Calculation of radioactive species transport in a TRIGA reactor

Nuclear Engineering and Design 262 (2013) 29–38 Contents lists available at SciVerse ScienceDirect Nuclear Engineering and Design journal homepage: www.elsevier.com/locate/nucengdes Calculation of radioactive species transport in a TRIGA reactor Daniela Mladin a,∗ , Mirea Mladin a,1 , Alexandru Toma a,1 , Cristian Dulama a , Ilie Prisecaru b,2 , Stefan Covaci a,1 a b Institute for Nuclear Research, Campului, 1, Mioveni 115400, Arges, Romania University “Politehnica” of Bucharest, Power Engineering Faculty, Nuclear Power Plants Department, Splaiul Independentei 313, Sector 6, Bucharest, Romania h i g h l i g h t s • • • • • We created a TRIGA facility model with CATHARE2. We calculated a source of fission products in postulated accident conditions. We calculated the source of Ar-41 during reactor normal operation. We modeled the transport and release of fission products and Ar-41 at reactor stack. Steady-state experimental Ar-41 volumetric activity is compared with the calculated activity. a b s t r a c t a r t i c l e i n f o Article history: Received 22 July 2012 Received in revised form 23 February 2013 Accepted 28 March 2013 The objective of the paper is to develop and use a model for radioactive species transport in the primary circuit and in the reactor hall of the Romanian TRIGA facility. CATHARE2 V25 code (Code for Analysis of Thermal–Hydraulics during an Accident of Reactor and Safety Evaluation) is used. CATHARE is developed by the French Atomic Energy Commission (CEA) and owned in partnership with three other French partners: EDF, AREVA-ANP and IRSN. The radio-chemical components in CATHARE2 include, besides activation products, four fission products with predefined characteristics (Kr-87, Xe-133, I-131, Cs-137). New radioactive species can be defined by the user, and the characteristics of the existing ones can be modified. The TRIGA model created comprises both the primary reactor circuit and reactor hall, involving water zones and non-condensable gas (air). Ventilation system is simulated by means of boundary conditions. Using the same facility model, two separate studies are performed with externally calculated sources: - fission product species transport and evacuation. This is done as PSA support studies, postulating core damage and volatile species release; - Ar-41 transport and evacuation. Argon activity at reactor stack is calculated for normal operation and compared to monitor readings. The paper describes also the calculation of the radioactive sources based on SCALE 4.4 in case of fission products, and using MCNP5 for Ar-41. © 2013 Elsevier B.V. All rights reserved. 1. Introduction ∗ Corresponding author. Tel.: +40 24 821 3410; fax: +40 24 826 2449. E-mail addresses: [email protected] (D. Mladin), [email protected] (M. Mladin), [email protected] (A. Toma), [email protected] (C. Dulama), [email protected] (I. Prisecaru), [email protected] (S. Covaci). 1 Tel.: +40 24 821 3410; fax: +40 24 826 2449. 2 Tel.: + 40 21 402 9511; fax: +40 21 402 9675. 0029-5493/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.nucengdes.2013.03.054 Romanian TRIGA reactor was commissioned in 1980 (first criticality was reached on November 17th 1979). There are two independent cores sharing the same pool: a high-flux 14 MW Steady State Reactor (SSR), research and materials testing reactor and an independent (from operational point of view) Annular Core Pulsing Reactor (ACPR). The SSR is a forced convection reactor cooled via a primary circuit with 4 pumps (2 pumps in operation at 14 MW) and 3 heat exchangers (2 exchangers in operation at 14 MW). The power is removed by a secondary circuit with cooling towers. The 30 D. Mladin et al. / Nuclear Engineering and Design 262 (2013) 29–38 Fig. 1. Dual core TRIGA ROMANIA. reactor was used for CANDU fuel testing, structural materials (steel, zircaloy) testing and isotopes production. The ACPR is a pulsed reactor cooled in natural convection. It can operate also in steady-state mode up to 500 kW. A view of the two cores is presented in Fig. 1. Characteristics of the Romanian TRIGA facility used in the calculation of radioactive species transport are given in Table 1. The CATHARE code (Code