A Radiation Monitoring System With Capability of Gamma Imaging and Estimation of Exposure Dose Rate

Transcription

1 IEEE TRANSACTIONS ON NUCLEAR SCIENCE, VOL. 49, NO. 3, JUNE A Radiation Monitoring System With Capability of Gamma Imaging and Estimation of Exposure Dose Rate Wanno Lee, Gyuseong Cho, and Ho Dong Kim Abstract A new radiation monitoring system, which has the capability of both gamma imaging and estimation of real-time exposure dose rate, was developed by an integration of a visual image and a dose image. It is composed of a portable gamma imaging system for visualizing gamma-source distribution and a dose conversion unit (DCU) for real-time dose conversion from gamma spectrum. Using Monte Carlo simulation, gamma-ray spectra for energies from 10 kev to 5 MeV with energy intervals of 10 kev were calculated for a cylindrical NaI (Tl) detector used in our developed system. These calculated spectra were compared with the measured spectra for several monoenergetic gamma rays using a multichannel analyzer. Based on the simulation data and the dose conversion coefficients obtained from ICRU 47, the dose conversion factors were also calculated by matrix calculation. The feasibility of the DCU using the dose conversion factors was evaluated by measurement of exposure dose rates as a function of distance between the gamma ray and detector for several standard sources. The exposure dose rates measured by the DCU are within an 11.5% deviation with analytical values based on gamma factor. They are also similar to the dose rates measured by a high-pressure ionization chamber with maximum difference of six. We finally visualized two-dimensional distributions of gamma source and measured the exposure dose rate simultaneously. I. INTRODUCTION APORTABLE gamma imaging system combined with a charge-coupled device (CCD) camera for radiation monitoring has been widely used for applications such as the preparation of an intervention in a hot cell or a control in the decommissioning of nuclear facilities. This system not only allows the operator to easily visualize the distribution of radiation source but also can distinguish two sources having different energies and radioactivities [1] [6]. A high-pressure ionization chamber (HPIC) has been generally used as an environmental radiation monitoring system (ERMS) because of the capability of the real-time exposure dose rate. Recently, solid-state detectors have been proposed for replacing the HPIC and adapted because of their simplicity in operation and maintenance and their capability of distinguishing man-made radiation from natural background [7] [10]. Manuscript received July 20, 2001; revised December 12, W. Lee and G. Cho are with the Radiation Detection and Medical Imaging Laboratory, Department of Nuclear Engineering, Korea Advanced Institute of Science and Technology, Taejon, Korea ( petor@cais.kaist.ac.kr). H. D. Kim is with the Korea Atomic Energy Research Institute, Taejon, Korea. Publisher Item Identifier S (02) Fig. 1. Configuration of our developed system, which is composed of gamma imaging system and dose conversion system. Although the portable gamma imaging system for radiation monitoring has several advantages, it is difficult to measure the real-time exposure dose rate. To address this problem, we propose a new radiation monitoring system by the application of dose conversion method used in the ERMS. II. MATERIALS AND METHODS A. System Description The portable gamma imaging system based on a scintillation detector consists of a pinhole collimator, a position-sensitive photomultiplier, an electronic circuit board, a digital signal and imaging processor, and a conventional CCD camera. We will not discuss the system because its principle and performance are well studied in many papers [1] [6]. The channeled pinhole collimator, which is one of the main differences between our gamma imaging system and the other system, was studied in our other paper [11]. Fig. 1 shows the configuration of our developed system, which integrates the gamma imaging system and the dose conversion system used in the dose conversion unit (DCU). The DCU will be described in the next section. B. Dose Conversion Method The exposure dose rate can be estimated from the measured gamma-ray spectrum, and its equation is followed by where is the measured spectrum and is the dose conversion factor depending on the detectors. and (1) /02$ IEEE

