The Japan Atomic Energy Agency (JAEA) operates the Fukushima Environmental Safety Center in Minamisōma City, Fukushima Prefecture, about 20 km north of the Fukushima Daiichi nuclear power plant, and is currently conducting a nationwide radioactive contamination survey and mapping project to help the surrounding area return to normal life after the accident [1-2]. As part of the mapping project, the JAEA is investigating radioactive contamination around Fukushima by conducting mobile radiation surveys using vehicles and aircraft, and updating contamination maps around Fukushima annually. This is an important task that is directly related to the development of environmental radiation detection methods and procedures in the Republic of Korea [3-4].
Surveys of radioactive contamination around the Fukushima area to obtain measurement data and perform analyses can provide an important reference for validating and complementing field gamma ray instruments and analytical methods developed by the Korea Atomic Energy Research Institute (KAERI). Therefore, the Environmental Instrumentation Team of KAERI in collaboration with the Remote Monitoring Group of the JAEA organized radiation surveys in highly radioactive areas around Fukushima. In this study, environmental radiation measurements were performed in two contaminated areas using three types of detectors: CZT, NaI(Tl), and LaBr3(Ce). Dose rate values were converted using dose rate conversion factors for the three types of detectors, and dose rate maps were prepared and compared.
2. Materials and Methods
2.1 Survey Area
Currently, a 5 km radius around the Fukushima accident area is set as a difficult-to-return zone, where decontamination of radioactive materials is still ongoing. As the areas where the measurements of this study were conducted are located inside the difficult-to-return zone, access was only possible with the cooperation of the Japanese government and the JAEA. Fig. 1 shows the location of the survey areas. In the current study, in situ measurements and walking surveys were conducted at site A and site B. Site A is in Okuma Village, a large area of forest and bushland, approximately 250 m across and 120 m long with gentle hills and private residences. To the north of the survey area is a two-lane road and some rice fields, and there is a small agricultural water reservoir in the left-center part of the survey area with surrounding forests. This is an area where the JAEA has conducted airborne surveys and environmental surveys every year since the accident and thus has many years of survey data. The red star in Fig. 2 marks the location of the in situ measurements and the altitude correction flight point for calibration of the airborne survey.
Site B is a public parking lot in Okuma village, where there are hotspot contamination areas with high radiation levels in certain areas. The parking lot area of site B is mostly asphalt pavement with a slight slope surrounded by forest. One hotspot contamination point is located in the northeast of site B, at the bottom of a slope where the parking lot is located. The hotspot is likely caused by long-term deposition of radioactive contaminants from rain along the sloped asphalt pavement. At site B, in situ surface measurements of the hotspot contamination point were taken as well as walking surveys of the entire parking lot. Fig. 3 shows the location of site B and the hotspot. Site B is also an area where the JAEA has conducted environmental surveys for many years.
2.2 Radiation Survey Instruments
MARK-A1 (Monitoring of Ambient Radiation by KAERI-Airborne 1) was developed as a lightweight instrument for rapid monitoring of contaminated areas with high dose rates after an accident using CZT detector . The advantage of CZT detector is that the material can be manufactured into different shapes, for example co-planar grids and small pixel detectors to optimize for differing applications. This therefore improves detection characteristics [6-12]. Also the CZT detector has a very high mass number (A), creating a very good counting efficiency – as the likelihood of interaction between the material with incident radiation is increased relative to other materials (e.g. air, water, and lead illustrated for comparison) and NaI(Tl) scintillators [9-11]. Despite these advantages, problems with the uniformity of the crystal structure, such as random grain boundaries, have limited its effectiveness by causing charge trapping – reducing the counting efficiency in certain applications and limiting the volume of the detectors to small sizes [6,12].
The system consists of a CZT (cadmium zinc telluride) detector with size of 15×15×7.5 mm3, from ZRF Ritec SIA(Latvia), a signal processing unit with MCA (multichannel analyzer) and pre-amplifier, a GPS (global positioning system), a voltage circuit and controller, a battery pack, and a Bluetooth interface unit connected to a PC on the ground. The CZT detector has the following specifications: bias voltage of 2,500 V or less, FWHM of 3.5% at 662 keV, and SHV connector type. In addition to the hardware components, a MARK-A1 dose conversion algorithm was developed to estimate ambient dose rates from the energy spectra measured during survey. Depending on the measurement environment, the MARK-A1 instrument can provide approximately 8 hours of continuous use on a full battery charge. The MARK-A1 is connected to a PC via Bluetooth and displays the measured spectrum in real time through the software developed from KAERI. The data is stored on the PC in real time and is also stored on a USB memory card plugged into the instrument. The reason for the USB storage is to ensure that no data is lost if the Bluetooth connection is unstable or disconnected. Fig. 4 shows images and schematics of the MARK-A1 components, and Fig. 5 shows the MARK-A1 system with CZT detector and dual check source. The total weight of MARK-A1 is below 1 kg, thus, the MARK-A1 can be easily carried by hand for the walking survey.