for Analysis of Thermal–Hydraulics during an Accident of Reactor and Safety Evaluation) (Lavialle, 2005) is developed to perform best-estimate calculations of pressurized water reactor accidents: PWR loss of coolant (large or small break, primary and secondary circuit), reactivity insertion, steam generator tube rupture, etc. It is developed in by the French Atomic Energy Commission (CEA) and it is owned by four partners: CEA, EDF, FRAMATOME-ANP and IRSN. CATHARE includes several independent modules that take into account any two-phase flow behavior: Table 1 Characteristics of TRIGA SSR. Name Characteristics Reactor tank Delay tank Pump Head Headtot Speed NPSH Total volume of water (including pool) Heat exchanger Water volume primary side Water volume secondary side Number of pipes (primary) Heat exchange surface Volume of reactor hall Ventilation system air flow (normal mode) Ventilation system air flow (emergency mode) 310 m3 110 m3 906 m3 /h (0.252 m3 /s) 35 m 42 m 1475 rot/min 5.4 m 480 m3 6.1 m3 9.1 m3 1262, I.D. = 20 mm 890 m2 18,500 m3 24,360 m3 /h 13,600 m3 /h • Mechanical non-equilibrium: vertical co- or counter-current flow, flooding counter-current flow limitation (CCFL), etc. Horizontal: stratified flow, critical or not critical flow co- or counter-current flow, etc. • Thermal non-equilibrium: critical flow, cold water injection, super-heated steam, reflooding, etc. • All flow regimes and all heat transfer regimes. In order to take into account these phenomena the CATHARE code is based on a two-fluid and six equation model with a unique set of constitutive laws. Various modules offer space discretization adapted to volumes (0D), pipes (1D) or vessels (3D) ready to assemble for reactor core and circuits description. In our case, the reactor hall, pool and primary circuit are modeled using volumes and pipes. Time discretization is fully implicit (semi-implicit for 3D) and enables solution stability to be achieved over a broad range of time step values. The maximum time step is up to the user and depends on the problem being solved. The present application to TRIGA primary circuit and reactor hall involves no heat generation. Thus, it is not challenging with respect to time step and solution stability. Concerning fission products treatment, implicit radio-chemical characteristics can be used for Kr-87, Xe-133, I-131 and Cs-137, or the user can define new components. The properties consist in halflife, activity per mass unit, Henry constant (for gases) and chemical properties for two-phase treatment: entrainment coefficients due to vaporization and condensation, time constants for gas into liquid dissolution and gas stripping from liquid. For other species, such as Ar-41, properties need to be defined in the input deck. In CATHARE, nuclides are treated individually, without possible chemical interactions. The paper uses the CATHARE2 V2.5 2 mod8.1 for two types of calculations: • Fission products transport in case of postulated accidents (underwater release and release in air) D. Mladin et al. / Nuclear Engineering and Design 262 (2013) 29–38 31 Fig. 2. Nodalization of TRIGA facility. • Argon 41 volumetric activity in the air evacuated at reactor stack in normal operation • Both analyses work with a hydraulic model of the TRIGA facility with external source of radiochemical species: Krypton, Xenon, Iodine and Cesium for the source term analysis, and Argon in the second case. Fission products source is computed from the inventory in the fuel and the temperature dependent release rate. Ar-41 source is computed from the (n,␥) in Ar-40 in the water inside the reactor core. The volumetric activity obtained is compared with readings from Ar-41 monitoring. Calibration of the noble gases monitor efficiency is also described. in terms of activity for components at different times from emergency shutdown) and cannot be applied to TRIGA fuel. Thus, the source for the four radio-elements included in CATHARE2 (Kr-87, Xe-133, I-131 and Cs-137) had to be calculated by other means and included in the defined flow of the SOURCE operator as an activity concentration per kg of gas or liquid. The same is true concerning Ar-41, only the way to calculate the source in the later case is different. Calculation of the radioactive sources is described in the next subchapters. 