2 1548 IEEE TRANSACTIONS ON NUCLEAR SCIENCE, VOL. 49, NO. 3, JUNE 2002 where is the detector response for a single incident gamma ray with energy and is the incident gamma-ray flux distribution. The exposure dose rate by substitution of (2) into (1) can be written as (3) According to [12], the integral in the bracket is equal to, and it represents the exposure dose rate due to a single incident gamma ray with energy. It is written by its definition as follows: (4) Fig. 2. A principle of dose conversion and the basic structure of the DCU. are the lower and the upper energy boundaries of the measured gamma-ray spectrum; they are set to 10 kev and 3 MeV, respectively [12]. There are usually two methods to calculate the exposure dose rate based on (1). The first method is the usage of a general spectroscopy system, consisting of a NaI(Tl) scintillation detector and a multichannel analyzer (MCA). However, it is not practical for real-time radiation monitoring because this method requires numerous calculations. The second method was proposed and developed by Moriuchi et al. It is a simple method to calculate the real-time exposure dose rate using a NaI(Tl) detector and a pulse-height analyzer [9]. Cho and other researchers also presented a similar method with Moriuchi to estimate the exact dose rate using a 2 2 in NaI(Tl) or a 3 3 in NaI(Tl) and the DCU [12] [14]. To measure the real-time exposure dose rate, we modified the DCU in an original way. Fig. 2 shows the principle of dose conversion through the gamma spectrum and the schematic structure of the DCU. While fixing the upper level discriminator (ULD) to the voltage level equivalent to 3 MeV, which is the level usually used in an environmental radiation monitoring system as the maximum energy range, the low-level discriminator (LLD) based on the inverse function is modulating. If the voltage level of the input signal is located between the ULD and the LLD, the DCU generates digital pulses, of which count rates are directly proportional to the exposure dose rate in air. The DCU is composed of a single channel analyzer (SCA), a read-only memory (ROM), a digital-to-analog converter, and a microprocessor. The inverse is stored in the ROM and is converted into a modulation waveform as a function of period and fed into the comparator within the SCA as LLD values. The important tasks of the system using the DCU are to calculate the exactly and to convert the inverse into the modulation wave function practically. C. Dose Conversion Factor The measured spectrum can be expressed by (2) where is the mass energy-absorption coefficient of air, is the electronic charge, and is the average energy spent to produce an ion pair in air. The values of for incident gamma-ray energies are given by ICRU 47 [15]. Finally, the integral in the bracket of (3) can be presented as Because cannot be practically obtained as a continuous value of all incident energy, we consider a discrete one. The dose conversion factor as a discrete expression is where is the index of the pulse-height interval related to the absorbed energy and is the index of the gamma-ray energy interval, meaning. Here we calculated the up to 5 MeV energy for conservation, but up to 3 MeV energy is only used for calculating. We calculated by the matrix calculation of the inverse, and. was finally calculated by interpolation of. Fig. 3 shows a Monte Carlo geometry for the calculation of the detector response; the shielding parts are not shown in order to explicate gamma-ray distribution after interaction with the collimator. The NaI(Tl) detector is a cylindrical type of 2-in diameter and 0.39-in height. Gamma rays enter the limited angle through the pinhole collimator, as shown in Fig. 3. We used the Monte Carlo -Particle Transport Code (MCNP), version 4B, for simulation [16]. D. Sampling Principle Using Modulation Wave Function For conversion from spectrum to dose with the DCU system, the energy-dependence dose conversion factor should be transferred into a modulation wave function as a function of time. On the assumption that is directly proportional to the absorbed energy in the detector, can be expressed by (5) (6) (7)

3 LEE et al.: RADIATION MONITORING SYSTEM 1549 Fig. 4. Sampling process of the pulse: (a) the modulation wave function as the LLD and (b) the passing probability of the input pulse when its height is proportional to the energy E. Fig. 3. Geometry of Monte Carlo simulation for calculating the detector response. Then, the exposure dose rate by substitution of (7) into (1) is rewritten by where is a constant. We use a value of 3 (V/MeV) for our system when is 3 MeV, and the maximum is set to 9V. When changes with the period time, the input pulses of the energy that are generated within the time for the period can pass the DCU because the input pulses during only this time exist between the fixed ULD and the various LLDs, as shown in Fig. 4. Therefore, the passing probability can be expressed by (13) where is the digital output pulse rate generated at the DCU, is a generation probability of the digital output pulse when the height of the analog input pulse is proportional to the energy absorption of the gamma ray, and is the dose conversion constant having units of R/hr/cps. This constant is calculated by (7) as (8) (9) (10) where is one and is the calculated value by the Monte Carlo method. The probability is finally presented by the dose conversion factor as (11) If we assume that the input pulses produced in a detector are a perfect random process, (8) could be accomplished by sampling the input pulses and allowing them to pass at a proportional rate according to their height. Fig. 4 shows the sampling principle of the input pulse by the modulation wave function as the LLD. On the assumption that the height of the input pulse is proportional to the absorbed energy, could also be described by (12) where is controlled by ROM (we determined its value as 3 ms through several experiments). If the input pulse of the maximum energy enters into the DCU, it always generates a digital pulse because the passing probability is one, according to (13). The energy dependence could be transferred into a time function by the conception of the probability as expressed above. The energy can be expressed by (13) in the following form: (14) Because the value of is equal to, this relation can be expressed by The modulation wave function inverse from (16). III. RESULTS AND DISCUSSIONS (15) (16) can be induced by the For the verification of Monte Carlo simulation results, we compared the measured spectra with the simulated results for several standard sources such as Co (0.122 MeV), Co (1.17 and 1.33 MeV), Cs (0.662 MeV), and Tc (0.140 MeV).

