- Email: [email protected]
Tritium distribution and chemical forms in the irradiated beryllium pebbles before and after thermoannealing
Fusion Engineering and Design 86 (2011) 2125–2128
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Tritium distribution and chemical forms in the irradiated beryllium pebbles before and after thermoannealing E. Pajuste ∗ , A. Vitins, G. Kizane, V. Zubkovs, P. Birjukovs Institute of Chemical Physics, University of Latvia, Kronvalda Blvd. 4, LV-1586 Riga, Latvia
a r t i c l e
i n f o
Article history: Available online 16 March 2011 Keywords: Beryllium Tritium inventory Chemical forms Structure
a b s t r a c t Beryllium pebbles are foreseen as a neutron multiplier in the tritium breeding blanket of the future fusion devices. Tritium inventory in the beryllium as a result of neutron-induced transmutations is a significant safety and technological issue for the operation of the breeding blanket. In this study, beryllium pebbles from 3 different irradiation experiments: BERYLLIUM, EXOTIC 8/3-13 and PBA, performed at High Flux Reactor HFR have been investigated. The distribution of tritium in the bulk of the pebbles and the abundance ratios of chemical forms of tritium T0 , T+ and T2 have been analysed before and after the different thermo-annealing experiments. In order to determine the abundance ratios of chemical forms, the method of chemical scavengers has been used. The structure analysis has been done by scanning electron microscopy. The main chemical form of the tritium localized in the irradiated beryllium pebbles is T2 , especially in the pebbles from PBA experiment. As a result of thermo-annealing, the abundances of the T2 and T0 change, T+ stays unaffected up to relatively high temperatures. The distribution of the tritium in the bulk is uneven – it increases rapidly at the centre of the pebble. Changes of pebble structure as a result of thermal treatment have been observed. © 2011 Elsevier B.V. All rights reserved.
1. Introduction
2. Experimental
Beryllium pebbles are foreseen as a neutron multiplier to the reference concept of the helium-cooled pebble-bed breeding blanket (HCPB) in the European Breeding Blanket Programme for the DEMO design [1]. Tritium inventory in the beryllium as a result of neutroninduced transmutations is a significant safety and technological issue for the operation of the breeding blanket. Tritium release rate from the irradiated beryllium at the different temperature regimes has been investigated previously [2,3]. Tritium release is governed by several mechanisms: diffusion, surface desorption and trapping. In order to describe and predict tritium behaviour during the thermo-annealing, it is necessary to estimate factors that may affect these mechanisms. In this study, neutron irradiated beryllium pebbles have been investigated. The distribution of tritium in the bulk of the pebbles and the abundance ratios of chemical forms of tritium T0 , T+ and T2 have been investigated before and after the different thermoannealing experiments. The structural changes of the samples, e.g. porosity, are also included in this study.
2.1. Samples
∗ Corresponding author. Tel.: +371 67033883. E-mail address: [email protected] (E. Pajuste). 0920-3796/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.fusengdes.2011.02.058
Three types of beryllium samples that have been irradiated in the experiments BERYLLIUM, EXOTIC 8/3-13 (Extraction of Tritium in Ceramics) and PBA (Pebble Bed Assembly) have been studied. All these irradiation programs have been performed in the High Flux Reactor (HFR) in Petten, the Netherlands. Description of the samples and irradiation conditions are summarized in Table 1.
2.2. Chemical scavenger and dissolution method In order to determine total tritium activity, bulk distribution and abundance ratios of chemical forms (T2 , T0 , T+ ) in irradiated beryllium pebbles, chemical scavenger and dissolution method has been used. Pebbles were dissolved in the solutions of 2 mol L−1 H2 SO4 and 2 mol L−1 H2 SO4 + 0.5 mol L−1 Na2 Cr2 07 in a special setup [2]. One hydrogen molecule evolved during beryllium dissolution corresponds to one beryllium atom. The dissolution rate of beryllium and hereby the thickness of dissolved layer can be calculated from the hydrogen measurements. The rate of hydrogen evolution was measured with a catarometer. In the solution of 2 mol L−1 H2 SO4 molecular and atomic tritium (T2 and T0 ) of the activities AT2 and AT0 respectively, present in a
2126
E. Pajuste et al. / Fusion Engineering and Design 86 (2011) 2125–2128
Table 1 Description of the samples.
