Radon sampling methodologies: A case for accurate, accessible measurements using household instruments

ABSTRACT

Radon is a prevalent carcinogenic gas and the leading cause of lung cancer in the United States besides smoking. As the residential environment is the primary source of radon exposure, accessible and accurate measurements of radon in this environment are essential. However, no radon monitors have been evaluated that are inexpensive enough for regular household use. In this study, we examine two household-grade, continuous monitoring devices, the Ecosense RadonEye and EcoQube. We compare them to two research-grade instruments, the Durridge Company Rad7 and the Rad Elec Inc. E-PERM. In our study, the Ecosense household radon monitors performed accurately and can be used by homeowners and researchers alike as an affordable and reliable radon sensor.

Implications: The ability of homeowners and renters to regularly monitor the radon levels inside their home is an important preventative health measure. However, low-cost instrumentation is needed that can provide accurate radon measurements. In this study, we show that the affordable Ecosense continuous monitors produce results that are in line with expensive research-grade instruments in a residential environment, over a range of concentrations. The Ecosense monitors may be suitable for home use, and they may provide a solution that can be used by policymakers and home-dwellers alike to improve regular radon monitoring in residences.

Introduction

Exposure to radon – including ²²²Rn, ²²⁰Rn, and their progeny – is the leading cause of lung cancer besides smoking (Copes and Scott 2007; US Environmental Protection Agency 2009) and the largest cause of radiation exposure in the U.S. outside of medical procedures (National Council on Radiation Protection and Measurements 1987). ²²²Rn is a colorless and odorless radioactive gas that decays (half-life 3.8 days) into radiation-emitting products (Porstendorfer and Reineking 1999; Porstendörfer 1994). ²²⁰Rn is even shorter lived (half-life 55 seconds) (Kanse et al. 2013). Radon enters homes by emission from soil and materials into the air through cracks and joints in the building, and it can accumulate to high concentrations in areas with reduced ventilation, particularly basements (Kearfott 1989; Li et al. 2022; Nazaroff 1992; White et al. 1992).

Radon has numerous cancer and non-cancer health effects. Inhalation of radon decay products causes irradiation of respiratory tract cells (National Council on Radiation Protection and Measurements NCRP 1999). Other studies found an association between residential radon exposure and cardiovascular disease (Nyhan et al. 2018, 2019), childhood leukemia (Lubin et al. 1998; Raaschou-Nielsen et al. 2008), breast cancer (VoPham et al. 2017), and hypertensive disorders of pregnancy (Papatheodorou et al. 2021). Mechanistically, radiation exposure from the decay of radon progeny can lead to oxidative stress and systemic inflammation after deposition in the lungs and migration throughout the body (Blomberg et al. 2020; Nel 2005; Taylor and Gercel-Taylor 2013; Thomashow et al. 2013).

As the home environment is the primary exposure source of radon, measuring and evaluating residential exposures to radon is essential for predicting the health effects of household exposure (United States Environmental Protection Agency 2003). Methods for radon monitoring include both passive and active devices (Carmona and Kearfott 2019; Keith et al. 2012). The period over which these devices measure can be long- or short-term. Radon levels in homes vary widely throughout the day, between days, and between seasons, so longer term measurements are important for accurately estimating average annual exposures under normal building conditions (Li et al. 2022; Steck et al. 2004). Among the devices capable of longer-term measurement, integrated devices provide an average measurement over an extended period, while continuous measurement devices provide frequent, often hourly, measurements. While longer-term measurements are important for quantifying exposure and estimating health effects, understanding the short-term variability is critical for identifying the cause of high radon levels in homes and the maximum radon exposure at given times of day. Hourly measurements allow for higher resolution monitoring, the ability to calculate the mean and standard deviation over short time periods, and analysis of peak values, which are all important metrics for looking at health endpoints. Hourly measurements also provide better precision for indoor air chemistry analyses.

Radon monitoring devices vary significantly in price, availability, sensitivity, accuracy, and response to environmental conditions, particularly humidity. Comparing across these different devices is valuable as it could save researchers and homeowners money and time if cheaper, easier-to-use instruments are available. For exposure studies, a sensor that is inexpensive is advantageous because it allows measurements to be made in multiple locations at the same time for a period of days or weeks.

Previous studies have compared the effectiveness of radon monitors (Burghele and Cosma 2013; Cardellini et al. 2016; Chung, Piersma, and Kearfott 2021; Gunning et al. 2016; Janik et al. 2010; Papp, Cosma, and Cucos 2016). However, many of these studies are limited by the fact that they do not utilize environmentally-relevant radon concentrations; they are conducted under laboratory conditions rather than in the field (or in subterranean chambers, with high levels of radon that are less relevant to residential environments); they do not intercompare integrated and continuous devices; and/or they compare over short time periods (24 hr).