2.2.2 MARK-B1 and -B2
The MARK-B series is a backpack-type measurement system developed to perform fixed surface measurements and backpack mobile surveys. For a mobile survey, the detector is mounted on a backpack and measures environmental radiation in a large area as the person moves. The measurement sensor used in the backpack is a scintillator. For the MARK-B1, a NaI(Tl) scintillator (3" × 3") was used, and for the MARK-B2, a LaBr3(Ce) scintillator (2" × 2") was used. A LaBr3(Ce) detector provides a better energy resolution than NaI(Tl) detector and short decay time of scintillation. However, the material is also hygroscopic, requiring an airtight chamber to avoid material degradation [13-14], with the further addition of weight to the system. Also, the scintillator has the disadvantage of containing an internal background due to lanthanide radionuclides in the scintillator crystal [13-15]. This poor peak shaping, however, does not represent a problem when mapping in post-disaster zones as Fukushima area, where radioactivity is almost entirely associated with the fission products of 131I, 134Cs, and 137Cs [1-2]. However, for mapping low-level radiation anomalies, a LaBr3(Ce) detector would not represent an appropriate detector choice.
The configuration of the backpack is shown in Fig. 6. The detector’s sensor, designed so that the sensor faces the ground when the backpack is worn, consists of a preamplifier at the bottom of the sensor, followed by the MCA (multi-channel analyzer). The backpack, like the MARK-A1, has a GPS sensor and a Bluetooth transmitter. The GPS sensor stores the GPS coordinates of the measurement points, and the energy spectrum data measured in the field can be checked in real time through a computer connected via Bluetooth communication. Like the MARKA1, the MARK-B Series’ dose conversion algorithm was developed to estimate the ambient dose rate from the energy spectrum measured during the survey. The MARK-B Series connects to a PC via Bluetooth, and the measured spectrum is displayed in real time using software developed by KAERI. As with the MARK-A1, data is stored in real time on your PC and also on a USB memory card plugged into the instrument. Depending on the measurement environment, the MARK-A1 instrument can provide approximately 8 hours of continuous use on a full battery charge.
The MARK-B series is also able to make in situ measurements on the ground. For this, the instrument part is removed from the backpack and fixed on a tripod with the detection sensor at a height of 1 m positioned perpendicular to the ground. The energy spectrum data of incident radiation are then saved over a period of time. As mentioned earlier, it is essential for the LaBr3(Ce) detector to eliminate background radiation effects from the lanthanide radionuclides of its scintillator crystal. The radiation energy spectrum measured with the detector inside a lead shield to measure the background of the scintillator itself is shown in Fig. 7. It is possible to detect peaks arising from the decay of 138La and gamma ray peaks emitted by the 227Ac nuclide and daughter nuclides of the scintillator crystal. The background dose rate of the scintillator crystal was calculated by dose rate spectroscopy from the measured energy spectrum, evaluated to be about 108 ± 9 nGy∙h−1 .
2.2.3 Energy Calibration and Background Measurement
Before full-scale radiation spectrum measurements in the radioactively contaminated areas around the Fukushima Daiichi nuclear power plant, energy calibration using calibration sources was required. For this, background measurements were performed inside and outside the JAEA office in Minamisōma, a decontaminated area, to first confirm the background radiation level. Inside the JAEA office, energy calibration of the detectors was performed using calibration sources provided by the JAEA; Fig. 8 shows a photo of the calibration measurement. Three calibration sources were used: 134Cs, 60Co, and 152Eu. Figs. 9, 10, and 11 show spectra measured with the MARK-A1, and -B1, -B2 detectors, respectively, measured with the calibration sources.
Outside the office, measurements were conducted for 30 min each. Fig. 12 is a photograph of the background radiation measurement outside the JAEA office, and Figs. 13, 14, and 15 are the background spectra obtained from the three detectors during a 30 min measurement. The energy spectrums show a 137Cs peak at 661.8 keV and a visible 40K spectrum at 1,460 keV. This indicates that 137Cs is present in the soil and buildings even though it has been decontaminated.
3.1 In Situ Measurement
3.1.1 Site A
The survey area is located about 4 km west of the Fukushima Daiichi nuclear power plant and includes a few private houses and attached fields. Behind the houses are dense forests and abandoned rice fields, which have not been decontaminated because the forest areas are difficult to access. At site A, we conducted in situ measurements for about 1 h. Since it was difficult to enter the area, we had to remove trees and grass to reach the measurement site using a weed whacker. Fig. 16 shows photos of the ground in situ measurement at site A.
