A Simple Method to Create Gamma-ray-source Spectrum for Passive Gamma Technique

Available online at www.sciencedirect.com

ScienceDirect Availableonline onlineatatwww.sciencedirect.com www.sciencedirect.com Available Energy Procedia 00 (2017) 000–000

ScienceDirect ScienceDirect

www.elsevier.com/locate/procedia

Energy (2017) 000–000 250–257 EnergyProcedia Procedia131 00 (2017) www.elsevier.com/locate/procedia

5th International Symposium on Innovative Nuclear Energy Systems, INES-5, 31 October–2 November, 2016, Ookayama Campus, Tokyo Institute of Technology, JAPAN TheMethod 15th International Symposium on District Heating and Cooling A Simple to Create Gamma-ray-source Spectrum for Passive Gamma Technique Assessing the feasibility of using the heat demand-outdoor a a b Tomooki SHIBAfor *, Shigetaka MAEDA , Hiroshi SAGARA , forecast temperature function a long-term district heat demand

Akihiro ISHIMIa, Hirofumi TOMIKAWAa I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc

a

a Japan Atomic Energy Agency (JAEA), Tokai, Ibaraki, Japan IN+ Center for Innovation, Technology band Policy Research - Instituto Superior Técnico,Japan Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal Tokyo Institute of Technology, Meguro, Tokyo, b Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France c Département Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France

Abstract A simple but sufficiently accurate method for constructing a gamma-ray source for photon-transportation calculations is developed Abstract for canisters containing fuel debris from the Fukushima Daiichi nuclear power plant. The new method couples a baseline spectrum evaluated using the ORIGEN2 code with a line spectrum obtained from fission products (FPs) of interest. One of the advantages heating networks commonly addressed in the as one of the most effective solutions decreasingThe the ofDistrict the method is its ability toare take bremsstrahlung X rays intoliterature consideration using the bremsstrahlung libraries offorORIGEN2. greenhouse gas emissions from to thecalculate building the sector. These of systems require high investments which are returned through the heat new gamma-ray source is used response a high-purity germanium (HPGe) detector, and these results are sales. Duewith to athe changed test climate conditions and building renovation policies, heatThe demand could decrease, compared benchmark of experimental measurements of irradiated fuel pins. shape in of the the future simulated gamma-ray prolonging the investment return period. spectrum agrees well with that of the spectrum obtained by experiment. Moreover, bremsstrahlung X rays are confirmed to have The effect main scope of this paper to assess because the feasibility of using the heat demand – outdoor temperature forscattering heat demand little on spectrum shapeisformation the baseline gamma-ray spectrum is mainly formed byfunction Compton of strong gamma fromofFPs such as located Cs-137 in andLisbon Eu-154. Gamma-ray a debris forecast. The rays district Alvalade, (Portugal), wassource used data as a are casecreated study.for The districtcomposition is consistedrecently of 665 evaluated JAEA, photon-transportation calculation is performed to evaluate the leakage from buildingsatthat varyand in subsequent both construction period and typology. Three weather scenarios (low, medium,gamma-ray high) and spectra three district arenovation canister containing debris for nuclear-material-accountancy purposes. scenariosfuel were developed (shallow, intermediate, deep). To estimate the error, obtained heat demand values were ©compared 2017 Thewith Authors. by Elsevier resultsPublished from a dynamic heatLtd. demand model, previously developed and validated by the authors. © 2017 The Authors. Published by Elsevier Ltd. committee of the 5th International Symposium on Innovative Nuclear Energy Peer-review under responsibility of the organizing The results under showed that when only weather change is considered, of error could be acceptable for Nuclear some applications Peer-review responsibility of the organizing committee of the the 5th margin International Symposium on Innovative Energy Systems. (the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation Systems. scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered). Keywords: Nuclear Power Station, fuel debris, passive The valueFukushima of slopeDaiichi coefficient increased on average within thegamma range technique of 3.8% up to 8% per decade, that corresponds to the decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and renovation scenarios considered). On the other hand, function intercept increased for 7.8-12.7% per decade (depending on the coupled scenarios). The values suggested could be used to modify the function parameters for the scenarios considered, and improve the accuracy of heat demand estimations. © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and *Cooling. Corresponding author: Tel.: +81-29-284-3456 E-mail address: [email protected]

Keywords: Heat demand; Forecast; Climate change 1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the organizing committee of the 5th International Symposium on Innovative Nuclear Energy Systems.