Manufacturer Production process Diameter, mm Main impurities Irradiation time Neutron fluence (E > 0.1 MeV) Irradiation temp. 4 He content Year of end of irradiation
PBA
EXOTIC 8/3-13
BERYLLIUM
NGK Insulators Ltd., Handa City, Japan Rotating electrode process (REP) 0.9–1.1 2300 ppm BeO, 300 ppm Mg 294 days (3–4) × 1025 m−2 420–820 K 300–600 appm 2004
Brush Wellman Inc., Cleveland, OH, USA Inert gas atomization (IGA) 0.1–0.2 3400 ppm BeO, 100 ppm Mg 449.8 days 2.70 × 1025 m−2 800–900 K 285 appm 2000
Brush Wellman Inc., Cleveland, OH, USA Fluoride reduction process (FRP) Approximately 2 3125 ppm BeO, 1200 ppm Mg 97.4 days 1.24 × 1025 m−2 780 K 480 appm 1994
Be sample transfer as T2 + HT into a flow of carrier gas, where the tritium activity released (1) was measured with tritium monitor TEM 2100A with a proportional gas flow-through detector DDH 32. AT,
gas, acid
= AT2 + AT0
(1)
T+ localized in a Be layer remains in the solution and the tritium activity AT sol. acid was measured with liquid scintillation method. 90% of H0 (T0 ) reacts with the scavenger Na2 Cr2 O7 in the solution of 2 mol L−1 H2 SO4 and remains in the solution. Activity of the tritium released into a gas phase and retained in the solution are the respective sums: AT,
gas, Cr(VI)
AT,
sol., Cr(VI)
= AT2 + x · AT0
(2)
= AT+ + (1 − x) · AT0
(3)
where value of x was found to be 0.1 (10%). The contents of T0 , T2 , T+ (Bq g−1 ) in a sample were determined separately from the corresponding differences in the activities by following equations: AT0 =
AT, gas, acid − AT, gas, Cr(VI) 1−x
(4)
AT2 = AT, gas, acid − AT0
(5)
AT+ = AT,
(6)
sol., acid
2.3. Structure analysis The microstructure of the samples was evaluated by means of scanning electron microscopy (Hitachi S-4800, 2 kV, 15 A). Samples were prepared by polishing first with SiC paper and then with diamond paste of 1 m particle size. 2.4. Thermo-annealing To describe and predict the diffusion process of tritium during thermo-annealing the estimation also of the resulting abundance ratio of tritium chemical forms is needed. Pebbles were heated in the special setup with and without action of fast electron irradiation (14 MGy h−1 , 5 MeV) and high magnetic field (1.7 T). Experimental setup, conditions and tritium thermo-desorption curves have been described previously [3]. 3. Results and discussion 3.1. Structure Structure analysis has been performed for the samples irradiated in the PBA experiment. Pebbles from BERYLLIUM experiment have been described elsewhere by other authors [4], whereas pebbles from experiment EXOTC 8/3-13 have not sufficient diameter (0.1–0.2 mm) for the polishing procedure.
Fig. 1. SEM image of polished cross-section of PBA pebbles: (a) not treated and (b) after treatment in high temperature (1260 ◦ C).
E. Pajuste et al. / Fusion Engineering and Design 86 (2011) 2125–2128
2127
Fig. 2. Abundance ratio of tritium chemical forms in the beryllium pebbles irradiated in experiments BERYLLIUM, EXOTIC 8/3-13 and PBA.