The energy spectra measured using MARK-A1, -Ba, and -Bs are shown in Figs. 17, 18, and 19, respectively. In the MARK-A1 data, peaks are barely visible in the high energy region above 1,500 keV but are well visible below this region. The measured spectrum of MARK-B1 shows a high count rate over a relatively large area, but the peaks are broad due to low energy resolution. For the MARK-B2 data, gamma-ray peaks are well represented due to high energy resolution, but the count rate in the high energy region above 1,500 keV is high due to internal background radiation.
All three detectors showed high count rates in the energy region of 662 keV and 796 keV, indicating high environmental contamination with radioactive cesium, 134Cs and 137Cs. Unusually, a high peak was seen at 1,334 keV, indicating high 137Cs radioactivity at the measurement point, which resulted in gamma-ray uranium ionization at 662 keV. The larger the size of the detector scintillator, the more likely it is to produce coincidence summation, and the more coincidence summation occurs, the more errors in the dose rate calculation occur.
The dose rates at the site A in situ measurement point calculated using the dose rate conversion factor are 5.1, 5.74, and 5.03 μGy∙h−1 from the MARK-A1, -B1, and -B2, respectively. The 15% higher dose rate of B1 than that of B2 and A1 is likely due to the effect of coincidence summation.
3.1.2 Site B Hotspot
Unlike site A where radioactive cesium is evenly distributed over a large area, site B has a hotspot of high contamination. The measurement results showed a spectrum with high contamination by 137Cs, similar to that at site A. Since the distance between the sites is within 1 km and more than 10 years have passed since the Fukushima nuclear accident, it can be inferred that the distribution of radionuclides is not significantly different. Fig. 20 shows pictures of the hotspot location and the in situ measurement at the hotspot location.
Analysis of the energy spectrum measured at the hotspot shows that the 662 keV peak due to 137Cs is measured with a very high count rate, and also that the high peak at 1,334 keV is due to coincidence summation of the instrument. The spectrum measured with the LaBr3(Ce) detector also shows a peak in the 605 keV region due to 134Cs. The 1,460 keV of 40K is almost identical, as it is a natural radionuclide and is homogeneously spread over the earth’s surface. In the high energy region, the coefficients are almost the same regardless of distance because there are no artificial nuclides that emit high-energy gamma rays, so the effect of internal background radiation is dominant. Dose rate analysis shows that B1 and B2 have significantly higher dose rates of 20.16 and 17.37 μGy∙h−1, respectively, compared to A1 at 10.34 μGy∙h−1, which is about half of B1. This is likely due to the fact that A1 conducted measurements at a distance of tens of centimeters from the hotspot.
3.2 Walking Survey
3.2.1 Site A
At Site A, the MARK-B1 and -B2 detectors were mounted on a backpack and a walking survey was conducted while wearing the backpack, while the MARK-A1 detector was held in the hand. The speed of the walking survey was maintained at a normal walking pace and the detector face was oriented parallel to the ground to ensure that it remained approximately the same as in the in situ measurement. The dose rate maps were drawn using dose rate values stored every two seconds. We traveled at an average speed of about 4 km∙s−1, although they slowed down in heavily forested areas where movement was difficult. The total measurement time was about 40 minutes. Fig. 21 shows photos of the walking survey at site A. The ambient gamma dose rate was calculated from the energy spectrum collected during a 2 s counting time, and then the dose rate was mapped along the GPS route to create a dose rate map. The MARK-A1 had a maximum dose rate of 8.98 μGy∙hr−1 and an average dose rate of 4.37 μGy∙hr−1, the MARK-B1 had a maximum dose rate of 11.27 μGy∙hr−1 and an average dose rate of 5.33 μGy∙hr−1, the MARK-B2 had a maximum dose rate of 9.95 μGy∙hr−1 and an average dose rate of 4.33 μGy∙hr−1. Table 1 shows the minimum, maximum, and average values by instrument for the walking survey. Similar to the in situ measurement, B1 shows a higher dose than B2, which is likely due to coincidence summation of the NaI(Tl) detector. The highest dose rates were measured near a small puddle at the left end of the survey area, which is believed to be a concentrated source of cesium from constant accumulation in the water. Figs. 22, 23, and 24 are dose rate maps from the MARK-A1, -B1, and -B2 walking survey data, respectively.