1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.

1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the organizing committee of the 5th International Symposium on Innovative Nuclear Energy Systems. 10.1016/j.egypro.2017.09.446

2

Tomooki Shiba et al. / Energy Procedia 131 (2017) 250–257 T. Shiba et al. / Energy Procedia 00 (2017) 000–000

251

1. Introduction Severe loss-of-coolant accidents at the Fukushima Daiichi Nuclear Power Station (1F) produced fuel debris in the reactor cores of Units 1–3. Since then, JAEA’s Integrated Support Center for Nuclear Nonproliferation and Nuclear Security has been promoting the development of passive gamma techniques, principally in collaboration with the Tokyo Institute of Technology. Some fission products (FPs) such as Ce and Eu seem to have very low volatility, even in high-temperature environments under severe accidents, and they are supposed to chemically coexist with fuel elements such as uranium and plutonium. Using our passive gamma technique, we first measure the gamma rays from such FPs and estimate their amount and burnup; then, we multiply the mass ratio of FPs and nuclear materials to obtain the mass of the nuclear material of interest. Feasibility studies have been performed based on experiences from Three Mile Island Unit 2 and using numerical [1-4] and material-science approaches [5]. Recently, the inventory of core materials in the 1F unit 1 was carefully estimated based on three-dimensional core-burnup and -depletion calculations with assumptions as reasonable as possible. Currently available “best estimates” are prepared by JAEA [6]to be utilized as fundamental data for further R&D of non-destructive-assay technologies for nuclear-materialaccountancy purposes [7]. Gamma-ray-source data should be developed for the debris composition based on the inventory, and a subsequent photon-transportation calculation should be performed to evaluate leakage-gamma-ray spectra according to the fuel debris. However, the creation of a line-spectrum source is considerably time consuming. In this paper, we develop a relatively simple but sufficiently accurate method for building up a gamma-ray source by coupling a baseline spectrum evaluated by the ORIGEN2 code [8] with several line spectra of interest. One of the advantages of this method is the capability of taking bremsstrahlung X rays into consideration utilizing the bremsstrahlung libraries of ORIGEN2. The new gamma-ray source is used to calculate the response of an HPGe detector, and the results are compared with a benchmark test of experimental measurement of irradiated fuel pins. After that, the gamma-ray-source data are developed for the debris composition based on “best estimates” using the new method, and a subsequent photon-transportation calculation is performed to evaluate the leakage-gamma-ray spectra from a canister containing fuel debris for nuclear-material-accountancy purposes. 2. Methodology 2.1. Coupling method The gamma-ray spectra from spent fuels cooled for several years comprise line-spectrum and baseline parts, which can be seen, for example, in Fig. 2–6 of Sato et al. [9] and in Fig. 3 of Uetsuka et al. [10]. The part containing the line spectrum is generated by strong gamma emitters, such as Cs-134, Cs-137, Eu-154, and Co-60, even after several years of cooling [10], and the part containing the baseline is formed of the following components: Compton scattering from strong gamma-emitting FPs as mentioned above, gamma rays from FPs whose peaks are not visibly prominent, especially in a logarithmic graph, and a bremsstrahlung continuum. Sagara et al. have created source-photon data lineby-line to perform a characteristic study of gamma-ray emission from hypothetical fuel debris [2]. However, the creation process of the source-photon data is considerably time consuming because too many line spectra need to be considered. The development process of the new method is relatively simple compared to the line-by-line creation of gammaray sources, but it is accurate enough for our purposes. In our source creation, the baseline spectra (divided into 18 energy groups) are evaluated with the ORIGEN2 code using a photon library without the FPs of interest. To build the line spectra of those FPs, we multiplied their activities by the gamma-ray intensities of each of their decays. Descriptions of the decay-gamma-ray energy and intensities can be obtained from the Evaluated Nuclear Structure Data Files available on the website for the National Nuclear Data Center [11]. Subsequently, both the 18-group baseline spectrum and the line spectrum were used as input data for the photon-transportation calculation performed using the MCNP5 code [12]. This new technique is hereafter called the “coupling method.” A schematic view of the calculation flow of the coupling method is shown in Fig. 1. One of the advantages of this method is that it enables bremsstrahlung X rays to be taken into consideration using the bremsstrahlung libraries of the ORIGEN2 code.

