrected for the dilution caused by random uncertainties in measuring radon concentrations. September 14, 2009 136 Comparative dosimetry of BEIR VI revisited. A.C. James, A. Birchall, and G. Akabani. Radiation Protection Dosimetry 108:3-26, 2004. • • • • • This paper confirms the BEIR VI Committee’s choice of K = 1 for application in their risk extrapolation model. K is a dimensionless factor that relates the risk to miners per unit exposure to that for an individual exposed in the home. Evidence suggests that most cancers are of monoclonal origin; that is, they originate from damage to a single cell. Sensitive targets are assumed to be the nuclei and cytoplasm of the basal cells and the secretory cells in the bronchioles, both located in the epithelial lining of the bronchial tree. Dose Conversion Coefficient (dose/WLM) averaged over all target nuclei in the lungs (bronchial, bronchiolar, and alveolar-interstitial regions) for the home environment, without smokers is 0.89 rad/WLM, and for smokers is 0.75 rad/WLM. EPA Assessment of Risks from Radon in Homes. D. Pawel and J. Puskin. Office of Radiation and Indoor Air, US EPA. June 2003. • • • • • • • • • • EPA risk assessment update based primarily on BEIR VI. Latest risk per unit exposure is 5.38E-4 fatal LC’s per WLM. This new risk estimate increases the estimated proportion of LC deaths from 8.5% (1992) to 13.4%. Previous risk per unit exposure was 2.24E-4 per WLM (EPA 1992) Of the 157,400 LC deaths in 1995, 21,100 (13.4%) were radon related. Estimated risk from lifetime exposure at 4 pCi/L: 2.3% for entire population, 4.1% for ES, and 0.73% for NS. Radon exposure accounts for 1 in 8 ES LC deaths and 1 in 4 NS LC deaths. Thus, the relative risk is higher in NS but the absolute risk is higher in ES. For exposure calculations use 70% occupancy factor and 0.4 equilibrium factor. Radon induced LC deaths tend to occur earlier than other LC deaths. The average radon induced LC death occurs at ~ 65 y compared to 72 y for all LC deaths. There are uncertainties in the estimates of risks from indoor radon; in fact, the BEIR VI committee identified 13 sources of uncertainty. September 14, 2009 137 Residential Radon and Risk of Lung Cancer, A Combined Analysis of 7 North American Case-Control Studies. Krewski, D, Lubin, J, et al. Epidemiology, Volume 16, Number 2, March 2005. • • • • To date 20 case-control studies of residential radon exposure and LC have been completed. The case-control studies to date cannot provide a definitive link between residential radon exposure and an increased risk of LC. Their results reflect a range of LC risks, including the possibility of no risk. The weight of evidence for radon carcinogenicity derives largely from underground miner studies. This study focused on the exposure time window of 5 to 30 years prior to diagnosis of LC. There was an 11% increase in LC at 2.7 pCi/L. Health Effects of Exposure to Radon, BEIR VI. Samet, J et al. National Academy of Sciences, 1999. • • • • • • • Based on 11 major studies of underground miners, which involved about 68,000 men, of whom 2,700 have died of LC. Most miners received radon exposures that were, on the average many times larger than those of people in most homes. BEIR VI central estimates are about 15,400 or 21,800 LC deaths per year attributed to radon among ever smokers and never smokers, depending on which model is used. Most radon-related deaths among smokers would not have occurred if the victims had not smoked. At low radon exposures, typical of those in homes, a lung epithelial cell would rarely be traversed by more than one alpha particle per human lifespan. Even allowing for a substantial degree of repair, the passage of a single alpha particle has the potential to cause irreparable damage in those cells that are not killed. The analysis of smoking and radon indicated a synergistic effect of the two exposures acting together, which was characterized as submultiplicative. Health Risks of Radon and other Internally Deposited Alpha-Emitters, BEIR IV. Fabrikant, J. I et al. National Academy of Sciences, 1988. • • • The Committee used a direct epidemiological approach instead of the dosimetric approach and used the data from four of the principle studies of radon-exposed miners. The model used was the modified relative-risk model. In this model radon exposures more distant in time have a smaller impact on the age-specific relative risk than more recent exposures. Cigarette smoking and exposure to radon progeny interact multiplicatively. September 14, 2009 138 • Estimate of lifetime risk of LC mortality due to lifetime exposure to radon progeny: 350 deaths/106 person WLM. Evaluation of Occupational and Environmental Exposure to Radon and Radon Daughters in the United States. Harley, N et al. NCRP Report no. 78, 1984. • • • • • • Following a latent period, the tumor rate is an exponentially decreasing function of time since exposure. Disease rate excess associated with a single exposure increases with age at exposure. Lung cancer is rare before the age of 40 years. Median age at LC among miners is about 60 yr in nonsmokers and 50 yr or older in smokers. The minimal time for tumor growth, from initial cell transformation to clinical detection, is 5 years. Derives a lifetime risk of LC of about 1.5 E-4 per WLM September 14, 2009 139 Epidemiology Section Epidemiology is the study of factors affecting the health and illness of populations. Epidemiology is the study of patterns of disease in human populations. The primary source of data for studying exposure to radon and its decay products and lung cancer has been epidemiologic studies of underground miners. It was these studies that were used to provide the downward extrapolation to provide the risk estimates for the residential exposures, even though the miners were exposed to higher exposures for shorter times. The mean exposure in the miner cohort was 164 WLM (NRC, 1999). A typical residential exposure is about 13 WLM (70 yrs, 1.25 pCi/L, 0.4 ER, 70% time indoors). However, it should be pointed out that there is an overlap of total cumulative exposure between the miner and residential exposures. Some of the lower exposures (~50 WLM) in the miners were similar to some of the higher exposures in homes. For instance, a homeowner exposed to 4 pCi/L for a 70 year lifetime accumulates ~41 WLM. The data for the miners consisted of 11 cohort studies. Lubin et al. (1994) performed an analysis of the combined data from the 11 cohorts and found that there was conclusive evidence that exposure to high levels of radon is associated with increased risk of lung cancer. The BEIR VI committee (NRC, 1999) then updated the miner studies by reviewing the current molecular and radiobiological basis of radiation effects on cells, examined the exposure differences in mines and homes (the K factor), reviewed the latest information on radon concentrations in U.S. homes and analyzed the properties of alpha particles and cellular interactions. With all this in mind, the committee concluded that 10-15% of the approximately 157,400 lung cancer deaths in the U.S. annually may be due to residential radon exposure. A growing list (20) of case-control residential radon exposure and lung cancer studies are providing additional evidence for the association of radon exposure and lung cancer. Krewski et al. (2006) provides a good review of the seven North American studies. The individual studies have limited statistical power and their results are inconsistent. However, more recent pooled analysis of various combined studies does provide evidence for a direct association between residential exposure to radon and lung cancer. This no longer necessitates the downward extrapolation from the miner studies to estimate residential risk. Many epidemiology studies assume that the exposure most relevant to the risk of lung cancer was the 30 years ending five years before the diagnosis of lung cancer. The latent period is the time from the initial radiological insult to the appearance of a clinically evident cancer. This time period was at least five years for lung cancer in the September 14, 2009 140 uranium miners. Additionally, this latent period may be different for smokers and nonsmokers, with nonsmokers having a longer latent period (Archer, 2004). BEIR VI report (NRC, 1999): -Examined 11 major studies of underground miners. - 68,000 men involved, of which 2,700 died of lung cancer - Majority of miners were smokers - Current EPA risks estimates are based on this miner epidemiology - The miner data showed an inverse exposure-rate effect. For a given dose or cumulative exposure, as the dose rate is lowered, the probability of carcinogenesis increases. However, this does not apply at the more typical residential exposures (~ less than 25100 WLM) where there is a very low probability of multiple alpha-particle traversals through a cell. “Of the residential case-control studies to date, 19 of 22 have shown a positive association with radon exposure and lung cancer; however, only five show a significant association. But no study shows a negative association.” Quote from R. William Field, Ph.D., University of Iowa, College of Public Health, at the 2008 International Radon Symposium, Las Vegas, NV. September 14, 2009 141 Tables I, II, III below with permission of senior author, Mustafa Al-Zoughool (Al-Zoughool, 2009). Table I. Types of epidemiological studies used to evaluate the risk of lung cancer due to radon exposure. Study type Target population Main purpose of the Method of radon study dosimetry Cohort studies Miners/occupational Determine the risk of Radiation exposure was exposure. lung cancer mortality in estimated using jobexposed miners exposure matrix (JEM) which provides exposure values for potential alpha energy from radon and its progeny in working level months (WLM). Case-control studies The general public/ Determine the risk of Year-long residential residential exposure. lung cancer in radon levels were residential settings. measured by a-track detectors and were used to estimate exposure in the 25 years prior to the index date. Pooled analysis of the Miners/occupational Obtain summary A summary of the cohort studies on exposures. estimates of the risk of WLM exposure was miners lung cancer in radonobtained for the total exposed miners using subjects using reported large sample size. exposure levels in the individual studies. Combined analysis of The general public/ Obtain accurate Available radon case-control studies residential exposure in estimates of lung measurements form Europe and North cancer risk from individual studies were America. residential radon used to estimate radon exposure by reducing exposure for the total uncertainty in radon individuals in all homes dosimetry. occupied over the past 5–30 years. Major findings/conclusions High levels of radon m exposure were associated with increased cancer risk. Most studies reported small insignificant association between residential radon exposure and lung cancer, some studies found negative association. A consistent linear relationship for cumulative radon progeny and lung cancer was observed in the range of miner exposures A significant increase in risk of lung cancer was associated with increased radon exposures with seemingly linear doseresponse relationship. September 14, 2009 142 Table II. Summary of the main characteristics and risk estimates of cohort studies. Study region (Ref number) Type of mine Exposed Person-years Non-exposed Lung cancer deaths Exposed Non-exposed Mean cumulative radon exposure in working level months † ERR/WLM† (95% confidence interval) ‡ Newfoundland, Canada: Extended study (Villeneuve et al. 2007) Fluorspar 88,842 NA* 191 62 378 0.0043 (0.0023, 0.0062) Germany (Grosche et al. 2006) Uranium 1 565 070 236 560 2201 187 241.1 0.0021 (0.0018, 0.0024) Czech Republic: Extended cohort (Tomasek 2002) Uranium 127 397 NA* 495 165 NA 0.026 (0.012, 0.041) France: Extended cohort (Laurier et al. 2004b) Uranium 50 034 6 338 85 45 71.3 0.006 (0.001, 0.012) France (Tirmarche et al. 1993) Uranium 39 487 4556 45 0 70.4 0.0036 (0.001, 0.013) Yunnan Province, China (Xuan et al. 1993) Tin 135 357 39,985 936 44 277.4 0.0016 (0.001, 0.002) W. Bohemia, Czech Republic (Tomasek et al. 1994) Uranium 10 3652 4,216 656 5 198.7 0.0031 (0.002, 0.006) Colorado Plateau (Hornung and Meinhardt 1987) Uranium 73 509 7403 292 2 595.7 0.004 (0.003, 0.007) Ontario, Canada (Kusiak et al. 1993) Uranium 319 701 61 017 282 2 30.8 0.0089 (0.005, 0.015) Newfoundland, Canada (Morrison et al. 1988) Fluorspar 35 029 13 713 112 6 367.3 0.0076 (0.004, 0.013) Malmberg, Sweden (Radford and Renard 1984) Iron 32 452 841 79 0 80.6 0.0095 (0.001, 0.041) Grants, New Mexico (Samet et al. 1991) Uranium 46 797 12 152 68 1 110.3 0.0172 (0.006, 0.067) Port Radium, Canada (Howe et al. 1987) Uranium 30 454 22 222 39 18 242.8 0.0019 (0.001, 0.006) Beaverlodge, Canada (Howe et al. 1986) Uranium 68 040 50 345 56 9 17 2 0.0221 (0.009, 0.056) Radium Hill, Australia (Woodward et al. 1991) Uranium 25 549 26 301 32 22 7.6 0.0506 (0.010, 0.122) Pooled analysis of 11 cohort studies: References 18–28 (Lubin et al. 1995) 907 459 242 332 2597 109 158.0 0.0049 (0.002, 0.010) *Non-exposed cohort was the general male population in the same region of the study; †Among radon-exposed miners; ‡ERR/WLM, excess relative risk/working level month. Excess relative risk expresses how much increase in the risk of the disease is due to exposure to a given agent. The ERR can be obtained by subtracting one from the relative risk. Working level month is a time-integrated exposure measurement, is the product of time in working months (170 hours) and working-level (WL). One WL equals any combination of radon progeny in 1 l of air that gives the ultimate emission of 130 000 MeV of energy of alpha particles. Consequently, 1 WLM corresponds to 2.08×10-5 J/m3 ×170 hours or 3.5×10-3 J-hours/m3. September 14, 2009 143 Table III. Major characteristics and findings of case-controls studies of residential radon and lung cancer. Cases/Controls Study (reference) North America New Jersey – I (Schoenberg JB 1992) New Jersey – II (Wilcox et al. 2007) Winnipeg (Letourneau et al. 1994) Missouri (Alavanja et al. 1999) Iowa (Field et al. 2000) Connecticut (Sandler et al. 2006) Utah-South Idaho (Sandler et al. 2006) Combined analysis of the above studies (Krewski et al. 2005) Europe Austria (Oberaigner W 2002) Czech Republic (Tomasek et al. 2001) Finland nationwide(Auvinen et al. 1996) Finland southern (Ruosteenoja et al. 1996) France (Baysson et al. 2004) Germany eastern (Wichmann et al. 2005) Germany western (Wichmann et al. 2005) Italy (Bochicchio et al. 2005) Spain (Barros-Dios et al. 2002) Sweden nationwide (Pershagen et al. 1994) Sweden never-Smokers (Lagarde et al. 2001) Sweden Stockholm (Pershagen et al. 1992) United Kingdom (Darby et al. 1998) Combined analysis of the studies in Europe (Darby et al. 2005) China Shenyang (Blot et al. 1990) Gansu (Wang et al. 2002) Estimated radon concentration* Bq/m3 Excess odds ratio† (95% confidence interval) 480/442 561/740 738/738 512/553 413/614 963/949 511/862 3662/4966 26 32 142 56 127 33 57 70 0.56 (-0.22, 2.97) 0.05 (-0.14, 0.56) 0.02 (-0.05, 0.25) 0.27 (-0.20, 1.53) 0.44 (0.05, 1.59) 0.02 (-0.21, 0.51) 0.03 (-0.20, 0.55) 0.11 (0.00, 0.28) 0.21 (0.03–0.51)‡ 183/188 171/713 881/1435 160/328 571/1209 945/1516 1323/2146 384/405 156/235 960/2045 258/487 196/375 960/3126 7148/14208 198 500 103 215 133 76 50 108 131 96 74 134 55 105 0.46 (< -0.046, >5.00) 0.09 (0.02, 0.21) 0.11 (-0.06, 0.31) 0.28 (-0.21, 0.78) 0.05 (-0.01, 0.12) 0.08 (-0.03, 0.20) -0.02 (< -0.18, 0.17) 0.14 (-0.11, 0.46) < -0.11 (<-0.11, 0.59) 0.10 (0.01, 0.22) 0.28 (-0.05, 1.05) 0.16 (-0.14, 0.92) 0.08 (-0.03, 0.20) 0.08 (0.03, 0.15) 0.16 (0.05–0.31)‡ 308/356 768/1659 85 223 -0.05 (<0.00, 0.08) 0.19 (0.05, 0.47) *Estimated average residential radon concentration in the 5–30 exposure time window; †The excess relative risk of lung cancer per 100 Bq/m3 increase in the time-weighted radon concentration; ‡After correction for random uncertainties in radon measurements. References: Al-Zoughool, M. and Krewski, D. Health Effects of Radon: A review of the literature. Int. Journal of Radiation Biology, Vol. 85, No. 1, January 2009. September 14, 2009 144 Archer, V.E., Coons, T, Saccomanno, G, Hong, D. Latency and the Lung Cancer Epidemic Among United States Uranium Miners. Health Physics, Vol. 87, No. 5, November 2004. Krewski, D., et al. A Combined Analysis of North American Case-Control Studies of Residential Radon and Lung Cancer. Journal of Toxicology and Environmental Health, Part A. 69:533-597, 2006. Lubin, J. H. Invited commentary: Lung Cancer and Exposure to Residential Radon. Am. J. Epidemiology 140:323-332, 1994. National Research Council. Committee on the Biological Effects of Ionizing Radiations. Board of Radiation Effects Research, Committee on Life Sciences, Health Effects of Exposure to Radon. Biologic Effects of Ionizing Radiation (BEIR VI). National Academy Press, 1999. September 14, 2009 145 Lung Cancer Section Leading cause of cancer-related deaths in the United States Smoking remains the predominant risk factor for lung cancer. However, only 10-20% of smokers develop lung cancer (Sophie, 2007). According to the U.S. Surgeon General, smoking is also associated with increased risk of at least 15 types of cancer, including lung cancer (PA Dept. Health, 2008). Cigarette smoking alone still accounts for approximately 30% of all cancer deaths in the United States, despite reductions in smoking prevalence. Most (80%) of these smokingattributable cancer deaths involve lung cancer; although, smoking causes other cancers (Jemal, 2008). Interestingly, cigarette smoke can deliver a significant radiation dose to the lung. Radiation “hot spots” may occur at bifurcations of segmental bronchi. The dose to these specific areas for adult smokers may be in the range of 0.8 to 1.0 rad/yr, and if a quality factor of 20 for alpha particles is applied this gives an annual dose equivalent of 16 rem. This is not directly comparable to the effective dose equivalent for radon progeny in that a tissue weighting factor is not available for inhaled cigarette smoke products (NCRP 95, 1987). Fifteen percent of lung cancers in men and 53% in women are not attributable to smoking; overall accounting for 25% of all lung cancer cases worldwide (Sophie, 2007). The US EPA estimates that radon exposure accounts for 1 in 8 lung cancer deaths in ever smokers, and 1 in 4 lung cancer deaths in never smokers (EPA, 2003). For the population as a whole the risk of a fatal lung cancer due to a lifetime exposure of 1 pCi/L is ~ 0.58%, or at the 4 pCi/L action level it is 2.3% (EPA, 2003). Adenocarcinoma is the most common form of lung cancer in never smokers (Sophie, 2007). Radon exposure is considered the second leading cause of lung cancer. Second-hand smoke is also a risk factor. Types of lung cancer: Small cell (oat cell carcinoma) and nonsmall cell Nonsmall cell accounts for 85-90% of lung cancer types and this is broken down into: Adenocarcinoma (40%) Squamous cell carcinoma (25-30%) Large cell carcinoma (10-15%) September 14, 2009 146 Small cell carcinoma accounts for 10-15% of all lung cancers. It is almost always caused by smoking (ACS). U.S. Statistics (2005): 90,139 men and 69,078 women died of lung cancer, for a yearly total of 159,217 (CDC) Symptoms of lung cancer: Shortness of breath Coughing that does not go away Wheezing Coughing up blood Chest pain Fever Weight loss Currently there is no major organization that recommends screening for early detection of lung cancer (Collins, 2007). However, a New England Journal of Medicine article in October 2006 found that lung cancer could be detected in 85% of patients in its earliest stage by the use of “annual low-dose CT screening (Henschke, 2006).” September 14, 2009 147 Map 1 (9) Prevalence of Cigarette Smoking - Adults - 2004 (age-adjusted, aged 18 years and over) 10% to <20% 20% to <21% 21% to <24% 24% to 28% No data Source: DATA2010 ...the Healthy People 2010 Database - March, 2008 Edition: Objective: 27-01a Map 2 (9) Lung and Bronchus Incidence Rates - 2004 (age-adjusted, per 100,000) 28.3 to <62.1 62.1 to <68.4 68.4 to <73.9 73.9 to 99.5 No data Source: U.S. Cancer Statistics Working Group. United States Cancer Statistics: 1999–2004 Incidence and Mortality Web-based Report. Atlanta: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention and National Cancer Institute; 2007. Available at: www.cdc.gov/uscs. Notes:Incidence rates (cases per 100,000) are age-adjusted to the 2000 US standard million population. All in situ cancer incidence cases are excluded. September 14, 2009 148 References: American Cancer Society Centers for Disease Control (CDC).Cancer Statistics, 1999-2005. Collins, L.G., Haines, C., Perkel, R., and Enck, R.E. Lung Cancer: Diagnosis and Management. American Family Physician, Volume 75, Number 1, January 1, 2007. Henschke, C.I. et. al. Survival of Patients with Stage I Lung Cancer Detected on CT Screening. New England Journal of Medicine, Vol. 355:1763-1771, No. 17, October 26, 2006. Jemal, Ahmedin, et. Al. Annual Report to the Nation on the Status of Cancer, 19752005, Featuring Trends in Lung Cancer, Tobacco Use, and Tobacco Control. Journal of the National Cancer Institute, Volume 100, Issue 23, December 3, 2008. National Council on Radiation Protection and Measurements. Radiation Exposure of the U.S. Population from Consumer Products and Miscellaneous Sources, NCRP Report No. 95, 1987. Pennsylvania Department of Health, Bureau of Health Statistics and Research, Statistical News. Volume 31, No. 3, May/June 2008. Pennsylvania Department of Health, Bureau of Health Statistics and Research. Reprinted from Statistical News, Vol. 31, No. 3, May/June 2008. All map data from CDC. Sophie, S., Schiller, J.H., and Gazdar, A.F. Lung Cancer in never smokers-a different disease. Nature Reviews, Volume 7, October 2007. U.S. EPA Assessment of Risks from Radon in Homes. Office of Radiation and Indoor Air, US Environmental Protection Agency, June 2003. September 14, 2009 149 Lung/Bronchus Cancer Incidence Rates by State State AL AK AZ CA CO CT DE DC FL GA HI ID IL IN IA KS KY LA ME MD MA MI MN MS MO Rate 77.0 70.7 55.4 53.5 50.8 64.5 80.0 56.9 73.9 72.8 54.2 57.8 69.3 76.1 65.2 65.2 99.5 79.9 79.1 53.2 69.7 71.4 58.6 77.5 77.4 State MT NE NV NH NJ NM NY NC ND OH OK OR PA RI SC SD TN TX UT VT VA WA WV WI WY Rate 61.9 64.9 72.6 70.9 62.1 45.3 63.5 72.9 56.1 72.9 78.5 68.4 68.4 73.5 69.6 59.0 82.2 65.9 28.3 62.8 67.6 67.1 85.6 62.6 53.6 Age-adjusted rate per 100,000 for both males and females. Age-adjusted to the 2000 U.S. Standard population. U.S. Average is 67.4. Source: Centers for Disease Control, National Program of Cancer Registries, U.S. Cancer Statistics. 2004. September 14, 2009 150 September 14, 2009 151 The above two pie charts from the National Council on Radiation Protection and Measurements (NCRP) shows the breakdown of radiation exposure to the U.S. population. The first pie chart shows the exposure from all significant sources, primarily background and medical. The two most significant aspects of this chart are the sevenfold increase in medical exposure, since the early 1980’s, primarily from CT scans and nuclear medicine, and the inclusion of thoron with the radon exposure. This first pie chart shows that radon/thoron contribute 37% of the total collective effective radiation dose to the general population. The second pie chart excludes medical exposures and just shows the background radiation exposures, of which radon-222 contributes 68%, and thoron-220 contributes 5%. References National Council on Radiation Protection and Measurements. NCRP Report No. 160, Ionizing Radiation Exposure of the Population of the United States, March 2009. September 14, 2009 152 A Brief History of the Rn-222 Occupational Limits Much of the federal guidance given below was based on studies starting in the early 50’s of uranium mines on the Colorado Plateau. The U.S. Public Health Service, primarily led by Duncan Holaday and his colleagues were the first group to raise concern about the potential health effects from exposure to radon decay products in the mines. There was already evidence coming from the “European Experience” where increased lung cancer rates were seen in the miners. However, there was great reluctance by the miners to take radon seriously, they were making good money. The mine operators were also reluctant to disturb operations. However, in spite of the reluctance, the Public Health Service was able to start getting into mines to take samples and have physical exams performed on many of the miners. It was too early yet to see any malignancies in the miners; however, the air samples were certainly alarming. One sample at a working face in a Utah mine showed 26,900 pCi/L, another at the entrance incline was 14,000 pCi/L. These samples are compared to what was seen in some German and Czechoslovakian mines with 1,000 and 1,500 pCi/L, respectively. It was becoming obvious that something had to be done. The Public Health Service estimated that a maximum allowable concentration of 100 pCi/L of radon would be safe in a mine. This was also the European standard. The only radiation standards at the time were those established by the NCRP in 1940. Finally, in the early 1950’s William Bale from the University of Rochester, and Dr. John Harley at the Health and Safety Laboratory found that it was not the radon but the radon decay products that were the significant health concern. It would be up to the Atomic Energy Commission to set the standard, see below. December 1968 the Federal Radiation Council (FRC) submitted three memorandums to the President concerning radiation protection guidance for federal agencies. The recommendations contained in the memorandums were based on FRC Report No. 8, “Guidance for the control of radiation hazards in uranium mining,” September 1967. The first memorandum was published in the Federal Register on August 1, 1967. The FRC considered exposure guidance of 36, 12, and 4 WLM per year. Based on a balance between risks to miners and exposure control capability in the mines they choose the 12 WLM per year limit. The second memorandum was published in the Federal Register on January 15, 1969. In this memorandum, the FRC gave guidance to federal agencies concerning underground uranium mining. They put forth eight recommendations, two of which are most important to this discussion. 1) Occupational exposure to radon decay products in underground mines shall not exceed 12 WLM in any consecutive 12-month period, and 2) The uranium mining industry is urged to continue to lower exposures so that the anticipated 4 WLM per year standard can be attained by January 1, 1971. September 14, 2009 153 In the May 25, 1971 Federal Register the Environmental Protection Agency (EPA) provided further guidance to federal agencies concerning underground mining of uranium ore. They concluded that 4 WLM per year was technically feasible, that the 4WLM standard would not have a severe impact on the uranium mining community and that a standard greater than 4 WLM would probably result in dosages greater than those permitted for other occupational exposure situations. This recommendation of 4 WLM per year was approved by the President and published in the January 15, 1969 Federal Register and was to become effective January 1, 1971. This date was later extended to July 1, 1971. Based upon the May 25, 1971 Federal Register announcement by EPA of the 4WLM/yr standard, public comments were received. The EPA responded to those comments as published in the July 9, 1971 Federal Register and concluded that no change would be made to the 4-WLM/yr standard. In the June 24, 1974 Federal Register, the Atomic Energy Commission (AEC) considered an occupational concentration value for Rn-222 decay products in their Table 1, Appendix B. The limit for Rn-222 (gas) would be replaced by a limit for the decay products since they are the major health hazard. This change would bring the limit to 4 WLM/yr as recommended by the EPA, which is about 1/3 of the then-current 10 CFR 20 value. This change was in conformance with the ICRP Publication 2, “Report of Committee II on Permissible Dose for Internal Radiation,” published in 1959, which recommended a limit on Rn-222 of 3E-8 µCi/ml (30 pCi/L) with decay products present. NCRP also recommended the same limit in their NBS Handbook 69, 1959. The AEC considered expressing the concentration of the Rn-222 decay products in terms of working level but rejected this because it would add a new unit to the table and add confusion. It was therefore proposed and amended that the current Appendix B, Table 1 limit for Rn-222 be deleted and a new line for Rn-222 decay products be added beneath the Rn-222 line. The limiting value for Rn-222 decay products would be 7E-8 µCi/ml (70 pCi/L). This limit is based on a one-week average. For this value, see Fed. Reg. Vol. 39 No. 122 Monday June 24, 1974. As published in the October 31, 1975 Federal Register, the AEC decided to express both radon and its decay products in conventional ways. Thus, the Federal Register announcement of June 24, 1974 was amended to show a Rn-222 concentration limit of 3E-8 µCi/ml (30 pCi/L). A footnote gave an alternate limit of 1/3 WL for decay products. This amendment became effective January 29, 1976. The Nuclear Regulatory Commission changed the averaging period for the Rn222 limit from one week to one year in the July 7, 1978 Federal Register. The Rn-222 limit of 30 pCi/L (0.33 WL) that became effective on January 29, 1976 did not appear in the 10 CFR 20 Appendix B table until the 1979 edition. Prior to that, it had been 100 pCi/L. September 14, 2009 154 The value of 30 pCi/L (0.33 WL) is still the current US NRC occupational exposure limit for radon-222. This value is also known as the DAC value, for Derived Air Concentration. The other value for radon-222 in 10 CFR 20 Appendix B, Table 1, is the ALI or Annual Limit of Intake, being 4 WLM for radon decay products. September 14, 2009 155 Radon and Geology Rocks most likely to cause radon problems and the uranium and radium sources they host. Table below from L.C.S. Gundersen and others. Rock Type Uranium and Radium Sources Black Shales, Lignite, Coal Uranium-bearing organic compounds; autinite, tyuyamunite Glauconitic sandstones Radium- and uranium-bearing iron oxides; heavy minerals Fluvial and lacustrine sandstones Roll-front deposits, which include uraninite, coffinite, pitchblende, secondary uranium minerals (tyuyamunite, carnotite, uranophane, and other uranyl vanadates); uranium and radium adsorbed onto organic material; iron and titanium oxides, placer deposits, which include heavy minerals. Phosphorite and phosphate Phosphate complexes; apatite Chalk and marl Phosphate complexes; apatite Carbonates Uranium and radium adsorbed onto ironoxide coatings; radium with organic material in soils; tyuyamunite, carnotite, and uranophane in karst and caves Glacial deposits Bedrock-derived clasts that comprise the glacial deposits are usually the principle source of radioactivity; uranium- and radium-bearing iron oxide and carbonate coatings on clasts are common Granites and granitic gneiss Heavy minerals; uraninite; brannerite; apatite, monazite, allanite Volcanic rocks Heavy minerals; uranosilicates Faulted rocks Heavy minerals; uraninite; uranium precipitated with hematite and titanium oxide; minerals found in uranium vein deposits Graphitic schist’s and gneisses, some of which are metamorphosed black shale and September 14, 2009 156 Rock Type siltstone Vein and vein-like deposits Uranium and Radium Sources Syenites, carbonatites, pegmatites Uraninite; other uranium minerals; heavy minerals Bauxite Heavy minerals Many kinds of uranium minerals; heavy minerals One scheme for the classification of radon-risk areas is based on the geological criteria as follows: Classification Criteria High-risk areas Rn in soil gas Conc. >than 1350 pCi/L Uranium-rich granites and gneisses. Contacts of oxidizing (red) and reducing (black) sedimentary rocks. Pegmatites (granitic) Alum shale Highly permeable soils Normal-risk areas Rn in soil gas conc. 270 to 1350 pCi/L Rocks and soils with low or normal uranium and radium content Soils with average permeability Low-risk areas Rn in soil gas conc. < 270 pCi/L Rocks and soils with very low uranium content. Limestone Sandstone Basic and ultrabasic, such as serpentinite igneous and volcanic rocks Soils with very low permeability, such as water saturated C.R. Cothern and J.E. Smith, Jr. Environmental Radon. Environmental Science Research, Volume 35. Plenum Press, 1987. As with most any classification system there will be exceptions to the generalizations in the above table. September 14, 2009 157 EPA’s Map of Radon Zones A map(s) of the US on a county-by-county basis identifying areas of highest radon potential (>4 pCi/L). Designed for federal, state and local governments to design and assist in outreach programs and resource management. Also to be used for targeting local municipalities and builders for the incorporation of radon-resistant new construction practices. Not to be used to make predictions about individual homes. The Radon Zone map is based on the work done by the USGS in cooperation with the US EPA to provide a radon potential for the US. That work resulted in ten separate booklets, for instance “The Geologic Radon Potential of EPA Region 3”, which includes PA, DE, WV, VA, and MD. The map assigns each of the 3141 counties in the US to a radon potential zone, Zone 1, Zone 2 or Zone 3. • • • Zone 1 counties have predicted average indoor screening level > 4 pCi/L Zone 2 counties have predicted average indoor screening level ≥ 2 pCi/l and ≤ 4 pCi/L Zone 3 counties have predicted average indoor screening level < 2 pCi/L Note some important words above; they are predicted - foretell on the basis of observation, experience or scientific reason. They are average - thus some will be higher and some lower. They are for indoor radon levels. They indicate screening levels, which implies lowest livable level of home. The USGS identified 360 separate geologic radon provinces for the US. These provinces were then categorized by five factors considered most important in assessing radon potential: existing indoor radon concentrations, geology, aerial radioactivity, soil parameters and foundation types. Observations from US Radon map: In general, coastal areas do not have high radon potential. Radon potential does vary throughout the US. Thus, we do not need the same level of effort in Texas as we do in Pennsylvania. September 14, 2009 158 NURE: National Uranium Resource Evaluation This project was sponsored originally by the U.S. Atomic Energy Commission and then later by the Department of Energy. It systematically evaluated the uranium resources of the United States. The project ran from 1974 to 1980. The project included water and stream-sediment sampling, rock sampling and analysis, airborne radiometric and magnetic surveying, geologic mapping and subsurface geologic investigations. The data from the NURE mapping was organized onto the USGS National Topographic Map Series 1:250,000 scale. Of most interest to this program are the aerial radiometric data. More than 1.1 million line-miles of surveys were flown to determine the presence of potassium, uranium and thorium. In the western U.S. flight-line spacing was 3 miles and in the eastern U.S. flight-line spacing was 6 miles. Flight lines were either east-west or north-south, and at a nominal ground clearance of 400 feet. A gamma-ray spectrometer pointed toward the ground detected the 1.764-MeV gamma ray from Bi-214. Bi-214 is one of the progeny of radon-222. The key information gained from the NURE program was the identification of areas favorable for uranium exploration and mining, with grades of at least 0.01 percent U3O8 . This information ended up on color-coded maps of the U.S. and the individual states. The maps showed a color scale typically from 0.5 to 5.5 ppm eU, where the e means radiometric equivalent uranium. Radiometric equivalent, because it was not measured directly but calculated. R.T. Peake considered rocks with >3 ppm uranium as being of high radon potential. Rocks with less than 1.5 ppm uranium may be considered low, and rocks with uranium between 1.5 and greater than 2.5 may be considered intermediate. How do the NURE data relate to radon? The NURE data give an indication of nearsurface uranium concentrations. The uranium concentration can be mathematically converted to radium-226 concentrations in picocuries per gram by the conversion: 1 ppm eU is equal to 0.33267 pCi/g Ra-226. Now with the radium concentration we have the direct parent of radon. Additionally, the radium content of the soil and rocks is one of the four primary factors to consider for an area’s radon potential, the other three being emanating power, soil permeability and soil moisture. A radium content of less than 0.5 pCi/g is generally considered low, moderate concentrations are in the range of 0.5 to 1.0 pCi/g, and high concentrations > 1.0 pCi/g. The NURE data are most useful for radon prediction when augmented with geologic and soil data. The NURE data typically do not provide useful information on a small-scale basis. Peake, R.T. Radon and Geology in the United States. Radiation Protection Dosimetry, 24: 173-178, 1988. September 14, 2009 159 Introduction to Soils Soils have two major volume fractions. The solid fraction consists mainly of mineral grains of a wide range of sizes and also includes a small amount of organic matter. The void fraction consists of fluid usually water and gas (similar in composition to air). The void fraction is also known as the soil porosity. The volume fraction of water is often called the moisture content. Soils are classified according to the size distribution of the solid grains. Clays - grain size < 2 micrometer, porosity 0.6 Silt - grain size 2 - 60 micrometer, porosity 0.5 Sand - grain size 60 - 2000 micrometer, porosity 0.4 Moisture Content Moisture content is a very important factor for radon emanation and migration in soil. For a well-drained soil, the void volume contains water in the smaller pores and air in the larger pores. Capillary water increases the radon emanation fraction by absorbing the recoil energy of the newly formed atom. However, this water does not increase the resistance of the soil to airflow to a great degree since it is the larger pores that make the dominant contribution to airflow. Permeability The velocity of fluid flow through the soil pores in response to the pressure gradient. The importance of permeability in relation to indoor radon arises from its very broad range of values. Range of permeability’s: higher (gravel) 10-7 m2 to lower (clays) 10-16 m2; permeability is also described in terms of a rate (inches/hr): Impermeable--------------<0.0015 in/hr Very slow-----------------0.0015 to 0.06 Slow ----------------------0.06 to 0.2 Moderately slow---------0.2 to 0.6 Moderate------------------0.6 to 2.0 Moderately rapid---------2.0 to 6.0 Rapid-----------------------6.0 to 20.0 Very rapid------------------- > 20.0 September 14, 2009 160 Since convective flow rate increases with increasing permeability, and since the radon entry rate increases with the convective flow of soil air into the structure, the potential for radon entry is expected to increase monotonically with permeability for large-grained soils. According to R.T. Peake soils with high permeability (>6 in/hr) may contribute to elevated indoor radon, even when the soil radium concentration is low. Convective vs. Diffusive flow and permeability Diffusive flow through soil predominates when permeability’s are less than ~ 1E7 cm2, and convective flow predominates when the permeability’s are greater than ~ 1E-7 cm2 (Tanner, 1964). This can lead to radon diffusion lengths of less than 1 cm for low-permeability, saturated soils to about 1 m for low-permeability, dry soils, and to as much as ~ 5 m by convective flow for higher-permeability soils. Thus, well-drained soils such as those found on hilltops may yield high indoor radon. Thus, on-site measurements or published data of soil permeability may be useful for identifying buildings with high radon concentrations. Radon Production in Soil This is a function of the radium content of the soil, which in turn depends on the radium content of the rocks from which the soil was formed. Emanation Coefficient or Emanation factor - Only a fraction of the radon generated in soil ever leaves the solid grains and enters the pore space of the soil. This fraction is known as the emanation coefficient or emanation fraction. Typical range of values for the emanation coefficient are 0.05 to 0.7 for soil. The emanation coefficient is largely controlled by the size of the mineral grains and the distribution of radium within the material. Where the radium is disseminated throughout the volume of the grain, the emanation coefficient is low; when radium is coating the mineral grains near the pore space, the emanation coefficient is high. During radon production, a radium atom decays via alpha decay into a radon atom. This radon atom has a recoil range of about 0.02 to 0.07 micrometers in common minerals; 0.1 micrometer in water, 63 micrometers in air. Now, there are three outcomes after this radium decay: 1. The radon atom stops in the fluid-filled pore space 2. The radon atom leaves one mineral grain and is trapped in another grain. 3. The radon atom begins and ends its recoil within a single grain. September 14, 2009 161 Important: Only radium atoms within the recoil range of the surface generate radon atoms that have any possibility of escaping the grain. Combining numerous factors suggests that radon release from the soil is maximal when the soil is moist. In summary, if we were going to try to characterize a geographical area for radon potential several factors we would consider are: 1. 2. 3. 