In this study, we compare four different instruments collocated in a home basement in Framingham, Massachusetts: two low-cost, commercially available household sensors (RadonEye and EcoQube, Ecosense, Haebong-ro, Danwon-Gu, Ansan-city, Gyeonggido, South Korea). These are compared to one research-grade electrostatic precipitation instrument (Rad7, Durridge Company Inc., 524 Boston Road, Billerica, MA 01821) and a pair of electret passive environmental radon monitors (E-PERM, Rad Elec Inc., Frederick, MD, USA).

Methods

In accordance with the U.S. Environmental Protection Agency radon sampling protocol, all samplers and devices were placed at least 2 feet above the floor, at least 3 feet from any doors or windows, and at least 4 inches away from any indoor surface. Samplers or devices were located a minimum of 4 to a maximum of 20 inches apart. Samplers and devices and their performance characteristics are summarized in Table 1.

Table 1. Radon devices compared in this study.

Manufacturer	Device	Type	Price ($)	Accuracy	Sensitivity (CPH per 1 Bq/m^3)	Range (Bq/m^3)	Measurement Period (measurement rate)	Features
EcoSense	Radon Eye	Pulsed Ion Chamber	174	<±10%	0.81	7–3700	1-hour moving average (every 10 minutes)	Bluetooth
EcoSense	EcoQube	Pulsed Ion Chamber	170	<±10%	0.81	7.4–3700	1-hour moving average (every 10 minutes)	Wi-Fi, Bluetooth
Durridge	Rad−7	Electronic Alpha Spectrometer	7,500	<±5%	0.78	4–750,000	1 hour (2 min − 24 hour)
Rad Elec	E-PERM	Electret Ion Chamber	3,000^a	<±5%		14–9,000	2–10 days

^a$15 is the estimated per sample cost. The cost for this study was $3,000 (voltage reader with reference electrets, $2,400; 2 E-PERM test kits, $340; 13 electrets at $20 each).

The RadonEye and EcoQube are radon measurement devices using pulsed ionization chamber technology that provides continuous, hourly (60-minute moving average) measurements. Integrated radon measurements were made using the E-PERM, which is a passive electret-based method. The E-PERM consists of an electret (a dielectric material carrying an electrical charge) mounted on an ion chamber with a filtered diffusion barrier. The initially charged electret in the E-PERM is neutralized by ions generated from radon decay in the ion chamber, causing the electret voltage to drop. This voltage drop is proportional to the time-integrated progeny concentration (Copes and Scott 2007). After sample collection, the electret voltage drop was measured with a voltage reader, which was calibrated using references provided by the manufacturer. Other reference methods using charcoal canisters or charcoal liquid scintillation vials were not suitable because they are not reliable under conditions of elevated relative humidity expected in the test homes. The RAD7 is an active electronic radon detector that measures alpha particles by spectrometry using a passivated implanted planar silicon detector to measure the energy of each alpha particle.

For comparison of the integrated device (E-PERM) with EcoQube and Radon Eye, the dates of sampling ranged from 3/6/2021 to 7/21/2022, spanning 502 days. The average duration for the E-PERM measurement was 4.5 days, with a range of 2.9 to 9.1 days. Two E-PERM instruments were used in each location, and the average of the 2 measurements were reported. The hourly values for the continuous instruments were averaged for comparison to the integrated instrument. The EcoQube #3 (EQ#3) vs. Radon Eye 1868 (RD1868) comparison used data from 9/9/21 to 7/26/2022, spanning 320 days. The comparison of the RD 1868 and EQ #3 to the RAD7 used measurements from 9/21/2021 to 7/4/2022, spanning 286 days. Instrumentation was added at different points depending on availability.

We took the 12-hour average for each device to compare among the continuous monitors. This helps control for noise and the fact that each device was not started precisely at the same time. The range of time spread between collocated device measurements was 0 to 14 minutes apart.

Comparison across devices was performed using linear regression (Python v. 3.9.7 using NumPy²) and characterized by the slope, intercept and R² of the regression.

Results and discussion

The RadonEye and EcoQube sensors both use pulsed ion chambers. These sensors have several similarities but also differences that make them suitable for use in different situations once their performance has been validated. The display on the RadonEye shows the numerical concentration for the previous 10-minute interval as well as the hourly and daily levels, whereas the EcoQube display consists of an LED light where the color indicates the concentration range (green for <100 Bq/m³, yellow for 100-148 Bq/m³ and red for >148 Bq/m³, the EPA action level for recommending radon remediation in homes). The EcoQube requires a Wi-Fi connection to transmit concentrations, while the RadonEye does not. We have studied the field performance of both devices in a residential environment where the radon levels have a wide range of concentrations.