Large voids became visible in the bulk of most of the examined PBA pebbles after polishing (Fig. 1). These voids are generated during the cooling phase of the fabrication process and according to the production method patent have been developed to prevent beryllium swelling and failure when irradiated with neutrons (helium produced in the nuclear reactions is stored in this void instead of beryllium lattice) [5]. Nevertheless, existence of such technical void increases tritium retention since it may act as a pressurized sink for tritium trapping [6]. Cross-section diameter of these voids ranges from 100 to 250 m that corresponds to approximately 0.5% of the pebble volume. Porosity of the PBA pebbles before thermal treatment was not observed since resolution of the method for these particular samples allows seeing pores larger than 500 nm, but according to literature [7,8] at the temperatures below ∼400 ◦ C size of helium inclusions are less than several tens of nm. Thermo-annealing experiment at the temperature of 720 ◦ C has been performed and at this temperature visible cracks are formed and the porosity starts to appear. After heating of the sample at high temperature (up to 1260 ◦ C) structure changes drastically – large radial cracks and high porosity appear (Fig. 1(b)). This is a result both of thermal stresses (cracks) and the coalescence of small helium inclusions into the larger ones. 3.2. Tritium chemical forms and distribution in pebbles irradiated in different experiments Tritium produced in the neutron induced transmutation of beryllium can diffuse into the lattice or can be trapped by structure traps (such as intragranular He bubbles, closed porosity and grain boundaries) or it may react with BeO to form Be(OT)2 with the formation energy of −0.7 eV (at the standard temperature, pressure) [9]. The abundance ratio of tritium chemical forms in the beryllium pebbles was measured for samples from all three irradiation experiments (Fig. 2). Large fraction (up to 96%) of the tritium accumulated in irradiated beryllium pebbles was found to be in the molecular form. We assume that it coexists with the helium in the gas inclusions. Large fraction of gaseous species might be trapped also in the technical void in the bulk of the pebbles that might explain the highest T2 content in the pebbles from the PBA experiment (fabricated by the REP method). Presence of molecular tritium in the pebbles is undesirable. In order to extract T2 from the metallic matrix it is required T2 to dissociate into T0 (dissociation energy for hydrogen −4.52 eV). In the investigated samples fraction of atomic tritium was in range from 1 to 32%. Atomic tritium can either exist as interstitial or can be trapped in the vacancy-based defect of the beryllium [10]. Atomic tritium can diffuse without any transformation – energy is required only for passing through the diffusion energy barrier. The role of molecular and atomic tritium ratio is obvious if one
Fig. 3. Distribution of the chemical forms of tritium in beryllium pebbles after thermo-annealing experiments, where R – radiation of accelerated electrons 5 MeV, 14 MGy h−1 , MF – magnetic field 1.7 T, and T – temperature: (a) 280 ◦ C and (b) 500 ◦ C.