4. Radium content of soil Temporal state of the soil, particularly moisture Soil permeability Weather - temperature, wind, rain fall References: Duval, J., Otton, J., Jones, W. Radium distribution map and radon potential in the Bonneville Power Administration Service Area. USGS Open File Report 89-340, 1989. Gundersen, L.C.S., Schumann, R.R., Otton, J.K., Dubiel, R.F., Owen, D.E., and Dickinson, K.A. Geology of Radon in the United States. Geological Society of America, Special Paper 271, 1992. Nazaroff, W., and Nero, A. Radon and its Decay Products in Indoor Air. John Wiley and Sons, 1988. Peake, R.T. Radon and Geology in the United States. Radiation Protection Dosimetry, 24:173-178 (1988). Tanner, A. B. Radon Migration in the Ground. A Review, in Adams, J.A.S., and Lowder, W.M., eds., The Natural Radiation Environment: Chicago, University of Chicago Press, pp 161-190, 1964. September 14, 2009 162 Radon Decay Products (RDP) RDP: Po-218 → Pb-214→ Bi-214→ Po-214→ Pb-210→ Bi-210→ Po-210 Synonymous with Radon Decay Products are “Radon Daughters” and “Radon Progeny.” Note: The Health Physics Society no longer uses or permits the use of the term “Radon Daughters” in its keywords or publications. Radon Decay Products are the radionuclides that follow the decay of radon-222 in the uranium-238 decay series. Radon Decay Products are principally distinguished from radon by their chemical activity; whereas, radon-222 is chemically inert for all practical purposes. RDP’s are originally formed as positive ions or neutral atoms. First decay product Po-218 is formed in positive state 80% of time. Single, positive charge is assumed. Positive charge acquired by stripping electrons by departing alpha or as a result of recoil. Probably no negatively charged Po-218. Po-218 undergoes ~ 1 trillion collisions until it thermalizes (nsec), it may still have a charge after thermalization. It finally does become neutral by scavenging electrons from its recoil path. Second decay product Pb-214, results from alpha decay of unattached or attached Po218. Pb-214 may remain attached or due to recoil energy (100 KeV) become unattached. Third decay product Bi-214, results from beta decay of Pb-214. Bi-214 typically remains in attached state since recoil energy of beta is only a few electron volts. This is not sufficient to promote detachment. Radon Decay Products are classed into size distributions: -Unattached Fraction (ultrafine aerosol mode)- 0.5 to 5.0 nm in diameter. -Unattached Fraction mostly consists of “free” Po-218 atoms. A few water molecules may also be attached. (Hopke, 1992) -Unattached Fraction due to high diffusivity and high deposition rate in the tracheobronchial region delivers a higher dose per unit exposure than attached fraction. -Unattached Fraction comprises about 5% of total Radon Decay Products in houses. (UNSCEAR, 2000) -Unattached Fraction dose conversion factor around 10 to 20 rad/WLM -Unattached fraction has diffusion coefficient of 0.054 cm2/sec. September 14, 2009 163 -Attached Fraction- 20 to 500 nm in diameter. -Attached Fraction dose conversion factor around 0.2 to 1.3 rad/WLM. -Attachment process is due to two main mechanisms; classical diffusion and gas kinetics The fate of the decay products: Decay, Attachment, Deposition, Recoil and Resuspension. Radioactive Decay is a physical characteristic unique to each radionuclide and not affected by circumstances, such as temperature, pressure, or chemical environment. Attachment is the process whereby an unattached radon progeny atom or cluster, strikes and sticks to an aerosol particle. Most radon progeny activity will be found attached to particles in the size range 100 to 200 nm in diameter. Deposition describes the process whereby attached or unattached progeny stick to surfaces exposed to the air. The term “plateout” is often used for this phenomenon. The deposition onto surfaces is described by the deposition velocity. All to be said regarding this is that the deposition velocity for the unattached progeny is about 100 times greater than that of the attached progeny. Resuspension occurs when a previously deposited radon progeny (unattached or attached) recoils from the surface back into the air space. References: Nazaroff, W.W. and Nero, A.V. Radon and its Decay Products in Indoor Air. John Wiley & Sons. 1988. Cavallo, A., Hutter, A., and Shebell, P. Radon Progeny Unattached Fraction in an Atmosphere Far from Radioactive Equilibrium. Health Physics, 76(5):532-536; 1999. Hopke, P.K. Some Thoughts on the “Unattached” Fraction of Radon Decay Products. Health Physics, 63(2): 209-212; 1992. United Nations Scientific Committee on the Effects of Atomic Radiation. UNSCEAR 2000, Volume 1: Sources. Hopke, P.K. Harmonization of Radon Jargon and Units. Health Physics, Volume 64, page 459, 1993. September 14, 2009 164 Radon Variability That radon concentration does vary with time is well documented. Over short periods of time a change in concentration of a factor of five is not uncommon. One NJ residence had basement radon variations of a factor of 15 over a two-month period (Hernandez, et al., 1984). One interesting comment that Allan Tanner made while performing a bibliography search was that “nearly all papers discuss the effects of meteorological variables. Owing to nonideal conditions in the various test areas and to the understandable difficulty of isolating the effects of different variables, the results of these investigations do not present a very coherent picture; the variable inferred to dominate in one investigation is found to be of little significance in another.” This comment was made in 1964, however. Four factors that influence radon concentrations indoors are properties of the building material and ground; building construction; meteorological conditions; and occupant activities. Building materials can have varying concentrations of radiumbearing material and different diffusion properties. The ground surrounding the structure can have varying radium content, different porosity’s and permeability’s. The structure can be slab-on-grade, full basement, crawl space or some combination. A floating slab may exist or the pour may be monolithic. Wind, rain, temperature and barometric pressure all factor into affecting indoor concentrations. Operation of HVAC systems and window opening would probably be the major factors due to occupants. The sources of radon in the home environment are soil gas (90%), domestic water and underground wells (8-9%) and building materials (1-2%). There is a wide range of variability in the radon source materials that can add to the variability of the radon concentration in the home. The source material variability in decreasing order is ground water (1000X), soil and geologic substrate (20X or more), meteorological conditions (10X if short-term measurements are used), building materials (5X in most cases) and ventilation rate (usually less than 3X) (Mueller Associates, 1986). Meteorological factors are often interrelated in their influence on indoor radon concentrations. Some researchers have suggested that lower air exchange rates may cause higher radon concentrations. However, the forces that cause higher air exchange rates (winds and temperature differences) also cause an increase in the soil gas entry and may cause higher indoor radon concentrations. Soil moisture and its effect on radon emanation runs the gamut. At very low moisture content the radon atom has little opportunity to come to a stop in the open pore space, and it may become embedded in the mineral, moist soil conditions seems to lead to optimal emanation, and high soil moisture causes decreased diffusion and again low emanation. September 14, 2009 165 Even “general weather effects” can have substantial effects on short-term radon measurements. These weather effects should be considered as factors, which can compromise short-term measurement results (Hoffman, 1995). Radon Entry Dynamics For a given radon entry rate, the indoor concentration depends on the ventilation rate. Infiltration, the uncontrolled leakage of air, is the dominant mechanism for ventilation when windows and doors are normally closed. In typical U.S. housing stock, rates range from 0.3 to 1.5 ACH, but new construction techniques have reduced this value to 0.1 ACH. However, the broad range of indoor radon concentration is due primarily to differences in radon entry rate as compared to differences in ventilation rates. This as been confirmed in numerous studies where a weak correlation has been found between air exchange rate and indoor radon concentration. Radon entry into homes is via diffusion from soil and building materials, via off gassing from radon in water and via pressure-driven flow of soil gas. Pressure-driven (advective) flow is the predominant entry mechanism. This flow is driven by depressurization of the below-grade portion of the building relative to the soil. Temperature differences, wind speed, and barometric pressure changes may all induce this depressurization. At least one author (Minkin, 2008) believes that pressure-driven entry does not totally explain radon entry, particularly during the times when stack effect would be reversed as during warm weather. He has proposed an entry mechanism of thermodiffusion, whereby a temperature difference can induce mass transport. Soil gas entry is through cracks and gaps in the foundation structure, and these openings typically provide little resistance to flow. Soil permeability to air is therefore the major factor governing indoor radon concentrations. Soil-gas entry is essentially proportional to permeability (Garbesi, et al 1999). Recent research has suggested that diffusion can be a significant contributor to indoor radon, particularly at low indoor concentrations where advection may only account for 20% of the radon entry (Renken et al., 1995). Soils: Nazaroff and Nero 1988 Soil and rock act as the main source for generation of radon. The radium content of soil reflects the radium content of the rocks from which the soil formed. However, some locations can have soils that have been transported from distant locations and therefore the underlying rock type may not reflect the prevailing soil. The average radium-226 content of Pennsylvania surface soils is about 1 pCi/g. Movement of radon within the soil is restricted to several meters or less; therefore, a foundation would not draw radon from very distant locations. Two processes; convective flow and diffusion govern the exchange of gas between the soil and the atmosphere. Convective flow is due to pressure differences September 14, 2009 166 between the soil and the atmosphere. These pressure differences can be induced by soil and air temperature differences, barometric pressure and wind movements. The upper few inches of soil may show significant diurnal temperature changes resulting in convective flow; however, most gas exchange occurs via diffusion (Brady, 1974). Some important characteristics of soil that are discussed below are grain size, permeability, porosity and moisture content. These characteristics have a large influence on radon transport within the soil. The soil is divided into two major volume fractions; the solid fraction composed mainly of mineral grains and the void fraction (also known as soil porosity), which usually consists of water and air. The volume fraction of water is called the moisture content. A soil is saturated when the moisture content equals the porosity. Soil porosities are commonly in the vicinity of 0.5. Grain sizes range from clays at <2 µm, to silt at 2-60 µm, to sand at 60-2000 µm. Moisture content is a very important factor for radon emanation and migration in soil. For a well-drained soil, the void volume contains water in the smaller pores and air in the larger pores. Capillary water is held in the small pores and in a film around the surface of soil particles. This capillary water increases the radon emanation fraction by absorbing the recoil energy of the newly formed radon atom. However, this water does not increase the resistance of the soil to airflow to a great degree since it is the larger pores that make the dominant contribution to airflow within the soil. Permeability describes how readily a fluid can flow through a soil. It relates the fluid flow through the soil pores to the pressure gradient. Its importance in the study of radon arises from the very broad range of permeability’s found in soil. Common soils range from 10-8 m2 (clean gravel) to 10-16 m2 (clay). Larger grained soils generally have higher permeability’s. At the low end of this range, molecular diffusion is the dominant process, at the upper end convective flow is the dominant transport process. Diffusivity, owing to random molecular motion, is the tendency for a substance to migrate down its concentration gradient in a material. The term used to describe this flux is the diffusion coefficient. In a porous medium (soil), it is a property of the fluid (radon) in the pores. The movement of radon from soil to the atmosphere appears to be primarily due to molecular diffusion. The radon diffusion coefficient in air is 1.2E-5 m2/sec; the radon diffusion coefficient in soil of low moisture content is 1E-6 m2/sec. Water plays an important role in influencing the radon diffusion coefficient in soil. In saturated soil, the radon diffusion coefficient may be reduced to 2E-10 m2/sec. This value is so much lower than for that in air that we can view the effect of water on the radon diffusion in soil as blocking a fraction of the available pore space. Emanation coefficient is the fraction of radon generated that leaves the solid grains and enters the pore volume of the soil. Only radium atoms within the recoil range of the surface (0.02-0.07 µm for common minerals) have any possibility of escaping the mineral grain; maybe around 25%. Radon atoms in the deeper regions of the crystals are unavailable to the pore space without the development of a large internal surface, such as September 14, 2009 167 may result from chemical corrosion, weathering, or intensive fracturing on a microscopic scale (Tanner, 1964). Moisture content of soil has a large impact on the emanation coefficient. A radon atom entering a pore space partially filled with water has a high probability of stopping in the water, and from there it readily transfers (~0.1 sec) to air in the pore. This suggests that radon release from soil, combining emanation and transport, is maximal when soil is moist. One important consideration must be kept in mind when considering the above soil characteristics and radon transport. The presence of a house may influence the spatial distribution of soil moisture and, thereby, the emanation and migration of radon. The house acts as an umbrella to precipitation whereby the soil surrounding the house is more affected than the soil beneath the structure. A wind of 6.7 mph can induce a pressure difference across the soil and substructure on the windward side of about 2 Pa. This pressure difference varies as the square of the wind speed; therefore, much larger pressure differences are possible. Since wind-induced pressures can fluctuate rapidly, do they have time to be transmitted through the soil? This depends on the soil permeability. For this wind-induced pressure to be transmitted 1 meter through the soil in clay takes 10 days, in silt it takes 30 minutes and in gravel it takes 0.01 seconds. In addition to wind speed, wind direction can also affect radon entry in the absence of other factors. A definite trend indicates that high wind speeds produce a depletion of radon concentration in soil gas down to 44” (Kraner, 1964). This brings up a good point. Someone calls you up and asks about a radon test they had performed during some windy conditions. Did the wind have any effect on my results? From the above it would seem to depend on soil permeability, which we obviously don’t know. Thus, it is not easy to answer the question. Besides its effect on soil, temperature would also act to produce a stack effect in the building since a pressure difference exists across any vertical wall separating air masses of different temperature. The temperature creates a convective loop that carries air into the building near the ground and out of the building toward the top of the structure. Stack effect is primarily a function of temperature difference and height of building. It has been demonstrated that temperature differences associated with extreme weather conditions can generate pressure differences of several pascals, which contribute to radon driving forces and indoor radon concentrations (Al-Ahmady, 1994). There is a strong correlation between the indoor-outdoor pressure differences and the indooroutdoor temperature differences. It is the temperature difference, which causes air volume movements, which consequently cause the pressure difference. Regarding radon exhalation rates from soil Kojima and Nagano, 2000 found that wind, which is often associated with decreasing barometric pressure, had a significant effect; whereas, temperature had only a minor effect on soil exhalation rates. The wind September 14, 2009 168 velocity induces negative pressure difference by lowering surface pressure, the effect results in the increasing upward flow of soil gas. Compared with the pressure changes associated with wind and temperature differences, the magnitude of barometric pressure change is large, with excursions from the long-term average routinely exceeding 100 Pa. Over a three-year period of measurement Hoffman, 1995 found barometric pressure swings as much as 336 Pa. However, this only leads to soil gas flow into the building if this large pressure difference causes a sustained pressure difference between basement and the pore air of nearby soil. A barometric pressure drop alone does not appear to affect basement radon concentration as much as a barometric pressure drop plus rainfall (Harley, et al., 1984). Kraner et. al., 1964 described the changing barometric pressure effects using a piston analogy of the atmosphere. This produced a short-range displacement of soil gas, moving under the influence of a pressure differential between the atmosphere and soil gas at depth. The effects of precipitation showed radon-laden soil-gas increases at depth owing to a “capping effect,” in which the moisture significantly reduces vertical porosity of the surface layers. This effect continues while the soil is extremely moist (Kraner, 1964). Well-frozen ground appears only to reduce the soil-gas flux. A 40% reduction in flux was seen with ground frozen to 6” compared to average summertime values (Kraner, 1964). For a house with a basement on relatively permeable soil, and the surface of the soil frozen, a barometric pressure change could lead to a flow of soil gas that is funneled through the basement as a result of the reduced permeability of the frozen soil. A similar situation might exist for buildings immediately following a heavy rain. Harley noted that enough rain to “plug” the soil surface may inhibit surface release and lead to a buildup of radon at depth. Tanner 1964 The three factors most commonly observed to have pronounced effects on radon concentration and exhalation are rainfall, freezing and snow cover. Both transport and diffusion are affected. With heavy rainfall, the soil gases near the surface tend to be displaced upward, carrying radon with them and increasing the exhalation rate temporarily. Thereafter, the reduced diffusion coefficient and reduced permeability of the wet ground restrict migration by both mechanisms. Exhalation of radon is markedly reduced, with a commensurate increase of radon in the soil. September 14, 2009 169 Fleischer et al. 1983 Summer radon levels are usually substantially different from winter. The reasons for this variability include increased ventilation of the homes during summer (open windows) and lack of operation of central heating systems that circulate air. Grainger et al., 2000 found that on the Isle of Man (UK) radon concentrations were highest during the winter months indoors and highest during the summer months outdoors. No explanation was given. Majborn 1990, in Denmark, found the “normal” seasonal variation of highs in winter and lows in summer in a study of 10 single-family houses. He found a strong positive correlation between average indoor-outdoor temperature difference and indoor radon. He also found that radon in basements during winter was about 35% higher than radon in basements during the summer. Growing information on karst geology areas suggests this to be another source of variability for radon occurrence. The tendency in these areas is for a reverse of the “normal” pattern of winter high and summer low radon occurrence. However, even in karst areas not all homes are affected by this “reversal effect.” Which homes are affected may depend on their degree of connection to the karst solution cavities and their location in relation to topographic features (Smithard, 2009 and McNees and Roberts, 2004). Arvela et al. 1994 Radon in Eskers in Finland. Eskers are long narrow steep-sided ridges formed by glacial streams. The soil is permeable sand and gravel, which allows significant soil gas flow. In these areas air above 0 degrees C flows out of the ground in the winter and into the ground in the summer. Indoor radon is affected by permeable soil, subterranean airflows and wind effects hitting the slope. During the summer, the ambient air is warm and the subterranean air is cool, thus the airflow is from top of the slope to bottom. During the winter, the ambient air is cold and the subterranean air is warmer, thus the airflow is from bottom of slope towards top. Large errors can occur in predicting indoor radon if you measure the homes at the bottom of the slope in the wintertime (underestimate by factor of 3-4) and the homes at the top of the slope in the summertime (2-10% of annual average). Thus, radon variability is complex; permeable soil, wind effects; location on slope and temperature, all affect variability; however, temperature is the dominant factor affecting airflows in the eskers and the annual variations in the indoor radon concentration. September 14, 2009 170 Valen et al., 2000 Found the same effect in Norway in a glacial valley. There is a movement of relatively warm soil air towards the higher areas during winter giving rise to high radon content in the ground in the topographical elevated areas, while the lower areas are aerated. In summertime, the process is reversed, giving rise to high radon content in the lower parts of the area. Nero et al, 1983 No correlation was seen between radon concentration and air change rate in three groups of houses. Differences of radon concentration among the group are due to source strength. However, for a given source strength, the indoor concentration can be expected to depend largely on the ventilation rate. This correlation does not hold up in the real world, because there are so many other variables; source strength, soil characteristics, foundation type, house characteristics, ventilation rate and meteorological conditions. Nazaroff et al, 1985 In a house with a crawl space, a modest drop in barometric pressure and a period of heavy rain caused the indoor radon and crawl space radon to rise to its highest level during a 5-week measurement period. The rain may be acting in one of two ways; it could act by funneling the radon from the soil into the crawl space: with heavy rain, the permeability of the soil surrounding the house is greatly reduced while the permeability of the soil beneath the house remains unchanged; as the barometric pressure falls, soil gas then flows into the crawl space at a higher rate then it does out of the soil surrounding the house. The alternative explanation is that the downward movement of water through the soil may act like a piston and displace the radon, which then flows into the crawl space. Soil moisture: Radon emanation from soil grains and its transport through interstitial spaces are significantly affected by soil moisture content. Radon exhalation from a completely dry soil was at its lowest due to the reduced emanation coefficient, and was at its lowest for saturated soil, due to its low diffusion coefficient. Thus, it would seem that somewhere in between you would have the highest exhalation rate. Other factors affecting radon exhalation from soil are meteorological variables (barometric pressure, wind speed, relative humidity), vegetative cover, pressure effects, particularly the pressure difference between pore space air and outside air. A 1% difference in barometric pressure results in a 60% change in the exhalation rate. Temperature effects produce two types of behavior in soils; the first type is the widely observed diurnal variations that result from competition of convective flow due to temperature differences in soil from day to night and from turbulent mixing in the atmosphere, which leads to an increase in exhalation in daytime and a reduction at night. The second type of behavior results from the direct heating of the soil (Collé et al., 1981). Stranden et al., 1984 also found a temperature effect on soil radon exhalation, with an September 14, 2009 171 increase in exhalation with increasing temperature. Physical adsorption of gases on solids is known to be temperature dependent; therefore, an increase in temperature caused a decrease in the adsorption of radon on the soil grains, with an increase in exhalation. This effect is not as significant as the moisture effect on exhalation. A good value to quote for radon exhalation from soil is 0.43 pCi/m2/sec. However, meteorological variables play only a minor role in soil radon exhalation compared to the soil characteristics. Stranden et al., 1984 Three major effects of soil moisture causing an increase in radon exhalation; the direct recoil fraction of the emanation power is increased when there is a fluid present in the internal pores of the material, the fluid may hinder adsorption of radon gas on internal surfaces of the material, and with a soil moisture content gradient in the sample, active transport of radon on water molecules may take place. On the other hand, water present in the internal pores reduces the diffusion of radon out of the material. The radon diffusion coefficient for water is 0.00001 cm2 sec-1 compared to the radon diffusion coefficient for air, which is 0.01 cm2 sec-1. Thus, up to a certain point of soil moisture content the increasing effects are dominating. After an optimum moisture content the reduced diffusion due to the water will dominate, and exhalation will decrease. Schubert, M. et al., 2002 Radon at the soil-air interface. Again, soil moisture is found to be a crucial factor in radon variation, and soil moisture is affected by meteorological conditions. The highest radon concentrations at the soil-air interface, and at depth of 5 cm, was found in early morning hours, and lowest values emerge in the afternoon. These highs and lows are due to temperature gradient at air-soil interface and to wind speed. The highest wind speeds appear at about noon and in the early afternoon and the lowest ones at about midnight and in the early morning hours. During the night and early morning hours, the temperature difference between soil and air is positive, that is the soil is warmer than the air directly above it, thus an upward directed convective radon flux enhances the overall diffusive transport leading to the early morning maximum. As soon as the outside air temperature is higher than the soil-gas temperature the situation is reversed, because the convective flux is now downward and this reduces the upward diffusive flux, leading to the minimum radon concentrations in early afternoon. This occurs only at the soil-air interface and at very shallow depths. Below ~ 30 cm no diurnal effects are seen. Levesque, B. et al. 1997 894 Quebec residences show a lognormal distribution and summer values are lower than winter values. Concentrations found in basements are clearly higher than those found on first floors. First floor to basement ratio is 0.59, i.e. basement 10 pCi/L September 14, 2009 172 first floor 5.9 pCi/L. They did not see any difference between first floor and second floor. Two other factors contributed to increased radon; building on hilly terrain and a fireplace increased radon in basements. Fisher et al., 1998 found in Iowa homes that the first floor/basement ratio is slightly different for one- and two-story homes; 0.61 and 0.53, respectively. For twostory homes, the second floor was about equal to the first floor, with a second floor/first floor ratio of 1.02. Lenzen, M. et. al. 1999 Showed a direct connection between radon-222 concentration and the earth tides at a period of 12.4 hours, showing a positive correlation. This relation may be due to a compression of the pore space due to the tidal compression, causing a porous flow of radon-bearing gas across the rock-air interface. This was observed in a gypsum mine in Luxembourg. Diurnal tides run on a 12-hour cycle with a high and low. Marley, 2001 Investigated several homes on British Isles. Found seasonal variability contrary to most, where radon was higher in the summer than in the winter by a factor of two to three times. This study found radon variability is primarily dependent on barometric pressure, vapor pressure and wind variation (including direction relative to the building concerned), with barometric pressure being the primary factor. Hintenlang, et al., 1992 It is typically assumed that radon entry is due to pressure-driven flow with indoor pressure being lower than sub-slab pressure. Circumstances can occur where the highest concentration of indoor radon will correspond when the house is under neutral pressure conditions. A semidiurnal variation of barometric pressure is a well documented result of atmospheric tides resulting from solar heating and coriolis forces (an apparent force that as a result of the earth’s rotation deflects moving objects) on the earth. These barometric pressure changes are not related to changing meteorological conditions. This oscillation of barometric pressure produces a natural pumping action on soil gas into houses by inducing small indoor/sub-slab pressure differentials. This mechanism provides a means of pumping radon into a house without depressurizing the interior of the house relative to the outdoors. Additionally, since the indoor/outdoor pressure equalizes so quickly there is little infiltration of outdoor air. September 14, 2009 173 Steck, 1992 This author found that a significant spatial variation can exist within a house. The most significant spatial variation was found in basements, most likely due to point sources. Much less variation was found on first and second floor. Marley, 1999 Radon variability can to some degree reflect occupancy of a building. Consider a school HVAC operation. Unit ventilators typically come on an hour or two prior to start of class and then shut down around 4 PM during the weekdays. On the weekends, the units may not operate at all. Intermittent operation of the AC system reduced radon by a factor of four, and operation of the central heating (CH) reduced radon by 40%. The operation of the CH had the effect of increasing the internal temperature with a corresponding small increase in pressure. If these two mechanical devices were not present, the radon variability would be more directly and consistently determined by general atmospheric conditions. In the summer, the AC air is more dense then outdoor air. In theory, flow-reversal due to a gravity-driven outflow of air may reduce radon influx. However, this pressure difference is very small and could be easily overshadowed by meteorological factors (Hoffman, 1995). We have not mentioned anything about progeny; however, mechanical devices would have a pronounced effect on progeny distribution. Matthews et al., 1990 found that operation of heating and air-conditioning systems in two unoccupied houses resulted in a three-fold increase in transport rate of tracer gas from crawl space to living area. Nazaroff et al., 1985 has shown that fireplace operation can lead to house depressurization and increased radon entry but that effect can be masked by geological factors affecting the availability of radon in the near-surface soil. Chittaporn, et al. Showed that variability of basement radon was associated with air exchange rate. The lowest air exchange rate was during highest outdoor temperature (summer) and the basement radon was the highest. Conversely, colder outdoor temperatures yielded larger pressure differences yet lower basement radon concentrations. It was also found that drops in barometric pressure caused transient radon surges into the basement. Cohen, et al., 1988 A very large sample size showed winter radon about 60% higher than summer radon for the living areas of the home. Spring and fall radon were about 40% higher than summer radon. The winter/summer ratio (living area) for PA was 1.86. Basements September 14, 2009 174 were about 2.5 times higher than other rooms, with this difference being larger in the summer and smaller in the winter. Miles, 1998 A 2000 house survey in the UK showed a clear pattern of high indoor radon in winter months and lower radon during summer months. A similar result was found in Swedish homes. However, in both cases, some houses showed just the opposite effect, but this was only in a small number of homes. Hans, et al., 1985 A study of homes (68) in Butte, MT presented some limited data showing daily cycle of radon with indoor minimums during daylight hours and maximums during nighttime hours. This cycle was more pronounced in the summer and less pronounced in the winter. The occurrence of the large cycles was caused principally by alteration of the ventilation rates. Hans, et al., 1986 EPA Report Found the low to occur during the warmer months, corresponding to increased ventilation rates and possibly due to less radon in the soil gas due to more exhalation to the atmosphere. Ground freezing lowers radon exhalation rates, which increases belowgrade concentrations available for transport and diffusion into homes. The maximum indoor concentration was found to occur in December and the minimum in August. The study included twenty homes in Butte, MT. Borak, et al., 1989 A survey of 110 homes in Fort Collins, CO found radon in summer to be about 40% of the annual average, and radon in the winter was about 1.7 times higher than the annual average. The radon concentration in basements was two times higher than on ground floors; however, there were no differences between ground floors and second floors. Porstendorfer, et al., 1994 Outdoor radon also shows a diurnal cycle, with its highest activity during the night and early morning hours when the atmosphere is most stable. At noon and early afternoon, the mixing of the lower atmosphere is strongest and radon concentration is lowest. Indoor radon is mainly influenced by source strength and air change rate, and both of these factors can change with meteorological conditions, occupant activities, mechanical ventilation and heating systems. September 14, 2009 175 Conclusions A lot of information has been presented. Sometimes there are differences from one study to another, and sometimes one variable is looked at more closely than another. The above information has come from research studies, conducted usually with relatively small sample size and at different locations in the country and throughout the world. There can certainly always be exceptions to the conclusions drawn regarding radon variation from one house to another and under different meteorological and occupant conditions. Below is a general compilation of the variables that affect radon variability. Soil moisture: The moisture content of soil and its affect on radon exhalation run from dry soil where exhalation is reduced, to saturated soil where exhalation is also reduced, to moist soil where radon exhalation is optimal. Wind: Wind can affect both the structure and the soil-gas concentration. Wind can have a depressurizing effect on the basement, with soil-gas entry being dependent on soil permeability. Wind can also induce convective flux of soil gas out of the soil causing a depletion zone. Temperature: Temperature, like wind, can affect both the structure and soil-gas concentrations. Inside/outside temperature difference across the building shell results in a stack effect. A type of “stack effect” also operates in soil where in the early morning, when the soil is warmer than the air, there is an upward convective flow. The opposite also occurs when air is warmer than the soil. Increased soil temperature would tend to decrease adsorption of radon on soil grains and cause increased exhalation. Barometric Pressure: A decrease in barometric pressure allows more radon to easily exhale from the soil surface. However, this drop may not affect the basement radon concentration unless the pressure drop is sustained long enough to affect the soil pore air space below the foundation. A barometric-pressure drop plus rain seems to affect basement radon more than just a barometric-pressure drop. Rain: Enough rain to saturate soil, after some time, reduces surface exhalation and leads to increased radon concentration at depth. Frozen Ground: Primarily reduces soil-gas flux, but does not stop flux completely. Seasonal Variation: Generally, indoor radon is higher in winter and lower in summer. Diurnal Variation: The indoor radon concentration tends to be higher during the night and early morning and lower radon during noon to early afternoon. House Level: Basement concentration is about twice what is found on the first floor. First floor and second floor are about equal. The basement to first floor ratio does change slightly with season. September 14, 2009 176 This paper identified about thirteen factors that can affect radon variation in the soil and house environment. The thirteen factors being soil moisture content, soil permeability, wind, temperature, barometric pressure, rainfall, frozen ground, snow cover, earth tides, atmospheric tides, occupancy factors, season and time of day. One can see the complexity of understanding and studying radon variability in homes. References Al-Ahmady, K. and Hintenlang, D. Assessment of Temperature-Driven Pressure Differences with Regard to Radon Entry and Indoor Radon Concentration. AARST International Radon Symposium; 1994. Arvela, H., Voutilainen, A., Honkamaa, T. and Rosenberg, A. High Indoor Radon Variations and the Thermal Behavior of Eskers. Health Physics 67:254-260, 1994. Borak, T., Woodruff, B. and Toohey, R. A Survey of Winter, Summer and AnnualAverage Rn-222 Concentrations in Family Dwellings. Health Physics 57, 465-470, 1989. Brady, N. The Nature and Properties of Soils, 8th Edition. MacMillan Publishing Co., Inc. 1974. Chittaporn, P. and Harley, N. Indirect Measurement of Sub-floor Radon Using Passive Monitors. Cohen, B. Variation of Radon Levels in U.S. Homes Correlated with House Characteristics, Location, and Socioeconomic Factors. Health Physics 60, 631-642, 1991. Collé, R., Rubin, R., Knab, L. and Hutchinson, J. Radon Transport Through and Exhalation from Building Materials: A Review and Assessment. U.S. Department of Commerce, National Bureau of Standards. 1981. Fisher, E., Field, W., Smith, B., Lynch, C., Steck, D. and Neuberger, J. Spatial Variation of Residential Radon Concentrations: The Iowa Radon Lung Cancer Study. Health Physics 75, 506-513, 1998. Fleischer, R., Mogro-Campero, A. and Turner, L. Indoor Radon Levels in the Northeastern U.S.: Effects of Energy-Efficiency in Homes. Health Physics 45, 407412, 1983. Garbesi, K., Robinson, A., Sextro, R. and Nazaroff, W. Radon Entry into Houses: The Importance of Scale-Dependent Permeability. Health Physics 77, 183-191, 1999. September 14, 2009 177 Grainger, P., Shalla, S., Preece, A. and Goodfellow, S. Home Radon Levels and Seasonal Correction Factors for the Isle of Man. Physics Medical Biology 45. 22472252. 2000. Hans, J.. Lyon, R. and Israeli, M. Temporal Variation of Indoor Radon and Radon Decay Product Concentrations in Single Family Homes. Health Physics Society, Midyear Symposium. 