4 1550 IEEE TRANSACTIONS ON NUCLEAR SCIENCE, VOL. 49, NO. 3, JUNE 2002 Fig. 5. Comparison between the measured spectrum by NaI(Tl) scintillation detector with MCA and the simulated spectrum by Monte Carlo method for Cs (662 kev). Fig. 7. Dose conversion factor G(E) as a function of incident gamma-ray energies from 10 kev to 3 MeV. Fig. 6. Three-dimensional detector response function calculated by the Monte Carlo simulation from 10 kev to 5 MeV with 10-keV energy intervals. Fig. 8. Comparison between calculated and measured dose rates as a function of source detector distance for Cs. Fig. 5 shows the measured and the simulated spectra for the Cs gamma-ray source. Two results are almost equal in the all energy regions except two parts of low energy and Compton edge. It seems that these differences are due to background and scattered gamma rays with the shielding materials. The trends of the other standard sources are also similar to Fig. 5. However, these effects are not significant because this developed system does not use the MCA and the data of the low-energy regions below 60 kev. After confirming our MCNP simulation method based on comparison process, we calculated the according to the change of gamma-ray energies from 10 kev to 5 MeV with 10-keV energy intervals. The calculated was shown in Fig. 6. Fig. 7 shows, which was calculated by the inverse and the dose conversion coefficient. increases for the energies above 60 kev, as expected. Although the calculated showed an inverse trend curve below 60 kev, it is not significant in our dose conversion system because all gamma-ray energies below 60 kev are shielded by aluminum cap of NaI(Tl) detector. Based on the trends of the calculated for energies above 60 kev, the modulation wave function was directly induced by the inverse. We used this function for estimating the exposure dose rate. Fig. 8 shows the measured and the calculated exposure dose rate for Cs source. The measurement data for the various distances by the NaI(Tl) detector-based DCU agree with analytical values using gamma factor. All our measurement data are underestimated when they are compared with the analytical values. The maximum difference between the calculated and the measured was 11.5% when the Cs source of 5.4 was located in the distance of 12 cm from the detector. The exposure dose rates measured by both our development system and the HPIC were in good agreement within 6% deviation when measured separately in the same conditions and the same standard sources. The two-dimensional color gamma image and its real-time exposure dose rate for a point source are shown in Fig. 9 when the activity of the Cs is 1 and the distance from the source to the detector is 120 cm. The difference between the measured and the calculated dose rate is about 7.3%.

5 LEE et al.: RADIATION MONITORING SYSTEM 1551 Fig. 9. The fusion gamma imaging and the real-time exposure dose rate using our developed system. IV. CONCLUSION We calculated the dose conversion factor using Monte Carlo method for the NaI(Tl) detector. From this result, the dose conversion system using the DCU was developed to access the exposure dose rate for radiation monitoring. The exposure dose rates of the developed system show good agreement with analytical values and experimental values using the HPIC for several standard sources. By the integration of the dose conversion system based on the DCU and the conventional gamma imaging system, we developed the new radiation monitoring system with the capability of gamma imaging and estimation of the real-time exposure dose rate. REFERENCES [1] R. H. Redus, V. Nagarrkar, L. J. Cirignano, W. McGann, and M. R. Squillante, A nuclear survey instruments with imaging capability, IEEE Trans. Nucl. Sci., vol. 39, pp , Aug [2] R. H. Redus, M. Squillante, J. S. Gordon, P. Bennett, and G. Entine, An imaging nuclear survey system, IEEE Trans. Nucl. Sci., vol. 43, pp , June [3] S. V. Guru and Z. He, A portable gamma imaging system for radiation monitoring, IEEE Trans. Nucl. Sci., vol. 42, pp , Aug [4] R. Redus, M. Squillante, J. Gordon, G. Knoll, and D. wehe, A combined video and gamma ray imaging system for robots nuclear environments, NIM Phys. Res. A 353, pp , [5] O. Gal and F. Jean, The CARTOGAM portable gamma imaging system, IEEE Trans. Nucl. Sci., vol. 47, pp , June [6] O. Gal, C. Izac, F. Laine, and A. Nguyen, CARTOGAM: A portable gamma camera, NIM Phys. Res. A 387, pp , [7] H. Hayakawa, M. Ohnishi, and H. Shimada, Measuring additional dose rate contributed by nuclear plants, Health Phys., vol. 64, no. 3, pp , [8] S. Moriuchi and L. Miyanaga, A spectrometric method for measurement of low-level gamma exposure does, Health Phys., vol. 12, pp , [9], A method of pulse height weighting using the discrimination bias modulation, Health Phys., vol. 12, pp , [10] J. S. Jun, C. Y. Yi, H. S. Chai, and H. Cho, Calculation of spectrum to dose conversion factors for an NaI(Tl) scintillation detector using the response matrix, J. Korea Phys. Soc., vol. 28, no. 6, pp , [11] W. Lee et al., Monte Carlo simulation for pinhole collimator design in radiation monitoring system, in Proc. KARP Spring Meet., 2000, pp [12] G. Cho, H. K. Kim, H. Woo, G. Oh, and D. K. Ha, Electronic dose conversion technique using a NaI(TL) detector for assessment of exposure dose rate form environmental radiation, IEEE Trans. Nucl. Sci., vol. 45, pp , June [13] J. Park, A study on the design of the digital SCA and DCU for the environmental radiation monitoring post, Masters thesis, Sogang Univ., Korea, pp , [14] H. G. Lee, A study on the environmental radiation monitoring system post control, Masters thesis, Sogang Univ., Korea, pp , [15] ICRU, Measurements of dose equivalents from external photon and electron radiation,, ICRU Rep. 47, Apr [16] LANL Group X-6, MCNP-4B General Monte Carlo code N -particle transport code version 4B, LA M, Mar