3. Calculation of the fission products source 2. TRIGA facility model The nodalization of the problem is depicted in Fig. 2. The reactor hall and reactor pool were modeled using Volume type module from CATHARE2, while the delay tank and pipes are modeled with Axial components (Mladin et al., 2013). The reactor hall has two boundary conditions for inlet and outlet of air, simulating the air circulation performed by the ventilation system. The model represents the primary circuit components in a simplified manner. The water volume is preserved but some components have been collapsed: two pumps at nominal power are represented by a single pump with mass flow rate of 500 l/s given by the combined action of the two primary pumps, the two heat exchangers are represented by only one, composed of inlet volume, individual thin tubes (1262 each of the two heat exchangers) and outlet volume. Since reactor core is not modeled, no heat transfer was considered and consequently there was no need for the secondary system. The first aim of the model was to calculate the fission products and Ar-41 transport in the primary system, reactor hall, and the evolution of their concentration in different zones and in the exhaust air flow rate. The fission products release at core damage is simulated by means of a radio-chemical components source (SOURCE operator) at the elevation of the core inside the volume representing the pool. CATHARE2 is an evolved instrument capable of calculating the source of radio-elements coming from clad rupture of the defined fuel. Unfortunately, the release from fuel model in CATHARE is dedicated exclusively to PWR (UO2 vertical fuel, inventory predefined An average TRIGA LEU bundle was modeled using SAS2H module from SCALE 4.4 (Hermann and Parks, 1998). One of the main utilizations of this module is to generate radiation and heat sources for depleted fuel. For each time dependent burnup composition, SAS2H does 1-D neutronic transport calculation using XSDRNPM (Greene and Petrie, 1998) for the assembly using a procedure with two distinct cell models. The first calculates the elementary cell of the fuel surrounded by water, obtaining the neutronic cross sections averaged on the cell neutron spectrum. The second model represents a larger cell from an infinite grid, model that can represent the TRIGA bundle. A schema of the SAS2H input model created for determining the radioactive inventory in TRIGA bundle is given in Fig. 3. Neutronic spectrum inside fuel from the extended cell model is used to determine the cross sections for nuclides in the specified burnup composition. Neutronic cross sections resulted from transport calculations at each burnup time step are used in point calculations with ORIGEN-S (Hermann and Westfall, 1998) which produces the burnup dependent composition for the next transport calculation. This sequence is repeated for the entire burnup history given by the user in the input deck. We have introduced a simple and conservative burnup history of the reactor: 14,500 MWD released without interruption (no ‘cooling’) by a 29 fuel bundle core (500 MWD per fuel bundle). The results of the SAS2H calculation are presented in Table 2, which gives the activity and mass of the four volatile fission products that are included in the CATHARE2 model. The last column in Table 3 gives the estimated releases, calculated as described below. 32 D. Mladin et al. / Nuclear Engineering and Design 262 (2013) 29–38 Table 2 Inventory per bundle and fission product sources for an underwater release. Whole core release in 100 s (100% for noble gases and 25% for I and Cs). Fission product Mass/bundle (kg) Release/core with 29 bundles (kg) Source in liquid phase (kg) Source in gas phase (kg) Kr (Kr-87) Xe (Xe-133) I (I-131) Cs (Cs-137) 4.45E−04 1.49E−04 1.31E−04 2.97E−02 7.36E−04 2.46E−04 5.41E−05 1.23E−02 0 0 4.82E−05 1.09E−02 7.36E−04 2.46E−04 5.90E−06 1.34E−03 Fig. 3. Regions of the TRIGA bundle model for SAS2H. Work described in (Baldwin et al., 1980), produced by General Atomic Company, gives the correlation used to calculate the release of fission products from TRIGA fuel, both gaseous and volatile metals: 4 = 1.5 × 10−5 + 