1985. Hans, J. and Lyon, R. Seasonal Variations of Radon and Radon Decay Product Concentrations in Single Family Homes. U.S. Environmental Protection Agency, EPA 520/1-86-015, August 1986. Harley, N. and Gerrilli, T. Factors Controlling Indoor Radon Levels, Annual Report to the U.S. Department of Energy, May 31, 1984. Hernandez, T., Ring, J. and Sachs, H. The Variation of Basement Radon Concentration with Barometric Pressure. Health Physics 46, 440-445, 1984. Hintenlang, D. and Al-Ahmady, K. Pressure Differentials for Radon Entry Coupled to Periodic Atmospheric Pressure Variations. Indoor Air 2, 208-215, 1992. Hoffman, R. Radon Contamination of Residential Structures: Impact of the “Weather Effect” on the Short-Term Radon Test. AARST International Radon Symposium; 1995. Kojima, H. and Nagano, K. Dependence of Barometric Pressure, Wind Velocity and Temperature on the Variation of Radon Exhalation. AARST International Radon Symposium, 2000. Kraner, H., Schroeder, G. and Evans, R. Measurements of the Effects of Atmospheric Variables on the Radon-222 Flux and Soil-gas Concentrations, pp 191-215. In The Natural Radiation Environment, Eds. John A.S. Adams and Wayne M. Lowder, 1964. Levesque, B., Gauvin, D., McGregor, R., Martel, R., Gingras, S., Dontigny, A., Walker, W., Lajoie, P. and Letourneau, E. Radon in Residences: Influences of Geological and Housing Characteristics. Health Physics 72, 907-914, 1997. Majborn, B. Seasonal Variation of Indoor Radon Concentrations. U.S. EPA, The International Symposium on Radon and Radon Reduction Technology, EPA/600/990/005c, January 1990. Marley, F. Investigation of Atmospheric, Mechanical and Other Pressure Effects Influencing the Levels of Radon and Radon Progeny in Buildings. Health Physics 77, 556-570, 1999. September 14, 2009 178 Marley, F. Investigation of the Air Pressure Characteristics Influencing the Variability of Radon Gas and Radon Progeny in Domestic Vernacular Buildings. Health Physics 81, 57-69, 2001. Matthews, T., Wilson, D., Thompson, C., Monar, K. and Dudney, C. Impact of Heating and Air Conditioning System Operation and Leakage on Ventilation and Intercompartment Transport: Studies in Unoccupied and Occupied Tennessee Valley Homes, Journal of the Air and Waste Management Association, 40, 194-198, 1990. McNees, J. and Roberts, S. Alabama Karst Variability Study. International Radon Symposium. Newport, RI. September 2004. Miles, J. Mapping Radon-Prone Areas by Lognormal Modeling of House Radon Data. Health Physics 74, 370-378, 1998. Minkin, L. and Shapovalov, A. Indoor Radon Entry: 30 Years Later. Iranian Journal of Radiation Research, 6(1):1-6, 2008. Mueller Associates. Indoor Air Quality Environmental Information Handbook: Radon. Prepared for U.S. Department of Energy, DOE/PE/72013-2, January 1986. Nazaroff, W. and Doyle, S. Radon Entry into Houses Having a Crawl Space. Health Physics 48, 265-281, 1985. Nazaroff, W., Feustel, H., Nero, A., Revzan, K., Grimsrud, D., Essling, M; and Toohey, R. Radon Transport into a Detached One-story House with a Basement. Atmospheric Environment, 19, 31-46, 1985. Nazaroff, W. and Nero, A. Radon and Its Decay Products in Indoor Air. John Wiley & Sons, Inc. 1988. Nero, A., Boegel, M., Hollowell, C., Ingersoll, J.,and Nazaroff, W. Radon Concentrations and Infiltration Rates Measured in Conventional and Energy-Efficient Houses. Health Physics 45, 401-405, 1983. Porstendorfer, J., Butterweck, G. and Reineking, A. Daily Variation of the Radon Concentration Indoors and Outdoors and the Influence of Meteorological Parameters. Health Physics 67, 283-287, 1994. Schubert, M. and Schulz, H. Diurnal Radon Variations in the Upper Soil Layers and at the Soil-Air Interface Related to Meteorologocal Parameters. Health Physics 83, 9196, 2002. Smithard, J. Factors Affecting Seasonal Variation in Radon. Environmental Radon Newsletter, Summer 2009, issue 59. Health Protection Agency, UK. September 14, 2009 179 Steck, D. Spatial and Temporal Indoor Radon Variations. Health Physics 62, 351-355, 1992. Stranden, E., Kolstad, A. and Lind, B. Radon Exhalation: Moisture and Temperature Dependence. Health Physics 47, 480-484, 1984. Tanner, A. Radon Migration in the Ground: A Review. Pp 161-190. In The Natural Radiation Environment, Eds. John A.S. Adams and Wayne M. Lowder, 1964. Valen, V. and Soldal, O. Variations in Radon Content in Soil and Dwellings in the Kinsarvik Area, Norway, are Strongly Dependent on Air Temperature. AARST International Radon Symposium, 2000. September 14, 2009 180 Radon in the Workplace OSHA’s primary mission is to provide for the safety of the American workers. OSHA regulations do not apply to the residential environment. OSHA Ionizing Radiation Regulations: 29 CFR 1910.1096 Exposure Limit for Rn-222 in a restricted area: 100 pCi/L (this is the value found in the 1970, 10CFR20, Appendix B, Table 1, Column 1, for occupational exposure). Exposure limit is based on average conc. for 40-hour week. May be proportionally increased or decreased, depending on the 40-hour work week. The employer shall perform surveys and measure the concentration of radioactive material present, where employees are exposed to said radioactive material. Posting Requirements: A room with concentration in excess of 100 pCi/L, or an occupied room with average conc. during occupancy > 25 pCi/L, shall be conspicuously posted, “Caution, Airborne Radioactivity Area.” Personnel monitoring equipment is required in restricted areas if the employee is likely to receive in any calendar quarter a whole body dose in excess of 0.31 rem (25% of the calendar quarter limit of 1.25 rem). Personnel monitoring would also require that the employer maintain records of personnel radiation exposure. Types of Areas: Unrestricted Area: < 3 pCi/L Where an employee continuously present would not receive in any one hour a dose in excess of 2 mrem or a dose in any seven consecutive days of greater than 100 mrem. Where airborne radioactive material does not exceed the limits found in 10 CFR 20, Appendix B, Table 2, Column 1, 1970 for effluent releases). This value may be averaged over a period of not greater than one year. September 14, 2009 181 Restricted Area: > 3 pCi/L. Any area for which access is controlled by the employer for purposes of protection of individuals from exposure to radiation or radioactive materials 1910.1096(a)(3). No individual in a restricted area may receive, in any period of one calendar quarter, a dose in excess of 1.25 rem to the whole body, head and trunk, active blood forming organs, lens of eyes, or gonads; 18.75 rem to the hands and forearms, or feet and ankles; 7.5 rem to the skin of the whole body 1910.1096(b)(1). These values may be exempted if during any calendar quarter the dose to the whole body does not exceed 3 rems; and the dose to the whole body, when added to the accumulated occupational dose to the whole body, shall not exceed 5(N-18) rems, where “N” is the individuals age in years. The employer must also maintain past and current exposure records 1910.1096 (b)(2) (i,ii,iii). No one under 18 years of age may receive in one calendar quarter a dose in excess of 10% of the limits specified above (1.25, 18.75, and 7.5 rem) in a restricted area. No employee shall be exposed to airborne radioactive material in an average concentration in excess of the limits specified in Table 1 of Appendix B to 10 CFR Part 20, in a restricted area. For radon-222 this is 100 pCi/L averaged over a 40-hour workweek 1910.1096(c)(1). No one under 18 years of age may be exposed to airborne radioactive material in an average concentration in excess of the limits specified in Table II of Appendix B to 10 CFR part 20, in a restricted area 1910.1096(c)(2). For radon-222 this is 3 pCi/L averaged over a period not greater than one week. An employee who enters a restricted area who receives or is likely to receive a dose in any calendar quarter in excess of 25% of the values specified above (1.25, 18.75, and 7.5 rem) shall wear personnel monitoring equipment for the purpose of measuring the dose received 1910.1096(d)(2)(i). Radiation Area: > 5 mR/hr High Radiation Area: > 100 mR/hr Airborne Radioactive Area: >100 pCi/L or > 25 pCi/L (occupied area). This area shall be conspicuously posted, “Caution, Airborne Radioactivity Area.” September 14, 2009 182 Diagnostics Diagnostics (our definition): A series of questions, observations and measurements designed to assess the cause of a problem. Visual Inspection Homeowner Questions Building Material Surface Radon Flux Radon in Water Grab Samples PFE Mapping Air-Flow Measurements Blower-Door Tests Photos Potential Failure Modes: Inadequate PFE, Untreated Sources, Reentrainment, Fan failure, Radon in Water, Building Materials, Outdoor Radon. One of main entry points of radon into homes is via the wall/floor joint. Studies have shown that once the gap width exceeds 0.5 mm (~1/32”), there is no longer a significant increase in radon entry, with increasing gap width (UNSCEAR, 2000). Forced air distribution systems can influence radon in at least two ways (Turk, 1988): 1. Leaky returns in basement have been observed to depressurize basement as much as 10 Pa. 2. They can transfer large amounts of basement radon to the upper floors. The soil temperature surrounding the building foundation significantly influences (positively and negatively) the pressure difference that drives radon into buildings (Turk, 1988). September 14, 2009 183 Stack Effect: The movement of air into and out of buildings, driven by the buoyancy of air. Buoyancy occurs due to a difference in indoor-to-outdoor air density resulting from temperature and moisture differences. The movement of air in the building can be either up or down. The driving force for air movement is the pressure difference between the inside and outside of the building. That pressure difference can be calculated with the following equation: Ps = 0.52 PH (1/To – 1/Ti) where Ps = Total pressure difference caused by stack effect, inches of water column. P = Ambient pressure, psia H = Building height, ft To = Absolute temperature outside, in Kelvin Ti = Absolute temperature inside, in Kelvin The neutral pressure plane is where the inside and outside pressures are equal. The neutral pressure plane can shift up or down depending on how the leakage openings of the building are distributed, top to bottom. Infiltration occurs below the neutral plane and exfiltration occurs above it. Interestingly, in the summer, when the outside air temperature is higher than that inside the pattern of pressure differences and air flow is the reverse of that during the winter. Infiltration occurs at the upper floors, and exfiltration at the lower levels, with air flowing downward within the building. The stack effect is much reduced (Wilson, 2004). Pressure Field Extension (PFE): The maximum distance from a suction point where the sub-slab volume remains depressurized to a magnitude capable of preventing soil-gas flow into the building (Hintenlang, 1991). PFE is a function of applied pressure at the suction point, pit size and sub-slab permeability (Hintenlang, 1991). In soils with poor communication (low permeability) the pressure field may take minutes to hours to extend to its fullest point (Hintenlang, 1991). The creation of the suction pit, immediately below the suction pipe, has been empirically demonstrated to enhance pressure field extension (Hintenlang, 1991). If you double the applied pressure at a suction point you will approximately double the pressure at a given test hole some distance away. However, you will not appreciably increase the distance of the pressure field. To do this you most likely need to add another suction point (Hintenlang, 1991). September 14, 2009 184 Radon diffusion through concrete can be a source of indoor radon. Radon can diffuse through intact concrete at a rate of about 1/10,000,000 to 1/100,000,000 per meter squared per second. Diffusion may be more significant in low-permeability soils, where there is less air-driven seepage through cracks (Rogers, 1994). On the Subject of U-tubes I recently had a useful conversation with a mitigator looking for help with diagnostics on a home he was working on. This contractor made the comment that he was looking at the U-tube and then making a decision on what to do with the current fan he was using. This got me thinking about U-tubes. First, U-tubes should not be used to make decisions about fan choice. The primary function of the U-tube is as an indicator device, primarily for the homeowner. The contractor should mark the U-tube reading just after installation and note this on the label. The homeowner then has a reference value with which to compare future readings. Let’s look at the two extremes: 1. You have a U-tube reading that is very “high.” Let’s say 2” WC. This does not necessarily mean that you are using the wrong fan. It could mean that that the pipe has fallen down and embedded in the soil and cut off all of your flow. In this case, you obviously don’t need a different fan, you just need to remove the blockage. It could also mean that you are dealing with very low-permeability soil. This would result in very low flows and high vacuum readings. This may suggest a higher vacuum fan, but other things should also be done with pit size and additional suction points. 2. Now we have a U-tube reading that is very “low.” Let’s say 0.1” WC. This could mean that you have very high-permeability soil. This would probably mean that you want a fan to move more air. You could have leaks in the piping system or the foundation. This would result in high airflows and low vacuum readings. Obviously sealing the leaks would help with this problem. You could have short-circuiting from a suction point through a footer at the walkout side of a basement. This would suggest that you want to move the suction point to a more central location. September 14, 2009 185 So, as you can see there can be multiple reasons for high or low U-tube vacuum readings, and some of them have nothing to do with the fan. These examples are good. You could elaborate further by pointing out that changes in the u-tube reading over time do not necessarily indicate whether the changes in conditions, which have resulted in the u-tube change, are favorable or unfavorable in terms of system effectiveness. What I like to focus on when explaining the usefulness of looking at the u-tube is that there is no particular relationship between the u-tube reading and the effectiveness (radon control) of the system(or of the PFE values) . This is true not only from one system to another, but for a particular system from one time to another. The strength and extent of the pressure field is far more indicative of potential effectiveness, and that is the pressure value mitigators should focus on, whether in system design or troubleshooting. (Personal communication, Jack Hughes, Southern Regional Radon Training Center) References: Hintenlang, D.E. and Barber, J.M. Practical Engineering Analysis for Indoor Radon Mitigation Using Sub-slab Depressurization Techniques. Radiation Protection Management, Vol. 8, No. 1, pp. 63-77, 1991. Rogers, V.C., K.K. Nielson, R.B. Holt and R. Snoddy. Radon Diffusion Coefficients for Residential Concretes. Health Physics, Vol. 67, No. 3, Sept. 1994. Turk, B.H, Prill, R.J. and Sextro, R.G. Intensive Radon Mitigation Research: Lessons Learned. The 1988 Symposium on Radon and Radon Reduction Technology, October 17-21, 1988, Denver, CO. United Nations Scientific Committee on the Effects of Atomic Radiation, UNSCEAR 2000 Report, Volume 1: Sources, Annex B. Wilson, A.G. and Tamura, G.T. Stack Effect in Buidlings. Canadian Building Digest, CBD-104, 2004. September 14, 2009 186 Fan Selection The fan must be able to maintain an appropriate suction by producing an adequate airflow in the system. Typical airflows for residential applications range from ~ 20 to 100 cfm. The ideal situation is to have a tight foundation, with a good aggregate layer below the slab, and a fairly impermeable soil around the foundation. In this situation, fairly low airflows may be required to control soil gas entry. This ideal situation would probably require a very low system to produce the necessary airflow. The fan is the integral part of the sub-slab depressurization system providing the work of moving air from one location (under the slab) to another location (the atmosphere). This movement of air creates a low-pressure area under the slab relative to the basement. Fan selection involves comparing fan operating characteristics with the performance requirements (required airflow) and resistance characteristics of the system. One does not want to consider just flow or just pressure when deciding on a fan. The use of a fan curve and system curve supplied by the manufacturer will allow for appropriate fan selection. However, few mitigators take the time to produce a system curve. Be aware that the manufacturer fan curves are derived under laboratory conditions. At least one experienced mitigator believes that the quantitative diagnostics is by far the most appropriate approach to fan selection and total system design. This leaves one with two other options for fan selection; manufacturer-supplied qualitative information and field experience. The manufacturer’s information may state, “good for low-flow applications,” “prepiped new construction,” “good communication and small foot print,” etc. Field experience comes from installations in a given area and your success with a certain fan. This method may provide for good radon control, but it may not provide for the most appropriate system, i.e. those that produce the required indoor radon reduction without creating unnecessary system or building operating costs, or creating other safety hazards or detrimental conditions (Hughes, 2009). Fan selection continued; Manufacturer’s Information Dave Kapturowski of RadonAway: For good communication, where you are ventilating the soil and typically moving less than 100 cfm of air the RP or XP series fans should fit the bill. All of the current RP and XP fans will move this amount of air. Since they should all do the job, you may want to also consider the physical size of the fan and its power consumption to narrow your choice. XP 151, XP 201, XR 261, RP 140, RP 145. 3” pipe OK For good communication, but where you may encounter flows greater than 100 cfm, such as a walkout basement or basements where foundation openings cannot be sealed, then go with the RP 260 or RP 265. 4” pipe preferred A very good choice for RRNC would be the low-power RP 140, it is very quiet. September 14, 2009 187 For tight soil and poor communication, where you may be moving less than 20 cfm of air, you are trying to establish a pressure field, where more static pressure is needed. You most likely will also have multiple suction points. GP201, GP301, GP401, GP501. RadonAway Manufacturer Data Fan Max. Pres./Flow Power Consumption RP140 RP145 RP260 RP265 0.5”/134 cfm 2”/173 cfm 1.5”/275 cfm 2”/327 cfm 14-20 W 37-71 W 52-72 W 86-140 W XP151 XP201 XR261 1.6”/180 cfm 1.9”/125 cfm 1.8”/230 cfm 45-60 W 45-60 W 65-105 W GP201 GP301 GP401 GP501 2”/82 cfm 2.5”/92 cfm 3”/93 cfm 4”/95 cfm 40-60 W 55-90 W 60-110 W 70-140 W All above fans ETL listed for outdoor use. Fantech Manufacturer Data HP2133 ----- Low power and low flow, good communication and low radon HP2190 ----- Similar to HP190, but smaller housing HP175 ------- Good communication, lower radon, small footprint, RRNC, lower wattage than HP190 HP190 ------- Good overall choice, moderate radon, multiple suction points, medium footprint HP220 ------- High radon, high flow, multiple suction points, large footprint, moderate communication September 14, 2009 188 Fan Max. Pres./Flow Power Consumption HP2133 HP2190 HP175 HP190 HP220 0.84”/134 cfm 1.93”/163 cfm 1.66”/151 cfm 2”/157 cfm 2.4”/344 cfm 14-20 W 60-85 W 44-65 W 60-89 W 85-152 W All HP series UL listed for outdoor use Final method of fan selection is to generate a “system curve” and then plot this on graph paper that has multiple fan curves already plotted. See which fan curve intersects the system curve at the appropriate position. Ideally, you would pick a fan for which the system curve intersected the fan curve roughly in the middle of the fan curve. The basis for this selection is that the fan would be operating at a comfortable point, and could handle increases or decreases in flow over time that may occur as the soil dries out or becomes saturated. The above method may be more complex, but it provides for a much more “appropriate” selection and may save contractor costs by using smaller fans and homeowner costs in utility bills over the many years of operation. One would also need a pitot tube, several magnehelics or a digital micromanometer. Below are presented some anecdotal information from some local radon mitigation contractors on their experiences with fans: Please be aware that these approaches are obviously qualitative. Field Experience 1 • • • I use a lot of RP-145’s and RP-140”s. Poured concrete foundations and stone under slab RP-140 works well Wattage should be considered. Over ten years costs can add up. Field Experience 2 • • Basically use two fans RP-145 and RP-140. For RRNC RP-140 is good choice, unless you encounter a lot of airflow that can’t be sealed. September 14, 2009 189 • • • • • For existing homes will often perform diagnostics during install. Cut 5” hole, put RP-145 on hole, measure PFE at distant points. If measured PFE is > 5 Pa, then figure RP-140 will give about 2.5 Pa, which should be enough. If < 5 Pa, use RP-145. Often can get a 50-Pa differential pressure with RP-140. Sometimes just take a chance and guess; however, fan choice ideally depends on diagnostics. Very occasionally (10-12/yr) will use HP-220 or RP-265 for high-flow applications. Only very occasionally will use GP-501. Seldom need to draw 4” of vacuum. Too expensive. Field Experience 3 • • • • • • • • • First dig hole to see what is under slab. Use two fans; XP-151 and GP-501. Lots of gravel and loose fill use XP-151. Tight soil use GP-501. The ultimate determination is the reading on the U-tube. Sometimes for lots of airflow and higher vacuum use RP-265. For RRNC, with sub-slab loop use XP-151. When doing the installation it’s a seat of the pants thing. The bottom line is the confirming test. Sometimes I will change a fan to get to the right number. Field Experience 4 • • • • • • • • • • • Use XR-261 94% of time, existing homes and RRNC. Use RP-265 2% of time, where I need more airflow than XR-261. Use GP-501 0.5% of time, where low permeability fill and/or finished basement. Where you can’t place suctions points where you want. Use GP-401 3.5% of time, usually older homes, and > 10 pCi/L. Crews rate sub-slab permeability on scale of 1 to 10. 1 is damp clay, 10 is abundant, clean large gravel. XR-261: 6 to 10 RP-265: 6 to 10, large area, high air flow GP-401: 2 to 5, tight substrate GP-501: 1, very tight substrate GP-501 not often used by experienced mitigator Noise: XR-261*, RP-265***, GP-401**, GP-501***** Some Fan Facts September 14, 2009 190 Maximum fan pressure is often stated two ways; maximum fan pressure and maximum fan operating pressure. Maximum fan pressure is the reading on the u-tube when the fan is not moving any air. This will cause the fan to run hotter and shorten life. Maximum fan operating pressure is the reading on the u-tube when about 20 cfm of air is moving through the fan. This airflow will help extract heat from the fan. Every fan should be operating below its specified maximum fan operating pressure. If a fan is operating at its maximum fan pressure, steps should be taken to introduce air into the system. Within a certain range for each fan model, fan power consumption is determined by the amount of air the fan is moving. The more air moving through the fan the more power required. When there is no air moving through the fan blades, the motor is coasting. When given a range of fan power consumptions (i.e. 60 to 90 watts), this means that at maximum suction the fan will consume 60 watts, because it is doing the least amount of work, and at maximum flow it will consume 90 watts, because it requires more power to move air. One can get a very rough idea of fan power consumption from the u-tube, and the above relation. Say a fan is rated from 0 to 2” WC, and 60 to 90 watts. If the u-tube was showing zero, there was basically no vacuum and the fan would be moving the most amount of air; therefore, the power consumption would be closer to 90 watts. If the utube was showing 2” WC, there would be very little airflow and the fan would be consuming closer to 60 watts. Stacking fans, that is running them in series, can be done in field applications when additional airflow is required and higher vacuum is needed to produce that airflow. When you stack two fans you will double the maximum static pressure that one fan would produce, but the actual static pressure developed and the airflow increase will vary. You may only increase the airflow a small amount. This is because airflow is dependent on other things besides the fan; such as, sub-slab material, foundation openings and suction-point location. Stacking fans can be useful in dealing with tight soil. When stacking two fans you want to use identical fans. Using different fans can cause cavitation problems and nonlaminar flow problems, both of which can cause inefficiency problems. Also, leave about 15 pipe diameters of separation between the two fans that will be stacked. This also helps preserve laminar flow. Condensation can be the single most significant factor in fan failure. According to Dave Kapturowski of RadonAway the amount of condensation produced is dependent on the climate, soil conditions, ducting above the fan and the ducting material. In fact, Dave has calculated that under certain conditions the typical residential system can produce about one gallon of water per day due to condensation. September 14, 2009 191 How to deal with condensation: Mount the fan in a location that minimizes the length of ducting above the fan, use the least amount of pipe in unconditioned spaces, consider using schedule 40 piping, or even insulated piping, be careful not to create any low spot in your piping network, assure that there is adequate slope on piping for condensation to drain back to the soil and use a condensate bypass mechanism for condensation to drain down and around the fan. Fantech has recently (6/09) announced a new fan with a condensate bypass built in. References: Hughes, J. Personal Communication. 2009 Kapturowski, D. Radon Today. Fall Issue 2004 Kapturowski, D. Radon Today. Winter/Spring Issue 2005 Kapturowski, D. Radon Today. Summer Issue 2005 September 14, 2009 192 Radon Fan Comparison List Fan Duct Dia. (in.) HP CFM @ 0”WC “WC @ 0 CFM 0.019-0.027 0.050-0.095 0.070-0.095 0.115-0.188 0.138-0.209 0.064-0.100 0.087-0.141 0.053-0.065 0.060-0.080 0.060-0.087 0.053-0.080 0.073-0.121 0.080-0.148 0.080-0.188 0.094-0.174 0.201-0.362 0.141-0.261 0.241-0.429 134 173 275 327 510 215 230 125 180 125 82 92 93 95 ~90’s 110 40 53 0.8 2.1 1.8 2.5 2.2 2.1 1.8 1.2 1.6 2.0 2.0 2.6 3.4 4.2 4.39 (flat box) 18 (high suction) 27 (high suction) 50 (high suction) 134 134 163 151 157 344 122 167 148 214 263 263 289 289 408 429 649 134 263 0.84 0.84 1.93 1.66 2.01 2.46 0.87 1.6 0.79 1.15 1.58 1.58 2.32 2.32 2.14 (high flow) 2.48 (high flow) 2.58 (high flow) 1.26 1.58 MANUFACTURER MODEL WATTS RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RadonAway RP140 RP145 RP260 RP265 RP380 XR161* XR261 XP101* XP151 XP201 GP201 GP301 GP401 GP501 GP500 HS2000 HS3000 HS5000 14-20 37-71 52-72 86-140 103-156 48-75 65-105 40-49 45-60 45-66 40-60 55-90 60-110 70-140 70-130 150-270 105-195 180-320 4 4 6 6 8 Fantech Fantech Fantech Fantech Fantech Fantech Fantech Fantech Fantech Fantech Fantech Fantech Fantech Fantech Fantech Fantech Fantech Fantech KTA HP-2133 LV-2133 DC HP-2190 HP175 HP190 HP220 FR100 FR110 FR125 FR140 FR150 FR150 DC FR160 FR160 DC FR200 FR225 FR250 ECL452 KTA150 DC 14-20 14-20 60-85 44-65 60-85 85-152 13-19 58-75 15-18 45-60 54-72 54-72 103-130 103-130 106-128 111-152 146-248 110 54-72 4 4 4 4 4 6 4 4 5 6 6 6 6 6 8 8 10 6 0.019-0.027 0.019-0.027 0.080-0.114 0.061-0.090 0.080-0.114 0.114-0.204 0.044-0.056 0.078-0.100 0.020-0.024 0.060-0.080 0.097-0.125 0.097-0.125 0.130-0.174 0.130-0.174 0.142-0.172 0.149-0.204 0.196-0.333 0.148 0.097-0.125 AMG AMG AMG AMG AMG AMG AMG AMG Spirit Maverick Hawk Prowler Legend Eagle Fury Force 20 75 75 130 150 160 175 302 3 4 6 3 6 3 8 4 0.027 0.114 0.114 0.174 0.201 0.215 0.235 0.188 121 221 295 163 353 124 544 240 0.9 1.88 1.6 2.71 2.6 4.0 2.48 5.51 RAM/GAM Eng. RAM/GAM Eng. RAM/GAM Eng. RAM/GAM Eng. RAM/GAM Eng. RAM II 24VDC RAM 8 24VDC Mini RAM 115VAC RAM II 115VAC Grand RAM 230VAC 4 2 4 4 6 0.051 0.060 0.107 0.107 0.335 195 70 124 195 430 1.9 8.0 1.2 1.6 2.6 Rosenberg * Rosenberg * Rosenberg * Rosenberg * Rosenberg * Rosenberg * Rosenberg * R100 R125 R150 R160 R200 R200L R250 * Models no longer sold. Note: Check manufacturer for exact specs. 6 4 4 3 3 3 3 3 3 inlet/2 out 3 inlet/2 out 3 inlet/2 out 38 max 40 max 20 max 81 max 250 max 50 50 90 90 125 210 330 0.067 0.067 0.121 0.121 0.167 0.281 0.442 Rosenberg lists their watts as input watts. Prepared by Matt Shields, PA DEP 1933/09 Revised 1 atmosphere = 1.013255x10^6 dynes/cm^2 14.696 psia 29.921” Hg 1013 mbar 760 Torr 760 mm Hg 406.782” WC 10.3355 meters WC 33.899 ft WC 101,325 Pa 14.696 psia = 406.782 inches WC 1 psia = 27.680 “ WC 1” WC = 249.09 Pa 1” WC = 2.5 mbar 1” WC = 0.074” Hg 1 mbar = 0.4” WC 1 psia = 6894.7 Pa 4 pCi/l = 0.00002 ppm = 0.02 ppb = 148 Bq/m3 194 Post-Mitigation Radon Data Year-long ATD Data Below is a compilation of PA Radon Division long-term ATD data obtained from our Remedial Program. Our Remedial Program sends out a charcoal and an ATD to homeowners who have provided us proof of having installed an active radon mitigation system. The charcoals are analyzed in our Radon Division lab, and the alpha tracks are analyzed by the alpha-track lab. For this current data set, all of the ATD results were from Landauer. The homeowners are told to initially expose the charcoal in the basement, and if that result comes back less than 4.0 pCi/L, then expose the ATD for one full year. Some do not expose the ATD for a whole year, but most do. This compilation of data is from 1999 to 2008. Mean +/- 1 SD Median Sample Size Range Post-mitigation, year-long, radon test results Basement First Floor 2.1 ± 1.9 pCi/L 1.0 ± 1.4 pCi/L 1.5 pCi/L 0.5 pCi/L 435 158 0 to 14.7 pCi/L 0 to 11 pCi/L The above table clearly shows that on the average the active radon mitigation systems are doing a very good job of radon reduction. Actually, for the case of a lognormal distribution the median is the more appropriate statistic. The sample size is fairly decent for this type of data. The first floor/basement ratio shows 1.0/2.1 = 0.476, which is relatively close to the value we previously determined from the Radon Analyzer data of 0.5. The interesting aspect is that the 0.5 value (radon analyzer data) was mostly from nonmitigated homes; whereas, these data are all post-mitigation data, and the ratio stays very similar. A review of the range values above shows some rather high results for year-long post-mitigation data. From our data we have no information on the installed system and what has transpired in the home, except that is was an active ASD. For the basement results the data show that 88.4% of the results are below 4.0 pCi/L, and 11.5% above. The first-floor data show 96.2% of results below 4.0 pCi/L, and 3.8% above. We do not know what transpired with the results greater than 4.0 pCi/L. It is presumed that homeowners would have called the contractor back for necessary modifications. September 14, 2009 195 The above data impliy that most active ASD systems can reduce basement radon concentrations to the range of 1.5 to 2.0 pCi/L, and that the systems are initially effective (<4 pCi/L) about 90% of the time. We will attempt to do some further data analysis with pre- and post-mitigation charcoal results and from the pre- and post-mitigation data in Oracle, in the future. Dr. Daniel Steck (Steck, 2008) of St. John’s University, MN provides some similar data for post-mitigation radon concentrations in Minnesota homes. The data below come from his Table 2. Landauer Radtrack ATD’s were used. The measurement period for these ATD’s was one-half of the winter season and the spring. Primary Site Average 0.84 pCi/L Median 0.3 pCi/L Sample Size 132 The primary site is composed of 90% basement and 10% first-floor readings. This data set shows both average and median values less the PA Radon Division data from above. Brodhead (Brodhead, 1995), in a survey of nationwide mitigation contractors, found that 94% of survey participants (n= 226) who deployed an alpha-track detector had post-mitigation results less than 4.0 pCi/L, and 70% of the survey participants had results less than 2.0 pCi/L. However, measurement location is not given and the measurement period is for three months. References Brodhead, B. Nationwide Survey of RCP Listed Mitigation Contractors. Proceedings of the 1995 International Radon Symposium, Nashville, TN. Steck, D. Post-mitigation Radon Concentrations in Minnesota Homes. Proceedings of the 2008 International Radon Symposium, Las Vegas, NV. September 14, 2009 196 Ambient Radon Factors affecting ambient concentrations: Snow cover- likely to depress radon gas release into the atmosphere (Steck, 1999). Soil moisture content- 40 to 60% can produce high radon emanation and diffusion from soil (Steck, 1999). Wind direction- This would tend to suggest a change of source with changes in wind direction (Steck, 1999). Soil cover or lack of cover. i.e. bare tilled soil. The air permeabilities of plowed soils are 4 to 1000 times greater than corresponding compacted samples (Steck, 1999 and Ball, 1981). Seasonality- maximum in winter (dry season) and minimum in summer months (wet season), in Brazil. Other data from Japan shows similar pattern. U.S. data show minima during spring, and maxima in late summer and fall (Gesell, 1983). Seasonality effects are primarily related to precipitation, relating to soil moisture, and prevailing winds. Atmospheric conditions- early morning atmospheric temperature inversions lead to a stable atmosphere that restricts vertical mixing. This results in maximum ambient radon during these times. With sunrise, solar radiation heats the ground and causes lower atmospheric warming. This increases vertical mixing, with a decline in radon concentration in the afternoon (Magalhaes, 2003). Local geology- outdoor concentrations can be correlated with different concentrations of radon in soils and uranium and its progeny in rocks. Height above ground- Some (Price, 1994) saw indistinguishable radon values at 0.5, 1.0, and 2.0 meters, others (Moses, 1960 saw a gradient from ground level (0.032 m) and a height of 0.97 m, with values decreasing with increasing height. Soil characteristics such as grain size, mineralogy, porosity and permeability affect how much radon enters the soil gas. There is a mild correlation between soil-gas concentration and radon in ambient air (Price, 1994). Soil-gas concentration- Concentration of radon in air is governed by the source term, which is the concentration in the soil (Gesell, 1983). Barometric Pressure- A 1% fall in barometric pressure can double the emanation rate, which in turn would increase the radon concentration close to the ground surface (NCRP, 1975). September 14, 2009 197 Ambient radon concentration and elevation Radon enters the atmosphere at the soil-air interface. Radon concentration in the atmosphere is governed by the source term and dilution factors. Atmospheric radon exhibits a vertical concentration profile, being highest at the soil-air interface and being immeasurably low in the stratosphere. Over Land vs. Over Oceans Due to the low radium content of the oceans, the ambient radon will always be lower over the oceans than over the land surfaces (NCRP, 1975). Ambient radon concentration over oceans have been measured at 0.001 pCi/L (Wilkening, 1990). Daytime vs. Nighttime Overall averages tend to show that nighttime concentrations are a few times higher than those existing during the day, primarily due to atmospheric stability (NCRP, 1975). Maximum and Minimum ambient Radon-222 concentrations D. Steck found high of 1.5 pCi/L (Measurements in Iowa and Minnesota) D. Hopper found high of 1.1 pCi/L during the national survey. Fisenne and Harley found a low of 0.015 pCi/L in NYC. UNSCEAR 2000 shows a low of 0.001 pCi/L “Typical” Ambient Radon-222 Values D. Hopper established a U.S. median value of 0.39 pCi/L J. Price found median for Nevada of 0.4 pCi/L D. Steck found geometric mean of 0.67 pCi/L for IA and MN. M. Magalhaes found mean Rn-222 EEC of 0.3 pCi/L for Rio de Janeiro T. Borak found geometric mean of 0.4 pCi/L in Ft. Collins, CO. UNSCEAR 2000 reports 0.27 pCi/L as compilation of numerous studies. Ambient Radon-220 Concentrations Very limited data. UNSCEAR 2000 reports value of 0.27 pCi/L similar to the Rn-222 value. D.Steck also shows value of 0.27 pCi/L from IA and MN. M. Doi found 0.25 pCi/L in Chiba City, 50 km east of Tokyo N. Harley measured 0.4 for Central Park and 0.48 for Northern NJ. NRE VII September 14, 2009 198 Radon-222 Equilibrium ratio in outdoor air UNSCEAR 2000 uses rounded value of 0.6, but with a range from 0.2 to 1.0. References Ball, B.C., Harris, W. and Burford, J.R. A laboratory method to measure gas diffusion and flow in soil and other porous materials. J. Soil Science 32:323-333, 1981. Gesell, T.F. Background Atmospheric Radon-222 Concentrations Outdoors and Indoors: A Review. Health Physics, Vol. 45, No. 2, 289-302, 1983. Magalhaes, M.H. Radon-222 in Brazil: an outline of indoor and outdoor measurements. J. of Environmental Radioactivity, 67 (2003), 131-143 Moses, H., Stehney, A.F. and Lucas, H.F. The Effect of Meteorological Variables upon the Vertical and Temporal Distributions of Atmospheric Radon. J. Geophysical Research. 65:1223-1238, 1960. NCRP 45, Natural Background Radiation in the United States, Nov. 15, 1975. Price, J.G., Rigby, J.G., Christensen, L., Hess, R., LaPonite, D.D., Ramelli, A.R., Desilets, M., Hopper, R.D., Kluesner, T. and Marshall, S. Radon in outdoor Air in Nevada. Health Physics, Vol. 66, No. 4, 433-438, 1994. Steck, D. and Yassin, S. Variation of atmospheric radon and radon progeny in central North America. Proc. 5th Int. Conf. On Rare gas Geochemistry, 107-116, 1999. This is for the above four bullets. Wilkening, M. Radon in the environment: Studies in environmental science. No. 40. Amsterdam, Elsevier, 1990. September 14, 2009 199 Radon Statistics The data below are from an unpublished report titled A Statistical Report of Pennsylvania- Radon-222, April 2007. These data were compiled from the PA DEP, Radon Division Radon Analyzer (Lewis, 2007), which is a web-enabled, multidimensional database that presummarizes all of the data from radon testing records into predefined logical categories such as zip code, test values, measurement location in building, house type, etc. Table 1 below shows the evolution, at least from 1990, of the different measurement methods over time. One can see that the activated charcoal (AC) starts out in 1990 with a significant lead, only to be surpassed by the short-term electrets in the mid to late nineties. Table 1 Measurement Method Results by Year Year CR AT AC ES EL LS 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 All Msmt Types 33,351 37,308 41,666 43,956 48,179 57,744 64,138 65,600 81,666 75,026 74,091 75,988 84,473 75,416 1,771 1,795 2,273 3,260 2,526 2,988 4,998 6,199 6,006 10,754 13,129 14,195 16,210 17,071 613 4,127 2,711 2,250 1,404 731 1,014 838 844 1,108 1,212 967 828 1,027 22,365 18,239 19,103 18,177 24,828 29,640 26,917 22,582 34,778 24,184 20,174 21,666 22,709 21,991 8,562 13,029 17,313 19,866 18,969 24,026 30,701 33,207 37,302 35,999 32,188 31,144 35,567 24,835 40 114 266 394 444 356 457 521 511 471 347 271 274 150 --4 --9 8 3 51 2,253 2,225 2,510 7,041 7,745 8,885 10,342 Total 858,602 103,175 19,674 327,353 362,708 4,616 41,076 Qualifications: Table 1 uses all house types, all building locations, all counties, and years 1990 to 2003. Table 2 shows the number of measurements in a particular house location and the associated average radon concentration. The average for the third floor of 4.28 pCi/L is the most difficult to explain. It may be due to the relatively small sample size (n= 1601). September 14, 2009 200 Location Basement First Floor Second Floor Third Floor Slab-on-grade Above Crawl Space Table 2 Sample Size by Measurement Location Average Result (pCi/L) Sample Size 599,580 156,502 30,930 1,601 44,525 6,385 7.11 3.6 2.82 4.28 4.73 2.5 Qualifications: Years 1990 to 2003, all counties, all house types, all measurement types. The results from basements and first floors may not be simultaneous measurements from the same homes. Table 3 shows the distribution of our measurement data by house type. One will note the large sample size (n=273,327) for the “Unknown” building type. This is due to the radon reports being sent to the Department, where for some reason the tester or lab has not annotated a house type. We are forced to classify it as “unknown.” Table 3 Number of Measurement Results by Housing Type Building Type Number of Results Percent Avg. pCi/L 2-Story House 3-Story House Ranch Split Level Bi-Level Cape Cod Townhouse Contemporary Raised Ranch Commercial Bldg. Public Bldg Unknown 355,356 41.6% 5.4 31,656 3.7% 8.5 65,015 7.6% 6.6 26,931 3.2% 4.4 10,994 1.3% 5.6 16,965 2.0% 5.2 55,254 6.5% 2.8 8,161 0.9% 6.4 2,100 0.25% 6.0 6,926 0.8% 3.4 334 0.04% 3.7 273,327 32% N= 853,019 Qualification: All counties, all years 1990-2003, all measurement types, all-building locations. September 14, 2009 201 Table 4 shows both basement and first-floor measurements by month of year. These data are presented below in a graphical format. Table 4, Average Radon by Month, Basement and First Floor Month Bsmt Avg Count 1st Fl Avg. Count 1st fl/Bsmt January February March April May June July August September October November December Avg or Total 9.3 8.1 7.6 7.3 6.6 6.1 5.5 5.9 7.0 8.0 8.6 8.9 7.4 40,435 42,944 53,670 52,277 49,948 46,620 46,029 43,388 40,424 49,284 47,533 35,856 548,408 5.1 4.9 4.4 3.7 3.7 3.2 3.0 3.2 3.8 4.5 5.2 5.1 4.1 12,630 13,812 17,355 17,210 17,157 16,079 14,829 14,232 12,339 13,686 13,310 10,897 173,536 0.55 0.6 0.58 0.51 0.55 0.53 0.53 0.55 0.53 0.56 0.60 0.57 0.56 Qualifications: All short-term test methods included (AC, CR, LS, ES). All single-family house types included plus unknown category. For 1st-floor average, first floors and slab-on-grade were used, second floor was not used. The basement and firstfloor measurements are not necessarily simultaneous measurements from the same house. Radon Concentration verse Month Radon Analyzer Data 10 9 8 7 6 Bsmt 5 1st Fl 4 3 2 1 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month September 14, 2009 202 Table 5 shows a breakdown of data by two six-month seasons. These two time periods were chosen only because Rhode Island had done a similar analysis, and we then had some data with which to compare. They also represent warm and cold weather periods, at least in Pennsylvania. Table 5 Residential Data, Basement and First Floor, Grouped by Season Season Sample Size Average (pCi/L) April thru September October thru March 348,240 335,565 5.72 7.63 Table 6 gives a finer resolution on the Table 5 above. Table 6 Residential Data, Basement and First Floor, Grouped by Season Season Sample Size Average (pCi/L) Jan, Feb, Mar Apr, May, Jun Jul, Aug, Sep Oct, Nov, Dec 172,600 187,051 161,189 162,965 7.56 5.95 5.45 7.72 Now if we look at the extremes of winter (7.56) to summer (5.45) months we see a percent difference of 32%, almost identical to the October to March six-month period of Table 7. We then see much less of a difference if we look at adjoining seasons. Spring (5.95) to summer (5.45) shows an 8.7% difference and fall (7.72) to winter (7.56) shows only a 2.1% difference. Winter (7.56) to spring (5.95) shows a 23.8% difference and summer (5.45) to fall (7.72) shows a 34.5% difference. References Lewis, R. A Statistical Report of Pennsylvania, Radon-222. Unpublished report. Pennsylvania Department of Environmental Protection, Bureau of Radiation Protection, Radon Division, 2007. September 14, 2009 203 State Radon Data We had data provided to us by two of the larger charcoal testing laboratories in the country. One of the labs provided the data in the breakdown as seen in the below tables, the other lab provided just raw data for each state by date and result. Due to time constraints, we used just the one data set with the already presummarized data. The below four tables are an arrangement of radon testing data as provided by this private testing laboratory. Obviously, this laboratory is supplying test kits to all 50 states and the District of Columbia. Data qualifications are 2- to 7-day charcoal, basement and first-floor results, all states, all seasons, years 2000 to 2009 and a total sample size of 718,615. It should also be understood that this data set is not a random distribution. Many states may even use this lab to conduct specific surveys in concentrated areas of their state with known high radon concentrations. This may tend to bias these data high. Also being a private laboratory there may be a large proportion of sales directly to private homeowners, with the potential for improper testing protocols being followed. The testing data were broken down into four tables simply by sorting the data on (1) sample size for each state (Table 1), (2) average radon concentration for each state (Table 2), (3) the maximum test result (Table 3) and (4) by the percent of test results greater than or equal 4.0 pCi/L (Table 4). The column of interest is highlighted in red. Sample size for the states shows a large range, and could be due to very low