compares the tritium release from EXOTIC 8/3-13 and PBA pebbles. From the EXOTIC 8/3-13 pebbles tritium release is starting at much lower temperatures and full detritiation is reached much faster (the diameter difference definitely also has an impact). Chemically bonded tritium T+ was found to be in range from 3 to 12%. Highest concentration of T+ was in the pebbles from the experiment EXOTIC 8/3-13. According to the chemical analysis these pebbles have the highest beryllium oxide content. Oxide could be formed as a result of the chemical interaction with lithium ceramics during the irradiation (signs of Si were found on the surface of these pebbles). Due to large specific surface oxide content could be increasing also during the storage in air. Tritium retained in the form of beryllium hydroxide stays immobile unless temperature required for dehydration of Be(OT)2 is reached (complete dehydration is reached at temperatures above 950 ◦ C) [11]. Bulk distribution of tritium in the pebbles has been determined for the pebbles of BERYLLIUM and PBA experiments. Distribution was calculated by measuring tritium activity in the gas phase released during the dissolution, therefore distribution of chemically bonded tritium T+ was not determined (more likely most of it is in the surface oxide layer and in the oxide inclusions in the bulk). For both types of samples distribution of tritium was similar–low concentration in the first micrometers from the surface, more or less uniform distribution in the bulk and in some pebbles a sharp peak somewhere in the bulk. The peak might indicate the position of the large void that was found during the structure investigations. 3.3. Evolution of tritium chemical forms at the thermo-annealing Changes of the abundance ratio of tritium chemical forms during the thermal treatment of sample at different conditions, such as fast electron irradiation and high magnetic field have been done for the samples from EXOTIC 8/3-13 experiment. Treatment has
2128
E. Pajuste et al. / Fusion Engineering and Design 86 (2011) 2125–2128
been performed at temperatures 280 ◦ C for 3 h and 500 ◦ C for 1 h (Fig. 3). The results show that the amount of chemically bonded tritium changes insignificantly after the thermal treatment of pebbles. At low temperature (280 ◦ C) only content of atomic tritium has decreased, other forms stayed immobile. Heating at higher temperature decreases the T2 content because of the molecule dissociation into the mobile T0 and also the formation of cracks and open porosity where molecular tritium may migrate. Dissociation of molecular tritium is also the reason the atomic form content virtually stays at the same level. Under action of fast electrons the radiolysis of T2 takes place and therefore the increase of atomic tritium content. The highest decrease of molecular tritium and tritium total content in samples is observed under the simultaneous action of temperature, irradiation and magnetic field. This phenomenon could be explained by the spin-spin transformations (radiolysis in high magnetic field). In the dissociation (radiolysis) process correlated radical pair is formed with anti-parallel electron spins, whereas magnetic field changes them to parallel and probability of recombination of the radicals decreases [12]. Therefore more tritium is in a mobile T0 form. 4. Conclusions Structure analysis revealed large technical voids in the bulk of the beryllium pebbles produced by REP method. Existence of the void may play crucial role in the tritium retention. The main chemical form of the tritium localized in the irradiated beryllium pebbles is T2 . High content of this immobile form might significantly extend the time and therefore increase the cost of detritiation method based on thermo-annealing process. Determination of the chemical forms of tritium before and after thermal treatment in different condition gives the possibility to estimate the processes occurring during the thermoannealing.
Acknowledgments This work was performed within the EFDA Goal Oriented Training Programme “Breeding Blanket Development for Fusion Reactors – EUROBREED”. The support of L’Oréal Latvian Fellowship for Women in Science 2010 with the support of the Latvian National Commission for UNESCO and the Latvian Academy of Sciences is also gratefully acknowledged. References [1] M. Gasparotto, L.V. Boccaccini, L. Giancarli, S. Malang, Y. Poitevin, Demo blanket technology R&D results in EU, Fusion Engineering and Design 61–62 (2002) 263–271. [2] J. Tiliks, G. Kizane, A. Vitins, E. Kolodinska, E. Rabaglino, Magnetic field effects on tritium release from neutron-irradiated beryllium pebbles, Nuclear Technology 159 (2007) 245–249. [3] J. Tiliks, A. Vitins, G. Kizane, V. Tilika, E. Kolodinska, S. Kaleja, B. Lescinskis, Effects