3.6 × 103 · e−1.34×10 /T (1) where T is the fuel temperature (K), stating that, though originally developed for HEU fuel, it is applicable to LEU as well. For temperatures lower than 400 ◦ C releases are recoil-dominated and the first term of sum in (1) is controlling. For higher temperatures (>400 ◦ C), releases are mainly governed by a diffusion-like process and the second term becomes the dominant contributor. Release fractions given by (1) assume failed or ruptured cladding. The final Safety Analysis Report (General Atomic Company, 1974) indicates 940 ◦ C as the fuel limit temperature when cladding temperature may be at the same value as the fuel. The limit is based on calculations of the stress on clad incolloy material due to temperature dependent hydrogen dissociation from ZrH1.6 . For 940 ◦ C, using (1) we get a release fraction value of 5.7% which is the release fraction entering the results in Table 2 that describes the inventory in gap per TRIGA bundle (25 fuel pins) and release from the damaged core. The source of fission products assumes 100% of the TRIGA core fuel damage in case of Loss of Flow Accidents (LOFA) and 80% in case of Loss of Coolant Accident (LOCA). Using results from Mladin (2003), we estimated roughly the time for the release process as being 100 s. This was considered to be the time interval between hot pin failure and low power density pins failure in a LOFA without scram accident, taking into account that release from a broken clad is almost instantaneous for gases and volatile nuclides found in the fuel pellet – clad gap. In case of LOCA, the timing is different. According to the calculations with RELAP5 (The RELAP5 Development Team, 1995) (see Fig. 4), the fuel temperature in the maximum loaded pin (power of hot pin to power of average pin-ppf-equal to 1.92) reaches the safety limit taken as damage threshold (940 ◦ C) at about 6500 s from the reactor scram, while the average pin will be at the same temperature value at about 8000 s. The number of fuel elements which can fail in this case is smaller than in case with LOFA with no scram, because a smaller power loading than the average (ppf = 1) will not lead to a temperature excursion above 940 ◦ C. At this value of the damage threshold we estimate 80% of the core with failed cladding, because the outer fuel bundles will very likely preserve cladding integrity. In case of LOCA, the fission products source is defined in Table 2. Kr-87 is considered representative for all Krypton radioactive isotopes and its mass is the sum of the masses of Kr-83m, Kr85, Kr-85m, Kr-87, Kr-88 and Kr-89. Xe-133 represents all Xenon radioactive isotopes and is the sum of the masses of Xe-131m, Xe133, Xe-133m, Xe-135. In case of Iodine, represented by I-131, the list is: I-131, I-132, I-133, I-134, I-135. For Cesium (Cs-137): Cs134, Cs-135, Cs-137. We neglected from counting isotopes of the four elements which are very short lived (T1/2 < 1 min). Also, the lists do not mention isotopes accumulated in very small quantities. The calculations were done in the following assumptions: - one hundred percent of the noble gases in the fuel-clad gap are released; - twenty-five percent of the Iodine and Cesium are released from the fuel elements, the remainder being considered deposited on the relatively cool cladding. For an underwater accident only 10.9% of the release is considered gas (10% assumed to form organic compounds that escape pool water and 1% of the balance undissolved in the pool water) (General Atomic Company, Table 3 Inventory per bundle and fission product sources for release in air. Release from 80% of core in 1500 s (100% for noble gases and 25% for I and Cs). Fission product Mass/bundle (kg) Gas release from 80% of the core (kg) Kr (Kr-87) Xe (Xe-133) I (I-131) Cs (Cs-137) 4.45E−04 1.49E−04 1.31E−04 2.97E−02 5.89E−04 1.97E−04 4.33E−05 9.82E−03 Fig. 4. LOCA-evolution of maximum fuel temperature in maximum loaded element (ppf = 1.92) and average fuel element (ppf = 1) (Mladin, 2003). D. Mladin et al. / Nuclear Engineering and Design 262 (2013) 29–38 Fig. 5. Instantaneous and integral ( int) stack release of Krypton at LOFA with emergency ventilation (-Em) and normal ventilation (-Nor). 