radon activity in a particular state, or a very active state happening to use this particular lab for numerous types of in state surveys. September 14, 2009 204 Table 1, State Ranking by Sample Size State HI MS LA DC AR AK OK NV AZ ME VT NJ SD TX DE AL IN WA NM OR CA WV RI ND MD MT NE UT ID MA WY KY CT SC MO FL GA NH TN VA KS NY WI CO NC IL OH IA PA MN MI N= 18 170 244 302 322 373 397 837 1148 1192 1222 1245 1586 1973 2114 3050 4133 4167 4425 4882 5129 5136 5302 6285 6817 7397 7444 7444 7455 8160 8539 9623 10010 10689 11010 11205 12843 13853 13961 15558 15857 25634 35087 36169 36775 37427 44059 47800 57709 79421 85010 AVG 0.1 1.4 0.6 1.9 3 6.4 2.3 2.7 2.3 6.7 3.6 3.1 8.1 2.2 2.4 3.9 4.6 5.3 3.8 3.7 1.6 6.6 4.4 6 5.1 7.3 5 4.4 6.1 4.1 5.4 8.2 3.9 2.9 4.6 1.8 2.5 5.1 4.3 3.2 5.3 4.1 6.2 6.1 3.5 5 8.5 6.4 8.5 4.7 3.4 September 14, 2009 MAX 0.7 11.8 21.4 22.5 75.1 273.1 59.3 72.8 73.4 321.2 110.1 71.3 279.9 145 46.9 1383.9 616.6 260.6 103.4 121.2 166.9 1228.8 229.7 181.5 355.4 2574.3 202.9 542.3 357.1 183.4 240.2 334.9 309.5 79.3 249.6 342.7 90.4 278.8 367.8 165.5 104.5 318.6 686.9 1618.3 1167.3 826 844.2 870.6 747.2 849.7 458 < 4.0 pCi/L 18 159 239 265 276 268 350 693 984 685 933 948 663 1813 1697 2423 2647 2927 3125 3519 4653 3233 3552 3055 4555 4056 4080 4918 4722 5822 4849 5170 7292 8359 7094 10089 10503 9287 9660 11766 8646 19387 19538 18976 27229 22012 22223 23512 31639 45396 62955 4-9.9 pCi/L 0 8 3 31 31 59 26 107 121 313 198 218 548 105 360 432 1049 673 1012 923 370 1083 1283 2271 1495 2072 2421 1844 1660 1608 2630 2277 1968 1775 2814 882 1958 2976 3000 2926 4999 3873 10333 11627 7184 10790 12929 14987 13693 26724 16731 10-19.9 pCi/L 20-49.9 0 3 1 4 7 18 11 25 31 123 59 67 261 29 52 117 317 340 219 323 87 491 315 743 462 829 766 500 611 512 796 1215 511 456 805 161 314 972 905 673 1760 1507 3561 4092 1786 3625 5279 6916 6736 6093 4078 0 0 1 2 5 21 9 10 10 49 25 10 96 9 5 58 111 171 60 110 11 252 132 189 250 363 172 163 363 176 235 759 198 86 267 56 65 502 335 169 416 698 1330 1265 497 909 2688 2223 4195 1094 1112 50-100 0 0 0 0 3 5 1 2 2 18 5 2 11 6 0 19 8 43 8 6 7 58 17 20 38 57 3 12 78 31 20 165 31 13 25 14 3 88 57 23 35 136 230 159 53 74 617 137 1080 86 101 >100 0 0 0 0 0 2 0 0 0 4 2 0 7 11 0 1 1 13 1 1 1 19 3 7 17 20 2 7 21 11 9 37 10 0 5 3 0 28 4 1 1 33 95 50 26 17 323 25 366 28 33 < 4.0 pCi/L 100% 94% 98% 88% 86% 72% 88% 83% 86% 58% 76% 76% 42% 92% 80% 79% 64% 70% 71% 72% 91% 63% 67% 49% 67% 55% 55% 66% 63% 71% 57% 54% 73% 78% 64% 90% 82% 67% 69% 76% 55% 76% 56% 53% 74% 59% 50% 49% 55% 57% 74% 205 Table 2, State Ranking by Average Radon State HI LA MS CA FL DC TX OK AZ DE GA NV SC AR NJ VA MI NC VT OR NM AL CT MA NY TN RI UT IN MO MN NE IL MD NH WA KS WY ND ID CO WI AK IA WV ME MT SD KY OH PA N= 18 244 170 5129 11205 302 1973 397 1148 2114 12843 837 10689 322 1245 15558 85010 36775 1222 4882 4425 3050 10010 8160 25634 13961 5302 7444 4133 11010 79421 7444 37427 6817 13853 4167 15857 8539 6285 7455 36169 35087 373 47800 5136 1192 7397 1586 9623 44059 57709 AVG 0.1 0.6 1.4 1.6 1.8 1.9 2.2 2.3 2.3 2.4 2.5 2.7 2.9 3 3.1 3.2 3.4 3.5 3.6 3.7 3.8 3.9 3.9 4.1 4.1 4.3 4.4 4.4 4.6 4.6 4.7 5 5 5.1 5.1 5.3 5.3 5.4 6 6.1 6.1 6.2 6.4 6.4 6.6 6.7 7.3 8.1 8.2 8.5 8.5 September 14, 2009 MAX 0.7 21.4 11.8 166.9 342.7 22.5 145 59.3 73.4 46.9 90.4 72.8 79.3 75.1 71.3 165.5 458 1167.3 110.1 121.2 103.4 1383.9 309.5 183.4 318.6 367.8 229.7 542.3 616.6 249.6 849.7 202.9 826 355.4 278.8 260.6 104.5 240.2 181.5 357.1 1618.3 686.9 273.1 870.6 1228.8 321.2 2574.3 279.9 334.9 844.2 747.2 < 4.0 pCi/L 18 239 159 4653 10089 265 1813 350 984 1697 10503 693 8359 276 948 11766 62955 27229 933 3519 3125 2423 7292 5822 19387 9660 3552 4918 2647 7094 45396 4080 22012 4555 9287 2927 8646 4849 3055 4722 18976 19538 268 23512 3233 685 4056 663 5170 22223 31639 4-9.9 pCi/L 0 3 8 370 882 31 105 26 121 360 1958 107 1775 31 218 2926 16731 7184 198 923 1012 432 1968 1608 3873 3000 1283 1844 1049 2814 26724 2421 10790 1495 2976 673 4999 2630 2271 1660 11627 10333 59 14987 1083 313 2072 548 2277 12929 13693 10-19.9 pCi/L 20-49.9 0 1 3 87 161 4 29 11 31 52 314 25 456 7 67 673 4078 1786 59 323 219 117 511 512 1507 905 315 500 317 805 6093 766 3625 462 972 340 1760 796 743 611 4092 3561 18 6916 491 123 829 261 1215 5279 6736 0 1 0 11 56 2 9 9 10 5 65 10 86 5 10 169 1112 497 25 110 60 58 198 176 698 335 132 163 111 267 1094 172 909 250 502 171 416 235 189 363 1265 1330 21 2223 252 49 363 96 759 2688 4195 50-100 0 0 0 7 14 0 6 1 2 0 3 2 13 3 2 23 101 53 5 6 8 19 31 31 136 57 17 12 8 25 86 3 74 38 88 43 35 20 20 78 159 230 5 137 58 18 57 11 165 617 1080 >100 0 0 0 1 3 0 11 0 0 0 0 0 0 0 0 1 33 26 2 1 1 1 10 11 33 4 3 7 1 5 28 2 17 17 28 13 1 9 7 21 50 95 2 25 19 4 20 7 37 323 366 < 4.0 pCi/L 100% 98% 94% 91% 90% 88% 92% 88% 86% 80% 82% 83% 78% 86% 76% 76% 74% 74% 76% 72% 71% 79% 73% 71% 76% 69% 67% 66% 64% 64% 57% 55% 59% 67% 67% 70% 55% 57% 49% 63% 53% 56% 72% 49% 63% 58% 55% 42% 54% 50% 55% 206 Table 3, State Ranking by Maximum Radon Value State HI MS LA DC DE OK NJ NV AZ AR SC GA NM KS VT OR TX VA CA ND MA NE RI WY MO WA AK NH SD CT NY ME KY FL MD ID TN MI UT IN WI PA IL OH MN IA NC WV AL CO MT N= 18 170 244 302 2114 397 1245 837 1148 322 10689 12843 4425 15857 1222 4882 1973 15558 5129 6285 8160 7444 5302 8539 11010 4167 373 13853 1586 10010 25634 1192 9623 11205 6817 7455 13961 85010 7444 4133 35087 57709 37427 44059 79421 47800 36775 5136 3050 36169 7397 AVG 0.1 1.4 0.6 1.9 2.4 2.3 3.1 2.7 2.3 3 2.9 2.5 3.8 5.3 3.6 3.7 2.2 3.2 1.6 6 4.1 5 4.4 5.4 4.6 5.3 6.4 5.1 8.1 3.9 4.1 6.7 8.2 1.8 5.1 6.1 4.3 3.4 4.4 4.6 6.2 8.5 5 8.5 4.7 6.4 3.5 6.6 3.9 6.1 7.3 September 14, 2009 MAX 0.7 11.8 21.4 22.5 46.9 59.3 71.3 72.8 73.4 75.1 79.3 90.4 103.4 104.5 110.1 121.2 145 165.5 166.9 181.5 183.4 202.9 229.7 240.2 249.6 260.6 273.1 278.8 279.9 309.5 318.6 321.2 334.9 342.7 355.4 357.1 367.8 458 542.3 616.6 686.9 747.2 826 844.2 849.7 870.6 1167.3 1228.8 1383.9 1618.3 2574.3 < 4.0 pCi/L 18 159 239 265 1697 350 948 693 984 276 8359 10503 3125 8646 933 3519 1813 11766 4653 3055 5822 4080 3552 4849 7094 2927 268 9287 663 7292 19387 685 5170 10089 4555 4722 9660 62955 4918 2647 19538 31639 22012 22223 45396 23512 27229 3233 2423 18976 4056 4-9.9 pCi/L 0 8 3 31 360 26 218 107 121 31 1775 1958 1012 4999 198 923 105 2926 370 2271 1608 2421 1283 2630 2814 673 59 2976 548 1968 3873 313 2277 882 1495 1660 3000 16731 1844 1049 10333 13693 10790 12929 26724 14987 7184 1083 432 11627 2072 10-19.9 pCi/L 20-49.9 0 3 1 4 52 11 67 25 31 7 456 314 219 1760 59 323 29 673 87 743 512 766 315 796 805 340 18 972 261 511 1507 123 1215 161 462 611 905 4078 500 317 3561 6736 3625 5279 6093 6916 1786 491 117 4092 829 0 0 1 2 5 9 10 10 10 5 86 65 60 416 25 110 9 169 11 189 176 172 132 235 267 171 21 502 96 198 698 49 759 56 250 363 335 1112 163 111 1330 4195 909 2688 1094 2223 497 252 58 1265 363 50-100 0 0 0 0 0 1 2 2 2 3 13 3 8 35 5 6 6 23 7 20 31 3 17 20 25 43 5 88 11 31 136 18 165 14 38 78 57 101 12 8 230 1080 74 617 86 137 53 58 19 159 57 >100 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2 1 11 1 1 7 11 2 3 9 5 13 2 28 7 10 33 4 37 3 17 21 4 33 7 1 95 366 17 323 28 25 26 19 1 50 20 < 4.0 pCi/L 100% 94% 98% 88% 80% 88% 76% 83% 86% 86% 78% 82% 71% 55% 76% 72% 92% 76% 91% 49% 71% 55% 67% 57% 64% 70% 72% 67% 42% 73% 76% 58% 54% 90% 67% 63% 69% 74% 66% 64% 56% 55% 59% 50% 57% 49% 74% 63% 79% 53% 55% 207 Table 4, State Ranking by Percent Radon Results greater than or equal 4 pCi/L State HI LA MS TX CA FL OK DC AZ AR NV GA DE AL SC VT NJ VA NY MI NC CT OR AK MA NM WA TN RI NH MD UT MO IN ID WV IL ME MN WY WI NE PA MT KS KY CO OH IA ND SD N= 18 244 170 1973 5129 11205 397 302 1148 322 837 12843 2114 3050 10689 1222 1245 15558 25634 85010 36775 10010 4882 373 8160 4425 4167 13961 5302 13853 6817 7444 11010 4133 7455 5136 37427 1192 79421 8539 35087 7444 57709 7397 15857 9623 36169 44059 47800 6285 1586 AVG 0.1 0.6 1.4 2.2 1.6 1.8 2.3 1.9 2.3 3 2.7 2.5 2.4 3.9 2.9 3.6 3.1 3.2 4.1 3.4 3.5 3.9 3.7 6.4 4.1 3.8 5.3 4.3 4.4 5.1 5.1 4.4 4.6 4.6 6.1 6.6 5 6.7 4.7 5.4 6.2 5 8.5 7.3 5.3 8.2 6.1 8.5 6.4 6 8.1 September 14, 2009 MAX 0.7 21.4 11.8 145 166.9 342.7 59.3 22.5 73.4 75.1 72.8 90.4 46.9 1383.9 79.3 110.1 71.3 165.5 318.6 458 1167.3 309.5 121.2 273.1 183.4 103.4 260.6 367.8 229.7 278.8 355.4 542.3 249.6 616.6 357.1 1228.8 826 321.2 849.7 240.2 686.9 202.9 747.2 2574.3 104.5 334.9 1618.3 844.2 870.6 181.5 279.9 < 4.0 pCi/L 18 239 159 1813 4653 10089 350 265 984 276 693 10503 1697 2423 8359 933 948 11766 19387 62955 27229 7292 3519 268 5822 3125 2927 9660 3552 9287 4555 4918 7094 2647 4722 3233 22012 685 45396 4849 19538 4080 31639 4056 8646 5170 18976 22223 23512 3055 663 4-9.9 pCi/L 0 3 8 105 370 882 26 31 121 31 107 1958 360 432 1775 198 218 2926 3873 16731 7184 1968 923 59 1608 1012 673 3000 1283 2976 1495 1844 2814 1049 1660 1083 10790 313 26724 2630 10333 2421 13693 2072 4999 2277 11627 12929 14987 2271 548 10-19.9 pCi/L 20-49.9 0 1 3 29 87 161 11 4 31 7 25 314 52 117 456 59 67 673 1507 4078 1786 511 323 18 512 219 340 905 315 972 462 500 805 317 611 491 3625 123 6093 796 3561 766 6736 829 1760 1215 4092 5279 6916 743 261 0 1 0 9 11 56 9 2 10 5 10 65 5 58 86 25 10 169 698 1112 497 198 110 21 176 60 171 335 132 502 250 163 267 111 363 252 909 49 1094 235 1330 172 4195 363 416 759 1265 2688 2223 189 96 50-100 0 0 0 6 7 14 1 0 2 3 2 3 0 19 13 5 2 23 136 101 53 31 6 5 31 8 43 57 17 88 38 12 25 8 78 58 74 18 86 20 230 3 1080 57 35 165 159 617 137 20 11 >100 0 0 0 11 1 3 0 0 0 0 0 0 0 1 0 2 0 1 33 33 26 10 1 2 11 1 13 4 3 28 17 7 5 1 21 19 17 4 28 9 95 2 366 20 1 37 50 323 25 7 7 > 4.0 pCi/L 0% 2% 7% 8% 9% 10% 12% 12% 14% 14% 17% 18% 20% 21% 22% 24% 24% 24% 24% 26% 26% 27% 28% 28% 29% 29% 30% 31% 33% 33% 33% 34% 36% 36% 37% 37% 41% 43% 43% 43% 44% 45% 45% 45% 46% 46% 48% 50% 51% 51% 58% 208 Table 5 below gives a summary of the four tables above. It is provided for states to compare their data to this large data set. We then provide Table 6 as another means of comparison with the states ranked by percent of readings greater than 10 pCi/L. State N= Avg. Max. U.S. 718608 5.2 2574 Table 5 %<4 % 4-9.9 63 25 % 10-19.9 % 20-49.9 8.3 3.1 % 50-100 0.5 %> 100 0.17 %>4 37 Table 6 State % >10 pCi/L State % > 10 pCi/L HI 0% UT 9% LA 1% MN 9% MS 2% NY 9% DC 2% TN 9% CA 2% MO 10% FL 2% IN 11% DE 3% MD 11% TX 3% NH 11% GA 3% AK 12% AZ 4% IL 12% NV 4% WY 12% AR 5% NE 13% SC 5% WA 14% OK 5% KS 14% VA 6% ID 14% MI 6% WI 15% NJ 6% ND 15% AL 6% CO 15% NC 6% WV 16% NM 7% ME !6% VT 7% MT 17% CT 7% IA 19% RI 9% OH 20% MA 9% PA 21% OR 9% KY 23% SD 24% U.S. 12% September 14, 2009 209 State Radon Rankings This analysis of state radon potential is based on the private lab data provided to us. Qualifications for this lab data are found in the State Radon Data section of this report. We present these data cautiously in that we know there can always be problems or disagreements in what to use and how to use it for the purpose of ranking. Please remember that this ranking is designed to give an indication of which states have the most severe radon problems in terms of the total number of people affected, as well as the magnitude and percentages of radon occurrences. Conversely, the states with the least severe radon problems can also be considered. We took the data and then ranked the states using nine categories; average radon concentration, maximum radon value, percent of test results greater than 4 pCi/L, percent of test results in the 4 to 10 range, in the 10 to 20 range, in the 20 to 50 range, in the 50 to 100 range, in the greater than 100 range and a final category that considered state average radon and state population. After the state ranking in each category, we then assigned a number, starting with 50 for the “worst” state. For instance, Montana had the highest radon measurement result from this data set and they were given 50 points. Colorado had the next highest maximum radon value and they were given 49 points. Where there was a tie between states, each state received the same score. We then assigned a score for all 50 states plus the District of Columbia. After each state was given a score for each of the nine categories, we then simply added up all the scores for each state. The state with the highest number was considered the state with the most “severe” radon problems; conversely, the state with the lowest score was considered the state with the least “severe’ radon problems. Below is the tabulation of state rankings. State Hawaii Louisiana Mississippi District of Columbia Delaware Oklahoma Nevada South Carolina Arizona Arkansas Georgia California Texas Florida Vermont New Jersey Virginia New Mexico Oregon September 14, 2009 Score 132 145 153 160 183 199 202 205 207 211 214 215 221 227 228 242 248 251 265 State North Carolina New Hampshire New York Wyoming Utah Tennessee Missouri Nebraska Kansas Washington North Dakota Maryland Indiana Idaho Minnesota Maine West Virginia Illinois Montana Score 296 300 301 308 308 309 315 319 325 329 332 332 334 339 340 354 356 358 371 210 State Connecticut Rhode Island Alabama Massachusetts Michigan Alaska Score 276 278 280 285 286 291 State South Dakota Wisconsin Iowa Colorado Kentucky Pennsylvania Ohio Score 372 380 384 388 397 411 422 We have also received radon testing data from ten state radon coordinators: California, Colorado, Idaho, Minnesota, Nevada, New York, Ohio, Pennsylvania, Utah and Wyoming. Since we had these data, we decided to compare them to the private-lab data from the four tables above. Sometimes the comparisons may not be most appropriate since the qualifications on each data set are different. We also do not know the source of the results from the state radon coordinators; it could be any one of a number of private labs or possibly a state radiation lab. These data may be most useful to those state radon coordinators who supplied these data. The data qualifications for the state-provided data are immediately below each table, where supplied by the state radon coordinator. Range OH-State data* OH- Lab data <4 51.4% 50.4% 4 to 10 27.8% 29.3% 10.1 to 20 12.2% 12% 20.1 to 50 7% 6.1% 50.1 to 100 1.1% 1.4% >100 0.53% 0.7% N= 134,833 44,059 *All house levels, all Measurement types, all seasons, 1986-2007. Range NV-State data* NV-Lab data <4 75% 83% 4 to 10 18.8% 12.7% 10.1 to 20 4.2% 3% 20.1 to 50 1.3% 1.2% 50.1 to 100 0.2% 0.2% >100 0.2% 0% N= 2,274 837 * Lowest Floor, Short-term only, 2003 to 2008 September 14, 2009 211 Range NY-State data* NY-Lab data <4 66.1% 75.6% 4 to 10 18.3% 15.1% 10.1 to 20 8.6% 5.9% 20.1 to 50 5.5% 2.7% 50.1 to 100 1.1% 0.5% >100 0.3% 0.1% N= 39,715 25,634 * Charcoal, Basement and First Floor, 1990 to 1999. Range MN-State data* MN-Lab data <4 59.3% 57% 4 to 10 33% 33.6% 10.1 to 20 6.5% 7.6% 20.1 to 50 1.2% 1.4% 50.1 to 100 0.1% 0.1% >100 0.0003% 0.04% N= 41,123 79,421 * Basement and First Floor, 1990 to 2000 Range WY-State data* WY-Lab data <4 64% 56.7% 4 to 10 26% 30.7% 10.1 to 20 7% 9.3% 20.1 to 50 2% 2.8% 50.1 to 100 0.3% 0.2% >100 0.1% 0.1% N= 18,574 8,539 * All levels, 1990 to 2009, short-term Range CA-State data* CA-Lab data <4 91.4% 90.7% 4 to 10 6.5% 7.2% 10.1 to 20 1.3% 1.7% 20.1 to 50 0.5% 0.2% 50.1 to 100 0.14% 0.13% >100 0.09% 0.02% N= 14,282 5,129 *Short- and long-term, basement and first floor, 1990 to 2000 September 14, 2009 212 Range UT-State data* UT-Lab data <4 66% 66% 4 to 10 24% 24.7% 10.1 to 20 7% 6.7% 20.1 to 50 2.2% 3% 50.1 to 100 0.16% >100 0.09% N= 12,925 7,444 *1994 to 2009, basement and main floor. Range ID-State data* ID-Lab data <4 66.8% 63.3% 4 to 10 21% 22.2% 10.1 to 20 7.2% 8.2% 20.1 to 50 4% 4.8% 50.1 to 100 0.7% 1.0% >100 0.3% 0.3% N= 3,551 7,455 *1990 to 2000, basement and first floor. Range CO-State data* CO-Lab data <4 4 to 10 10.1 to 20 20.1 to 50 50.1 to 100 >100 N= 51.5% 32.2% 11.7% 3.9% 0.5% 0.1% 61,767 *Years 2005 to 2008 52.4% 32.1% 11.3% 3.5% 0.4% 0.13% 36,169 Range PA-State data* PA-Lab data <4 61% 55% 4 to 10 23% 24% 10.1 to 20 9% 12% 20.1 to 50 5% 7% 50.1 to 100 1% 2% >100 0.4% 0.6% N= 878,600 57,709 *Short- and long-term, basement and first floor, all house types. September 14, 2009 213 Range MI-State data* MI-Lab data <4 75% 74% 4 to 10 19% 19.7% 10.1 to 20 4.7% 4.8% 20.1 to 50 1.4% 1.3% 50.1 to 100 0.1% 0.1% >100 0.04% 0.03% N= 96,353 85,010 *Basement, first floor, other floors, activated carbon, short-term, 1993 to 2009. Acknowledgements We wish to thank and acknowledge the following individuals for supplying the above data. From Air Chek Shawn Price and Michael DeVaynes, from Michigan Sue Hendershott, from Colorado Chrystine Kelley, from Idaho Megan Keating, from Utah John Hultquist, from California George Faggella, from Wyoming Steve Melia, from Minnesota Andrew Gilbert, from Ohio Chuck McCracken, from Nevada Susan Roberts and from New York Jerry Collins. September 14, 2009 214 EPA State Rankings The data in the below table are based on US EPA State Radon Residential Survey. The survey covered single-family detached homes, multi-unit structures and mobile homes. Testing used short-term tests, in lowest-livable level, during winter heating seasons. The states are ranked by the two categories shown; percent of homes above 20 pCi/L, and percent of homes above 4 pCi/L. State Source PA IA NY MT NJ ND NH OH CO NE RI ME WY IN KY MN MD WA MA TN UT VA ID DC CT VT IL NV KS MO AK WI NM EPA/State EPA/State State EPA/State State EPA/State State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State State EPA/State EPA/State State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State September 14, 2009 % of Homes Greater than 20 pCi/L 7.9 7.5 5.1 4.7 4.6 4.3 3.7 2.8 2.7 1.9 1.9 1.9 1.8 1.5 1.5 1.4 1.4 1.3 1.3 1.3 1.3 1.2 1.1 1.0 0.9 0.9 0.8 0.8 0.7 0.7 0.6 0.6 0.6 State IA ND NE MN MT CO PA NY NJ OH NH WI IN WY MA KS NM RI ID ME KY MO CT VT MD TN WV UT VA DC IL MI NV % Homes Greater than 4.0 pCi/L 71.0 60.7 53.5 45.4 42.2 41.5 40.5 32.8 32.5 29.0 27.4 26.6 26.5 26.2 22.7 22.5 21.8 20.6 20.3 20.0 17.1 17.0 16.5 15.9 15.9 15.8 15.7 14.0 13.9 13.0 12.0 11.7 10.2 215 State Source WV MI NC AL SC TX AZ CA MS GA FL OR OK LA HI SD DE EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State EPA/State State State EPA/State EPA/State EPA/State None State September 14, 2009 % of Homes Greater than 20 pCi/L 0.5 0.4 0.3 0.3 0.3 0.2 0.1 0.1 0.1 0 0 0 0 0 0 Not available 0 State WA AK GA NC AZ AL AR FL SC TX OR OK CA MS LA SD DE % Homes Greater than 4.0 pCi/L 8.6 7.7 7.5 6.7 6.5 6.4 5.0 4.0 3.7 3.6 3.6 3.3 2.4 2.2 0.6 Not available Not available 216 PA Radon Analyzer Data vs. UK Data Comparison of two data sets, United Kingdom and U.S. (PA), shows radon variability versus month of the year. The two data sets are in different units, pCi/L and Bq m-3, and the UK data are just from living rooms, which I assume to be first-floor measurements. These two graphs do show a very similar pattern of radon variability over time from two different parts of the world. PA Radon Analyzer Data Radon Concentration Vs. Month, 1990-2003 10 n= 858,602 9 8 Radon Conc. (pCi/L) 7 6 Bsmt 1st Fl. 5 4 3 2 1 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month UK Radon Data Living Room Rn Conc. 30 n=2000 25 Radon conc. (Bq/m3) 20 15 10 5 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month September 14, 2009 217 References Lewis, R. A Statistical Report of Pennsylvania- Radon-222, PA Dept. of Environmental Protection, Bureau of Radiation Protection, Radon Division, April 2007. Unpublished Report. Wrixon, AD, Green BMR, Lomas, PR, Miles, JCH, Cliff, KD, Francis, EA, Driscol, CMH, James, AC and O’Riordan, MC. Natural Radiation Exposure in UK Dwellings, NRPB R-190, 1988. September 14, 2009 218 International Radon Section Table 1. Domestic radon concentrations and Action Levels in different countries Country Average radon Action Level concentration (Bq/m3) in homes (Bq/m3) Australia Belgium Czech Republic Finland Germany Ireland Israel Lithuania Luxembourg Netherlands Norway Poland Russia Sweden Switzerland United Kingdom European Community USA Canada * * 140 123 50 60 * 37 * * 51-60 * 19-250 108 70 20 * 46 * 200 400 200 400 250 200 200 100 250 20 200 400 * 400 1000 200 400 150 200 Taken from the web site for WHO. (As of 5/29/2009) September 14, 2009 219 Comparison of International Radon Action Levels Country Existing Dwellings New Buildings Canada 5.4 pCi/L Finland 22 pCi/L 5 pCi/L Germany 8 pCi/L 8 pCi/L Ireland 5 pCi/L 5 pCi/L Norway 22 pCi/L 5 pCi/L Sweden 11 pCi/L 4 pCi/L Spain 11 pCi/L 5 pCi/L Switzerland 5 pCi/L United Kingdom 5 pCi/L 5 pCi/L United States 4 pCi/L 4 pCi/L http://www.co.jefferson.co.us/health/health_T111_R42.htm September 14, 2009 220 10 CFR 20 Appendix B, Table 1 & 2 Concentration in air and water above natural background Radon-222 Year 1959 Maximum permissible concentrations For 40 hour week For 168 hour week (Occupational Exposure) (Nonoccupational Exposure) MPCw MPCa MPCw MPCa ----- Taken from ICRP 2, 1959 Year 1970 30 pCi/L 10 pCi/L The subscripts “w” and “a” stand for water and air. Table I Restricted Areas Col. 1 (air) Col. 2 (water) 100 pCi/L ----- ----- Table II Unrestricted Areas Col. 1 (air) Col. 2 (water) 3 pCi/L ----- Table I based on exposure to conc. specified for 40 hours in any 7-day period. Table II Conc. may be averaged over period not greater than 1 year. 1975 100 pCi/L ----- 3 pCi/L ----- Table I based on exposure to conc. specified for 40 hours in any 7-day period. Table II Conc. may be averaged over period not greater than 1 year. 1977 100 pCi/L ----- 3 pCi/L ----- Table I based on annual average Table II Conc. may be average over a period not greater than 1 year. 1979 30 pCi/L ----- 10 pCi/L ----- Table I based on annual average Table II Conc. may be average over a period not greater than 1 year. 1980 30 pCi/L ----- 3 pCi/L ----- Table I based on annual average Table II Conc. may be average over a period not greater than 1 year. September 14, 2009 221 1985 30 pCi/L ----- 3 pCi/L ----- Table I based on annual average Table II Conc. may be averaged over a period not greater than 1 year. Taken from NRC 10 CFR20, App. B MPC values based on ICRP 2 switched to DAC values based on ICRP26/30 in 1991. Year 1998 Table 1 Occupational Values Inhalation Col. 1 (oral) Col. 2 (ALI) Col. 3 (DAC) ----4 WLM 30 pCi/L or 0.33 WL Table 2 Effluent Conc. Col. 1 (air) 0.1 pCi/L Col. 2 (water) ----- Table 1 Based on annual average Table 2 Based on annual average 2003 ----- 4 WLM 30 pCi/L or 0.33 WL 0.1 pCi/L ----- Table 1 Based on annual average Table 2 Based on annual average The tables above show the evolution of the radon-222 occupational standard over time. The first table is from the International Commission on Radiological Protection, ICRP 2, 1959. This data also occur in National Bureau of Standards (NBS), Handbook 69, August 1963. The two other tables are both from Nuclear Regulatory Commission (NRC), 10 CFR 20, Appendix B. These three tables cover a time period of 46 years, from 1963 to the current, since the standards in 2003 are still in effect today. The tables obviously do not cover all years; they show a representation of limits over time, as well as the time periods upon which the limits were based. The interesting observation from the tables is the change in the maximum permissible air concentration value from 30 pCi/L to 100 pCi/L, than back to 30 pCi/L. The earliest NBS table uses the concept of the Maximum Permissible Concentration (MPC) for its limiting values in the table. The MPC values are based on occupational values, above background, that would limit a whole body dose to less than 5 rem, or 15 rems to the lung (the critical organ for radon). Radon-222 is interesting in that the decay products are considered present in the state of equilibrium typical of that attained in ordinary air. A quality factor of 10 was used for alpha particles in the NBS tables. September 14, 2009 222 The column in the first table listed as “168 hour week” is for the nonoccupational exposure. It is analogous in the other two tables to “Unrestricted Areas,” and “Effluent Concentrations.” The two NRC tables are divided into two basic columns, one for occupational exposures (Restricted areas), and one for nonoccupational exposures (Unrestricted areas). The occupational exposures would occur on the job site on any NRC licensed facility where the radioactive materials exist. The nonoccupational exposures are applicable to the assessment and control of radioactive dose to the general public. In the early seventies, the NRC also used the concept of the Maximum Permissible Concentration (MPC) to describe the limiting amount of radioactive material to which individuals could be exposed. They changed over to the current concept of the DAC and ALI in 1991. The ALI or Annual Limit of Intake is an amount of radioactive material ingested or inhaled, which would result in either (1) a committed effective dose equivalent (CEDE) of 5 rems or (2) a committed dose equivalent (CDE) of 50 rems to an organ or tissue. The values of DAC or Derived Air Concentration are limits intended to control chronic occupational exposures. The DAC is the concentration of radioactive material in air and the time of exposure to that radionuclide in hours. For the particular case of Rn-222 the most current version of 10 CFR 20 lists two values for both the DAC and ALI, one with decay products removed and one with decay products present. For the case of parent radon-222 plus decay products present at 100 percent equilibrium, the values are as listed in the above table for the year 2003. The DAC value then assumes that the worker is immersed in the pure parent plus all of the decay products in equilibrium. Not to add more confusion to the tables above, but we can also mention the U.S. Department of Energy (DOE) and its occupational dose limits as found in 10 CFR 835. It also publishes Derived Air Concentration values for numerous radionuclides; of interest to this manual are the Rn-222 and Rn-220 values. It has recently updated (effective July 9, 2007) both values based on the dose conversion convention of 0.5 rem per WLM, found in ICRP 65, Protection Against Radon-222 at Home and at Work (ICRP, 1994). The Rn-222 progeny annual exposure limit went from 4 WLM to 10 WLM, and the Rn220 progeny limit went from 12 WLM per year to 30 WLM per year. The corresponding DAC values are 80 pCi/L or 0.83 WL for Rn-222 and 10 pCi/L or 2.5 WL for Rn-220. Even these increases above the previous values still carry the same risk as 5 rems of total effective dose equivalent, according to ICRP 65. September 14, 2009 223 References ICRP Publication 2, Report of Committee II on Permissible Dose for Internal Radiation, 1959. Pergamon Press. ICRP Publication 65, Protection Against Radon-222 at Home and at Work, 1994, Pergamon Press. NCRP Report 22, Maximum Permissible Body Burdens and Maximum Permissible Concentrations of Radionuclides in Air and in Water for Occupational Exposure. National Bureau of Standards Handbook 69, August 1963. September 14, 2009 224 Other Radon Reference Manuals EPA, Radon Facts, August 1987 Nature and Extent of Radon Measuring and Mitigating Radon Radon Action Program State Contacts Reference EPA, Radon Reference Manual, September 1987 EPA, Radon Facts, April 1993 Health Radon Surveys Mitigation New Construction Proficiency Programs Environmental Indicators Cooperative Partners Public Information Miscellaneous U.S. Public Health Service Agency for Toxic Substance and Disease Registry (ATSDR) Toxicological Profile for Radon December, 1990 Public Health Statement Health Effects Chemical and Physical information Production, Import, Use, and Disposal Potential for Human Exposure Analytical Methods Regulations and Advisories References Glossary World Health Organization (WHO) Indoor Air Quality: A Risk-based Approach to Health Criteria for Radon Indoors April 1993 Radon in the Environment Sub-Group: Risk Estimates Sub-Group: Policy Issues Sub-Group: Risk Communication Summary, Conclusions, and Recommendations September 14, 2009 225 FACTORS ASSOCIATED WITH RESIDENTIAL RADON TESTING INTENTIONS AMONG KENTUCKY HOMEOWNERS Gwendolyn H. Rinker, PhD, ARNP, University of Kentucky, Lexington, KY Abstract Kentucky leads the nation in both lung cancer incidence and percent smoking. In addition, many of the counties have average radon levels above the Environmental Protection Agency recommended action level. Guided by the Precaution Adoption Process Model, the purposes of this study were to: 1) examine the relationship between synergistic risk perception (radon plus tobacco smoke) and intention to test for radon; and 2) investigate the association between situational factors, smoking status, and intention to test for radon. A cross-sectional study design was utilized to survey two groups: 1) a convenience sample of those requesting radon test kits through the Kentucky Radon Program; and 2) randomly selected homeowners from five Kentucky counties. Test kit requests from the random group were then tracked. Results from multivariate logistic regression analysis examining factors associated with radon testing intention will be presented. This study will be useful in planning interventions to increase radon testing in Kentucky. This research was supported by: 1) dissertation research awards from the Clean Indoor Air Partnership, University of Kentucky, and the Delta Psi Chapter of Sigma Theta Tau International Honor Society for Nursing; and 2) a pre-doctoral traineeship from the Rural Cancer Control Program, University of Kentucky, funded by the National Cancer Institute. 226 The purpose of this study was to examine factors that are associated with radon testing intentions among Kentucky homeowners. Exposure to radon, a naturally occurring gas derived from the decomposition of uranium in the ground, is the second leading cause of lung cancer, and is associated with an estimated 15,400 to 21,800 cases of lung cancer cases in the United States each year (National Research Council, 1999). Lung cancer is the second most commonly diagnosed cancer and has the highest mortality rate of all cancers (National Cancer Institute, 2007). Primary prevention of radon-related lung cancer is accomplished by first testing for radon levels and then installing radon mitigation systems to ventilate radon from homes with high radon levels (Environmental Protection Agency, 2009). Radon risk reduction is particularly relevant in Kentucky, which has the nation’s highest lung cancer incidence rate (101.2 versus 69.1 per 100,000) (National Cancer Institute, 2008); the third highest percentage of adults who smoke (25.2 versus 18.4 percent) (Centers for Disease Control and Prevention, 2008); and counties with radon levels greater than 4 picocuries per liter (pCi/L), the action level designated by the U.S. Environmental Protection Agency (EPA) (Environmental Protection Agency, 2009). According to the EPA, Kentucky has 30 “Zone One” counties, defined as counties with predicted indoor radon levels of 4pCi/L or higher, and most of the remaining counties have predicted levels between 2 and 4 pCi/L (Environmental Protection Agency, n.d.). Further analysis of residential radon values from the Kentucky Radon Program data describes even more counties above the 4 pCi/L level than those reported by the EPA (Kentucky Cabinet for Health and Family Services, 2004), with 46% of residential radon values greater than 4 pCi/L (Clay Hardwick, Kentucky Radon Program, personal communication, February 2, 2009). 227 Public awareness of radon is generally high, with 52-82% reporting having heard of radon (Environics Research Group, 2007; Gregory & Jalbert, n.d.; Halpern & Warner, 1994; Wang, Ju, Stark, & Teresi, 2000). These same authors report that despite increasing public awareness of radon, only 8-15% have tested or considered testing for radon. The gap between radon awareness and testing presents a challenge to public health professionals attempting to decrease lung cancer risk. This study addresses that gap by focusing on the factors that influence the progression through the process of the adoption of intention to test for radon (see Figure 1), as described in the Precaution Adoption Process Model (PAPM). Previous studies have examined the correlates of radon testing intentions and behaviors utilizing the PAPM. Factors positively correlated with deciding to test include: perceived risk likelihood, knowing others who had tested (Sandman & Weinstein, 1993; Weinstein & Sandman, 1992a), and perceived susceptibility (Sandman & Weinstein, 1993). Perceived effectiveness of mitigation (Weinstein, Sandman, & Roberts, 1991) and perceived difficulty of radon mitigation have not been shown to impact the decision to test for radon (Weinstein, Sandman, & Roberts, 1990; Weinstein et al., 1991). Correlates of intentions to test for radon include: perceived community radon risk, perceived personal susceptibility, perceived severity, perceived community concern, number of known radon testers known by an individual (Weinstein et al., 1991). Correlates of test kit orders include: perceived personal susceptibility, perceived severity of illness from radon, and intention to test for radon (Weinstein et al., 1991). Other factors that have been positively associated with radon testing include: education (Wang et al., 2000), income (Halpern & Warner, 1994; Hill, Butterfield, & Larsson, 2006), female gender (Halpern & Warner, 1994), the presence of children living in the home (DiPofi, 228 LaTour, & Henthorne, 2001), home ownership (Hill et al., 2006), and younger age (Halpern & Warner, 1994; Wang et al., 2000). Synergistic Risk Perception Synergistic risk perception, one’s assessment of risk from the combination of radon and smoking, was also measured in this study. The rationale for the inclusion of synergistic risk perception measurement is two-fold. First, the risk factors of radon and tobacco smoking are related, with more radon-related lung cancers occurring in individuals with a history of smoking (National Research Council, 1999). Although radon is a risk for both smokers and nonsmokers, those who have smoked and have had radon exposure are at a higher risk of developing lung cancer because of the multiplicative or synergistic interaction between tobacco smoke and radon on lung cancer risk. Second, there is evidence that perception of the synergistic risk between tobacco smoke and radon may predict changes in smoking behaviors and influence radon risk reduction behaviors. One group studied the impact of a radon and smoking educational intervention, synergistic risk perception and personality characteristics on smoking behaviors among smoking households who had requested radon test kits and found that those with who were more conscientious and who had higher synergistic risk perception were more likely to reduce indoor smoking (Hampson, Andrews, Barckley, Lichtenstein, & Lee, 2000). An extension and partial replication of this study demonstrated that households with radon levels ≥ 4 pCi/L were more likely to institute a new household smoking ban at the 12 months follow-up than households with radon levels < 4 pCi/L (Lichtenstein et al., 2008). Both a video and telephone counseling for smokers about the combined risk of radon and smoking was positively associated with requests for radon information, and those in the telephone counseling intervention group who had high 229 screening radon values were more likely to perform repeat radon tests and attempt to mitigate their homes (Lichtenstein et al., 2008). There is a need for further study regarding the influence of synergistic risk perception on radon risk reduction behaviors. If synergistic risk perception were found to predict both radon and tobacco related risk reductions, interventions which increase the perception of these combined risks could have significant import for lung cancer prevention. The purposes of this study were to: a) determine differences in demographics and reasons for not testing between those who intend to test and those who do not; b) investigate whether perceived susceptibility, synergistic risk perception, synergistic knowledge, perceived severity, social influence, and smoking status influence intention to test for radon, controlling for age, gender and education; and c) assess the relationship between the stage of the Precaution Adoption Model for testing and the order of new test kits by those in a random sample of homeowners in selected Kentucky counties. The Precaution Adoption Process Model (PAPM), a stage-based model that identifies phases along the route to protective health action, guided the study aims and hypotheses, and is further described in Figure 1. Hypothesis 1: Female gender, younger age, and higher socioeconomic status will be related to radon testing intentions. Hypothesis 2: Perceived severity, perceived susceptibility, synergistic risk perception, synergistic knowledge, social influence, and smoking status will be associated with radon testing intentions. Hypothesis 3: The proportion of those in the random sample who request a free test kit will be higher among those in PAPM stages 5 or 6 than in stages 1 through 4. 230 Methods Design The study had two phases: 1) a cross-sectional, non-experimental design in which data were collected via mailed survey from two subsamples of Kentucky, and 2) a prospective design in which survey participants in the random sample were invited to request a free radon test kit (see Figure 2). Sample Sampling for this study was done in two stages and included two subgroups (see Figure 2). The first subgroup was comprised of a stratified random sample of 40 property owners from each of five Kentucky counties (N = 200). The accessible population of five Kentucky counties was selected using the following criteria to ensure variability: 1) counties with differences in the amount of radon testing done in the past; 2) rural and urban counties; 3) counties from different regions of the state; and 4) counties with diversity in lung cancer incidence, percent adult smokers, and average radon values. The five Kentucky Counties chosen for this study were: Hancock, Henderson, Louisville-Jefferson, Rowan, and Whitley (see Table 1). The random sample from these five counties was selected via public access property roles available on the internet or hard copy from the county property valuators. Randomization was completed using the random number generator in SAS (SAS, Institute, & Inc, 2004). For the first subgroup, a radon test request coupon was attached to the mailed survey. The coupon and survey were matched by a code. The second subgroup was a convenience sample of 143 individuals who had requested radon test kits from the Kentucky Radon Program of the Kentucky Department for Public Health, Cabinet for Health and Family Services between January to May 2009. Those included 231 in this sample were 121 individuals from a variety of Kentucky counties who had been on a waiting list to receive test kits following two radon television spots (one in the LouisvilleJefferson county in January, 2009 and one in Lexington-Fayette county in February, 2009), as well as 22 additional individuals who requested test kits during the month of May, 2009. Of the 343 of the surveys mailed, 122 were returned (overall response rate of 35.6%). Of those who responded, 55 (45.1% of the total) were from the random sample (a group response rate of 27.5%), and 67 (54.9% of the total) were from the convenience sample (a group response rate of 46.9%). Four individuals from the random sample returned the coupons requesting a free test kit but did not return a completed survey and were excluded from the analysis. One individual in the random sample did not live in the county from which she was randomly selected and was excluded. Measures The 33 question survey consisted of a six pages containing mostly categorical, multiplechoice questions. Dependent variables included: 1) intention to test, operationalized by the stage of PAPM; and 2) the number of new radon test kits requested (random sample only). Independent variables include: perceived susceptibility, synergistic risk perception, knowledge of synergistic risk, perceived precaution effectiveness, perceived severity, social influence, and smoking status. Covariates included: age, gender, race, education, home ownership, length of residence, presence of a ground floor or basement, presence of children under age 10 living in home, age of residence, and reasons given for not testing for radon. 232 Intention to test An intention to test for radon was defined as a decision to perform a radon test, as measured by the respondent’s stated plan to perform the test. The outcome variable of intention to test for radon was measured by assessing the stage of PAPM into which each respondent falls. Those who responded affirmatively (“I have decided to test” or “I have already tested”) to the question: “What are your thoughts about testing for radon?” indicated stages 5 and 6 of PAPM, the intention to act (Weinstein et al., 1991). The variable was dichotomized as those with and without testing intentions (stages 5-6 and stages 1-4 of PAPM, respectively). The rationale for combining stages 5 (intention to test) and 6 (already tested for radon) are justified in this study, because those who had already tested were also planning to retest for radon, as evidenced by a request for a radon test kit from the Kentucky Radon Program. New test kit orders The randomly selected subgroup, or those homeowners who were randomly selected to receive the study survey, were invited to request a free radon test kit from the Kentucky Radon Program. The outcome variable “new test kit orders” was defined by whether participants in this group did or did not request a free radon test kit. Perceived susceptibility Perceived susceptibility, one’s perception about the likelihood or degree of risk from radon, was measured by the sum score of two items from a three item risk perception scale developed by Weinstein et al. (Weinstein, Lyon et al., 1998). The two items measured the respondent’s perceived likelihood of having radon in ones’ own home and in the home of someone else in the community. Perceived susceptibility, as measured in this study, can be equated to radon risk perception. 