252

Tomooki Shiba et al.Procedia / Energy00 Procedia (2017) 250–257 T. Shiba et al. / Energy (2017) 131 000–000

3

Fig. 1 Calculation flow of the coupling method

2.2. Benchmark test for HPGe-detector response We used a cut mixed oxide (MOX) fuel pin irradiated in the Joyo experimental fast reactor as a benchmark test. The gamma-ray measurement was conducted in the fuel monitoring facility at the Oarai research and development center of JAEA. The specifications of the fuel pin are shown in Table 1. The fuel pin was placed in a stainless-steel (SS) container. The gamma rays were measured using an HPGe GC2020-7600-2002C manufactured by CANBERRA [13]. There was an approximately 2-m-thick lead shield between the fuel pin and the HPGe. A schematic view of original experimental setup is shown in Fig. 2 (a). Since it was quite difficult to obtain precise statistics from the MCNP calculation due to the thick lead shielding, we decided to simplify the model to only a point source and a detector crystal as a simplified setup (Fig. 2 (b)). The F8 pulse-height tally of MCNP5 with a Gaussian-energybroadening (GEB) option was used to simulate the response of the HPGe detector. In the GEB option, the full-widthat-half-maximum (FWHM) was as follows: (1)

���� � � � �√� � �� � ,

where � is the incident-gamma-ray energy in MeV. The coefficients a (MeV), b (MeV1/2), and c (MeV-1) were provided by the user. The coefficients were decided with reference to a previous experiment measuring some peaks of FPs in an irradiated fuel pin. We used MCPLIB04 as the photon library and el03 as the electron library [12]. Cs134, Cs-137, Eu-154, and Co-60 were selected for consideration as line spectra for the first trial, referring to Uetsuka et al. [10]. Table 1 Nominal specifications of the MOX fuel pin Fuel pin (JU0001) Loading to core (YYYY/MM/DD)

2002/12/7

Removal from core (YYYY/MM/DD)

2006/11/30

Burnup (MWd/t)

53500

EFPD

265.23

Measurement (YYYY/MM/DD)

2012/12/3

Pin diameter (mm)

4.63

Pin length (mm)

65

Cladding thickness (mm)

0.7

4

Tomooki Shiba et al. / Energy Procedia 131 (2017) 250–257 T. Shiba et al. / Energy Procedia 00 (2017) 000–000

253

Fig. 2 (a) schematic view of the original experiment setup; (b) simplified simulation setup (point-source case)

3. Benchmark-test result and remodelling Fig. 3 shows a measured gamma-ray spectrum (in blue) and a simulated detector response (in orange). Note that the measured gamma-ray spectrum is drawn as a bolder line for comparison, and the simulated gamma spectrum is normalized to the height of Compton edge of Cs-137 from the experiment. There are visible discrepancies in the following points, circled in red: 1. 2. 3. 4.

In the low-energy region, the simulated spectrum is overestimated; In the multiple-scattering region by Cs-137 around 0.6 MeV, the simulation is underestimated; Some peaks are missing in the simulated spectrum; The overall trend of the simulated baseline spectrum above approximately 1 MeV is underestimated.

To overcome these discrepancies, we considered the following countermeasures to tune the simulation setup: 1. 2. 3. 4.

Placement of an Al plate to stop the low-energy gamma rays, which actually existed as the end cap of the detector in the experiment; Placement of a 1-cm-thick lead plate to force scattering of some photons before entering the detector; Inclusion of the line spectra of Sb-125, Rh-106, Pr-144, and Mn-54, as suggested in Sagara et al. [2], to add more prominent peaks to the simulation; Change from a point source to a volume source to correct attenuation effects with reference to the actual pin size.