33 Fig. 6. Instantaneous and integral ( int) stack release of Xenon at LOFA with emergency ventilation (-Em) and normal ventilation (-Nor). 1974). Thus, only 2.725% from the I and Cs content of the fuel-clad gap forms the CATHARE gas source for these elements, the rest (22.275% of the gap content) being introduced as liquid source; - for fuel damage while in air, the release for noble gases (Xe and Kr) is 100% of the fuel-clad gap inventory, and for the I and Cs the release is considered 25%. 4. Calculation of the radio-elements transport Using the CATHARE2 model described above, we used the capabilities of the code to calculate the concentration of each fission product in different zones of the TRIGA facility, pool, primary lines, delay tank and reactor hall. There is a common VOLUME type element for the reactor hall and the reactor pool, the water level (10 m) in the lower part – which has different dimensions than the upper part – separating the two zones. It was necessary to fill the upper (gas) part with non-condensable (air), using the boundary conditions simulating the ventilation, as a preliminary state. After acquiring this, the source was opened and the concentrations (activities) were tracked inside the system and at the outlet air junction of the reactor hall. It should be mentioned that fission products are treated in CATHARE2 as pure species (i.e. no chemical interactions) and are introduced as gaseous or liquid sources at the location of reactor core in the model. Fig. 7. Instantaneous and integral ( int) stack release of Iodine at LOFA for normal ventilation (-Nor) (100% efficiency of the filters in emergency ventilation case). present the instantaneous and integral releases from the reactor stack to the environment for an underwater release corresponding to a LOFA with emergency ventilation and also with normal operation ventilation (Fig. 8). 4.1. Underwater release Two basic series of results are presented, with a residual flow rate in the primary circuit (50 l/s): • normal operation of the ventilation system: air flow rate is 24,360 m3 /h, • ventilation system in emergency mode: air flow rate is 18,500 m3 /h, efficiency of the filters is 100% for Iodine and 10% for Cesium. The residual flow rate (50 l/s) has a meaning for water mixing when forced circulation given by the primary pumps suddenly becomes unavailable, for instance during a (LOFA) or Loss of Power Supply (LOPS) initiating events. Its value is generic (10% of the normal flow, constant during the transient) although it is expected that the main pumps flow will vanish rapidly after LOFA, with only the emergency pump possibly still operating.In all cases, the duration of the transient after fission products source opening was 10,000 s. Figs. 4–7 Fig. 8. Instantaneous and integral ( int) stack release of Cesium at LOFA for emergency ventilation (-Em) and normal ventilation (-Nor) (10% efficiency of the filters in emergency ventilation case). 34 D. Mladin et al. / Nuclear Engineering and Design 262 (2013) 29–38 Fig. 9. Activity of Iodine in water at LOFA – pool and delay tank. Fig. 11. Instantaneous and integral ( int) Krypton stack release at LOCA with emergency ventilation (-Em) and normal ventilation (-Nor). The activity of I-131 and Cs-137 in the primary circuit water is presented in Figs. 9 and 10 respectively. It is due to the liquid phase source introduced for these two fission products. The fission products source accumulates rapidly inside the pool and the activity reaches an equilibrium with the activity of water in the other primary circuit components. The delay tank is taken here for illustration, and the activity equalizes, for the flow rate of 50 l/s, in about 6000 s. 4.2. Release in air Figs. 11–14 present the evolutions of the four fission products in case of Loss of Coolant Accident (LOCA). The maximum of instantaneous releases for all fission products is at 1500 s (end of release), the origin being the moment of fission products source opening which coincides with the damage for the maximum loaded fuel group (ppf = 1.92). The circuit here is empty (residual liquid phase), the fission products source is placed at the same elevation as before (1 m above the bottom of the pool) but the release is gaseous, inside the non-condensable (air). The