233 Synergistic risk perception Synergistic risk perception is the view that the relative risk of tobacco smoke and radon combined is more hazardous than smoking one pack a day without radon exposure. Synergistic risk perception has been defined as the way in which the public perceive their overall risk based on the interaction or combination of several health risks (French, 2002). Synergistic risk perception was measured by single question, comparing the risk of smoking combined with radon with the risk of smoking one pack of cigarettes without radon exposure. The question asked respondents to rate the risk of the combination or smoking and radon combined compared with the risk of smoking alone using a five point scale, with response options from “much less risky” (1) to “much more risky” (5). Those who responded that the radon and smoking combination was “more risky” than smoking alone indicated their belief in an “additive” interaction, and those who indicated the combination was “much more risky” indicated a multiplicative or synergistic interaction between radon and smoking. This measure of synergistic risk perception was similar to a relative risk measure previously developed by Hampson et al. (Hampson, Andrews, Lee, Lichtenstein, & Barckley, 2000), which was a seven point scale asking participants to rate the combined risk of radon and smoking compared to the risk of smoking one pack of cigarettes without radon exposure (“many times less risky than smoking” to “many times more risky than smoking”). Hampson et al. (2000) demonstrated a multiplicative model (i.e., respondents assessed the combined risk of radon and smoking as many times greater than the risk from smoking alone) using this measure. Knowledge of synergistic risk Knowledge of synergistic risk was defined as knowledge of the combined hazards of radon and smoking or secondhand smoke, and was measured by the sum of four true/false items 234 (1= correct response, 0 = incorrect response; range 0-4). Higher knowledge scores indicated more knowledge about the combined effects of radon and tobacco smoke (both active smoking and secondhand smoke were included). This measure was modified from an existing four item scale measuring knowledge of synergistic health risk, with the permission of the original author (Hampson, Andrews, Lee et al., 2000). Perceived precaution effectiveness Perceived precaution effectiveness was defined as the belief that radon mitigation is a successful method for decreasing radon risk, and was measured using two questions: 1) how hard it is to reduce radon to a safe level in homes that have problems (1 = not very difficult to 4 = very difficult); and 2) how much does reducing radon levels reduce the chances of getting sick (1 = would not reduce the risk to 4 = would reduce the risk completely). These questions were modified from those previously used (Weinstein et al., 1991). Perceived severity Perceived severity was defined as the belief that radon-related illness would be serious. It was operationally defined by a stated belief that such an illness would be serious or very serious, as measured by one question: “How serious would an illness caused by radon be?” (Weinstein et al., 1991). Social influence Social influence was conceptually defined as the belief that others in one’s community are concerned about radon and by the experience of peers in dealing with radon, and is measured by two questions, which ask respondents how many people they know who have tested, and how concerned about radon they believe others to be (Weinstein et al., 1991). 235 Smoking status Current, former, and never smokers. Current smoking status was assessed using two questions: (1) Have you smoked at least 100 cigarettes in your entire life? (yes/no), and (2) Do you now smoke cigarettes every day, some days, or not at all (Centers for Disease Control and Prevention, 2007). Current smokers were those who had smoked at least 100 cigarettes and currently smoked every day or some days. Former smokers were those who had smoked at least 100 cigarettes but who did not smoke at all at the time of the survey. Never smokers were those who had not smoked at least 100 cigarettes in their lifetime. Questions assessing active smoking status were derived from the 2008 Behavioral Risk Factor Surveillance Survey (BRFSS) (Centers for Disease Control and Prevention, 2007). Exposure to secondhand smoke. Exposure to secondhand smoke was assessed via selfreport of exposure to tobacco smoke in the home or workplace. Three questions modified from the Nurses’ Health Survey (Speizer, 1982) were utilized in order to determine previous or current exposure to tobacco smoke in the home, and current exposure in the workplace. Although selfreport of smoking status may result in misclassification bias, it will provide a beginning understanding of how synergistic risk perception varies with different levels of tobacco smoke exposure. Situational factors Situational factors were defined as factors that are perceived to facilitate or create barriers to radon testing, and were measured by asking participants to indicate reasons for choosing not to test for radon (items listed in Table 4) (Weinstein et al., 1991). Situational factors may have the greatest impact on the transition between deciding to act and acting (Weinstein & Sandman, 1992b). 236 Procedures Internal Review Board (IRB) approval was obtained from first the Kentucky Cabinet for Health and Family Services and then the University of Kentucky Medical IRB and the survey was pilot tested with six homeowners prior to the research study. Perceived length, understandability, and feasibility of the survey instrument were assessed via written comments. The few changes that were suggested were implemented. A revised survey instrument was then submitted to and approved by both IRBs. Research study packets containing the surveys were mailed in two batches: 1) surveys and coupons for a free radon test kit were mailed to a random sample of 200 individuals from a purposive sample of 5 Kentucky counties (see Table 1), and 2) survey packets were included with radon test kits and mailed to a convenience sample of 143 individuals who requested radon test kits from the Kentucky Radon Program. The second batch was mailed through the Kentucky Radon Program, Kentucky Cabinet for Health and Family Services, and was sent out over a period of 4 weeks during May, 2009. Several survey methods suggested by Dillman (2009) were used to increase response rates. First, an incentive in the form of a two dollar bill was attached to each survey. Second, approximately ten days after the surveys were mailed, a reminder postcard was sent to each person included in the sample (Dillman, Smyth, & Christian, 2009). Data Analysis Descriptive statistics were conducted, including frequency distributions or means and standard deviations, as appropriate to the level of measurement. These univariate analyses were carried out by the whole sample, subgroup (i.e., convenience and random samples), and by 237 outcome (i.e., those with and without testing intentions). Bivariate analysis was done between independent variables and intent to test for radon. In particular, the Rao-Scott chi-square test of association (for nominal explanatory variables and covariates), Mann-Whitney U test (for ordinal variables), or two-sample T-test (for continuous variables) was used to determine differences between those who intend to test and those who don’t. Explanatory variables and covariates in the study included: perceived susceptibility, synergistic risk perception, synergistic knowledge, perceived precaution effectiveness, social influence, perceived severity, situational factors, smoking status, gender, race, education, and income. Multivariate logistic regression analysis was performed to examine the variables that were associated with intention to test. Data analysis was conducted using SAS (SAS Institute Inc., 2004) with an alpha level of 0.05 throughout. Given the data were obtained using a complex survey design, SAS procedures appropriate for this type of design, including SURVEYFREQ, SURVEYMEANS, and SURVEYLOGISTIC, were used. Results Sample characteristics Overall, participants lived in 17 different Kentucky counties, with 45.1% of the total sample residing in Jefferson County. The mean age of the total sample was 52.1 years and the majority was female with an annual household income of at least $50,000, college graduate, Caucasian, and non-smoker (see Table 2). All but two respondents reported being Caucasian: one Asian and one Hispanic (both in the convenience sample). Approximately 44% of the participants had smoked at some point in their lifetime, although only 8% were current smokers. The number of years current or former smokers had smoked averaged 17.8 (s.e.m. 2.0). Almost one third (36 or 31.9%) of respondents reported some current secondhand smoke (SHS) exposure 238 either at work or at home. The mean number of children aged 10 or less per household was 0.4 (s.e.m. = 0.1), with 26.2% (32) of the total sample having children 10 years and younger at home. The mean number of years lived in their current residences was 13.0 years (s.e.m. = 1.1), and the average age of the participants’ residences was 31.4 years (s.e.m. = 2.3). Most reported having a ground floor or basement in their residence (99 or 81.8%). All but one respondent reported owning their home or residence. Comparisons of convenience and random samples There were no significant differences between the convenience and random samples in terms of age, gender, income, education, race, or smoking status (see Table 2). The random sample came from 5 counties, as described in the methods above. The convenience sample resided in 15 different counties, with the majority (62.7%) living in Jefferson county. Radon awareness. There was a significant difference between the convenience and random samples in radon awareness (see Table 2). The random sample had more individuals who were unaware, or who had never thought about radon testing. Over half of the participants had heard of radon via television. Other common sources of information about radon included newspaper, magazine, radio, and family, friends, co-workers, neighbors. The random and convenience samples were similar in their sources of radon information, except, unlike the convenience sample, several in the random sample reported having heard about radon at work, and some reported never having heard of radon. One individual from the convenience sample reported being a licensed home inspector with certification to test for radon. There were no significant differences in knowledge of synergistic risk between those in the random and convenience samples (t = 0.78, p = 0.4). 239 Reasons for obtaining a test kit. A majority of the total sample (57.0%) reported obtaining the radon test kit because it was free; there was a significant difference between the random and convenience samples in the number who reporting obtaining the kit because it was free (χ2 = 9.51, p < 0.01). Twelve respondents obtained a test kit because a family member had lung cancer, 8 reported having a friend with lung cancer, and 5 respondents reported a personal history of lung cancer. Suggestions from friends, neighbors, family members, and health care providers all resulted in test kit orders. Three respondents from the total sample reported obtaining a test kit after having considered doing so for some time. For example, one participant wrote, “After being knowledgeable about radon for years, I decided to get a test kit!” Another reported that he had “heard of radon for years” and stated “[I] can’t remember why I finally decided to get [a] test kit except it was free and concern if [I have a] high level [of radon I] will get lung cancer if don’t do something about it. A third participant stated, “radon is out of [sight] out of mind…I should have done it [tested] but haven’t.” Other reasons given for obtaining a kit include real estate transactions (2), an internet search (1), and “did not know radon was a health problem.” Reasons for not testing for radon (i.e., situational factors). Over a third (38%) of the total sample reported that they did not test because they did not believe they had a radon problem in their home (see Table 4). Six respondents cited lack of knowledge (in the write-in ‘other’ response) as the reason they did not test. One wrote that he was “too illiterate.” Others wondered if the test would give a “true result” and if an elevated result would make it “harder to sell our house.” One responded, “…just another abstract, intangible thing to be stressed out about…ignorance is bliss!” 240 Comparisons by intention to test There were no significant differences in mean age between those who planned to test and those who did not (t = 0.41, p = 0.7). There were no differences in gender, income, race, smoking status, or the presence of children under age 10 in the home (see Table 3) between those with and without testing intentions. There was, however, a significant difference in education between those who planned to test and those who did not. There was also a significant difference between groups in those reporting the presence of a ground floor or basement in their homes (χ2 = 4.30, p = 0.04). There was no statistically significant difference between those who tested and those who did not in the mean number of years respondents reported living in their current residences (t = 1.03, p = 0.3), or in the mean residence age (t = -1.94, p = 0.1). Over half (69 or 63.3%) of those who had heard of radon indicated intentions to test for radon. One respondent who had mitigated (and therefore already tested for radon) did not plan to test again. Three respondents indicated that they had already mitigated but planned to (re)test for radon. Thus, it can be assumed that all of those in PAPM stage five or six had testing intentions, justifying the use of a dichotomous outcome variable. Income (<$50,000 versus ≥ 50,000) was significantly related to education (< college graduate versus ≥ college graduate) (χ2 = 6.61, p = 0.01); therefore, education was selected to represent socioeconomic status, excluding income from the regression model. Race was also eliminated from the regression since there were only 2 respondents who were not Caucasian. Due to a marginal association between presence of a ground floor or basement in one’s home and education level (χ2 = 2.99, p = 0.08), both variables were not included in the regression, so as to minimize the likelihood of multi-collinearity. Education was selected from these two variables, 241 as it conceptually would be more likely to be related to the outcome (i.e., intention to test for radon). Perceived susceptibility. Perceived susceptibility was significantly higher in the group that planned or had already tested for radon than in the group not planning radon testing (z = -4.37, p < 0.001). The two components of perceived susceptibility, perceived likelihood of a radon problem in one’s home and the perceived percentage of radon problems in one’s area, both had significant differences between those with and without testing intentions (see Table 4). Synergistic risk perception. Synergistic risk perception was significantly different in the group that planned to test or had already tested, compared with those in the other PAPM stages (z = -2.18, p = 0.03). There were no significant differences in synergistic risk perception between ever smokers and never smokers (z = -0.81, p = 0.2), current smokers and others (z = -1.50, p = 0.1), former smokers and others (z = 0.10, p = 0.9), or between those who were or were not currently exposed to secondhand smoke (at home or at work) (z = -0.80, p = 0.4). When synergistic risk perception was dichotomized by those who indicated a multiplicative interaction (i.e., those who answered the combination of radon and smoking is “much more risky” than smoking alone) versus all other responses, there was a significant difference in the frequency of those reporting a multiplicative interaction among those who reported testing intentions versus those who did not (χ2 = 6.94, p = 0.01). This dichotomized synergistic risk perception variable was used for the logistic regression. Synergistic knowledge score. The average knowledge score for the combined group was 3.0 (s.e.m. = 0.1), with no significant differences between those with and without testing intentions (t = -1.49, p = 0.1). There was a significant relationship between synergistic risk perception and knowledge score (F = 3.98, p < 0.01). 242 Perceived precaution effectiveness. Perceived precaution effectiveness, was measured by two ordinal level variables, perceived effectiveness of mitigation in decreasing radon risk, and the perceived ease of mitigation. Ease of mitigation was not significantly different between the groups (z = 1.46, p = 0.1). Perceived effectiveness of mitigation was significantly different between those who intended to test and those who did not (z = -2.29, p = 0.02). However, because the effectiveness of mitigation variable had only three categories, it could not be entered into the regression as an ordinal variable, and it could not be used as a dichotomous variable (i.e., would not take away risk versus would partially or completely take away risk from radon). Only six individuals in the total sample (5.0%) believed that mitigation was not effective in reducing risk from radon. Thus, due to limited variability in the responses, perceived precaution effectiveness was not included in the regression model. Social influence. Community concern was collapsed from a four categories to three by combining “concerned” and “very concerned” into one category, since one of the cells was zero, limiting the comparison between groups. Community concern, both as a 3 category variable and as a dichotomous variable, was significantly different between groups based on testing intentions. The average number of known households who had tested for radon and the knowing someone who had tested was significantly different between those who planned to test and those who did not (t = -2.45, p = 0.02). College graduates and those with higher perceived susceptibility, perceived severity, and those who knew others who had tested for radon and perceived that others in their community were concerned about radon had higher odds of testing intentions (see Table 5). Those who had graduated from college had 3.7 times higher odds of planning to test for radon than those who did not plan to test (OR 3.67, 95% CI 1.10-12.27). The perception of radon-related illness as 243 being serious or very serious was associated with > 8 times higher odds of planning to test for radon (OR 8.25, 95% CI 2.46-27.73). Knowing others who tested and a perception of community concern about radon were also associated with radon testing intentions. There was no association between radon testing intentions and synergistic risk, smoking status, gender, or age. When current exposure to secondhand smoke was used as a proxy for smoking, this variable was not associated with radon testing intentions. Request for free radon test kit in random sample and PAPM stage Most (40 or 72.7%) of participants in the random sample requested a free radon test kit. There was not a significant relationship between requesting a free test kit and reporting testing intentions (χ2 = 2.29, p = 0.1). Discussion College graduates were more likely to have radon testing intentions than those with less education, although there were no differences in age, gender, income, race, smoking status, the presence of children under age 10 in the home, or synergistic knowledge between those with and without testing intentions. All but two of the participants were Caucasian, so the effect of race on testing intentions could not be analyzed in this study. Perceived susceptibility, perceived severity, knowing others who tested for radon, and perceived community radon concern were all positively associated with radon testing intentions. Synergistic risk perception was not associated with radon testing intentions in the regression analysis. Contrary to the study hypothesis, females were not more likely to plan to test for radon than males. There were also no significances differences between males and females in perceived susceptibility or synergistic risk perception. 244 Intention to test for radon Perceived susceptibility. As hypothesized, perceived susceptibility was significantly higher in the group that planned or had already tested for radon. The two components of perceived susceptibility, likelihood of a radon problem in one’s home and the percentage of radon problems in one’s area, were both significantly different when those who planned to test were compared to those who did not. Similarly, both the belief that there was not a radon problem in one’s home and the belief that there was not a radon problem in one’s area were given as reasons for not testing significantly more frequently in those not planning to test for radon than among those with testing intentions. This was an expected finding, since perceived susceptibility has been previously cited as a prerequisite to radon testing intention and behavior (Duckworth, Frank-Stromborg, Oleckno, Duffy, & Burns, 2002; Weinstein, 1988; Weinstein & Sandman, 1992a). It is important to note that more participants reported that they did not “think I have a problem [their] in my home” than that there were “no problems in [their] my area” as reasons for not testing (36.1% and 23.8%, respectively). This finding is congruent with another study reporting 43% of respondents cited perceived lack of a radon problem in their home as a reason for not testing for radon (Kennedy, Probart, & Dorman, 1991), and others that report that respondents rate environmental hazards as riskier for others than for themselves (Park, Scherer, & Glynn, 2001). This finding points to an important consideration for future intervention. Communication about risk from radon should not only seek to increase general public knowledge about radon risk, but also provide geographic-specific information regarding risk in one’s geographic area. This finding is confirmed by a study demonstrating that individuals who 245 were aware that they lived in a high radon area were more likely to test for radon (Wang et al., 2000). Perceived severity. Those who perceived that the illness caused by radon would be serious or very serious were more than eight times more likely to plan radon testing than those who thought it was not serious or somewhat serious. This finding is congruent with previous work demonstrating a significant correlation between perceived illness severity and radon testing intentions (Weinstein et al., 1991). Synergistic risk perception. Contrary to the study hypothesis, synergistic risk perception was not significantly associated with testing intentions. Most of the respondents had an inaccurate understanding of the combined risks of radon exposure and smoking, and more of them rated the combination as “more risky” rather than as “much more risky.” The former describes an additive interaction between smoking and radon, and the later describes a more accurate multiplicative or synergistic interaction. Further, when the variable was dichotomized for the regression (“much more risky” = 1, other responses = 0), respondents with the correct understanding of synergistic risk (i.e., the combination of radon and smoking are much more risky than smoking alone) were not more likely to intend to test for radon. The lack of a perceived multiplicative interaction between radon and smoking has been previously cited in the literature (Hampson et al., 1998; Hampson, Andrews, Lee et al., 2000). It is unclear whether this lack of perceived multiplicative interaction is more a function of incomplete knowledge of the sample or the degree of accuracy of the measures than of the attitudes and beliefs related to risk. For example, a study by Bonnin-Scaon et al. indicates that participants who where retested, subsequent to education on the synergy of radon and smoking, moved from a sub-additive (i.e., the combined risk of radon and smoking is less than the risk 246 from smoking) to a multiplicative model of risk perception (Bonnin-Scaon, 2002). In this study, synergistic risk perception and knowledge score were related, and knowledge score was not significantly different between those who did and did not intend to test. These findings lend credence to the premise that the lack of an accurate understanding of synergistic risk is related to incomplete knowledge. Others, however, have reported that participants gave different synergistic risk estimates for themselves than for others in their home or in their neighborhood, regardless of the same actual risk (Hampson et al., 1998). Unrealistic optimism, where individuals are more concerned with health risk for society at large than for themselves personally, has been reported in relation to risks such as contaminated drinking water, AIDS, heart disease, and radon (Park et al., 2001). For example, some do not test their homes for radon because they believe they are less at risk than those around them, regardless of their actual risk. Further study is needed to examine the combined effect of both knowledge and optimistic bias on the measurement of synergistic risk perception. Synergistic risk perception was not found to be significantly different based on smoking status with neither current nor former or ever smokers, nor with those currently exposed to secondhand smoke. Having ever smoked at least 100 cigarettes in one’s life was not associated with radon testing intentions or synergistic risk perception. A larger sample size with more smokers is needed to assess for differences in synergistic risk perception between current smokers and nonsmokers, as there were only 12 current smokers in this study. Several studies have reported less concern about radon by smokers than nonsmokers (Kennedy et al., 1991; Mainous & Hagen, 1993), and suggested that health education strategies be focused on smokers. 247 Increasing the concern of smokers about radon, and specifically about the combined risk of radon and smoking is an important area for future intervention research. Social influence. A number of individuals (35.2%) reported hearing about radon from family members, friends, neighbors, or co-workers. Knowing others who tested for radon and having a perception that community members are concerned about radon were both associated radon testing intentions. The actions or accounts of others, such as family, friends, or neighbors, may influence individuals to transition from being unengaged with radon as a health hazard (stage two of PAPM) to deciding to take action (Weinstein, 1988; Weinstein & Sandman, 2002). Media. The role of the media, especially television, on increasing knowledge about radon and prompting test kit orders is an important finding in this study. Over half of the total sample and convenience sample had heard of radon through television. The convenience sample consisted of individuals who requested a kit following two local television spots on radon, so this finding is unsurprising. However, almost half (47.3%) of the random sample had also heard about radon by television. Other media sources radon information included magazine, newspaper, and radio. Media may be an important venue for dispersing the joint smoking and radon risk message. Working to decrease smoking rates, household smoking, and residential radon exposure together would be a cost effective manner to address major lung cancer risk factors, thereby decreasing lung cancer incidence in Kentucky. Free test kits. Many of the respondents reported that they obtained a radon test kit because it was free. While this study does not address the number of these respondents who will follow through with testing and mitigation if indicated, it does point out an important issue for state radon programs. Offering a free test kit may be a valid strategy for promoting or prompting residential radon testing or at least the order of a test kit. For those who have already decided to 248 test for radon, interventions that increase the ease of radon testing (i.e., through the provision of information and/or a test kit) are more effective at increasing test kit orders than interventions that increase risk perception (Weinstein, Lyon, Sandman, & Cuite, 1998; Weinstein et al., 1990, 1991). Further research is needed to find effective ways to motivate those obtaining free test kits to return the test kit and then mitigate if radon levels are elevated. Implementing policy that requires radon testing, reporting, and mitigation as part of real estate transactions has been suggested as one way of increasing the effectiveness of programs than distribute free test kits (Field, Kross, & Vust, 1993). Test kit orders Those in the convenience sample had all requested radon kits prior to the receipt of their survey packets, and thus, their intention to test for radon is assumed. However, radon testing intentions (i.e. stage of testing) were not significantly associated with radon test kit orders when the analysis included only those in the random group. There are several possible explanations for this. First, some respondents may have requested the test kit simply because it was free, even if they did not plan on using the kit. This is suggested by the significant relationship between those who reported obtaining the test kit because it was free and those who ordered the kit (χ2 = 4.04, p = 0.04). Second, there may have been temporal issues that are impossible to decipher. This is because the coupon for the free test kit was included in the survey packet due to time constraints and as approved by the Kentucky Cabinet for Health and Family Services Internal Review Board. Since the coupon insertion in the research packet was a sort of intervention, it is difficult to determine if the PAPM stage of testing reported by the respondent was the same stage the respondent was in when the coupon for the free kit was returned. Weinstein noted that direction of causality for factors related to or predictive of test kit orders would be clearer if the 249 test kit is not offered until after the survey is returned (Weinstein et al., 1990). This is an important consideration for future prospective studies. Limitations Selectional bias is a possible limitation in this study. The Kentucky Radon Program does not currently have the ability to obtain test results from tests purchased from commercial sources, such as home improvement stores. Therefore, the study sample, particularly those in the convenience subsample, may not represent the general population of individuals who elect to test their homes for radon, which may decrease the generalization of the study findings. Selection bias may also result if the individuals who chose to participate in this study are different than those who did not. For example, study participants in the random sample were more likely to have graduated from college and had higher incomes than most in their respective counties (see Table 1). It is also important to note the small number of individuals who were current smokers in this study. The primary reason that this sample had a higher socioeconomic status and lower smoking rate than the general population is that only homeowners were selected for inclusion in this study, and those higher a socioeconomic status are more likely to be homeowners (Haurin, Herbert, & Rosenthal, 2007). Although survey methods were used to increase response rates (Dillman et al., 2009), the relatively low response rate limits the generalization of the study findings. Those who chose to respond to the survey may have had more knowledge about radon and would have been more familiar with radon testing than non-responders, which may have biased survey responses. Two additional factors in the sample bias the study results. First, there were five participants who reported having lung cancer, which was 4.1% of the sample. Individuals with a personal or family history of lung cancer may have been more likely to be selected for this 250 sample because they were more likely to request a test kit from the Kentucky Radon Program or more likely to respond to the survey packets sent to those randomly selected from the five counties. Individuals with a history of exposure (i.e. to lung cancer) may respond to risk information differently than others (Smith & Johnson, 1988). Second, although all of those receiving a radon test kit were given a copy of the EPA publication “A Citizen’s Guide to Radon,” those in the convenience sample were given this information in the same mailing as the research survey packet, as per standard protocol of the Kentucky Radon Program. Although participants were instructed to answer the survey questions according to their thoughts about radon prior to receiving the research survey packet, this informational brochure may have impacted their survey responses. A final limitation of this study is its cross-sectional design. Although this study provides important information about associations between study variables and intentions to test for radon, it is unable to provide predictions about directional causality. Further research utilizing prospective design is needed to determine the impact of study variables, particularly synergistic risk perception, on the transition between stages of the adoption of radon testing behavior. Conclusion There are three areas in which this study contributes to existing knowledge. First, this study demonstrates that lack of perceived susceptibility, primarily the lack of belief that there is a radon problem in one’s home, contributed to decisions not to test for radon. Second, synergistic risk perception (i.e. the perception about the combined risk of tobacco smoke and radon exposure) was not significantly related to radon testing intentions, and most respondents did not correctly understand the synergistic relationship between smoking and radon. There were no differences in synergistic risk perception or testing intentions based on smoking status. Further 251 research is needed to investigate the influence of synergistic risk perception on radon testing behaviors in current and former smokers. Third, this study identifies factors related to radon testing intentions among Kentucky homeowners, including social influence, the prompt of a free test kit, and the impact of media outreach on increasing radon testing. These findings will be utilized to increase residential radon testing and decrease lung cancer risk among Kentucky homeowners. Acknowledgments This study was a collaborative effort with the Kentucky Radon Program, Kentucky Cabinet for Health and Family Services. Many thanks to my dissertation committee co-chairs, Dr. Ellen Hahn and Dr. Mary Kay Rayens, for their guidance in this project. 252 Figure 1. Application of the Precaution Adoption Process Model to Radon Risk Reduction 253 Figure 2. Factors associated with residential radon testing intentions: Study protocol 254 Table 1. Purposive sample of five Kentucky counties for random group County Rural 2000 Region Received Percent 1 Census of Radon Adult Population Kentuck Outreach in Smoking 2 y past few (2000years3 2004)4 Radon Averag e pCi/L (20002004)3 Lung Cancer Incidence (20002004)5 Median Household Income6 Percent college graduates7 Hancock Yes 8,392 Western No 18.1% 3.00 115.13 $47,558 8.1% Henderson No 44,829 Western No 33.7% 1.84 92.96 $41,692 13.8% Louisville- No Jefferson 693,604 Central Yes 25.2% 6.43 99.54 $43,677 24.8% Rowan Yes 22,094 Eastern No 27.6% 1.39 87.42 $34,278 21.9% Whitley Yes 35,865 Eastern No 33.2% 2.11 121.45 $27,424 13.4% 1 Economic Research Service Rural-Urban Commuting Areas (rural defined as RUCA codes 4-10) http://www.ers.usda.gov/data/ruraldefinitions/KY.pdf 2 U.S. Census Bureau, 2000 3 Kentucky Radon Program 4 BRFSS, 2000-2004 5 Kentucky Cancer Registry, 2000-2004 age-adjusted incidence rate 6 U.S. Census Bureau, 2007 7 Percentage of individuals aged 25 or older with a Bachelor's degree or higher, U.S. Census Bureau, 2000 255 Table 2. Descriptive statistics and group comparisons between random and convenience samples: Percentages and frequencies (N = 122) Total1 Convenience Random Test Sample Sample Statistic2 (n = 67) (n = 55) χ2 = 0.00 Gender Female Male Annual Household Income < $50,000 ≥ $50,000 Education Level < College Graduate ≥ College Graduate 62.2% (74) 37.8% (45) 61.2% (41) 37.3% (25) 60.0% (33) 36.4% (20) χ2 = 1.53 28.2% (31) 71.8% (79) 20.9% (14) 68.7% (46) 30.9% (17) 60.0% (33) χ2 = 2.14 49.2% (58) 50.8% (60) 41.8% (28) 55.2% (37) 54.5% (30) 41.8% (23) 98.3% (117) 1.7% (2) 95.5% (55) 3.0% (2) 96.4% (53) ------- 10.0% (12) 90.0% (108) 35.0% (42) 65.0% (78) 55.0% (66) 45.0% (49) 9.0% (6) 89.6% (60) 35.8% (24) 62.7% (42) 53.7% (36) 44.8% (30) 10.9% (6) 87.3% (48) 32.7% (18) 65.5% (36) 54.5% (30) 43.6% (24) Race Caucasian Minority Smoking Status Current Yes No Former Yes No Never Yes No How much heard about radon Nothing 11.5% (13) A little bit 28.7% (35) Some 49.2% (60) A great deal 11.5% (14) χ2 = 0.13 χ2 = 0.12 χ2 = 0.01 χ2 = 12.15* 3.0% (2) 25.4% (17) 59.7% (40) 11.9% (7) 9.7% (11) 32.7% (18) 36.4% (20) 10.9% (6) 1 Sum may not equal N (122) and percentage may not equal 100% due to missing values Group comparisons based on Rao-Scott chi-square test * p < 0.05 2 256 Table 3. Descriptive statistics and group comparisons between those who planned to test and those who did not: Percentages and frequencies (N = 122) Total1 Planning Not Planning 2 to Test to Test (n = 69) (n = 53) Test Statistic3 χ2 = 0.07 Gender Female Male Annual Household Income < $50,000 ≥ $50,000 Education Level < College Graduate ≥ College Graduate 62.2% (74) 37.8% (45) 62.3% (43) 36.2% (25) 58.5% (31) 37.7% (20) χ2 = 1.70 28.2% (31) 71.8% (79) 21.7% (15) 71.0% (49) 30.2% (16) 56.6% (30) χ2 = 8.69** 49.2% (58) 50.8% (60) 36.2% (25) 60.9% (42) 62.3% (33) 34.0% (18) 98.3% (117) 1.7% (2) 95.7% (66) 2.9% (2) 96.2% (51) -------- Race Caucasian Minority Smoking Status Current Yes 10.1% (12) 10.2% (7) No 91.9% (102) 88.4% (61) Former Yes 35.0% (42) 30.4% (21) No 65.0% (78) 61.8% (47) Never Yes 55.0% (66) 58.0% (40) No 45.0% (54) 40.6% (28) Current secondhand smoke exposure at home or work Yes 31.9% (36) 27.5% (19) No 68.1% (77) 68.1% (47) Children under 10 in household Yes 26.2% (32) No 73.8% (90) χ2 = 0.02 9.4% (5) 88.7% (47) 39.6% (21) 58.5% (31) 49.1% (26) 49.1% (26) χ2 = 1.17 χ2 = 0.93 χ2 = 0.69 32.1% (17) 56.6% (30) χ2 = 0.00 26.1% (18) 73.9% (51) 26.4% (14) 73.6% (39) 1 Sum may not equal N (122) due to missing values “Planning to test” indicates those in Precaution Adoption Process Model stages 5 and 6 3 Group comparisons based on Rao-Scott chi-square test * p < 0.05; ** p < 0.01; *** p < 0.001 2 257 Table 4. Reasons given for not testing for radon: Percentages, frequencies, and comparisons between groups (N = 122) __________________________________________________________________________________________________________ Total Planning Not Planning Rao-Scott to Test to Test Chi-Square Test (n = 69) (n = 53) ____________________________________________________________________________________________________________ Radon risk is exaggerated 10.7% (13) 11.6% (8) 9.4% (5) 0.17 No radon problems in my area 23.8% (29) 13.0% (9) 37.7% (20) 9.81** Don’t think I have a problem in my home 36.1% (44) 21.7% (15) 54.7% (29) 13.73*** Neighbor’s readings are low 0.8% (1) -------1.9% (1) Will wait to see what others in the community find 9.8% (12) 4.3% (3) 17.0% (9) 5.27* Not interested 9.0% (11) 7.2% (5) 11.3% (6) 0.57 Getting rid of radon would be too hard/expensive 23.8% (29) 27.5% (19) 18.9% (10) 1.35 Didn’t know it was possible to test 14.8% (18) 10.1% (7) 20.8% (11) 2.57 Don’t know what testing method is best 21.3% (26) 15.9% (11) 28.3% (15) 3.00 Don’t know how to get a radon test kit 26.2% (32) 20.3% (14) 34.0% (18) 2.74 Costs too much to test 16.4% (20) 14.5% (10) 18.9% (10) 0.37 Results won’t be kept confidential 3.3% (4) 2.9% (2) 3.8% (2) 0.06 Haven’t gotten around to it 25.4% (31) 27.5% (19) 22.6% (12) 0.44 Takes too long to get test results 4.9% (6) 2.9% (2) 7.5% (4) 1.34 Already decided to test or have tested 29.5% (36) 50.7% (35) 1.9% (1) 35.04*** Other reason 9.0% (11) 7.2% (5) 11.3% (6) 0.57 ____________________________________________________________________________________________________________ Participants were asked to select all responses that were applicable, so percentages do not equal 100%. * p < 0.05; ** p < 0.01; *** p < 0.001 258 Table 5. Factors associated with radon testing intentions among Kentucky homeowners (N = 122) OR 95% Confidence Interval Age 1.00 (0.99, 1.00) 1.71 * (0.51, 5.77) Education ≥ College Graduate < College Graduate 3.67 * (1.10, 12.27) Current smoking Yes No 4.15 * (0.65, 26.61) Perceived susceptibility 1.82 (1.12, 2.96) Synergistic risk perception Yes No 3.01 * (0.90, 10.02) Perceived severity Serious or very serious Not serious or somewhat serious 8.25 * (2.46, 27.73) Know others who have tested for radon Yes No 6.31 * (2.09, 19.04) Perceived community concern about radon Yes No 3.68 * (1.16, 11.69) Gender Female Male * Reference Group 259 References Bonnin-Scaon, S., Lafon, P., Chasseigne, G., Mullet, E., & Sorum, P. C. (2002). Learning the relationship between smoking, drinking alcohol, and the risk of esophageal cancer. Health Education Research, 17(4), 415-424. Centers for Disease Control and Prevention. (2007). 2008 BRFSS questionnaire. Atlanta: Centers for Disease Control and Prevention. Centers for Disease Control and Prevention. (2008). BRFSS prevalence data-tobacco use 2008. Retrieved August 4, 2009, from http://apps.nccd.cdc.gov/BRFSS/list.asp?cat=TU&yr=2008&qkey=4396&state=All. Dillman, D. A., Smyth, J. D., & Christian, L. M. (Eds.). (2009). Internet, mail, and mixed-mode surveys: The tailored design method (3rd ed.). Hoboken, NJ: John Wiley & Sons., Inc. DiPofi, J. A., LaTour, M. S., & Henthorne, T. L. (2001). The new social marketing challenge to promote radon testing. Health Mark Q, 19(1), 79-90. Duckworth, L. T., Frank-Stromborg, M., Oleckno, W. A., Duffy, P., & Burns, K. (2002). Relationship of perception of radon as a health risk and willingness to engage in radon testing and mitigation. Oncology Nursing Forum, 29(7), 1099-1107. Environics Research Group. (2007). 2007 National radon and indoor air quality survey (No. HC POR-06-94). Ottawa, Ontario, Canada: Health Canada. Environmental Protection Agency. (2009). A citizen's guide to radon: The guide to protecting yourself and your family from radon (No. EPA 402/K-09/001). Washington, D. C.: Environmental Protection Agency. Environmental Protection Agency. (n.d.). Map of radon zones for Kentucky. Retrieved June 30, 2009, from http://www.epa.gov/radon/states/images/kentucky.gif. 260 Field, R. W., Kross, B. C., & Vust, L. J. (1993). Radon testing behavior in a sample of individuals with high home radon screening measurements. Risk Analysis, 13(4), 441447. French, D. P., Gayton, E. L., Burton, J., Thorogood, M., & Marteau, T. M. (2002). Measuring perceptions of synergistic circulatory disease risk due to smoking and the oral contraceptive pill. Risk Analysis, 22(6), 1139-1151. Gregory, B., & Jalbert, P. P. (n.d.). National radon results: 1985 to 2003. Washington, D.C.: Environmental Protection Agency. Halpern, M. T., & Warner, K. E. (1994). Radon risk perception and testing: Sociodemographic correlates. Journal of Environmental Health, 56(7), 31-35. Hampson, S. E., Andrews, J. A., Barckley, M., Lichtenstein, E., & Lee, M. E. (2000). Conscientiousness, perceived risk, and risk-reduction behaviors: A preliminary study. Health Psychology, 19(5), 496-500. Hampson, S. E., Andrews, J. A., Lee, M. E., Foster, L. S., Glasgow, R. E., & Lichtenstein, E. (1998). Lay understanding of synergistic risk: The case of radon and cigarette smoking. Risk Analysis, 18(3), 343-350. Hampson, S. E., Andrews, J. A., Lee, M. E., Lichtenstein, E., & Barckley, M. (2000). Radon and cigarette smoking: Perceptions of this synergistic health risk. Health Psychology, 19(3), 247-252. Haurin, D. R., Herbert, C. E., & Rosenthal, S. S. (2007). Home ownership gaps among lowincome and minority households. Citiscape: A Journal of Policy Development and Research, 9 (2), 5-52. 