The reconstructed simulation model is shown in Fig. 4. The sizes of the SS container, fuel pin, Al plate, HPGe, and Cu cold finger are exact.

254

Tomooki Shiba et al. / Energy Procedia 131 (2017) 250–257 T. Shiba et al. / Energy Procedia 00 (2017) 000–000

5

Fig. 3 Gamma-ray-spectrum comparison (point-source case)

Fig. 4 Reconstructed simulation model

Fig. 5 shows a comparison between the measured gamma-ray spectrum (in blue) and the re-simulated detector response (in orange). Although the discrepancies disappeared, the simulation was overestimated in the multiplescattering region because the multiple-scattering effect caused by the lead plate was too strong; thus, this design could still benefit from further fine-tuning for improved accuracy. Nevertheless, the design has already been well tuned to a visible level of accuracy. Table 2 presents peak-height comparisons of prominent peaks between the measurement and simulation. Peak heights agreed within 30% with a few exceptions; we can see that the peaks of Rh-106 were underestimated, indicating systematic bias. Co-60 is overestimated due to the lack of impurity-analysis data used in the ORIGEN2 input data for the first place. Cs-134, Cs-137, and Eu-154 agreed well within 10% with the exception of the 1.37-MeV peak of Cs-134. It is worth noting that the peaks should be examined more accurately by comparison of the peak area of the region of interest in future studies.

6

T.Tomooki Shiba etShiba al. / Energy 00 (2017) 000–000 et al. / Procedia Energy Procedia 131 (2017) 250–257

255

Fig. 5 Gamma-ray-spectrum comparison (re-modeled case)

Table 2 Peak-height comparison Nuclide Sb−125

γ ray energy (MeV) 4.28E−01

Measurement

Simulation

Difference*

9.65E+03

5.99E+03

−38%

4.63E−01

6.91E+03

6.11E+03

−12%

6.01E−01

4.15E+03

4.38E+03

6%

6.36E−01

2.91E+03

3.17E+03

9%

6.71E−01

5.68E+02

4.91E+02

−14%

5.63E−01

2.36E+03

2.58E+03

9%

5.69E−01

3.30E+03

3.34E+03

1%

6.05E−01

1.31E+04

1.25E+04

−5%

7.96E−01

9.10E+03

9.56E+03

5%

8.02E−01

1.03E+03

9.87E+02

−4%

1.04E+00

1.71E+02

1.83E+02

7%

1.17E+00

2.05E+02

2.16E+02

6%

1.37E+00

2.29E+02

2.88E+02

26%

Cs-137

6.62E−01

4.10E+05

3.74E+05

−9%

Rh-106

6.22E−01

9.74E+03

5.77E+03

−41%

1.05E+00

1.07E+03

7.21E+02

−33%

1.13E+00

3.08E+02

2.31E+02

−25%

6.92E−01

3.34E+02

3.08E+02

−8%

7.23E−01

1.34E+03

1.29E+03

−4%

7.57E−01

4.49E+02

4.04E+02

−10%

9.96E−01

6.50E+02

6.87E+02

6%

1.00E+00

9.74E+02

9.09E+02

−7%

1.27E+00

1.46E+03

1.51E+03

4%

6.97E−01

7.23E+02

5.29E+02

−27%

Cs−134

Eu-154

Pr-144

Tomooki Shiba et al. / Energy Procedia 131 (2017) 250–257 T. Shiba et al. / Energy Procedia 00 (2017) 000–000

256 Mn-54 (AP**)

8.35E−01

8.32E+02

1.04E+03

25%

Co-60 (AP)

1.17E+00

5.85E+02

1.63E+03

179%

1.33E+00

5.23E+02

1.52E+03

192%

7

Difference = (Simulation – Measurement) / Measurement. ** AP = Activation Product *