Comparison of the continuous monitoring instruments (EcoQube, EQ #1 and Radon Eye, RD 1868) with the integrated device (EPERM) is shown in Figure 1. The regression lines relating EQ#1 and RD 1868 to EPERM had slopes of 0.933 and 0.983, respectively. The coefficients of determination were 0.93 and 0.88 for EQ #1 yellow for 100 and RD 1868, respectively. Some variability in slope, intercept and correlation is due to the fact that the measurements took place in different locations for much of the testing span, and only collocated for a few of the measurements. As such, a different range and distribution of concentrations resulted.

Figure 1. Continuous (hourly) versus integrated EPERM radon device measurements (Bq/m³ (National Council on Radiation Protection and Measurements 1987). The hourly devices are Ecosense EcoQube (EQ #1) (blue) and Ecosense Radon Eye (RD 1868) (red). Coefficient of determination (R²) values are 0.93 and 0.88 for EQ #1 and RD 1868, respectively. The sample size for EQ#1 is n = 47 and for the RD 1868 is n = 59.

The continuous instruments comparison against the research-grade continuous monitor, RAD7, is shown in Figure 2. Both instruments compared well to the RAD7, with slopes of the linear regressions for RD 1868 and EcoQube #3 (EQ#3) against the RAD7 of 1.049 and 1.112, respectively. The coefficients of determination were 0.92 and 0.93 for the RD and EQ#3, respectively.

Figure 2. Comparison of (a) Radon Eye (RD 1868) against RAD 7 (R² = 0.92) (n = 186) and (b) EcoQube (EQ #3) against RAD7 (R² = 0.93) (n = 186), 12-hour average (Bq/m³) (National Council on Radiation Protection and Measurements 1987).

The collocated Ecosense instruments, Radon Eye and EcoQube, show very good agreement (Figure 3). Consistent results are expected from two instruments measuring at the same time in the same environment (basement). The slope of the regression line fit to the plot of EQ#3 vs. RD 1868 was 1.046. The coefficient of determination was 0.95.

Figure 3. Radon Eye 1868 (RD 1868) versus EcoQube #3 (EQ#3) (R² = 0.95) (n = 590), 12-hour collocated measurements in the basement (Bq/m³) (National Council on Radiation Protection and Measurements 1987).

The Ecosense continuous monitors are in good agreement with both the more expensive research grade instruments and with the integrated monitor in a residential environment, over a range of concentrations. These low-cost sensors performed well, even under basement conditions where the relative humidity was elevated, and because they are relatively inexpensive it is practical to deploy them in field studies where sensors are deployed in multiple locations at the same time, for periods of days or weeks. These qualities also make the Ecosense monitors suitable for home use.

Author contributions

P.K. and J.L. conceived of the project. J.L. collected the samples. R.A.S. and P.K. performed calculations and analysis. All authors were involved in writing the manuscript. All authors read and approved the final version of the manuscript.

Data access

R.A.S. and P.K. had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are openly available in Figshare at http://doi.org/10.6084/m9.figshare.21365307.

Additional information

Funding

This publication was made possible by USEPA grant RD-835872, by the National Institute of Environmental Health Sciences [R21 ES029637], and by the National Institute of Environmental Health Sciences, National Institutes of Health [T32 ES007069]. Its contents are solely the responsibility of the grantee and do not necessarily represent the official views of the USEPA. Further, USEPA does not endorse the purchase of any commercial products or services mentioned in the publication.

Notes on contributors

Rebecca A. Stern

Rebecca A. Stern is a postdoctoral researcher in the Department of Environmental Health at the Harvard T.H. Chan School of Public Health in Boston, MA.

Joy Lawrence

Joy Lawrence is a research associate in the Department of Environmental Health at the Harvard T.H. Chan School of Public Health in Boston, MA.

Jack M. Wolfson

Jack M. Wolfson is a research associate in the Department of Environmental Health at the Harvard T.H. Chan School of Public Health in Boston, MA.

Longxiang Li

Longxiang Li is a postdoctoral researcher in the Department of Environmental Health at the Harvard T.H. Chan School of Public Health in Boston, MA.

Petros Koutrakis

Petros Koutrakis is a professor in the Department of Environmental Health at the Harvard T.H. Chan School of Public Health in Boston, MA.