261 Hill, W. G., Butterfield, P., & Larsson, L. S. (2006). Rural parents' perceptions of risks associated with their children's exposure to radon. Public Health Nursing, 23(5), 392399. Kennedy, C. J., Probart, C. K., & Dorman, S. M. (1991). The relationship between radon knowledge, concern and behavior, and health values, health locus of control and preventive health behaviors. Health Education Quarterly, 18(3), 319-329. Kentucky Cabinet for Health and Family Services. (2004). Kentucky radon zone map. Retrieved June 30, 2009, from http://chfs.ky.gov/NR/rdonlyres/F4045C71-D661-468693F2-F25003397B6D/0/KentuckyCountiesMap1.pdf. Lichtenstein, E., Boles, S. M., Lee, M. E., Hampson, S. E., Glasgow, R. E., & Fellows, J. (2008). Using radon risk to motivate smoking reduction II: Randomized evaluation of brief telephone counseling and a targeted video. Health Education Research, 23(2), 191-201. Mainous, A. G., 3rd, & Hagen, M. D. (1993). Public perceptions of radon risk. Family Practice Residency Journal, 13(1), 63-69. National Cancer Institute. (2007). SEER Cancer Statistics: Cancer of the Lung and Bronchus. Retrieved October 10, 2007, from http://seer.cancer.gov/statfacts/html/lungb.html?statfacts_page=lungb.html&x=15&y=16. National Cancer Institute. (2008). Incidence rates table: Lung & bronchus cancer. Retrieved August 4, 2009, from http://statecancerprofiles.cancer.gov/cgibin/quickprofiles/profile.pl?21&047. National Research Council. (1999). Health effects of exposure to radon: BEIR VI. Washington, D.C.: National Academy Press. 262 Park, E., Scherer, C.-W., & Glynn, C.-J. (2001). Community involvement and risk perception at personal and societal levels. Health, Risk and Society, 3(3), 281-292. Sandman, P. M., & Weinstein, N. D. (1993). Predictors of home radon testing and implications for testing promotion programs. Health Education Quarterly, 20(4), 471-487. SAS, Institute, & Inc. (2004). SAS 9.1.2. Cary, NC: SAS Institute Inc. Smith, V. K., & Johnson, F. R. (1988). How do risk perceptions respond to information? The case of radon. The Review of Economics and Statistics, 70(1), 1-8. Speizer, F. E. (1982). Nurses' health study questionnaire. Boston: Harvard Medical School. Wang, Y., Ju, C., Stark, A. D., & Teresi, N. (2000). Radon awareness, testing, and remediation survey among New York State residents. Health Physics, 78(6), 641-647. Weinstein, N. D. (1988). The precaution adoption process. Health Psychology, 7(4), 355-386. Weinstein, N. D., Lyon, J. E., Sandman, P. M., & Cuite, C. L. (1998). Experimental evidence for stages of health behavior change: the precaution adoption process model applied to home radon testing. Health Psychology, 17(5), 445-453. Weinstein, N. D., & Sandman, P. M. (1992a). A model of the precaution adoption process: Evidence from home radon testing. Health Psychology, 11(3), 170-180. Weinstein, N. D., & Sandman, P. M. (1992b). Predicting homeowners mitigation responses to radon test data. Journal of Social Issues, 48(4), 63-83. Weinstein, N. D., & Sandman, P. M. (2002). The precaution adoption process model. In K. Glanz, Rimer, B. K., & Lewis, F. M. (Ed.), Health behavior and health education: Theory, research, and practice (pp. 121-143). San Francisco: Jossey-Bass. Weinstein, N. D., Sandman, P. M., & Roberts, N. E. (1990). Determinants of self-protective behavior: Home radon testing. Journal of Applied Social Psychology, 20(10), 783-801. 263 Weinstein, N. D., Sandman, P. M., & Roberts, N. E. (1991). Perceived susceptibility and selfprotective behavior: A field experiment to encourage home radon testing. Health Psychology, 10(1), 25-33. 264 ENERGY LOSSES AND OPERATIONAL COSTS OF RADON MITIGATION SYSTEMS L. Moorman, Ph.D. Radon Home Measurement and Mitigation, Inc. Fort Collins, CO ABSTRACT Energy losses and operational costs of radon mitigation systems in typical locations within the various heating zones in the USA have been calculated that include a modeling of seasonal and daily temperature variations. Two types of energy losses are compared with the electrical energy consumption for various ventilators that will be presented graphically. These energy losses will also be compared with the losses for higher quality radon mitigation systems, where the yearly energy losses are designed to be minimal. Optimization does not lead to the same results for all locations. The results will also be presented as all inclusive formulas for operational energy losses and costs including altitude corrections. The applicability of these calculations will be discussed and the case will be made that the knowledge of these issues may help an individual radon mitigation company to improve its competitiveness. ENERGY LOSS TYPES RELEVANT TO RADON MITIGATION SYSTEMS Operational energy costs of radon mitigation systems are either ignored or merely partially taken into account when various mitigation systems are compared. The cost that is typically recognized is (1) the direct electrical cost to operate the ventilator, but consists also of (2) conductive thermal losses and (3) convective warm and cool air losses. In the winter due to the operation of a furnace the additional operational energy costs consist of thermal conduction through the boundaries of the building (referred to later as W2), and warm air lost by convective replacement of warm air with cold air from outside the building (W3). In the summer, when an air conditioner or other cooling system exists, additional energy losses consist of conductive thermal losses (S2) conductions through the boundaries of the building and cooled air lost by convective replacement of cooled air with warm air from outside the building (S3). In calculating the total energy burden the efficiencies of equipment to effectively heat the replaced air in the winter and cool air in the summer must be included, and when operational cost rates to the occupant are calculated typical utility energy cost rates have to be included. 265 The direct electrical yearly energy to operate the fan (1) is easiest to measure or estimat. Measuring pressure differences across the ventilator can be accomplished directly with a digital micro-manometer that can be combined with the information from the fan curves of the manufacturer. From this information the operational volumetric flow rate can be calculated for each section of pipe. The electrical energy costs of ventilators can be approximated by its rating. Although the actual operation point for a ventilator will be below this rated power, the operational electrical energy used will be close to the maximum energy rating when large air flow rates are observed. Conductive energy losses (W2 and S2) are hard to measure and exist in the winter as cooling of the bottom of the concrete slab by air movement caused by the system under the slab and thermal diffusion from the warm top to cold bottom of the slab (or membrane in case of a crawlspace) and vice versa in the summer. In general wooden residential buildings with floating slabs in the dryer climates are constructed with conductive losses much smaller than convective losses and it is estimated that the additional conductive losses fall within the uncertainty of the calculations of the convective losses that will follow. Convective energy losses (W3 and S3) during heating and cooling seasons generated by a radon system, are caused by gaps and openings in concrete slabs of basements, sump pit openings, and between improperly sealed membrane-to-foundation wall boundaries, and at overlaps of membrane sections. The effect of this is not negligible and is not often included in an energy calculation. There are two potential exceptions to radon systems causing additional convective energy losses. One exception (A) is when the additional convective air loss by the radon mitigation system from the building has the effect to reduce the loss of a fraction of air loss elsewhere from the building envelope for example by raising the zero pressure plain inside the building. This effect has been noticed in another study but Fig. 1: Energy (dashed) and Air flow (solid) diagram for a radon system extracting warm air from the house in the winter. Energy: Top of diagram, Air flow bottom of diagram, Inside vs. outside the house is represented by the left vs right side of the diagram. 266 and in that study did not completely reverse the effective flow in the upstairs [1]. Another exception (B) is when a fraction of the air loss that the radon system is exhausting is unconditioned air that entered the building before it had a chance to fully mix with warm air and reach a new thermal equilibrium [2], which can be the case in a cold crawlspace. In both cases the fraction of exhausted cold air mass does not contribute to additional energy losses compared to what the building would have suffered without the radon mitigation system. In this paper we will assume that both of these cases do not apply to the situation that will be discussed. The reason for this is that in most houses with a furnace or conventional boiler or heater, by Code, one or two required make-up fresh-air openings in the basement exist preventing back drafting of the furnace, but also allowing most of the additional evacuated air by the radon mitigation system to be drawn in at this location and be mixed with the interior basement air. Thus in the generic situation the air that entered mixes fully with inside air before it is removed from the house by the radon system. However residences entirely operating on electrical appliances, or Solar or Green houses with their specific energy circumstances may have to be investigated separately. In Figure 1 an energy-air-flow diagram is shown for a radon mitigation system in the winter. The bottom half of the diagram indicates the relevant air flow movement and top half the energy flow. In addition items in the left side of the diagram are events occurring inside the house and in the right side of the diagram outside of the house. The diagram indicates that the system draws in warm air from the inside of the building that bypasses the lowest slab of a house through its openings in the winter. Thus the first line of defense against convective heat losses is to seal all openings from the interior of the house through and around concrete slabs to the soil with a durable caulk. The caulk has to have enough elasticity to allow for small movements of components it is adhered to without breaking, such as in the case of floating slab to foundation wall joints. Therefore special caulks are applied and insulating foams are rarely used successfully because they do not have the elastic durability, Fig. 2: Energy/Air-flow diagram for a radon mitigation system extracting cool air from the house in the summer. 267 resistive porosity, and adherent quality we need for radon mitigation systems. A better sealing of these openings results in lower convective energy losses. Given the examples mentioned an effective estimate of 50% convective energy losses is a good first guess for calculation purposes in sub-slab situations with gravel as is prescribed in certain Codes. Finally a distinction must be made between heating and cooling losses when various energy sources are being used. In these calculation we assume that in the winter a normal gas fired furnace or boiler with 70% efficiency is used and in the summer a whole house air conditioning unit is used with moderate energy efficiency whose compressor does work and uses electrical energy. MEASURING AN AIR REMOVAL RATE FROM A BUILDING AND THE EFFECT OF SEALING THE SLAB An unfinished basement was constructed with floating slab and gravel as sub-slab material, and an interior drain tile pipe and all openings through the slab were appropriately sealed during this research. Openings included all expansion joints near floor and walls, control joints interior to the slab in 10x10 square foot sections, whether it had cracked yet or not, to provide protection for future cracking, openings around plumbing pipes, rough-in openings, and all additional cracks visible in the slab. The existing sump pit opening was sealed with a hard cover transparent polycarbonate material to accommodate future visual inspection. Figure 3 shows the results of localized, single point, measurements of the sub-slab difference pressure across the thickness of the concrete slab material at different distances from the radon extraction cavity that were made before the sealing process was initiated. After the sealing process was completed the measurements were repeated the base pressure drop at the cavity can be roughly translated into an effective flow rate through the radon system. Using the published ventilator curves from the manufacturer of this ventilator the effective air removal rate can be estimated. In this case an effective removal rate of 280 cfm was reduced to 143 cfm just by complete sealing of the slab. This means that an effective additional air removal rate of 137 cfm is taken from this basement by the radon mitigation system with this ventilator when the basement remained unsealed compared to when the basement was sealed. This derivation was rough and ignores all other parts of the mitigation system. A more sophisticated analysis [3] taking the friction of the equivalent length of the piping in this 1-branch system into account allowed a modeling of the proper sub-slab cavity pressures resulting in flow rates of 167 cfm and 97.8 cfm The resulting cavity resistance change that was used to model this was 6.65 10-6 min/cf and 3.65 10-5 min/cf. This demonstrates what size changes of cavity resistance can be accomplished by sealing alone. The total effective flow rate change was 69.2 cfm. This type of data should have a consequence for new home construction design. If RRNC building techniques require gravel (or gravel is chosen) and these systems do not require sealing of the slab at the same time, or sealing is not properly inspected by the municipal inspector, basements have been observed to be finished immediately during construction or later without appropriate sealing of the slab, despite new home building 268 Codes that require it. This will have the consequence that the building occupant will be burdened with additional indoor air losses of a similar magnitude as we just presented when radon systems are installed while proper sealing is no longer possible because openings and gaps are not reachable any more. Moreover, expansion joints of 40 year old homes have been observed to have been completely deteriorated, which will affect the operation of currently installed radon systems well into the future when such joints could not be sealed during installation. Techniques to seal expansion joints of finished basements are either very costly (remove drywall and plates to reach the joints) or messy (clean, vacuum and caulk the joint through the drywall) and satisfactory methods are not existing at this moment, but will be needed in the future when advanced diagnostic methods will show which houses have unacceptably high energy losses for the occupants living in them in the future. This is a current challenge for the radon industry to solve. This paper will address the question what magnitude of energy losses and costs can be expected in these situations. If control grooves in slabs are not caulked, or are sealed with the wrong caulk material that tends to harden too much and crack later, or grooves are not properly cleaned and vacuumed before caulking and caulk has been observed to peal loose with time, a significant additional operational air removal loss can be introduced by activation of passive radon removal systems, even when they are Fig. 3: Effect of sealing an unfinished basement on the radon pressures and system air removal flow from the building with large gravel and interior perimeter drain under slab. The reduction in air removal rates based on complete simulation of the system [1] shows that in this case a removal rate of 69.2 cfm air is prevented from the building, when complete sealing is applied before finishing the basement (Ventilator power is 150 W). 269 activated much later. An obvious advantage is that once the floor is sealed a ventilator with smaller power can be chosen to mitigate the same building, reducing air losses even further as well as reducing direct electrical costs and Air Conditional losses in the summer. As an additional benefit this causes a lower noise burden on the living area. If in the example of figure 3 all of the remaining air flow through the radon mitigation system came from the sub-slab material under the building that ultimately came from the surrounding outside soil a reduced energy loss for this radon removal system would have been introduced. As a second example we show in Figure 4 a similar situation with pea-size gravel and no interior perimeter drain. It can be seen that the subslab resistance for air flow is reduced and is not symmetric around the cavity because of the two data points near 27 ft distance that have quite different values. When sand and clay are used the heating losses are reduced even more because of the low porosity of the material; however when low resistance channels exist in the sub-slab materials cracks and expansion joints can be reached and can provide significant pathways with higher flow with similar concern for operational energy and cost losses for the system. Fig. 4: Effect of sealing an unfinished basement on the radon pressures and system air removal flow from the building with sand and pea gravel under the slab and with no interior perimeter drain. The reduction in air removal rates based on the complete systems analysis shows that in this case a removal rate of 28.2 cfm air is prevented from the building, when complete sealing is applied before finishing the basement (Ventilator rating is 83 W). RADON REMOVAL EFFECTIVENESS IN VERY LOW ENERGY LOSS SYSTEMS The two examples presented before indicate the importance of proper sealing when leaky slabs are encountered. Although improvements can be made in existing homes there is no substitute for being able to work the problem from the ground up. In the example we will look at in this section a Radon Risk Evaluation had been performed on 270 a location where the house was going to be constructed. The basement had been excavated and the concrete foundation walls poured. The Radon Risk Evaluation indicated that the finished house would have had a radon concentration of approximately 150 pCi/L (See Fig. 5). We were asked to install our most effective and energy efficient radon mitigation system that the owners preferably would like to see work as a passive system, to avoid any noises added to the house and surroundings. A double barrier passive system was installed that resulted after completion in a measured radon concentration (2 day short term test, closed house conditions) of approximately 25 pCi/L. During this time the plumber had left an 8 inch slit in the crawlspace membrane near the water heater. After repair and activation of the membrane very low levels were reached. A follow up measurement operating the system as a passive system measured a level of 5.2 pCi/L. Additional measurements with a Continuous Radon Monitor (femtotech CRM 510) were done by operating various fans and electronically operating fans at various Fig. 5: Radon Risk Evaluation and double barrier passive and active operation of high energy efficient combined ASD and membrane radon mitigation system. rotational speed. Pressures across the fan were measured whose flow rates had been compared with measurements in a bench set up elsewhere. From figure 5 it can be seen that the measured radon levels at different flow rates had a background. A theoretical fit based on a volume theory resulting in a hyperbolic shape showed a fit as shown that indicated a background of 0.85 pCi/L. This background was reached for an extra 271 powerful fan operating at its full power at 150 Watt which with the low frictions in this two-branch mitigation system was at flow rates of 150 cfm. Similarly we found that even at very small flow rates the radon level did not rise above 2 pCi/L even at a flow rate of as little as 20 cfm. The ventilator used was a 12 V-DC fan drawing a power of 3.2 Watt. The physical dimensions of the ventilator was a 2 inches diameter axial fan that was inserted inside the 4-inch PVC pipe on a perpendicular polycarbonate holder that allowed leakage for water around it through the pipe. From this graph we can conclude that with limited testing not much benefit seems to be indicated to run a 150 cfm 150 Watt commercial radon ventilator compared to a 3.2 Watt ventilator. However it must be kept in mind that the membrane system was hermetically double-sealed below the slab and a full interior perimeter drainpipe was inserted in the gravel under the double membrane, which treatment can only be done during new home construction. Thus the question is if it is possible to reduce energy losses through sealing and reducing the power of the ventilator while maintaining a high radon removal efficiency. A theoretical fit curve of the data indicates that to minimize the radon concentration a maximum power of the ventilator would be required. The theoretical fit shows a Fig 6 Short term radon tests in the house after installation of a high energy efficient radon mitigation system allowing a number of ventilators to run at different power. Ventilator powers range from 3.2 W (20 cfm) to 150 W (150 cfm). background of 0.85 pCi/L that is not affected by the flow rate. Its interpretation is that this background is caused by a combination of radon in the outside air around the building and the combined sources of building materials inside the building. The choice of maximum air flow through the system would add unnecessary energy losses and additional noise to the house to the extent that a homeowner may not want to 272 live with. This is an example in support of the fact that the requirement in the author’s opinion for a ‘higher quality’ radon mitigation system is not to uniquely ‘maximize’ the radon reduction, but to ‘optimize’ a number of variables. The four desired variables to optimize simultaneously are (1) maximizing radon removal, (2) minimizing energy losses, (3) minimizing the noise burden on the most sensitive living areas, and (4) minimizing visibility of the system. The latter is added such as not to affect curb appeal and visibility from a backyard and low visibility through the most frequently visited areas in a house. Systems installed in this way will be functional and avoid adding a burden to the house. When done well in new home construction by conscientious mitigators passive mitigation systems can be designed optimally in this sense and provide for an uncomplicated activation, if radon tests show the need later. A MODEL FOR EXTERIOR TEMPERATURES In order to calculate which energy losses a radon mitigation system adds to the utility costs of an occupied residence we must first define the baseline, which is the outside environment with which the house exchanges air. In the industry and academics a useful measure in terms of heating and cooling degree days has been used for decades [4]. Limitations of the method are that it only takes into account temperatures due to space heating, not when solar heating becomes important. In this Sense the CDD has limitations. It assumes the Heat Loss is proportional to the temperature difference. In most cases cooling is applied only in a few rooms which limits Fig. 7: Example of a match of a three parameter model that fits the seasonal cycle of temperatures for thirty year average monthly Heating Degree Days and Cooling Degree Days for Fort Collins CO. February’s shorter month causes a seeming irregularity. 273 the usefulness of the method. [5] A single heating degree day based on the reference temperature 65 degrees Fahrenheit is defined by the average temperature during one day to be 1 degree Fahrenheit below 65 degrees, relating to the assumption that the occupant would want to heat the residence during that day by an average of 1 degree to bring it from 64 to 65 oF. Five different Climate Zones are classified for the HVAC industry in which specific recommendations for use and operation of furnaces and air conditions may vary across the Climate Zones. Based on the information of thirty year averages of heating degree days (19712001) [6] and cooling degree days (1961-1990) [7] a model for exterior temperatures can be made with an average amplitude describing the seasonal temperature cycle, and a number of days shift from January 1, can be made that approximates best the temperature for each day during the year. The consistency of this three parameter fit with the heating degree days curve as well as the cooling degree curve is shown in figure 7. The curve can be captured remarkably well with these three parameters. On top of this seasonal cycle a daily cycle is chosen with a temperature amplitude based on average high-low temperature differences within the day. The resulting model can be considered a reasonable approximation of the average hourly temperatures that we can give with a total of only four parameters. Fig. 8 Hourly modeling of outside temperatures in Fort Collins CO where the daily highs and lows are taken 17 oF up and down from the average temperature (black line). Heating of the house is set at the point where the outside temperature is below 68 F (red line) and cooling starts when the outside temperature is above 72 oF (blue line). The daily modeling parameters for outside temperatures in Fort Collins, CO, throughout the year are exemplified in figure 8. The outside temperatures vary between the two extremal green curves with black curve being the daily average The red horizontal line in this figure indicates the outside temperature 68 oF below which heating 274 Table 1 Locations across various Climate Zones with Fit parameters to the 30 year averaged HDD and CDD data City Index KEY WEST SAN DIEGO CHARLSTON AP ATLANTA NASHVILLE NEW YORK C.PARK INDIANAPOLIS FORT COLLINS SIOUX FALLS SAINT CLOUD CARIBOU ANCHORAGE DILLON KW SD Charl. Atl Na NY Ind. FC SiFa SntCl Car An Dil State Climate HDD Zone FL CA SC GA TN NY IN CO SD MN ME AK CO 5 4 5 4 4 3 2 2 1 1 1 1 1 64 1063 1973 2827 3658 4744 5521 6587 7746 8812 9505 10470 11208 CDD Temperature Seasonal O ( F) variation O ( F) 4798 87 7 984 64.7 7.8 2266 66 17 1667 62 18 1616 59 21 1096 55 22 1014 52 23 571 48.5 25 744 44 27 417 43 30 131 39 28 0 36 22 0 35 21 Shift Daily (days) Ampl. O ( F) 210 5 216 10 199 17 200 17 202 17 204 17 200 17 200 17 197 17 200 17 201 17 194 17 206 17 of the inside air in the house is chosen. The blue horizontal line gives the temperature 72 oF above which cooling of the inside air in the house is chosen. The Climate Zones follow the classification for heating zones. HDD is an abbreviation of Heating Degree Days and CDD for Cooling Degree Days which are often chosen with a reference temperature of 65 oF. Classification boundaries are shown in Table 2. Temperature, Seasonal Variation, in degrees Fahrenheit, and Shift in number of days with respect to January 1, are the best three parameter fit that was found to simulate both HDD and CDD based on their 12 monthly average recorded values over several decades at each City’s location. Daily amplitude is a reasonable guess based on location, with locations near coasts and equator assumed to have smaller daily variations than all other locations. Table 2: The definition of Climate Zones in Cooling Degree Days and Heating Degree Days with a reference temperature of 65 oF is used in the heating and cooling industry. Locations in all zones are represented in our calculations, see Table 1. 1 2 3 4 5 Fewer than 2,000 CDD and more than 7,000 HDD Fewer than 2,000 CDD between 5,500 to 7,000 HDD Fewer than 2,000 CDD and between 4,000 to 5,499 HDD Fewer than 2,000 CDD and fewer than 4,000 HDD between 2,000 CDD and 4,000 CDD 275 NUMERICAL CALCULATIONS OF OPERATIONAL COSTS The outside temperature simulations at each location are used in an hour by hour method to numerically calculate the convective heating and cooling losses: "Q perhour = C p , m n!T (1) The heating energy needed is expressed in Joule and Cp,m is the molar heat capacity of air at constant pressure, n is the number of molecules measured in moles of air (Avogadro: 6.022 1023molecules/mol) and ΔΤ is the temperature difference in Kelvin we need to accomplish in heating or cooling the gas. The molar heat capacity, Cp,m at constant pressure for a diatomic ideal gas, which is appropriate for heating dry air at atmospheric pressure, can be expressed in Ru, the universal gas constant, whose numerical constant is 8.314 Joule/mol K, as follows: 7 C p , m = Ru =29.085 J/(Mol K) (2) 2 which can be converted in values we will recognize in other formulas later. The heat capacity at constant pressure can be converted to a volumetric quantity: C p ,m = 5.6815 10-6 kWhr/(ft3 oF) (3) 3o = 0.019375 Btu/(ft F) The later value is well known in the heating industry as approximately 0.02 Btu/(ft3 oF). Using the ideal gas law, PV=nRuT the heat added to an air mass of n mole in volume V that is drawn into the house and that is increased by temperature ΔΤ can be written as: 7 ' PV $ )Q perhour = C p , m n!( = % (4) "!T 2& T # This means that a volume of cold air drawn to the inside, V, every hour is the product of a fractional loss, f, and the volumetric rate R moved by the ventilator through any cross section of the main radon vent pipe, which when expressed in cubic feet per minute (cfm) must be multiplied with 60 to calculate the air volume per hour and with 0.30483=0.02831687, which is the volume of 1 cubic foot expressed in m3. The temperature difference as calculated per hour and converted from degrees Fahrenheit, t, to absolute units in Kelvin, T, can be written as 5& , t )# (5) .T = $ti - * to - t A cos(2/ '! 9% + 24 (" The average temperature T at which the energy is heated, in equation 4, is taken to be the freezing point of water, which is 273 Kelvin, and is a good approximation for the average temperature of the air when winter air is heated, introducing only a small error. The outside daily average temperature in degrees Fahrenheit is simulated by: t to = t Avg + tSeasonalA cos(2! ) (6) 24 Converting the expressed energy from Joule to kWhr requires a factor 2.78 10-7 kWhr/J. The energy efficiency of the method to generate a unit of air volume must be included by dividing through the efficiency of the device that we choose to use. For a gas fired furnace the efficiency, g, will be taken here to be 70%, as was also indicated in figure 1. 276 Thus putting all factors together we can calculate the heating energy needed per hour expressed in kWhr per hour, by writing equation 4 as: 7 . 60 / 0.028317 / fR + ( 5 . t +% 2.78 " 10!7 (7) 1Q perhour = , P ) & , ti ! (t0 ! t A cos(20 )) )# 2273 24 *$ 0.70 *' 9 Using the value for 1 atmosphere, P=101,325 Pa, and by combining the constant factors we find for the energy loss in kWhr per hour by a simple rearrangement of the factors: 60 % 5.681 $10#6 t "Q perhour = fR(ti # (t0 # t A cos(2! ))) (8) 0.70 24 where the efficiency 0.70 shall be replaced by the Coefficient of Performance when we will be considering cooling of air instead of heating later. The hourly energy rate loss could be easily analytically integrated if this formula was valid during the entire year but this would not be realistic. To simulate the behavior that is closer to reality we use a criterion that switches the furnace on and off depending on the outside temperature condition. In these calculations we use the criterion that for all outside temperatures below the ti=68 oF the furnace will maintain the house to this target interior temperature by heating the air that is drawn in from the outside. Similarly for all values above 72 oF in the summer our simulation describes to cool the outside air with an electrical air conditioner. In case we are cooling outside air entering the house in the summer we will replace the furnace efficiency 0.70 by the Coefficient of Performance. The Coefficient of Performance is dimensionless and indicates how much heat energy (in any unit) can be removed from the cold reservoir of the air conditioner per unit energy work done by the unit. This was also indicated in figure 2. A typical air conditioner used these days in homes (not the newer high efficient type) with an energy efficiency rating (EER) of 8 Btu/Whr will have a Coefficient of Performance (COP) of 2.343. This is the value we will be using in this paper yet it can be easily adapted if a different COP is known in the final presentation of the results. In Figure 9 we have displayed the daily energy losses due to the operation of a radon mitigation system calculated based on the hourly modeling of exterior temperatures in Fort Collins. The horizontal scale starts on January 1. This simulation shows that for this location the energy losses predominantly are from the heating, next the electric loss form running the ventilator and the energy losses through cooling are smallest. The energy losses are converted to dollar costs based on the cost rate for the utility applicable i.e. Natural Gas cost rate for low outside temperatures during heating periods, and Electrical cost rate for air conditioning cooling when outside temperatures are too warm. The current cost Rates for energy sources are given in the table 3, and were calculated from the billings of local energy providers for 2009: Future estimated 277 Fig 9.: Energy losses due to operation of a radon mitigation systems calculated based on the hourly modeling of exterior temperatures indicated in Fig. 7 for Fort Collins, CO (Ventilator: 150 W ventilator, 120 cfm,, Internal losses: 60 cfm). rates are including expected rate increases within the next decade. It should be clear that depending on location and cost rates the ranking of importance of the three components in Figure 8 for the occupant in dollars spent may vary. Using these Energy cost rates and the previously discussed convective loss efficiency of 50% (LE=0.5) and Furnace or boiler efficiency of 70% (FE=0.7) and for the AC unit an EER of 8 Btu/Whr (COP of 2.343) we can calculate the operational costs per year for various ventilator configurations and locations. Table 3: Local utility energy rates in 2009 in comparison to future expected rates that were used in these calculations Energy source Natural Gas Electricity Local Current Energy Cost Rate (2009) 3.0 c/kWhr 7.8 c/kWhr Future Estimated Energy Cost Rate 4.0 c/kWhr 12.0 c/kWhr A large variation of operational costs was found with location. Total radon mitigation operational costs were found to vary from less than $225 per year at the warmest regions to $500 per year in the coldest regions considered for the largest air-losses considered in the calculations and using heating by gas. The largest contributor to operational costs was generally the heating cost, except for the warmest regions. 278 Figure 10: Additional Energy Cost calculated by using hourly computer model following parameters and abbreviations defined in Table 1 and the future estimated energy rates from table 2. THE BEST FIT FORMULA ACROSS LOCATIONS AND SYSTEMS In general the three most significant contributions to the energy losses are the electrical power operating the fan continuously and the two convective losses causing outside air to be heated in winter and cooled in summer. Rather than the detailed derivation from first principles shown in the previous section we will look here to start the description from the point of view of proportionalities that can be expected on reasonably grounds and using the numerical data we calculated in the previous section we will try to obtain effective proportionality constants for the largest range of realistic parameters. When this is accomplished and the range of variables for which the simplified formula is a good approximation this formula can be used within its validity ranges to calculate additional results with a reasonable degree of accuracy, without having to resort to additional numerical calculations. However it must be kept in mind that there is no guaranty this formula will be describing all data points correctly with a single set of 279 effective parameters. On the other hand if we wanted to we could have derived a single set of exact parameters for each individual location. Since the ventilator operates throughout the year the energy loss over the year is proportional to the power, P, with a proportionality constant, a: (9) ! = aP If we draw a fractional volume rate of air, f, from the interior of the house and the rest from the soil under the house, given a certain air removal rate by the ventilator, R, the energy loss due to heating or cooling will on the average be linear with the rate at which the air is removed from the interior, fR. When heating of the air is needed the energy loss rate is approximated to be proportional to the number of heating degree days for any given location, H, and inverse proportional to the efficiency, g, with which this heating occurs. Furthermore the heating energy will be proportional to the mass in the air at a certain volumetric rate that is replaced, thus it will be proportionally to the density of the air at the altitude where the house is located, which is proportional to the barometric altitude formula. This can be described as an exponential factor, where in a good approximation the pressure drops a factor 2 due to an altitude increase of 5.5 km, for which the parameter L will be used. Defining a proportionality constant, b, that will be interpreted later, leads to the simple relationship for all excess connecting heating energy loss due to an operating ASD radon mitigation system in units of energy: A " =b HfR ! L 2 g (10) Similarly, the excess convection cooling energy loss is approximated to be proportional to the number of cooling degree days for any given location, C, and inverse proportional to the efficiency, e, with which this cooling is accomplished by the Air Conditioning unit. Defining an effective proportionality constant, c, that will be interpreted later, leads to the simple relationship for all heating losses in units of energy: A CfR ! L " =c 2 e (11) Thus the total energy loss to the occupant of the house due to the radon mitigation system per year can be described by adding the three contributions: A ! H C L " = aP + (b + c ) fR 2 (12) g e The energy make-up efficiency in terms of heating of the air was included in the simulations, thus the proportionality constant, b, does not include the efficiency factor for a natural gas furnace (g=0.70), indicating how much energy of the source energy it 280 takes to produce a unit energy of heated air. For example when an electric heating element would be used this parameter would be 100% (g=1.0). When using the air conditioner, the efficiency is equal to that of a reversed heat pump with efficiency larger than 1. For an air conditioner with an Energy efficiency Ratio (EER) equal to 8, the coefficient of performance (COP) is 2.343, which means that for every Joule electrical energy used by the air conditioner 2.343 kWhr of heat is transported from the interior of the building to the outside. Figure 11: Best fit for largest group of data by adjusting the defined effective proportionality constants. Key West and to a lesser degree Dillon are the only significant deviations when these effective values of the parameters are used. The electrical cost of operating the ventilator can now be calculated by multiplying the first factor containing the power directly with the electrical energy cost rate factor, ε. The cost rate factor for the heating energy loss term is given by the energy cost rate of the source energy,γ, which although it differs in case natural gas, electric, oil or propane are used, can be accounted for easily using the formula. In addition to the Future estimated energy cost rates shown in table 3 we have chosen the following set of parameters in the formulas consistent with our earlier numerical calculations: ε = $ 0.12 /kWhr γ = $0.04 /kWhr g = 0.70 e = 2.343 A=0 ft The only parameter that is not realistic in many locations is the altitude, A, which is chosen to be the value at sea level for all locations (1 atmosphere= 101,325 Pa). This is 281 done since the pressure in the numerical simulations we compare with in equation 7 is also taken tot be 101,325 Pa. We will comment on altitude effects later. The complete formula for the operational cost to the home owner or occupant of operating a radon mitigation system with using natural gas to heat the house is thus: A U = a"P + (b# ! H C + c" ) fR 2 L g e = uP + (vHfR + wCfR )2 ! (13) A L All quantities are known except for the proportionality constants, a , b and c. The proportionality constant a can exactly be calculated because the ventilator is running all the time throughout the year. By expressing the utilities rate costs, U , in $/yr and the power P in Watts, the proportionality constant is simply the number of 1000 hours in a year (khr/yr). This number is 8.759. Since both b and c are effective parameters, their value can be determined by fitting the theoretical behavior to the simulated numerical data. The proportionality constant b is the costs in dollars per year for a unit of air. This method on the simulated data leads to the following best fit values excluding for Key West and Breckenridge, the extremal data points. Energy parameters a b c Cost parameters u v (for natural gas) w Fit Values 8.759 0.009283 0.01760 Future Current 1.051 0.6831 0.0005380 0.0004035 0.0004570 0.0002971 Units (kWhr/yr)/W=k hr/yr kWhr/(HDD cfm yr) kWhr/(CDD cfm yr) ($/(W yr) $/(HDD cfm yr) $/(CDD cfm yr) Table 3: Effective parameters that fit data well, except for Key West. The energy costs across Zones for various air removal rates at the projected source energy rates are given as data points in the following figure: The parameters b, c, v, w can be reinterpreted by writing them in terms of the heat capacity of air at constant pressure at 1 atmosphere pressure, as was introduced in equation (3) in kWhr/ft3 oF. This can be done by factoring out a few trivial factors. Because we are converting the quantity from unit cfm which has the time unit minute in it to HDD which has the unit day in it, the number of minutes in a day has to be factored out and we the result are written in terms of effective, dimensionless f-parameters. For the energy parameters this is: b = f b " 24 # 60 # 5.6815 " 10!6 kWhr /(HDD cfm yr) c = f c " 24 # 60 # 5.6815 " 10!6 kWhr /(CDD cfm yr) 282 Similarly, for the operation cost parameters, it makes sense to factor out the cost rate and efficiency of the utilities energy conversion: v = f v ! 