4. Application to fuel debris from 1F unit 1 4.1. Geometry of fuel-debris simulation As an applicability test, a new gamma-ray source was developed using a coupling method based on the common set of source terms of the 1F unit 1 debris described by Nagatani et al. [7]. The subsequent simulation of the leakage gamma rays from the canister containing the debris was performed by MCNP5. We used MCPLIB04 as our photon library and el03 as our electron library. In this study, we considered the same HPGe detector as in sub-section 2.2 to obtain a good resolution for the spectrum. A lead block was used to correctly simulate the multi-scattering effect, as suggested by the analysis in section 3. A 1-cm-thick layer of water between the canister and detecting system was also assumed, as required in the common model of simulation [7]. The simulation setup for the debris case is shown in Fig. 6. As was mentioned in 2.1, ORIGEN2 is capable of choosing a bremsstrahlung library for the cases of assumed matrix materials UO2 and H2O, as well as the no-bremsstrahlung case. We conducted simulations using these 3 libraries for comparison.

Fig. 6 Setup geometry for the debris case

4.2. Simulation result Fig. 7 shows the results of the simulations comparing each spectrum. Note that we chose to show the result in the case of low-FP release with 10 years of cooling time. There is no visible shape difference between spectra, that is to say, the effect of the bremsstrahlung X ray has little effect upon formation of the spectrum shape. The reason for this is that the baseline spectrum of gamma rays is mainly formed by the Compton scattering of strong gamma rays from FPs such as Cs-137 and Eu-154. Hence, we conclude that the UO2 bremsstrahlung library is sufficiently accurate to be used in future discussion and analysis of the bremsstrahlung effect by fuel debris. 154Eu was clearly seen at 1.27 MeV; this is one of the most important nuclides for nuclear-material accountancy of fuel debris [2].

8

Tomooki Shiba et al. / Energy Procedia 131 (2017) 250–257 T. Shiba et al. / Energy Procedia 00 (2017) 000–000

257

Fig. 7 Gamma-ray spectra from fuel debris

5. Conclusion We developed a simple but sufficiently accurate method to build up a gamma-ray source for photon-transportation calculation from canisters containing fuel debris. The simulated shape of the gamma-ray spectrum agreed well with that obtained by the experiment. Moreover, we confirmed that bremsstrahlung X rays had little effect on forming the spectrum shape because the baseline spectrum of gamma ray was mainly formed by the Compton scattering of strong gamma rays from FPs, such as Cs-137 and Eu-154. The gamma-ray-source data were created for a debris composition newly evaluated in JAEA, and the subsequent photon-transportation calculation was performed to evaluate the leakage-gamma-ray spectra from a canister containing the fuel debris for nuclear-material-accountancy purposes. The peak of Eu-154 was clearly seen at 1.27 MeV, even after 10 years of cooling. Acknowledgements The authors are particularly grateful for the support provided by Mr. K. Okumura of Japan Atomic Energy Agency. References [1] H. Sagara et al., Proceeding of INMM 54th Annual Meeting (2013) [2] H. Sagara et al., J. Nucl. Sci. Tech., Vol. 51, No. 1, pp. 1-23, 2014, DOI: 10.1080/00223131.2014.852994 [3] H. Tomikawa et al., Proceeding of INMM 55th Annual Meeting (2014) [4] T. Shiba et al., Proceeding of INMM 56th Annual Meeting (2015) [5] A. Ishimi et al., JAEA-Testing 2014-002 (2014) [6] K. Okumura, to be decided how to refer [7] T. Nagatani et al., Proceeding of INMM 57th Annual Meeting (2016) [8] ORNL. ORIGEN2.2. CCC-371. Oak Ridge (TN): Radiation Safety Information Computational Center (2002) [9] S. Sato et al., CRIEPI Report, L14003 (2015) [10] H.Uetsuka et al., JAERI-Research 95-084 (1995) [11] National Nuclear Data Center (NNDC) 2016. Available at: http://www.nndc.bnl.gov [Accessed April 2016] [12] RSICC. Monte Carlo N-particle transport code system including MCNP5-1.60 and MCNPX-2.7.0 and data libraries: CCC-740. Oak Ridge (TN): Radiation Safety Information Computational Center; 2008. [13] GC2020-7600-2002C Germanium Coaxial Detector, CANBERRA Industries Inc. Available at: http://www.canberra.com/products/detectors/pdf/SEGe-detectors-C40021.pdf