released masses of the four fission products tracked are given in Table 3. The underwater emission of a gas source in CATHARE leads to gas being trapped at a certain extent in the residual gas phase of the pool (typically, in CATHARE2 for strongly subcooled liquid, the void fraction is 1.E−5). In case of release in air, the source will go entirely into the atmosphere of the reactor room. This is the explanation for larger releases in the environment in Figs. 11 and 12 for Krypton and Xenon compared Fig. 10. Activity of Cesium in water at LOFA – pool and delay tank. Fig. 12. Instantaneous and integral ( int) Xenon stack release at LOCA with emergency ventilation (-Em) and normal ventilation (-Nor). to Figs. 5 and 6 respectively, even if the core damage is only 80% in the later case compared to 100% in a LOFA without scram. In case of Iodine and Cesium the airborne source inside the reactor hall at LOFA is smaller than at LOCA, roughly by a factor of seven (2.725% compared to 0.8 × 25%). No release outside the reactor building is given in emergency ventilation cases for Iodine since the efficiency of the filters is considered 100%. For Cesium, the efficiency of the filters is 10%. Fig. 13. Instantaneous and integral ( int) stack Iodine release at LOCA for normal ventilation (-Nor) (100% efficiency of the filters in emergency ventilation case). D. Mladin et al. / Nuclear Engineering and Design 262 (2013) 29–38 35 Table 5 Calculated Ar-41 production in TRIGA fuel at 14 MW. Fig. 14. Instantaneous and integral ( int) Cesium stack release at LOFA for emergency ventilation (-Em) and normal ventilation (-Nor) (10% efficiency of the filters in emergency ventilation case). 5. Calculation of the Ar-41 source As the coolant water passes through the reactor core, Ar-41 will be produced through the reaction: 40 Ar(n,␥)41 Ar where the Ar-40 is dissolved in the water. If the water is saturated with Argon, the Ar40 concentration, S, as a function of temperature is given in Table 4. In MCNP (X-5 Monte Carlo Team, 2003), the Ar-41 concentration at 20 ◦ C was introduced in the water surrounding TRIGA fuel elements, over the active length of the fuel (55.88 cm), for all 29 fuel bundles in the core configuration presented in Fig. 15. The water inside the experimental locations was not considered because at the time when dosimetry records of Ar-41 activity during reactor operation were made, the locations were filled with experiments. As an MCNP5 tally, we requested the (n,␥) reaction rate on Ar40 in the cells that contain the water inside shroud for each of the 29 fuel bundles. The calculated Ar-41 atom density at maximum reactor power level (14 MW) in these cells is given in Table 5 together with the volume of water, because a few fuel elements are missing in some of the fuel bundles. The resulting Ar-41 source is 6.54 GBq/core, in the liquid phase. 6. Ar-41 transport The Ar-41 gas source is produced by outgasing of Ar-41 due to heatup of water when it passes through the reactor core. The designed inlet-outlet temperature difference is 7 ◦ C for TRIGA 14 MW steady state reactor. Interpolating in Table 6 it yields a gas fraction resulting from a factor of 0.11 applied to the liquid phase Ar-41 concentration inside the reactor pool. This is realized in CATHARE2 model by defining two sources of Ar-41 at the height where the core is placed. The first, in liquid phase, comprising initially all the Ar-41 produced by the (n,␥) reaction in Ar-40. At each time step, concentration of Ar-41 in the pool water Table 4 Saturated Ar-40 concentration in water (General Atomic Company, 1974). 3 Temperature ( C) S (atoms/cm of H2 O) 10 20 30 40 50 60 70 80 1.14E+16 0.94E+16 0.79E+16 0.69E+16 0.62E+16 0.56E+16 0.52E+16 0.48E+16 ◦ Fuel bundle Volume of water (cm3 ) Ar-41 density (at/cm3 of H2 O) L-38 L-42 L-61 L-09 L-44 L-49 L-24 L-05 L-46 L-32 L-08 L-02 L-45 L-10 L-35 L-47 L-40 L-39 F-56 F-55 F-52 F-51 F-59 F-50 F-53 F-57 F-58 F-60 F-54 1.90E+03 1.90E+03 1.90E+03 1.90E+03 1.90E+03 1.98E+03 1.90E+03 1.90E+03 1.90E+03 1.90E+03 1.90E+03 1.98E+03 1.90E+03 1.90E+03 1.90E+03 1.98E+03 1.90E+03 1.90E+03 1.90E+03 1.90E+03 1.90E+03 1.90E+03 1.90E+03 1.90E+03 1.90E+03 1.98E+03 1.90E+03 1.90E+03 1.90E+03 1.18E+05 1.21E+05 1.23E+05 1.33E+05 