24 " 60 " 5.6815 !10#6 ! 0.04 / 0.7 w = f w ! 24 " 60 " 5.6815 !10#6 ! 0.12 / 2.343 The resulting dimensionless f-parameters introduced in these equations are the various effective proportionalities close to unity. The following are the best result using the linear least square fitting method: Table 4: Least square fitted, best values of the dimensionless fit parameters that describe the numerical operational heating and cooling energy losses and costs. 1.134 fb fc 2.15 fv 1.15 fw 1.09 Figure 12.: Energy Cost formula compared to numerical simulation for various ventilator of various strengths and simulated air vent losses fans and scenarios. In addition to the data introduced in Fig. 10, three data points for New York were calculated for electric heat and the low power fan in Fort Collins. 283 These are displayed as dashed lines in the figures 11 for energy losses and 12 for energy costs. There are two reasons why the f-parameters are different from unity. This is because (1) the exterior temperature model takes into account daily variations, and (2) the switch-on temperatures of heating and cooling equipment was taken at 68 oF and 72 o F, respectively, which is different from he 65 oF for which the HDD and CDD values can be found in each location. Thus comparing with the HDD and CDD with reference base 65 oF at any location, we are working with somewhat misalligned variables. However this problem is taken care of by introduction of these f-parameters following from a least square fit across most of the HDD range.. A few additional data points are shown in Figure 12 both as numerically evaluated and using the formula, in order to see how accurate the formula works for a variety of situations. The three dashed data points were calculated by the simulation method for each of the three ventilator loss situations in New York and shifts are indicated to higher costs. In addition one data point for Fort Collins with the lowest energy fan introduced in chapter 3 is also shown in this Figure. Its’ volumetric rate was 20 cfm, but because of the special sealing applied, we estimate it to take no more than 5 cfm from the house. The energy efficiency of this system as a whole is shown to be impressively low compared to any other data point in the Figure. The formula was next applied for all data points and it can be seen that a good approximation of the formula to the three data points was obtained for electric heat. This shows reliability of the formula in predicting a variety of circumstances. The solid numerical data points are identical to the previous figure. ENERGY LOSSES RELATIVE TO ELECTRICAL COSTS In Figure 13 it is interesting to compare the ratio between heating and electrical Fig.13: The ratio of heating and electrical Costs across the HDD values. 284 cost to run the ventilator for each of the three considered mitigation scenarios: C b"HfR (14) Rh = h = Ce a!gP In general cooling contributes less than the electrical costs as shown in figure 14 and in zones 1,2 and 3 less than 40% of the electrical costs. In Caribou we see a value over 50%. Only in Key West we calculate a fraction above 60% of the electrical costs which has to be ignored due to its known inaccuracy of these parameters for Key West. Rc = C c cCfR = Ce aeP (15) Fig.14 The ratio of cooling and electrical costs across the explored HDD range Whereas heating contributes to most of the costs. For heating zones 4 and 5 with values under 2000 HDD the cooling can contribute equal or more than heating. In terms of the yearly total energy costs relative to the yearly electrical costs as shown in figure 15 for the midrange and more powerful ventilator considered we calculated a factor of 3 for Dillon, Colorado, but the moderate zones 2 and 3, this factor is calculated to be twice the electrical costs.. For the warmest regions, heating Zones 4 and 5 the total energy costs to electrical cost ratio is calculated to be generally smaller than a factor 285 2.5. The least powerful ventilator had the highest factors across the entire HDD range which were up to a factor 4.5 in the coolest climate of Zone 1. Rtot = 1 + Rh + Rc (16) Fig.15 The ratio of Total energy costs calculated and electrical Costs across the HDD values THE EFFECT OF ALTITUDE The effect of altitude on convective energy losses is directly related to the pressure at which the volume of air is extracted by the mitigation system from the house. The barometric altitude formula describes the pressure loss as a function of altitude due to a layer of air carrying all air above it while the layer itself is carried by all air below within the constant gravitational field of the earth. The exponentially decreasing formula with increasing altitude that results is known as the barometric altitude formula and can be written as: ! z P ( z ) = P0 e ! #z = P0 " 2 L (17) with β a known constant and P0 equal to 1 atmosphere at Sea level (at z=0 ft). The value of β allows us to write this formula alternatively as indicated with L approximately 5.5 km, the altitude increase where the pressure assumes half its value. As an example figure 16 indicates the relative change over the pressure range relevant to altitudes in most residential locations of the US. As an example for Fort Collins at 5100 ft altitude this value results in a relative pressure loss of 17.8%. Extreme locations, such as Twin Lakes, CO can go up to 41.6% loss of pressure compared to 1 atmosphere. From equation 4 we see that the heat added for the same volume at the same temperature at different altitudes is proportional with pressure, P, thus the same factor was taken into 286 account into the fitting formulas in equations 10-13 to describe the heat lost for a fixed volume rate fR at various altitudes. Fig 16: Relative pressure loss as a function of altitude following the barometric altitude formula for the altitude range of most homes in the USA. Fort Collins is at 5100 ft with 17.8% atmospheric pressure loss. Radon installations in Twin Lakes, CO were harder to accomplish than in Fort Collins. ENERGY LOSSES FOR ALTERNATIVE RADON MITIGATION SYSTEMS ERV/HRV systems are subject to similar type of heat losses as discussed, but the calculations here cannot be applied for the ventilation rates caused by these systems because in ERV/HRV systems a large fraction of the energy is recovered by an energy exchange medium, e.g. in the winter between a warm stream of air that is discharged out of the house to a cold stream of air that is pulled into the house. The air exchange medium makes the calculation we have done here not applicable on that type of situation. High energy efficient and electrostatic filtration to mitigate Radon Decay Products directly, not radon, do not exchange additional air with the outside environment, thus do not add additional energy losses beyond the already existing natural ventilation of the home. Thus from an energy efficiency point of view this technique maybe the most energy efficient method. However the number and power of ventilators may be very different from a radon ventilator in an ASD system and regular replacement costs of filters must be taken into account to determine operational costs. A more in depth analyses for each situation separately is necessary to reach a valid conclusion. 287 TOTAL AVERAGE COST COMPARISONS OVER LIFETIME OF SYSTEM Using a lifetime of a system of 40 years and for Fort Collins, Colorado which is in the middle of the HDD scale, the formula is used to calculate various components using equation 13 and the future variables for u , v, w from table 3. Durability of the ventilators is taken to have an average lifetime of 10 years. Installation costs are taken into account based on actual systems installed by us or others. In the histogram with the cost comparisons and power levels indicated along the base, the left four data each represent a system in Fort Collins without sealed slab or with finished basement that could not be sealed. The last data point represents the high energy efficient system presented as part of figure 6. The first data represent a system with the altitude correction. The second data represent a system without the altitude correction (sea level simulation), the effect is less than 19% because only a fraction of the energy losses involve air. It is clear that after 40 years the total cost per year to the home owners is lowest for the lowest power system, even when higher installation costs are included. In Figure 19 we have shown the cost development for the home owner including operational costs and ventilator replacement (proportionally taken into account) of each of these systems (calculated at sea level) accumulated over the years after installation. The open circles describe the cost tipping points where the occupant will have earned back their investment of the higher installation costs, after which it is clear that the system with lowest power ventilator will be most cost effective for the home owner. Radon levels are somewhat higher as indicated in figure 6. However even if higher Figure 18 Four systems are compared on their 40 year total costs to the consumer. The left four represent non-sealed systems, the right-most data represent a high energy efficient system (See Fig. 6) system hermetically sealed with a double membrane under the slab. Left two data differ only in that the actual altitude of Fort Collins (5100 ft) was taken into account in the calculation for operational costs of the first data point.. 288 power ventilators are used in the high efficiency systems they would not draw as much air from the basement as the system characteristics used in this calculation. From figure 19 we conclude that savings of radon mitigation systems are determined by how well they are installed, rather than any other parameter, specifically, they are not determined by the savings on installation costs. Figure 19 Comparison of accumulated costs in time to find out cross points for return on investment of a double membrane, completely sealed high energy efficient system compared to lower quality unsealed systems that have higher energy loss rates. CONCLUSIONS We considered various operational energy losses for radon mitigation systems. The convective heating, cooling losses and electrical costs were found to be the main energy losses. Measurements of air losses were discussed when gravel is applied below the concrete slab. It was concluded that substantial convective energy losses can be added to a house when proper sealing is not accomplished, or cannot be accomplished when the basement is finished. A case with an energy efficient system was discussed. It was concluded that although increased ventilation rates generate increased radon reduction, very satisfactory radon removal can be accomplished with extremely low power radon systems provided a high energy efficient passive radon mitigation system is installed during new home construction by conscientious mitigators. In order to perform numerical calculations for a variety of locations in the United States and across all climate zones, fit parameters for outside air temperature conditions describing a modulation with a yearly cycle and a daily cycle were evaluated for each location. Numerical calculations were performed of additional operational energy losses and costs employing realistic utility cost rates for the next decade. A large variation of operational costs was found as a function of location. Total radon mitigation operational costs for the midrange and highest power ventilators were calculated to vary from less than one and one half times, in the warmest locations, to up to three times the electrical costs in the coldest locations considered. The smallest power commercially available 289 ventilator considered had the largest loss ratios to electrical costs across all regions which reached up to four and a half in the coldest location. General formulas were derived with effective parameters to describe the numerical operational energy loss and cost data in an effective way across the largest possible fraction of the heating zones and ventilator powers. The effective parameters in the formulas derived by a least square fit method compared well with the numerical data except for the most extreme weather locations. The operational costs formula was also employed to look at alternatives such as electrical heating and high energy efficient systems and this was compared to a normal heating system using natural gas. These results evaluating the formula also compared well with direct numerical calculations. It was shown that altitude effects which had been left out of all numerical calculations on purpose can be taken into account by using the barometric altitude formula. The size of these effects was evaluated. Energy losses of alternative radon mitigation such as ERV systems were discussed, and it was concluded that the formulas derived here do not apply to HRV and ERV systems due to the nature how they recover energy. Similarly it was concluded that RDP mitigation systems cause less energy losses but add operational costs due to frequent filter replacements. A cost comparison over the lifetime of various systems in one location was made showing that over a forty year lifespan the costs saved by installing the highest energy efficient system during new home construction can be a four digit dollar amount. In this paper a case was made that for the highest quality radon mitigation systems one should look for the simultaneous “optimization” of four parameters, maximizing radon reduction, minimizing additional energy losses, minimizing noise effects and minimizing visual impact. REFERENCES 290 [1] B.Turk, J. Hughes. “Exploratory Study of Basement Moisture During Operation of ASD Radon Control Systems”, Revised 3/10/08 U.S. Environmental Protection Agency Indoor Environments Division Washington, http://www.epa.gov/radon/pubs/index.html [2] J. Bartholomew, Private Communication [3] L. Moorman, “Solving turbulent flow dynamics of complex, multiple branch radon mitigation systems”, AARST International Symposium Proceedings, 2008, Las Vegas, http://www.aarst.org/proceedings/2008 [4] J. Akauder, S Alvarez, G Johannesson, “Energy Normalization Techniques” p 59, in Energy Performance of residential buildings: a practical guide for energy rating and efficiency, edited by M Santamouris (2005), ISBN 1-902916-49-2 [5] J. E. Piper, “Operations and Maintenance Manual for Energy Management”. Jones E. Piper p. 314 (1999) ISBN 0-7656-0050-1 [6]Comparative Climatic Data", National Climatic Data Center, NOAA, 2001. Heating Degree Days, Normals 1971-2000 Years given by month; http://www.ncdc.noaa.gov/oa/climate/online/ccd/nrmavg.txt [7] Cooling Degree Days: Normals 1961-1990 Years given by month by NRCC (Northeast Regional Climate Center, Department of Earth and Atmospheric Sciences, Cornell), last updated 5/25/2000. http://www.nrcc.cornell.edu/ccd/nrmccd.html 291 2009 International Radon Symposium, St. Louis, MO CAN CAT LITTER BE A SOURCE OF INDOOR RADON ? 1 Michael E. Kitto1,2 and Traci A. Menia1 Wadsworth Center, New York State Department of Health, P.O. Box 509, Albany, NY 12201 2 School of Public Health, State University of New York, Rensselaer, NY 12144 Abstract Emanation of radon (222Rn) from several brands of cat (kitty) litter was measured with a continuous radon monitor. Radon emanating from the cat litters, encapsulated in an airtight container, produced equilibrium concentrations below 1.2 pCi/g. The measured radon flux was below 10 pCi/kg-hr for all the cat litters. Although each of the samples emitted a measurable amount of radon, the emanation is too small to raise indoor radon concentrations. The cat litters were measured with gamma-ray spectroscopy to identify and quantify the naturally occurring radionuclides in the samples. Secular equilibrium for the 238U and 232Th radioactive-decay series was evident. Introduction Radioisotopes of the three naturally occurring radioactive-decay series, headed by 232Th, 235 U, and 238U, are ubiquitous in soils. Although a different isotope of radon is formed in each of the radioactive-decay series, only 222Rn has a sufficiently long half-live (3.8 d) that it often occurs at elevated concentrations in indoor air. The 222Rn isotope, commonly and here referred to as radon, and its decay products have been linked by epidemiological studies to an increased risk of lung cancer, causing approximately 21,000 lung-cancer deaths (Lubin, 1997) annually in the United States (US). A majority of radon entry into a house occurs at the soil-foundation interface, with minor contributions from groundwater use, building materials, and outdoor air. In the past year, several homeowners and radon mitigators have inquired about radon emanation from cat litter, with the latter group noting increased radon levels in rooms containing a litter box. Given the health implications and the paucity of measurement data, goals of the 292 present study were to quantify the emanation of radon from cat litter and determine the associated gamma-ray activities. Methodology Radon Containers of cat litters that are commonly used in the US were obtained from local retailers (Table 1). All of the cat litters contained granules of clay (e.g, bentonite), and some contained a small portion of silica gel. Radon emanation from the litter samples was determined using the analytical method depicted in Figure 1 in a climate-controlled laboratory containing ~0.3 pCi/L of airborne radon. Samples of the cat litters, with weighs of 143-734 g, were individually sealed inside a 55-L airtight chamber to allow the emanating radon to ingrow to radioactive equilibrium in the enclosed air. Measurements of radon from the cat litters were completed at 1-h intervals using an AB-5 passive radon detector (PRD; Pylon Electronic Development Co. Ltd., Ottawa, Canada). The PRD is an alpha-scintillation counter with a background of 1.1 cpm and a high detection efficiency (1.2 cpm per pCi/L) for radon and its short-lived alpha-emitting progeny. The PRD was placed inside the chamber with each sample, and measurements were conducted up to 194 h. Radon emanated from the cat litter and diffused Table 1. Samples that were measured. Cat litter brand Aldi American Fare Arm & Hammer Arm & Hammer multiple cats Freshstep Price Chopper Scoopaway Tidycat Bentonite clay through the inlet filter of the PRD, to enter the measurement chamber. The equilibrium activity (A0) was determined by fitting measurement data from the ingrowth of radon in the chamber using the following: 293 T ln( 2 ) ' & # t1 AT = A0 $1 ' e 2 ! $ ! % " (1) where AT is the activity (pCi/L) at time T, and t1/2 is the half-life for radon (3.82 d). The PRD Fig.1. Photograph of the radon detection system used in the analysis of the cat litters. was calibrated in a certified radon chamber prior to use, and performance was monitored using a manufacturer-supplied 226Ra standard (Model 3150A). Gamma-ray emitters Activities of radioisotopes were determined in the cat litters using gamma-ray spectroscopic measurements. The samples were sealed in 0.5-L Marinelli containers and measured for µ1000 min each, using a high-purity germanium detector (HPGe; Canberra Industries Inc., Meriden, CT) in a low-background shield. The absolute detector efficiency was 131%, relative to a 3”x3” NaI(Tl) detector, and the resolution was 1.9 keV at 1333 keV. Spectra, collected for a 294 gamma-ray energy range of 46 to 1765 keV, and were analyzed by the Genie-VMS Spectroscopy System (Canberra Industries Inc.). The detector was shielded by 15 cm of lead, and the counting chamber was purged with nitrogen during the measurements to reduce the concentration of airborne radon-decay products from the room air. Results and Discussion Radon Activity due to ingrowth of radon inside the chamber was measured for each cat-litter sample. The PRD provided confirmation regarding radioactive equilibrium of the samples in the sealed chamber. The solid line drawn through each set of measurement data (Fig. 2) represents the activity expected based on the half-life of radon. In all cases, the observed radon ingrowth Fig. 2. Radon activities determined in samples of cat litters and bentonite clay. The solid lines represent the expected activities based on the 3.82-d half-life of radon. 295 coincided well with the expect (theoretical) ingrowth. The net count rates (unequilibrated) from the encapsulated cat litters ranged from 0.5 to 9.1 cpm. The radon concentration in the chamber (AT) was determined using the relationship: AT = (G ! B) CF (2) where G and B are the respective gross and background count rates (cpm), and CF is the calibration factor (1.2 cpm per pCi/L). If allowed to attain equilibrium, the concentrations in the chamber (A0) would have ranged from 0.7 to 13.3 pCi/L for the cat litters. By applying the chamber volume (55 L), we determined that the total equilibrium activity of radon in the chamber ranged from 38 to 730 pCi. Measurement of the sample masses allowed determination of the equilibrium mass-activity concentration, which varied from 0.3 to 1.2 pCi/g, which is similar to 238 U concentrations in soil. Determination of radon flux (F; pCi/kg-d) from the cat litters was determined using F= A0V! M (3) where V is the volume of the chamber (55 L), λ is the decay constant of radon (0.1814 d-1), and Fig. 3. Radon flux results of the measured samples of cat litter and bentonite clay. 296 M is the mass of the emanating material (kg). Using the equilibrated radon activities in the chamber, the flux from the cat litters ranged from 49 to 215 pCi/kg-d (Fig. 3). Thus, a litter box.containing 4.5 kg of cat litter will emit approximately 9-40 pCi/hr of radon into a room. Considering that a 20,000 L room filled with outdoor air (0.4 pCi/L) contains roughly 8,000 pCi of radon, the contribution from the cat litter would be undetectable Gamma-ray emitters The gamma-ray spectra contained well-defined peaks belonging to the 238U decay series (e.g., 214 Pb and 214Bi), 232Th decay series (e.g., 228Ac and 212Pb), 235U, and 40K. During secular equilibrium, the radioisotopes comprising each decay series have identical activities, thus any isotope will provide information on the activities of the entire decay series. Use of a single geometry (0.5-L Marinelli) allowed an intercomparison of the sample results. As shown in Figure 4, the cat litters contain 1.7 to 4.5 pCi/g of 238U and each of its decay products, a range similar to that found in soils. Due to low crustal abundances, isotopes from the 235U decay series Fig. 4. Gamma-ray activities determined in samples of cat litter and bentonite clay. 297 are 0.1-0.3 pCi/g in the samples. In nature, the activity of 235U is typically ~5% of the 238U activity, and our measurements confirm the 235U/238U ratios in the cat litters were 5-7%. Concentrations of the gamma-ray emitting isotopes from the 232Th decay series indicate that the cat litters contain 0.7-3.6 pCi/g of 232Th and each of its decay products (Fig. 4). The cat litters contained less activity of isotopes from the 232Th decay series than from the 238U decay series. The isotopes comprising the 232Th decay series have identical concentrations, suggesting radioactive equilibrium. With one exception, the activities of 40K in the cat litters were similar to concentrations of isotopes from the 238U series and 232Th series. Most of the cat litters contained below 7 pCi/g of 40 K, except for one that contained 18.7 pCi/g. Conclusions In conclusion, emission of radon from cat litters was determined by a radioanalytical method. The method demonstrated that radon emanates from the cat litters, but the equilibrium activities for the measured samples were below 1.2 pCi/g. Radon flux from the cat litters ranged from 49 to 215 pCi/kg-d. Gamma-ray activities for 40K and radioisotopes from the 238U- and 232 Th-decay series were determined in samples of cat litters used in the US. The majority of activity concentrations were similar to values found in soil. Results indicate that none of the cat litters contain sufficient radioisotope concentrations to confer a health issue. References Lubin JH, Boice Jr. JD. Lung cancer risk from residential radon: meta-analysis of eight epidemiologic studies. J. Natl. Cancer Inst. 89:49-57; 1997. 298 EVALUATION OF RADON´S EMISSION IN ORNAMENTAL ROCKS Gavioli, Y. S.1, Correia, J. C. G.1, Caranassios, A.1, Ribeiro, R.C.C. 1, Lamego, F. F. 2, Melo, V. P.2 1 Centro de Tecnologia Mineral -CETEM-RJ. Av. Pedro Calmon, 900, Ilha da Cidade Universitária, Rio de Janeiro - RJ. CEP 21941-590. Tel. (21) 3865-7276 / Fax : (21) 2260-9835. e-mail: ygavioli@cetem.gov.br 2 Instituto de Radioproteção e Dosimetria (IRD) - RJ Av. Salvador Allende s/n - Jacarepaguá - Rio de Janeiro - RJ – CEP - 22780-160 Tel: (21) 2442-1927 / Fax: (21) 2442-1950 Abstract The representative organizations of the sector of ornamental rocks had looked the Commission of Nuclear Energy by means of the Center of Mineral Technology Espirito Santo (CETEM-ES) requesting aid how much the evaluation of levels of radiation in Brazilian ornamental rocks, exported to U.S.A. for use as material coverings in residences. Exhalation of radon (Rn) from these exotic granites may expose the occupants to an increased risk of contracting lung cancer. The objective of this work was to evaluate the risk associated with the exhalation of radon of indoor covering plates in environments, by means of nuclear techniques and to use models of calculation of dose endorsed internationally. For in such a way, the characterizations radiometric and mineralogical of three types of silicatic rocks used for covering have been carried through, determining the rates of exhalation and concentration of radon´s activity in the chosen materials. Introduction The Brazilian industrial sector produces a great variety of granites, marbles, quartzite etc., reaching to all about 500 different types of rocks. Normally, these ornamental rocks are classified by corporate names, the producing state, the geologic origins and mineralogical composition. However, they are not classified by radiological activity. The study of radioelement´s concentrations in granites rocks is important for the classification granite petrography for application in civil construction. Granites are an example of rocks that contain a natural enrichment of uranium - 238 (238U); thorium 232 (232Th) and potassium - 40 (40K). The concentrations of these elements in the rock vary significantly from sample to sample. However studies have shown that the general petrography characteristic is closely correlated to the concentration of these radioactive elements (Whitfiel et al. 1959). In terms of surface mobility, uranium and potassium can easily be oxidized into water soluble compounds that can be leached from the surface matrix and deposited in sediments far from the origin of the rock. On the other hand, Thorium, which is less water soluble, is less susceptible to leaching and tends to stay within the rock matrix. Thus, over time this difference in solubility results 299 in the observed petrography characteristics of granite rocks and produce specific alterations in the ratios of natural radionuclides concentration: Th, U, K, Th/U and Th/K (Anjos et al, 2004). 1.1. Radioactivity in construction materials It is generally accepted that indoor residential gamma radiation exposure from construction materials comes primarily from the decay of 238U and 232Th, beyond 40K and their subsequent daughters.. For example in the 238U and 232Th decay´s series there are 16 daughters that release gamma radiation with emission greater that 10-3 fotons (Pink, 1997). However, only a small fraction of this gamma radiation actually results in dose to the occupant (Fernandes et al, 2004). 1.2. Radon in residences The primary means of radon entrance into residences is by means of cracks around foundations and drains, diffusion through floor coverings and other openings in the floor. Consequently, the Office of the General Surgeon and the Environmental Protection Agency (USEPA) recommend that radon tests be performed in all houses. In addition, radon exposure from water ingestion, in combination with radon inhalation, can also represent a significant risk to the population. This exposure combination exposes sensible cells in the respiratory and gastrointestinal tract results in increased risk of different types of cancers. However, the greater risk is to the lungs resulting in about 10,000 American deaths per year (HyperPhysics, 2009). Because granites are commonly used throughout the world in homes as ornamentation accents, counter tops and other interior coverings, radiation exposure concerns raised by the media gave rise to many baseless rumors about the overall safety of Brazilian granites. Consequently, Brazilian granite commercialization was impacted in some countries in Europe and Asia. Therefore, the relevance of this scientific study to address these misconceptions is both timely and important to Brazil. 1.3. Radon Radon is a colorless, odorless. tasteless gas, that is 7.58 times heavier than air and more than one hundred times heavier than hydrogen. It originates from either radium or thorium and is a member series of uranium decay series. These precursors of radon are found throughout the terrestrial crust, therefore all homes worldwide are potentially at risk. By virtue of being the only gas in the uranium decay series, radon has the potential to migrate from the soil or rock matrix and accumulating in residences, caves, mines and tunnels. The isotope 222Rn is an alpha emitter (T1/2 = 3,82 days), together with its not gaseous daughters 218Po and 214Po, are responsible for approximately 50% of the dose from natural ionizing radiation. Other isotopes of radon include 219Rn and 220Rn, which are products of 235U and 232Th, respectively. However, they have the very short half-life when compared with 222Rn, being of 219Rn of 3.96 seconds and of 220Rn of 55.6 seconds. Consequently Radon-222 is the only isotope capable of migration from the 300 soil and rock matrixes into homes (Chyi, 2008). Table 1.1, shows the Uranium-238 decay series (Craig, 2008). Table 1.1 Decay of uranium-238 ORIGINAL ELEMENT TRANSFORMED ELEMENT ThoriumThorium - 234 Uranium - 238 ThoriumThorium 234 Protactinium – 234 TIME OF HALFLIFE EMISSION 4,5 trillions of years One alpha particle 24,5 days One beta particle and one gamma ray Protactinium-234 Thorium – 230 269 thousandth years One beta particle and one gamma ray Thorium - 230 RadiumRadium – 226 83 thousandth years One beta particle and one gamma ray Radium - 226 Radon - 222 1590 years One beta particle and one gamma ray Radon - 222 * Polonium– 218 3,925 days One alpha particle Polonium - 218 Lead – 214 3,05 minutes One alpha particle Lead - 214 Bismuth – 214 26,8 minutes One beta particle and one gamma ray Bismuth - 214 Lead-210 to Polonium214 19,7 minutes One alpha particle or one beta particle and one gamma ray Polonium - 214 Lead – 210 150 micro second One alpha particle Lead - 210 Lead – 210 1,32 minutes One beta particle Lead - 210 Bismuth – 210 22 years One beta particle and one gamma ray Bismuth - 210 Polonium - 210 5 days One alpha particle and one gamma ray Polonium - 210 Lead - 206 ** 138 days One alpha particle and one gamma ray ** Is a steady isotope of lead. 1.4. Acceptable Limits of Gas Radon The primary dose from Radon and its daughters occurs from inhalation, which deposits its energy in the lungs. However, in accordance with the effective basic standards of protection against the ionizing radiations and the security of radiation sources (SAFETY SERIES In the 115), any exposure from natural sources should be excluded from regulation. The value used as an exclusion limit is, in accordance with the guide of security of the International Agency of Atomic Energy (SAFETY STANDARDS SERIES No. RS-G-1.7), 1 Bq/g for the series of 238U and 232Th, being the cited value to the precursor (“father”) of each series and 4 Bq/g for 40K. This value was determined based on the global distribution of the concentrations of activity of these radioisotopes in the earth’s crust. 301 If this reference value is used, it would not be necessary to take corrective action to reduce exposures.. Therefore, according to the references cited above I contend that commodities with 238U and 232Th with lower concentrations than the 1 Bq/g standard not be subject to the regulatory control for the purposes of radioprotection. Under these conditions, doses are not anticipated that exceed the limit of exemption of 0,3 mSvy-1. For activities that exceed the reference (e.g up to 10 times bigger) the regulatory authority will need to do a controlled evaluation on a case-by-case basis. The severity of the applied measures depending upon the level of the risk associated with the material. As for radon, in accordance with the Safety Series 115, action should be taken when the levels of annual average concentration will be between 200 and 600 Bq/m3 above 4.0 pCi/L in accordance with the guidance of US EPA. Actions of remediation are only justifiable at greater than 600 Bq/m3 or 20 pCi/L range where cost-benefit analyses should be considered,. Objective This work aims at to the evaluation of the rate of emission of radon gas from Brazilian ornamental rocks, in view of the health of the human beings and the influence of these on the economy of the country. Materials and Methods In this work, three types of granite samples known commercially as Crema Bordeaux, Mombassa and Golden, were analyzed. These samples were selected because of the high tax of exportation for use in international the civil construction For the evaluation, samples, in the form of non-resin, polished plates (15 cm x 30 cm x 2 cm) were selected. 3.1. Rocks in the Natural State For this study, sample preparation was not necessary. The testing apparatus consisted of a 20 L steel chamber and stamped removable covers for retention of the gas. Previously, the covers had been tested in order that confirm no leakage. Two plates of ornamental rocks, with known radiological potential were left in the box for eight days. To aid in mixing, a micro fan was used. Radon from the box was injected into a gas radon analyzer (Alpha Guard 2000 PRQ Genitron Instruments) in cycles of one and two hours. The in-growth of radon in the chamber was allowed to continue for a period of 7 half-lives. . It is possible to know the tax of gas emission on each sample of the chamber by means of equation 1. At = A0 (1- e-λt) (1) where λ is the constant of nuclide decline on question and A0 is the final value of the activity during 7 t ~ T1/2, approximately 27 days in the case of the gas Rn. The unit of 302 the final activity A0 is Bq m-3. This value multiplied for the constant of decline of the radon (λ = 2,724 x 101 s-1) and for the reason enters the volume of the container (V = 20,0 x 10-3 m3) and the area of the granite, F, allows to get the rate of exhalation of radon for unit of area of this, that is defined as the flow of set free radon of the surface of the analyzed material, and, in Bq m-2 s-1, that is, Becquerel per square meter per second, represented in equation 2. E = A0 λ (V / F) (2) In view of the comparison of the gas radon emission with the amount of radium in the samples the “final rate of emission of Rn for unit of mass” with unit in Bq kg-1 was calculated, using equation 3. CRn = (A0 V) / m (3) where m is the mass of the sample and A0 and V already had been defined in the Eqs. (1) and (2), respectively. 3.2. Worn out rocks For spectral gamma analysis, the samples were crushed, milled, screened and dried. For this analysis, the plates of the samples were crushed and then taken for milling. Dust from the milling process was controlled with aid of a sprayer (Fritsch). During the spraying, caution was used so that a majority of the samples would be between 0.177 mm and 0.149 mm after the screening. This size was considered optimal for gamma spectral analysis. Small amounts of sample in the container of the spray had been detritus of fish (titanium covered to prevent any type of contamination of the samples), and centrifuged at 400 r.p.m. for 2 minutes. After removal, the displaced samples were placed in a stack of bolters of 2,360 mm to remove the titanium balls of the spray, 0.177 mm and 0.149 mm, respectively. The samples were then put in a Ro-tap agitation device, for about 10 minutes. This allowed for total separation of the sample in the bolters. Although preference was given for samples between 0.177 mm and 0.149 mm, samples less than 0.149 samples were also used. Before mixing the samples were homogenized and the grain distribution was determined (e.g. what percentage was above or below 0,149 percentage mm). After the screening, the samples were ready for spectra-gamma analysis to allow for secular equilibrium, the samples were allowed placed in sealed containers for 30-days. A high efficiency gamma detector (HPGe Germanium) was used to measure the gamma emissions of 40K and the daughters of the series of U and Th. Counting time varied from 8 to 16 hrs depending on the activity (Bq/g) of the material. The results were then compared to the threshold values listed by the International Agency of Atomic Energy (AIEA). 303 Results and Discussions 4.1. Granulometric Analysis Table 1 indicates the results of the grain sized analysis of the Crema Bordeaux, Mombassa and Golden samples. Table 1: analysis of the Grain siz e in which Sample Granulometric Crema Bordeaux Mombassa Golden Fraction (mm) Mass (%) Mass (%) Mass (%) + 0,177 0 0 0 - 0,177 + 0,149 10,29 16,52 10,59 -0,149 89,71 83,48 89,41 Total 100 100 100 4.2. Gas Emission 222Rn Figure 1 represents the results of the test of evaluation of the rate of gas radon emission in the samples in its natural state, being able itself to observe that the rates for the samples Crema Bordeaux, Mombassa and Golden had been very low, 3,10 x 10-3 (Bq/m2/s), 1,60 x 10-3 (Bq/m2/s), 6,84 x 10-4 (Bq/m2/s), respectively. Figure 1: Rate of radon’s emission in each sample in its natural state 4.3. Radiometric analysis / Radiochemical 304 Figure 2 Radionuclide´s Distribution in the Samples Figure 3 Main Radionuclides Figure 2 shows the results of the analyses of the three samples. The determination of Frog, K, Th and gamma had been carried out after the 30 days and with a detector of germanium, as said previously in the methodology. The determination had been made by counting alpha and beta, in a proportional detector of low background, after chemical separations and the measures of Thorium for spectrometry with arsenazo III. Figure 3 shows the primary radionuclides measured in the samples. Conclusion Compared to international standards, the findings of this study indicate that radon emanation from these samples is very small. Therefore exposure risk from these types of granites would be minimal. References 305 Anjos, R. M., Veiga, R. T. Soares, A.M.A. Santos, J.G. Aguiar, M.H.B.O., Frascá, J.A.P. et al. Natural Radionuclide Distribution in Brazilian Commercial Granites, Radiation Measurements, 2004. Brodsky, A. Handbook of Radiation Measurement and Protection. CRP Press, available in: http://www.physics.isu/radiumf/natural.html, 1978. Chyi, L. L. Radon Testing of Various Countertop Materials Final Report, 2008. Craig Freudenrich, Ph.D., Marshall Brain. Available http://ciencia.hsw.uol.com.br/radonio1.htm. Access on: December 12th, 2008. in: Fernandes, H.M.; Rio, M. A. P.; Franklin, M.R. Impactos Radiológicos da Indústria do Fosfato. Série Estudos & Documentos, n. 56, ISSN 0103-6319, CETEM, 2004. HyperPhysics. Available in: http://hyperphysics.phyastr.gu.edu/hbase/nuclear/radon.html. Access on: January 20th, 2009. IARC - Agência Internacional de Pesquisa em Câncer. Office of the Surgeon General – OSG. Available in: http://www.surgeongeneral.gov/. Access on: February 25th, 2009. Rosa, R.. Exposição Potencial a Radiação Natural no Interior de Residências Devido ao Uso do Fosfogesso na Indústria da Construção Civil. [Tese de Mestrado]. Instituto de Biofísica Carlos Chagas Filho, Universidade de Federal do Rio de Janeiro. 122p., 1997. US EPA - United States Environmental Protection Agency. Available in: http://www.epa.gov/. Access on: January 21th, 2009. Whitfiel J. M.; Rogers, J. J. W.; Adams J. A. S..The relationship between the petrology and the thorium and uranium contents of some granitic rocks. Geochimica et Cosmochimica Acta 1959:17: 248-271. 306 MITIGATION APPROACH ON REDUCING THE NEUTRAL PRESSURE PLANE STAND-ALONE OR COMBINED WITH OTHER METHODS Georges A. Roserens Swiss Federal Office of Public Health, Radiological Risk Section, CH-3003 Bern, Switzerland ABSTRACT Radon risk reduction requires that mitigation systems should be cost-effective. Standard mitigation systems are well known, such as ASD, PSD, sealing, etc. Under natural conditions, the ground floor is subject to negative air pressure and the upper story to positive pressure. Such conditions are conducive to high radon concentrations as they encourage infiltration of the gas into the building through cracks, fissures and passages in the basement. Mitigation by reducing the neutral pressure plane is a simple method to reduce the radon concentration. To lower the neutral axis, we install small wall ventilators or passive ventilation through pressure differential balancing (in the window or in the wall). This method could also increase the efficacy of all other remediation systems. INTRODUCTION The neutral pressure plane is variable and depends on the inhabitant's behavior and the building type.. In Switzerland high radon concentration up to 1,000,000 Bq/m3 or 27,000 pCi/l at a depth of 3 meters in the soil has been measured. (Johner H. U. and Roserens G. A.1998). Radon entry routes: cracks, holes and gaps are vs: diffusion, water, buildings material, almost the major contributor to the radon concentration in dwellings. The quantity of air mixed with radon passes through a fissures or openings is mathematically complex to determine. 