3.2.2 Site B
At Site B, similar as site A the MARK-B1 and -B2 detectors were mounted on a backpack and a walking survey was conducted while wearing the backpack, while the MARK-A1 detector was held in the hand. A normal walking pace was maintained, and the detector face was oriented parallel to the ground to keep the survey almost identical to the in situ measurement. At site B, horizontal and vertical walking surveys were conducted according to an imaginary checkerboard grid for precise mapping. The ambient dose rates were calculated from the energy spectra collected during a 2 s counting period, and then the dose rates were mapped along the GPS path to create a dose rate map. The MARK-A1 had a maximum dose rate of 14.30 μGy∙hr−1 and an average dose rate of 3.35 μGy∙hr−1, the MARK-B1 had a maximum dose rate of 46.38 μGy∙hr−1 and an average dose rate of 6.16 μGy∙hr−1, the MARK-B2 had a maximum dose rate of 35.80 μGy∙hr−1 and an average dose rate of 7.56 μGy∙hr−1. Table 2 shows the minimum, maximum, and average values for each instrument in the walk survey. Figs. 25, 26, and 27 are dose rate maps of the MARK-A1, -B1, and -B2 walking survey data, respectively.
The maximum value of the dose rate was measured at the hotspot. A simple calculation shows that a year’s exposure to this spot would result in a radiation exposure of about 313 mSv, which is 15 times the Korean domestic worker safety standard. From the backpack survey results of site B, it was found that the survey could have a very large error depending on the speed of movement and the time taken. In the site A results, the average measured dose rate from B1 was higher than that from B2, but in this case B2 showed a higher average dose rate at site B. This can be seen in the dose rate maps, which show that the MARK-B1 was closer and spent more time in the hotspot. In the actual mobile survey, the backpack detectors traveled side-byside and performed the same survey; the error in the mobile survey data may be due to GPS errors. On the other hand, the maximum and average dose rates of MARK-A1 were much different from the other two instruments, which may be due to the fact that the MARK-A1 measurement was performed on a different day and the measurement route was quite different from the other two instruments, i.e., it did not pass close to the hotspot and the time spent in the hotspot was short. Table 2 shows the minimum, maximum, and average dose rate values by instrument from the site B walking survey.
4. Discussion and Summary
4.1 In Situ Measurement
In this study, fixed and mobile surveys were conducted in two areas around the Fukushima Daiichi nuclear power plant using three radiation detection instruments, CZT, NaI, and LaBr3. The trends were similar for all three instruments, but there were some differences in the dose rate values among the instruments. Table 3 shows the measured dose rate values at three locations from the in situ measurements. Among the results by instrument, only the hotspot in site B was significantly different, with errors of up to 20%. Except for the hotspot measurements, all other measurements showed similar results, which means that each instrument has sufficient reliability. The reason for the difference in CZT measurement results at the site B hotspot is that the radioactivity of the hotspot is very high, causing coincidence summation, and also that the measurement point was a little farther away from the hotspot than those of the other two instruments.
|Site A point
|Site B hotspot
4.2 Site A
Histograms were plotted for each of the three instruments to analyze the trends of the walking survey at site A. Fig. 28 shows histograms of the dose rate values for the MARK-A1, -B1, and -B2 instruments as a percentage probability. A similar trend can be observed for all three instruments. Fig. 29 shows box-and-whisker plots indicating that the mean and median of A1 and B2 are almost identical, while B1 is about 15% higher than the mean and median.
4.3 Site B
Similar to the previous section, a histogram was plotted for each of the three instruments to analyze the walking survey trends for site B. Fig. 30 shows histograms of the dose rate values for the MARK-A1, -B1, and -B2 as a percentage probability. It can be seen that the trends are similar for all three instruments. Fig. 31 shows box-and-whisker plots indicating that the mean and median of B1 and B2 are almost identical, while A1 is about 10% lower. In addition, there are many cases where the measured values of B1 and B2 are large, while the measured values of A1 are relatively small. This is because the MARKA1 measurement was performed on a different day and on a different measurement path from the other two instruments. In other words, the MARK-A1 did not pass close to the hotspot and the time spent near the hotspot was short, seemingly leading to the difference in results.
In this study, with the cooperation of the JAEA, we conducted detector comparison experiments and analyses in high radiation environments that are difficult to perform in Korean environments. The environmental radiation measurements were performed using three types of detectors, specifically CZT, NaI, and LaBr3, in two contaminated areas near the Fukushima Daiichi nuclear power plant. Dose rate maps were created for the three types of detectors and compared.
As a result, the trend of the three instruments was similar, but it was confirmed that there was a difference in dose rate by instrument. It was found that the large error was caused by the high radioactivity of cesium and differences in the measurement points in the hotspot area. Based on the results, it can be said that the three types of instruments have sufficient reliability and integrity even in a high radiation environment. In addition, the data obtained in this study can be used for further surveys or comparison with other methods of survey in the future, and can also be used to track dose rate changes by conducting future surveys in the same area.