1.23E+05 1.58E+05 1.44E+05 1.37E+05 1.53E+05 1.45E+05 1.22E+05 1.82E+05 1.34E+05 1.16E+05 9.57E+04 1.22E+05 1.01E+05 9.11E+04 6.52E+04 8.68E+04 9.09E+04 1.01E+05 9.04E+04 9.21E+04 7.88E+04 1.07E+05 6.49E+04 5.92E+04 5.32E+04 is retrieved and this source is decreased while the second source, gaseous, increases, simulating the outgasing of Ar-41 when water passes through the core. After some time, the sources reach the equilibrium, and the resulted Ar-41 activity at the air outlet boundary condition (reactor stack) is taken as the steady state value at the power level corresponding to the calculated Ar-41 source (see Fig. 16). The equilibrium activity is about 3.7E+4 Bq/m3 at 14 MW. This value will be compared with that resulted from Ar-41 monitoring. Two effects, not taken into consideration, should be mentioned. The first is the underestimation of the real Ar-41 concentration due to supplementary Ar-41 produced outside fuel bundles or in the top or bottom reflector of the fuel bundle. On the other hand, there might be a negative gaseous source of Ar-41 because of the possible redissolution of Ar-41 gas when primary circuit water is cooled into the heat exchangers. Neglecting such an effect will tend to overestimate the Ar-41 activity in the reactor hall and at exhaust stack. 7. Ar-41 monitoring At Romanian TRIGA, Argon radioactivity is monitored by means of a detection and measurement system for radioactive aerosols: the Gaseous Effluents Monitoring System (GEMS), simultaneously with other noble gases, radioactive Iodine and aerosols. The Operational Limits (OL) for gaseous effluents monitoring are calculated Table 6 Derived Emission Limits (DEL) for gaseous effluents. Radionuclide DEL (Bq/year) N-16 O-19 Ar-41 I-131 Sr-90 Ru-106 Ru-103 2.33E+12 5.83E+11 1.17E+14 1.75E+07 9.33E+07 7.00E+07 4.67E+07 36 D. Mladin et al. / Nuclear Engineering and Design 262 (2013) 29–38 Fig. 15. TRIGA core configuration for Ar-41 source calculation. starting from the Derived Emission Limits (DEL) (see Table 6) approved by the National Commission for Nuclear Activities Control (CNCAN) and presented in Table 7. 7.1. Calibration of the noble gases monitor efficiency In order to calculate the total efficiency of the measurement chain, the photopeak efficiency for the given measurement geometry was calculated with the ISOCS gamma spectrometry calibration code from Canberra Industries (see Fig. 17). Photopeak efficiency values were corrected with the peak-to-total ratio (P/T) specific to a NaI detector for the energy range of interest. The energy dependence of the P/T was fitted with a 3rd order dual logarithmic polynomial and is presented in Fig. 18. By applying the above algorithm, the values for total efficiency of the noble gases monitor were calculated and are presented in Table 8. Thus, the net count rate corresponding to the maximum released concentration limit of 1.483E+6 Bq/m3 is 556 cps. For determining the Ar-41 activity released at stack, the background is subtracted (approximately 21.5 cps) and the result is multiplied with 2222 which represents the inverse of the product of total Table 7 Operational limits. Radionuclid Operational evacuation limit (Bq/day) Operational concentration limit (Bq/mc) Ar-41 I-131 Ru-106 Ru-103 3.20E+11 4.79E+04 1.92E+05 1.28E+05 1,483,000 0.2 0.9 0.6 Fig. 16. Evolution of the calculated volumetric activity of Ar-41 in the gas evacuated from reactor stack. D. Mladin et al. / Nuclear Engineering and Design 262 (2013) 29–38 37 Table 8 Total efficiency of the noble gases monitor at different discrete energies. Energy (keV) εtot MGN 59.54 80.99 100 200 302.84 356 383.8 400 661.65 800 1000 1173.22 1332.49 1460.8 1500 0.020 0.021 0.022 0.022 0.020 0.019 0.019 0.019 0.016 0.015 0.015 0.015 0.015 0.015 0.015 Fig. 17. Photopeak efficiency for the noble gases monitor measurement chain. responding to an activity of 40.3 kBq/m3 was reached when the reactor power was maintained at 13 MW. 8. Conclusion Fig. 18. Energy dependence of the P/T for a NaI 3 × 3 detector. Ar-41 efficiency (0.015) and the volume of the enclosure (0.03 m3 ). In addition, the Ar-41 activity has to be multiplied by 1.25 for taking into account that only 80% of the exhaust flow rate is from the reactor hall. The recordings