1 & 2 #2 qV (inf =C d A$ 'p ! %) " n qV "inf =C inf (!p ) An important point to bear in mind, however, is that this quantity is related to the difference in pressure. 307 HVAC - HRV AND RADON Test and adjust your HVAC system. Depressurization should be avoided (more radon fx(Δp) and risk of backdrafting). If the building uses an HVAC system, it is necessarily to adjust it and reduce the negative air pressure or better to tend to have a slightly positive pressure 0 to1 Pa (0-to 0.00014 psi). MORE OR LESS AIRTIGHT HOUSES - WITHOUT HVAC - HRV Energy saving: buildings have an airtight shell. This prevents warm air from escaping and cold air from entering the home. The external envelope of a building should be as airtight as possible. Energy efficient house: make it airtight. For standard new houses measured: ca. 0.15 ACH Blower Door test for new green buildings: max. 0.6 ACH at 50 Pascals differential pressure neutral pressure plane after Natural ventilation: If you open a window at the lowest position you will lower the before neutral pressure plane. You can not justify the radon reduction with the dilution effect. In fact lowering the neutral pressure axis is much more important (reducing the vacuum effect). In winter the owners would open the window only few minutes daily, it is not enough to see a radon reduction in average. Drill a hole in the external wall, but you will get an uncontrolled quantity of air and cold, depending on the orientation of the wall, the wind etc.. (an unacceptable situation). To regulate the air quantity to lower the neutral axis, we use a small fan, blowing from outside to inside to control the air flow quantity (not to be confused with the house pressurization). 308 The airflow is managed between outside and inside. In the winter time it is not imperative to warm up the cold air because the quantity is minimal and sufficient to equilibrate the pressure difference. In many cases is also possible to install a simple grill in the window or in the wall to reach the same effect. Noise and dust: use a device with sound insulation to reduce the outside noise and an air filter. REDUCING THE NEUTRAL PRESSURE PLANE COMBINED WITH OTHER METHODS This method increases the effect of traditional mitigation such as: active soil depressurization, passive soil depressurization, crawlspace depressurization, crawlspace airing, hollow floor depressurization, hollow wall depressurization and basement depressurization. CONCLUSIONS Throughout the world, houses are different, climate is not the same and behavior may play an important role. It is very important to pursue and to analyze alternatives and to test the limits of these other possibilities. The focus is on developing low-cost mitigation systems. 309 310 311 WITH A WALL FAN WHEN JUST ONE AREA NEEDS TO BE REMEDIATED 3 (20-60m /h preheating not necessary) Fan switched off, vanes closed fermés 312 Stimulating radon safe building in radon prone areas by detailed scale radon hazard mapping André Poffijn*, Boris Dehandschutter, Etienne Noel, Stéphane Pépin 1 and Michel Sonck Federal Agency for Nuclear Control, Ravensteinstraat 36, B-1000 Brussels, Belgium. * Corresponding email: andre.poffijn@fanc.fgov.be ABSTRACT Radon is responsible for more than thirty percent of the radiation exposure of the Belgian population. In order to reduce the exposure to radon, the Federal Agency for Nuclear Control (FANC) has developed a radon action plan. The objectives of the action plan are to eliminate the high existing exposure situations in public buildings, workplaces and dwellings, and to limit exposures in new buildings. Radon-safe building techniques include the application of an anti-radon barrier at the soilslab interface and of an underlying permeable layer that can be actively or passively ventilated. A tool for decision-making, raising public awareness and aiding local governments to establish a radon safe building culture is the development of detailed scale radon hazard maps showing municipality scale variations in the geogenic radon hazard. Indoor and soil gas measurements and information on the subsurface are used to optimise the radon hazard maps and to assess the potential radon risk in building extension zones on the scale of the municipalities. The characterisation of the building extension zones in terms of radon hazard increases the efficiency of radon reducing policies. KEYWORDS Radon hazard, radon mapping, radon prone areas, radon safe building INTRODUCTION Radon is the most prevailing source of radiation exposure in the indoor environment (ICRP 60). The link between radon and lung cancer has been first recognised through miner cohort epidemiological studies (BEIR IV; BEIR VI; UNSCEAR). More recent case-control studies highlight the linear no-threshold relation between lung cancer risk and indoor radonconcentration in dwellings (Darby et al., 2005). In order to efficiently manage the radon exposure, most European countries have adopted a radon action plan, setting out the criteria, strategies and practical aspects of radon controlling activities. The Belgian radiation protection regulation (ARBIS) foresees the control of radon exposure in workplaces and in dwellings in radon prone areas. A national radon measurement campaign during the nineties highlighted the occurrence of radon prone areas in the southern part of the country (Poffijn and Vanmarcke; Zhu et al.). Ongoing indoor radon measurements allowed assessing the radon exposure of the Belgian population (Table 1). Radon safe building in the radon prone areas is essential to significantly reduce the collective exposure to radon in the course of house stock renewal. Radon mitigation is stimulated in the radon prone areas in buildings where high radon concentrations have been measured. Increasing public awareness is achieved through information sessions for the general population and training courses for building professionals. An important tool to stimulate new builders and architects to use radon safe building techniques is the use of municipality-scale radon hazard maps indicating the geogenic hazard for indoor radon problems on the scale of the municipality. The compilation of these radon hazard maps is discussed in this paper. 313 Table 1:Statistics for the indoor radon exposure in Belgium. Population Number of GM AM < 200 200-399 400-799 800 and more measurements Belgium 10502213 10447 59 57 97,8% 1,7% 0,4% 0,1% Wallonia 3413978 10108 84 82 95,6% 3,2% 0,9% 0,3% High-risk areas 376568 5605 137 169 19,9% 8,8% 4,3% Flanders 7088235 339 44 0,8% - - 67% 34 99,2% (GM: geometric mean; AM: arithmetic mean; all radon values in Bq/m³) METHODS Long-term (3 months) indoor radon measurements by solid state nuclear track detectors are used to track down buildings with high radon concentrations (for mitigation) and to map radon prone areas. In building extension zones outside the built-on areas, the indoor radon measurements are complemented by soil gas measurements and geological information. Soil gas is sampled at 90 cm depth and analysed by a scintillation counter. In-situ permeability is measured by a JOK-permeameter. The soil gas sampling strategy was based on an unclustered and regularly spaced coverage of the municipalities’ territory and a detailed analysis of building extension zones (Dehandschutter et al., 2008). All measurement points have been geo-coded, assigning geographical coordinates to each measurement point, as well as the local geological unit the point occurs on and introduced in a GIS Geodatabase. These geo-coded data are used to make radon hazard maps. Especially for building extension zones, soil gas measurements and geological information is used to assess the radon hazard complementary to the available indoor information. RESULTS In the radon prone areas of southern Belgium, about 5600 long-term indoor measurements have been collected, corresponding to a (clustered) sample density of about 10% of the house stock in these areas. Complementary to these indoor measurements, about 4000 soil gas measurements have been collected, leading to a sample density of about one per km² (Fig.1). Table 2: Statistics of the detailed measuring campaigns in the radon prone areas. Number of measurements AM GM Min Max SD GSD indoor 5605 169 137 6 4204 294 2,5 soil gas 4009 52 41 1 430 37 2,2 ( Indoor in Bq/m³; soil gas in kBq/m³; N: number of measurements; SD: standard deviation; GSD: geometric standard deviation) 314 Fig. 1: Indoor and soil gas measurements of the detailed campaigns in the radon prone areas 315 Mapping of the radon potential RP = (CA-1)/(-logk -10) as defined by Neznal (Neznal et al., 2004) gives a first indication of the radon hazard outside the built-on areas (Fig. 2a). However, the radon potential does not take into account the indoor radon measurements in the neighbourhood. Therefore, a combination of the indoor, soil gas (CA) and permeability (k) information is used to improve the assessment of the radon hazard (Fig. 2b). This combination is achieved by overlaying the up to its maximum value normalized radon potential and indoor radon grid and leads to values for the radon hazard ranging between 0 and 1. The results have been categorised in 5 classes (very low up to very high). This combination allows taking into account the most relevant indicators for the probability of high indoor radon (neglecting the building characteristics). (a) (b) Fig. 2: Indoor and soil gas radon with derived radon potential (a) and radon hazard (b). DISCUSSION In order to make the best use of all available parameters which are influencing the indoor radon, the geo-coded indoor and soil gas radon measurements, soil permeability and geological information (lithology) are combined into a single radon hazard map. The radon hazard map does not take into account the building construction characteristics. Since only indoor radon measurements from unmitigated buildings were considered, the indoor radon map can be seen as valid for the ‘average’ building in the area. Currently, the influence of the transfer factor from soil to house is being studied, and will be taken into account in the further development of the radon hazard maps. CONCLUSIONS The long term goal of the national radon action plan is to reduce the collective dose due to radon by reducing the average radon exposure of the population to levels that can be considered as optimized. Following the EU recommendations (90/143/Euratom), the design 316 level for new buildings is currently 200 Bq/m³. Based on the national radon map, radon prone areas have been determined where the probability to exceed the design level justifies the application of protective measures against radon in new buildings. In the radon prone areas, all relevant indicators for high indoor radon concentrations are used to assess the radon hazard. The data are combined into radon hazard maps which are used to stimulate new builders and local authorities to apply the most efficient protective measures against radon for new buildings in the radon prone areas. ACKNOWLEDGEMENT The detailed measurement campaigns have been carried out in collaboration with the Province of Luxemburg (Service d’analyse du Milieu intérieur, SAMI-Lux) and the International Bureau for Environmental Studies (IBES). REFERENCES 90/143/EURATOM. Recommendations of the Commission, of 21 February 1990, relative to the protection of the population against the dangers resulting from indoor radon exposure. Official Journal of the European Communities L 080 du 27/03/1990, p. 0026 – 0028. ARBIS. Algemeen Reglement op de bescherming van de bevolking, de werknemers en het leefmilieu tegen het gevaar van ioniserende straling - Règlement général de la protection de la population, des travailleurs et de l’environnement contre le danger des rayonnements ionisants (RGPRI). Arrêté royal du 20 juillet 2001. Moniteur belge, 30/08/2001, p. 2890929368. BEIR IV. Health risks of radon and other initially deposited alpha-emitters. US National Research Council Report, National Academy Press, Washington DC, 1988. BEIR VI. Health effects of the exposure to radon. US National Research Council Report, National Academy Press, Washington DC, 1998. Darby S. et al. Radon in homes and risk of lung cancer: collaborative analysis of individual data from 13 European case-control studies. British Medical Journal, v. 330, pp.223- 227, 2005. Dehandschutter, B., Poffijn, A., Ciotoli, G., Klerkx, J. Radon mapping using indoor, soil gas and geological data. IGC33, Oslo, 2008. The presentation can be downloaded from http://radonmapping.jrc.ec.europa.eu/ under ‘documents’. ICRP 60. Recommendations of the International Commission on Radiological Protection. International Commission on Radiological Protection, Publication 60, 1990. Oxford, UK, Pergamon Press,1991, 215 p. Neznal, M., Neznal, M., Matolin, M., Barnet, I., Miksova, J. The new method for assessing the radon risk of building sites. Czech Geo. Survey Spec. Paper 16, 48p, 2004. Poffijn A. and Vanmarcke H. The Indoor Radon Problem in Belgium. In: Indoor Air Quality and Ventilation, Selper Ltd, pp. 339-345, 1990. UNSCEAR. Sources and Effects of Ionising Radiation. Rapport to the General Assembly. United Nations, 2000. Zhu H.C., Charlet J.M. and Tondeur F. Geological Controls to the Indoor Radon Distribution in Southern Belgium. The Science of the Total Environment, v. 220, no. 2-3, pp.195-214, 1998. 317 Measurement of Granite Countertop and Other Building Materials Radon Emanation J.L. Alvarez, B.V. Alvarez, M. DeVaynes, S. Price Air Chek, Inc. 1936 Butler Bridge Rd Mills River, NC 28759 Abstract Granite countertops and other natural materials used in building construction may have concentrations of uranium series radionuclides that are higher than the local average background. Further, the concentration within the material may be highly variable and may contain inclusions that have concentrations much higher than the material average. Such materials have the potential to emanate radon to the extent that the radon concentration of indoor air is significantly increased. A study of radon emanation from polished granite sheets was performed and two simple tests were devised to determine the radon emanation rate. A generic calculation for the radon concentration potential increase was proposed, but is no substitute for direct measurement of the room air concentration. Introduction Concern for radon emanation in news and Internet reports prompted several studies of radon emanation from granite countertops and caused similar concern for other building structural and decorative materials (Kitto et al, 2008) (Brodhead, 2008) (Sundar et al.,2003). Uranium (U-238) and its subsequent progeny radon (Rn-222) are naturally occurring radioactive materials that that are found in nearly all natural materials at about 1 pCi/g, but with large variations by material and locality. Radon may emanate from the material at a rate dependent upon material properties and surface area. Radon emanation into a closed structure may increase the indoor radon concentration depending upon the exchange rate of the indoor and outdoor air. The U.S. Environmental Protection Agency (EPA) and other national bodies recommend or require that the homes and occupied structures maintain the indoor radon concentration below a specified limit. EPA recommends a limit of 4 pCi/L. Radon Emanation Theory Fick’s laws can describe the diffusion of one gas into another if the diffusing gas is sufficiently dilute. Fick’s first law in one dimension states that the flux of the gas across a surface is proportional to the product of the diffusion coefficient of the gas and the concentration gradient across the surface, J = "D J= D= C= dC , dx 1 Flux in molecules m-2 s-1 Diffusion coefficient in m2 s-1 ! Concentration in molecules m-3 318 x= Distance from the surface in m Fick’s second law describes the change in concentration with time, t, at a location, "C " 2C =D 2. "t "x 2 " x % Cx,t = C0erfc$ '. # 2 Dt & 3 The solution to eqn 2 is ! The function erfc is the complimentary error function and C0 is the concentration at the source surface. Eqns 2 and 3 show that the flux across the surface decreases over time as ! the concentration in the diffusion volume increases. Some authors call the reduction in flux back diffusion (Christopher et al., 1999). Eqn 3 is important for a closed system in which Cx,t approaches C0 at which point net diffusion across the surface approaches zero. An open system in which the air in the diffusing volume is well mixed and receives a continuous in-flow of clean air is of particular interest since most occupied volumes, where radon concentration may be of concern, fall into this category. The radon concentration of a well-mixed, open system having an air exchange rate, E, and receiving a constant source of radon atoms, S, each second can be described by dC S CE = " " #C . dt V V λ= 4 Radon decay constant in s-1 ! Eqn 4 integrates to #& + " ) t ' S $ &&1# e % V ( )) . C= E + "V % ( $E ' 5 The radioactive decay can be neglected in eqn 5 if the exchange rate is large relative to λV. For example, at one air change each hour, E = V, and for λ = 7.55 x 10-3 h-1, the ! decay constant is negligible in the denominator and exponent in eqn 5. The diffusion equations indicate that in a well-mixed, open system where the change from initial to final concentration is small, there is little change in flux across the surface, so within limit, the change in flux can be ignored. Materials and Methods Emanation rates for various materials were established in flow-through chambers using various materials and standards and two separate continuous radon concentration monitors (CRM). An accredited radon calibration laboratory calibrated the CRMs. One 319 radon emanation standard was obtained from National Institute for Science and Technology (NIST) and the other standards were materials that were evaluated by two other investigators. Fresh, radon-free air was drawn through the test chamber at a constant rate. The air was rendered radon-free by passing through an activated carbon filter. The filter was of sufficient volume that the break-through time was longer than 10 Ra-222 half-lives. The material emanation rate was calculated from the formula C= where C= S= E= S , E 6 Concentration in Bq m-3 ! emanation rate in Bq s-1 Source Volume exchange rate in m3 s-1 Eqn 6 is eqn 5 under equilibrium conditions when radioactive decay and change in flux are negligible in the flow –through system. The equilibrium concentration of the flowthrough system is equal to the emanation rate divided by the exchange rate, i.e., a volume of air at the equilibrium concentration replaces an equal volume at the same concentration. There are two advantages of the flow-through method over the closed volume method of measuring emanation rate. One is that equilibrium is reached quickly if the exchange rate is on the order of 10% of the total volume (90% equilibrium in 24 h). The other advantage is it is not necessary to know the volume of the chamber. It is necessary to measure the exchange rate (flow rate into the chamber) and to maintain the exchange rate low enough to obtain sufficient sensitivity for the radon concentration measurement. The flow rate in these measurements was made with a calibrated electronic mass flow meter. A NIST radon emanation source was used to calibrate all systems using the method of (Kotrappa et al., 2004). The source was rated at 372.7 Bq s-1. The characterized materials and the NIST source were used to test two methods for simple field measurement of emanation rate. The first method was an activated carbon packet below an overturned, shallow disposable dish1. The dish was sealed to the surface using KY Jelly (other potential sealants were potentially harmful to granite). The second method was a square of activated carbon cloth covered by a sheet of plastic2. The packet comprised 30 g of activated carbon in a 10x10 cm Tyvek envelope. The covering dish was made of plastic and had a volume of 1.1 L and enclosed a surface area of 0.05 m2. 1 2 Copyright AirChek, inc. Patent applied for. 320 The activated carbon cloth was 0.3 m2 and weighed 13 g. The carbon cloth was covered with Kraft paper of the same dimensions and taped to the surface with black electrical tape after covering with aluminum foil of the same dimensions. The collection method for the two systems is well-established with the best known example the Large Area Activated Charcoal Canisters (EPA 2008) collectors for radon emanation from soil. The concentration on the charcoal follows dC = S " #C , dt 7 which integrates to ! C= S (1# e# "t ). " 8 The tests of the methods showed that both methods closely followed eqn 8. The methods reach 90% equilibrium in 10 days. The tests need run only to the point where sufficient ! test. The capacity of the charcoal is sufficient to ignore back sensitivity is obtained for the diffusion from the charcoal. Results and Discussion The two field methods for measuring radon flux from a surface were calibrated against a NIST emanation source and both accumulator and flow-through methods of measuring radon flux. The agreement with a series of measurements was with a coefficient of variation near 5% for all devices. The data for the activated carbon packet system showed more scatter than the activated carbon cloth method. The smaller collection area of the activated carbon packet method caused the increased scatter. The results for the carbon packet method changed if the collection region was repositioned on a material that exhibited variation in exposure rate across the surface, while sufficient carbon cloth was used to cover the entire test surface. The scatter shown by the carbon packet method reveals the problem with making a single radon emanation measurement on a surface exhibiting variation in exposure rate. Several measurements are necessary to obtain a good estimate of the total emanation from the surface, if it is impractical to measure emanation from the total surface. It is left to the investigator define a ‘good’ estimate based on the requirements of and the intent of the measurement. This study was initiated to develop a practical method for determining the radon emanation rate from granite countertops to determine the potential radon concentration increase in a home from the installed granite. The concentration increase would be predicted using eqn 6 and an assumed air exchange rate. 321 For example, in the case of a flux of 10 Bq m-2 s-1 from 3 m2 of surface would be 30 Bq s1 . Given a room volume of 12 m3 and a room air exchange rate of 1.2 m3 h-1, the equilibrium radon concentration according to eqn 6 is 1500 Bq m-3. The example is for a single room, but in most cases, the whole house should be considered. The example is for an air exchange rate of one-tenth room volume each hours. Some commercial buildings may have exchange rates as high as 12 changes each hour, while newly constructed homes may be 0.1 air change or less each hour. While careful measurement of surface emanation rates can predict the equilibrium concentration of radon of a room, such a prediction is no substitute for measuring the actual concentration under normal circumstances. The usefulness of the methods reported here is to determine the source of radon, if concentrations above the limit are found. Conclusions Two simple techniques for measuring radon emanation were demonstrated that allow field determination of emanation rates without instrumentation and elaborate set up and system volume determination. References Kitto, M.E., Haines, D.K., DiazArauzo, H., Emission of Radon from Decorative Stone, AARST 2008. Brodhead, B., Measuring Radon and Thoron Emanation from Concrete and Granite using Continuous Radon Monitors and E-PERMs, AARST 2008. Sundar, S, B; Ajoy, K; Dhanasekaran, A; Gajendiran, V; Santhanam, R, Measurement of Radon Exhalation Rate from Indian Granite Tiles, AARST 2003. Chistopher, Y., Chao, H., and Tung, T.C.W, Radon Emanation of Building Material— Impact of Back Diffusion and the Difference between One-dimensional and Threedimensional Tests, Health Physics, Vol. 76, No. 6, pp 675-681, 1999. Kotrappa, P., Stief, L.R., and Volkovitsky, P., Radon Monitor Calibration using NIST Radon Emanation Standards: Steady Flow Method., AARST 2004. EPA Large Area Activated Charcoal Canisters, $0 CFR61, Appendix B, Method 115, 2008. 322 Uncertainties in the evaluation of the dose coming from radon in tourist caves Carlos Sainz Fernández*, Ismael Fuente Merino, Luis Quindós López, Jose Luis Gutierrez Villanueva, Jose Luis Arteche García, Luis Santiago Quindós Poncela Department of Medical Physics, RADON Group Faculty of Medicine, University of Cantabria c/ Cardenal Herrera Oria s/n, 39011, Santander, Cantabria, Spain *Corresponding author: e-mail: sainzc@unican.es Abstract Indoor radon poses a health risk confirmed by a variety of studies. High concentrations of this gas can be found in closed enclosures depending on several factors like activity of source term, permeability of the source/indoor air inter-phase, as well as the interchange rate of external air. Low ventilation rates together with the presence of 226-Ra in rocks, make the tourist caves a place where medium and high radon concentrations can typically be found. With the incorporation of EURATOM basic standards for radiological protection, into the national European legislations radon have been recognised as a health risk to be controlled in workplaces. The transfer of EURATOM standards to the Spanish legislation came out on Title VII (BOE 178/2001) where radiation coming from natural sources has an analogous role as radiation emitted from artificial ones. The most significant exposure to radon in show caves is received by tourist guides who work providing tours for the general public. The implementation of an efficient radiological protection system for this kind of workplaces must take into account that in some circumstances forced ventilation may alter the humidity inside the cave affecting some of the formations or paintings. For this reason the best option to protect workers from radon is a system based on limitation of exposure by restricting the amount of time spent in the cave. For doing so, the knowledge of main uncertainty sources in dose evaluation is of main importance. The principal variability sources in dose assessment are the integration intervals for radon measurements and the aerosol conditions related with the unattached fraction of radon progeny. The first factor takes into account the daily and monthly variations in radon concentrations and can not be fully addressed in caves with the seasonal correction factors proposed for the determination of radon doses in houses. On the other hand, mainly due to the low particle concentration usually present inside the caves, the unattached fraction of radon progeny can be higher than in other workplaces. This factor is significant when dose assessment is approached from the dosimetric respiratory track model of ICRP. In this work the results of radon measurements carried out monthly in different points of 7 caves located in the region of Cantabria (Spain) as well as estimations of the dose received by workers approached from different integration intervals are presented. Also the results concerning particle concentration and its relationship with unattached progeny fraction together with their implications in dosimetric calculations are discussed for some of the caves. KEYWORDS: radon, dose, tourist caves 323 1. Introduction When we breathe in air with a concentration of radon and radon progeny, several alpha radioactive decays take place inside the lung. This irradiation is responsible of about half of the annual average effective dose received by the human due to natural sources of radiation. [1]. Outdoor radon does not represent a significant health hazard because high concentrations are never reached. However, it becomes a problem when released into a closed or poorly ventilated enclosures like dwellings, buildings and also caves and mines. Indoor concentrations of radon and its short-lived progeny depend on several factors mainly related with the entry or production rate from various sources and the ventilation rate. Many workplaces both above and below ground may be affected by high radon concentrations. On its 1990 publication, the International Commission of Radiological Protection, ICRP recommended that exposure to high radon levels should be considered as occupational exposure and remedial actions have to be taken in such situations [2]. Concerning the Spanish situation regarding radiation coming from natural sources, in 2001 the Spanish law incorporated EURATOM basic standards for radiological protection, which include a request at the EC Member States to determine the working places on which exposure to natural radiation is significant. On Title VII (BOE 178/2001) radiation coming from natural sources plays the same role than radiation emitted from artificial sources. Tourist caves represent a type of workplace with particular environmental conditions that might be affected by high radon concentrations [3, 10]. In these places in which guides host visits for the general public, typical remedial actions like forced ventilation, sealing or reducing pressure in the source rock can not be used for conservation reasons. For example, forced ventilation could alter the humidity inside the cave thus affecting the paintings or geological formations that attract the tourists. So in most of the cases the only way to reduce radon exposure to guides and other workers is to apply a radiation protection system based on restrictions in the amount of time spent in the cave. The information needed for carrying out the abovementioned protecting actions is related with the specific characteristics of the cave concerning the behaviour of radon and it decay products. In order to perform a precise effective dose calculation, factors like unattached progeny fraction (fp), equilibrium factor (F) and particle concentration (Z) are of main importance. One of the specific characteristic of the caves is the high unattached fraction due to a particle concentration far below from the usual value in dwellings. The fp values can be higher than 0.1, for places with low ventilation and without additional aerosol sources, with Z < 4 103 particle cm-3 . In the present work, the results of radon measurements carried out monthly in different points of 7 caves located in the region of Cantabria (Spain), showed in Fig. 1, as well as estimations of the dose received by workers and general public are presented. Radon exposure values are calculated from different measurement integration intervals in order to stress its importance as a source of variability in dose assessment. Additionally the results concerning particle concentration and its relationship with 324 unattached progeny fraction together with their implications in dosimetric calculations are discussed. 2. Material and methods Radon measurements were carried out within 7 caves located in the region of Cantabria in the North of Spain (Fig. 1). Radon detectors were exposed monthly along 2008 and placed inside of each cave following criteria related with most probable risk situations for workers. The analysed points were those in which guides usually spent longer periods giving explanation to the public. CR39 track-etched detectors were used for integrating measurements. Every CR39 detector was fastened under the cap of a cylindrical polypropylene container 55 mm high and 35 mm diameter which prevents radon decay products and also 220Rn from entering. Then, only alpha particles from radon that has diffused into the container and from the polonium produced inside can strike the detector. After the exposure time an etching process is performed, and the radon concentration determined by counting the tracks in a given area. Accuracy and precision of this method has been tested in a National intercomparison exercise [11], as well as in a recent European intercomparison with good results (not yet published). Also particle concentration was also measured by means of a condensation particle counter (CPC ISI 3007). In this device, air is pumped at a rate of 100 cm3 min-1 and passed throughout a porous wick containing liquid isopropyl alcohol. After the exposure of the sample to the alcohol vapour, particles grow by condensation and can be detected optically with a laser light and a detection unit. With this system particle concentrations in the range of 0 to 500,000 particle cm-3 can be detected. Mean annual effective doses coming from radon inhalation have been estimated by using ICRP65 dose assessment methodology [4]. The dose conversion factor (DCF) used for radon exposure was 5 mSv per WLM at work and 4 mSv per WLM for the general public, and the dose assessment is usually performed by assuming an equilibrium factor of 0.4 and indoor occupancy 2000 hours per year. In the special case of workers in tourist caves it must be taken into account that the occupational time is significantly lower than 2000 hours per year. For this reason, doses have been estimated from the real distribution of time spent by tourist guides inside each cave. On the other hand, the singular atmospheric conditions characterized by high indoor humidity and very low particle concentrations makes more appropriate the use of an equilibrium factor of 0.6. From an alternative approach, the effective dose can be determined using the respiratory track model of ICRP 66 [5]. For doing so, the measurement of unattached fraction fp is essential. The dependence of the fp as a function of particle concentration Z can be approximated by the semi-empirical equation [6]: fp = 400/Z (cm-3) (Eq. 1) on the model of ICRP 66, DCFu for inhalation of the unattached short-lived radon progeny in mSv per WLM can be calculated from equation: 325 DCFu = 8.4 + 64 * fp (Eq. 2) 3. Results 3.1 Radon concentration It is known that radon concentration shows monthly variations in caves [8,12]. As an illustration, Fig. 2 represents the variation in radon levels at different sectors inside the cave of Las Monedas. As it can be seen, seasonal variations are clearer at specific sectors more influenced by external changes in meteorological parameters. In each cave, from the set of values obtained monthly at different points, the annual average radon concentration has been calculated. In order to assess the influence of the integration interval in the estimation of the annual effective dose, average radon concentrations were calculated from three and six month measurements. In Table 1, a comparison of three-month integrated values with the annual mean is showed. Taking as a reference value 1000 Bq m-3, which is the action level for radon concentration in workplaces established by IAEA in 1996 [7] it can be observed that all the studied caves present some points above this value, although all the annual means are below. It can be found significant relative differences as high as 90 % with the annual mean. In some cases, when the measurements are carried out in summer period, the annual exposition to radon can be overestimated by factors up to 2. On the other hand, mean values coming from winter or spring seasons can lead to an underestimation of similar magnitude. Fig. 3 illustrates the abovementioned differences. As it would be expected, integrated values obtained from six-month measurements fit better to the monthly averaged annual radon concentration, as it can be seen in Fig. 4. However, also in this approach, overestimations up to 75 % can occur when the measurements are carried out during the second half of the year, and underestimations up to 400 % can be observed when the measurements are performed during the first half of the year. 3.1 Dose assessment According to the above mentioned ICRP65 methodology and to the information available about the real time spent by each tourist guide at different sectors of the studied caves, the annual effective dose was estimated. As it can be seen in Table 2, only three workers received doses above the general public’s limit of 1 mSv per year. In the same way, Table 3 shows the dose coming from radon inhalation received by public in one typical visit for each cave. From ICRP’s human respiratory model point of view, the differences on aerosol conditions can modify the dose conversion factors. For the most usual aerosol conditions in homes of fp = 0.08 and equilibrium factor of 0.4, a DCF of 14 mSv per WLM has been obtained by Marsh et al. [9]. This DCF can significantly increase in caves, where particle concentration is very low and subsequently values of fp as high as 0.8 can be found. The uncertainties in the calculation of DCF can be high using the 326 dosimetric model because it involves the use of parameters like weighting factors for alpha particles and lung tissues which are difficult to determine accurately. In spite of this consideration, the great differences observed between the DCF’s obtained from both models show the main relevance on unattached fraction of radon progeny in the dose calculations. In the present work, particle concentration was measured monthly at different sectors inside the caves. Table 4 shows a summary of values together with the corresponding dose conversion factors for the unattached fraction (DCFu) according to dosimetric model. Particle concentrations as low as 100 cm-3 were found in some points, which mean DCFu’s from 3 to 8 times higher than that obtained from epidemiological approach. This would mean that in some work cases, radiological protection actions should be taken because doses up to 6 mSv per year could be possible. 4. Conclusions With this study the annual mean radon concentration have been determined from monthly measurements at different points of several tourist caves. The results indicate that for an accurate dose assessment for workers under realistic situations the most detailed knowledge of the indoor radon levels evolution is needed. Approximations made from usual longer integration intervals as 3 or 6 months can lead to over or underestimations of the dose received by people working in the caves. On the other hand, the extremely low particle concentrations found inside the caves can lead to higher doses coming from radon progeny inhalation than those received by people in workplaces with similar radon levels. Finally, in some cases particle concentrations lower than 400 cm-3 have been found which would mean an unattached fraction higher than 1. Maybe the relationship used in Eq. 1 can be questioned under special environmental conditions like those present in some tourist caves. This issue is now under analysis with new measurements of the unattached fraction in the studied caves. 327 5. References [1] United Nations Scientific Committee on the effects of Atomic Radiation UNSCEAR 2000 Report to the General Assembly with Annexes, New York, Vol. I: Sources, United Nations Publication, Sales No. E.00.IX.4. New York (2000) [2] Commission Recommendation of 21 February 1990 on the protection of the public against indoor exposure to radon. (90/143/EURATOM) [3] L. S. Quindós, P. Fernandez, C. Sainz, J. Gómez, Radon exposure in uranium mining industry vs. exposure in tourist caves, Rad. Prot. Dos., 111-1, 2004, 1-4 [4] International Commission on Radiological Protection. Protection against radon-222 at home and at work. Oxford: Pergamon Press; ICRP Publication 65; Annals of ICRP 23 (2). 1993 [5] International Commission on Radiological Protection. Human respiratory track model for radiological protection. ICRP, vol. 66. Oxford: Pergamon, 1994 [6] Porstendorfer J, Reineking A, Radon: characteristics in air and dose conversion factors, Health Phys. Vol.76, 3, 1999, 300-305 [7] International Basic Safety Standards for Protection against Ionizing Radiation and for the Safety of Radiation Sources, Safety Series No. 115, IAEA, Vienna (1996) [8] Fernández P.L; Gutierrez I; Quindós L; Soto J,. Natural ventilation of the paintings room in the Altamira cave. Nature, 321-6070, 1986, 586-588 [9] Marsh J, Birchall A, Butterweck G, Dorrian M, Huet C, Ortega X, Reineking A, Tymen G, Schuler Ch, Vargas A, Vessu G and Wendt J. Uncertainty analysis of the weighted equivalent lung dose per unit exposure to radon progeny in the home. Radiat. Prot. Dosim. 102 (3), 2002, 229-248 [10] Radiation Protection against radon in workplaces other than mines. Safety Reports Series No 33, IAEA, Vienna (2003) [11] National Intercomparison Campaign. Spanish Nuclear Safety Coujncil, CSN, Colección Documentos I+D. 12. 2004 [12] Lario J., Sanchez-Moral S., Cañaveras C., Cuezva S., Soler V., Radon continuous monitoring in Altamira Cave (northern Spain) to assess user’s annual effective dose. J. Env. Rad. 80, 2005, 161-174 Captions of figures and tables Figure 1: Location of the studied caves in the Cantabria region in Spain 328 Figure 2: Continuous radon measurements in the Castillo cave during a 10 days period. Figure 3: Continuous radon measurements in the Monedas cave during a 10 days period Table 1: Integrated radon concentration and mean annual effective dose in several places inside each cave. Figure 1 329 Figure 2 330 Seasonal Comparison 1600 1400 1200 ANNUAL Bq m-3 1000 AUTUMM WINTER 800 SPRING SUMMER 600 400 200 0 El Castillo Las Monedas Hornos de la Peña El Pendo Covalanas Cullalvera Chufín Cave Figure 3 Semestral vs. Annual 1200 1000 800 Bq m-3 ANNUAL 1st HALF 600 2nd HALF 400 200 0 El Castillo Las Monedas Hornos de la Peña El Pendo Covalanas Cullalvera Chufín Cave Figure 4 331 MEAN RADON CONCENTRATION (Bq m-3) CAVE ANNUAL AUTUMN WINTER SPRING SUMMER El Castillo 571 (40- 1600) 643 500 534 595 Las Monedas 775 (170-1700) 804 863 995 438 Hornos de la Peña 539 (80-5300) 817 121 123 1096 El Pendo 707 (80-1700) 337 342 683 1465 Covalanas 350 (25-1110) 332 500 413 157 Cullalvera 397 (20-1600) 564 60 281 684 Chufín 179 (25-530) 256 121 88 250 Table 1 332 Workers Tourist Guides Dose (mSv y-1) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 0.85 0.05 0.38 0.02 0.37 0.66 0.28 0.47 0.31 0.61 0.79 0.01 1.12 0.52 1.66 1.18 0.06 0.70 1.06 0.76 0.64 0.56 0.04 0.63 0.47 0.34 0.06 0.62 Cave El Castillo Las Monedas Hornos de la Peña El Pendo Covalanas Cullalvera Chufín Table 2 Dose to the Public in one visit (mSv) 0.0027 0.0031 0.0088 0.0037 0.0021 0.0023 0.0015 Table 3 333 Cave El Castillo Las Monedas Hornos de la Peña El Pendo Covalanas Cullalvera Chufín Mean particle concentration (cm-3) 868 (100-350) 1154 (600-3000) 3508 (440-19000) 1903 (400-5400) 1912 (370-16000) 3106 (950-12000) 3428 (1100-14000) DCFu (mSv WLM-1) 37,89 30,58 15,70 21,85 21,79 16,64 15,87 Table 4 334