for Ar-41 activity are presented in Fig. 19 together with the reactor power level for a period of 42 days in 2010. A plateau mean value for Ar-41 activity of about 36 cps cor- The paper presents a model for TRIGA reactor and investigation analyses of radio-chemical components behavior in the TRIGA facility with CATHARE2. For fission products analysis, it focuses on Kr-87, Xe-133, I-131 and Cs-137 considered as representative for the radioactive Kr, Xe, I and Cs, respectively. In this respect, the present work constitutes the support analysis for consequence analysis in the PSA project for Romanian TRIGA Reactor. The internal Initiating Events that can lead to fuel damage and fission products release in PSA Level 1 for TRIGA reactor are Loss of Flow, Loss of Power Supply or Loss of Coolant Accident. Two types of fission product sources are created: for underwater release (LOFA, LOPS without scram) and for release in air (LOCA) The analyses with CATHARE2 study the transport of radio-chemical components inside the primary circuit and reactor hall, and track their time evolution of concentration affected by disintegration, water entrainment and ventilation flowrate. CATHARE2, and consequently our calculations, do not take into account the chemical interactions of the released species, assuming the inventory existent inside the fuel gap at the moment of 60 14 13 12 50 11 10 9 8 7 30 6 5 20 4 3 10 2 1 0 09/16/2010 09/17/2010 09/18/2010 09/19/2010 09/20/2010 09/21/2010 09/22/2010 09/23/2010 09/24/2010 09/25/2010 09/26/2010 09/27/2010 09/28/2010 09/29/2010 09/30/2010 10/01/2010 10/02/2010 10/03/2010 10/04/2010 10/05/2010 10/06/2010 10/07/2010 10/08/2010 10/09/2010 10/10/2010 10/11/2010 10/12/2010 10/13/2010 10/14/2010 10/15/2010 10/16/2010 10/17/2010 10/18/2010 10/19/2010 10/20/2010 10/21/2010 10/22/2010 10/23/2010 10/24/2010 10/25/2010 10/26/2010 10/27/2010 10/28/2010 0 Fig. 19. Ar-41 activity recordings and reactor power level over 42 days in September–October 2010. PTER [MW] MGN02-T [CPS] 40 38 D. Mladin et al. / Nuclear Engineering and Design 262 (2013) 29–38 cladding rupture will be released only with respect to temperature. Also, aerosols deposition on surfaces inside the reactor hall is not considered. The same simplified TRIGA model is used to calculate the Ar-41 activity at stack in normal operation and compare with the measurement system recordings. The Ar-41 source is calculated using MCNP. Calculated and measured Ar-41 activity at reactor exhaust stack are in good agreement. References Baldwin, N.L., Foushee, F.C., Greenwood, J.S., 1980. Fission Product Release from TRIGA-LEU Reactor Fuels, GA-A16287. General Atomic Company, 1974. Safety Analysis Report of the Triga Steady-State Research/Materials And Testing Reactor, GA E-117-323. Greene, N.M., Petrie, L.M., 1998. XSDRNPM: A One-Dimensional Discrete-Ordinates. Code For Transport Analysis, NUREG/CR-0200, Rev. 6, Vol. 2, Section F3, ORNL/NUREG/CSD-2/V2/R6, September. Hermann, O.W., Parks, C.V., 1998. SAS2H: A Coupled One-Dimensional Depletion and Shielding Analysis Module, NUREG/CR-0200, Rev. 6, Vol. 1, Section S2. ORNL/NUREG/CSD-2/V2/R6. Hermann, O.W., Westfall, R.M., 1998. ORIGEN-S: SCALE System Module To Calculate Fuel Depletion, Actinide Transmutation, Fission Product Buildup And Decay, And Associated Radiation Source Terms, NUREG/CR-0200, Rev. 6, Vol. 2, Section F7, ORNL/NUREG/CSD-2/V2/R6, September. Lavialle, G., 2005. CATHARE V25 1 Users Manual, SSTH/LDAS/EM/2005-035. Departament D’Etudes des Reacteurs, Service de Simulation en Thermohydraulique, CEA Grenoble. Mladin, M., 2003. Analiza accidentelor de tip LOFA si LOCA la reactorul TRIGASSR, Institute for Nuclear Research, Internal Report no. 13.06.03 (limited access). Mladin, D., et al., 2013. Calculation of fission products transport in a pool type reactor using CATHARE2. U.P.B. Sci. Bull, Ser. C 75 (1), ISSN 1454-234x. The RELAP5 Development Team, 1995. RELAP5/MOD3 Code Manual Volume I: Code Structure, System Models, And Solution Methods, vol. II, NUREG/CR-5535 INEL95/0174, Volume I, June. 5-X Monte Carlo Team, 2003. MCNP – A General Monte Carlo N-Particle Transport Code, Los Alamos National Laboratory, LA-UR-03-1987, Version 5, April 24, vol. I and II (Revised 10/3/05, 2/1/2008).