of external energetic factors on tritium release from the EXOTIC 83/13 neutron-irradiated beryllium pebbles, Fusion Engineering and Design 84 (2009) 1842–1846. [4] A. Möslang, R.A. Pieritz, E. Boller, C. Ferrero, Gas bubble network formation in irradiated beryllium pebbles monitored by X-ray microtomography, Journal of Nuclear Materials 386–388 (2009) 1052–1055. [5] E. Ishitsuka, H. Kawamura, N. Sakamoto, K. Nishida, Process for preparing metallic beryllium pebbles, in: U.S. Patent (Ed.), Japan Atomic Energy Research Institute (JP) NGK Insulators Ltd. (JP) United States, 1999, p. 10. [6] F. Scaffidi-Argentina, M. Dalle Donne, H. Werle, Critical assessment of beryllium pebbles response under neutron irradiation: mechanical performance and tritium release, Journal of Nuclear Materials 258–263 (1998) 595–600. [7] L.L. Snead, Low-temperature low-dose neutron irradiation effects on beryllium, Journal of Nuclear Materials 326 (2004) 114–124. [8] I.B. Kupriyanov, R.R. Melder, V.A. Gorokhov, The effect of neutron irradiation on beryllium performance, Fusion Engineering and Design 51–52 (2000) 135–143. [9] F. Scaffidi-Argentina, Tritium and helium release from neutron irradiated beryllium pebbles from the EXOTIC-8 irradiation, Fusion Engineering and Design 58–59 (2001) 641–645. [10] M.G. Ganchenkova, P.V. Vladimirov, V.A. Borodin, Vacancies, interstitials and gas atoms in beryllium, Journal of Nuclear Materials 386–388 (2009) 79–81. [11] K.A. Walsh, Beryllium Chemistry and Processing, ASM International, Ohio, 2009. [12] H. Hayashi, Introduction to Dynamic Spin Chemistry: Magnetic Field Effects upon Chemical and Biochemical Reactions (World Scientific Lecture and Course Notes in Chemistry, vol. 8), World Scientific Publishing Company, 2004.
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Tritium distribution and chemical forms in the irradiated beryllium pebbles before and after thermoannealing E. Pajuste ∗ , A. Vitins, G. Kizane, V. Zubkovs, P. Birjukovs Institute of Chemical Physics, University of Latvia, Kronvalda Blvd. 4, LV-1586 Riga, Latvia
a r t i c l e
i n f o
Article history: Available online 16 March 2011 Keywords: Beryllium Tritium inventory Chemical forms Structure
a b s t r a c t Beryllium pebbles are foreseen as a neutron multiplier in the tritium breeding blanket of the future fusion devices. Tritium inventory in the beryllium as a result of neutron-induced transmutations is a significant safety and technological issue for the operation of the breeding blanket. In this study, beryllium pebbles from 3 different irradiation experiments: BERYLLIUM, EXOTIC 8/3-13 and PBA, performed at High Flux Reactor HFR have been investigated. The distribution of tritium in the bulk of the pebbles and the abundance ratios of chemical forms of tritium T0 , T+ and T2 have been analysed before and after the different thermo-annealing experiments. In order to determine the abundance ratios of chemical forms, the method of chemical scavengers has been used. The structure analysis has been done by scanning electron microscopy. The main chemical form of the tritium localized in the irradiated beryllium pebbles is T2 , especially in the pebbles from PBA experiment. As a result of thermo-annealing, the abundances of the T2 and T0 change, T+ stays unaffected up to relatively high temperatures. The distribution of the tritium in the bulk is uneven – it increases rapidly at the centre of the pebble. Changes of pebble structure as a result of thermal treatment have been observed. © 2011 Elsevier B.V. All rights reserved.
1. Introduction
2. Experimental
Beryllium pebbles are foreseen as a neutron multiplier to the reference concept of the helium-cooled pebble-bed breeding blanket (HCPB) in the European Breeding Blanket Programme for the DEMO design [1]. Tritium inventory in the beryllium as a result of neutroninduced transmutations is a significant safety and technological issue for the operation of the breeding blanket. Tritium release rate from the irradiated beryllium at the different temperature regimes has been investigated previously [2,3]. Tritium release is governed by several mechanisms: diffusion, surface desorption and trapping. In order to describe and predict tritium behaviour during the thermo-annealing, it is necessary to estimate factors that may affect these mechanisms. In this study, neutron irradiated beryllium pebbles have been investigated. The distribution of tritium in the bulk of the pebbles and the abundance ratios of chemical forms of tritium T0 , T+ and T2 have been investigated before and after the different thermoannealing experiments. The structural changes of the samples, e.g. porosity, are also included in this study.
2.1. Samples
∗ Corresponding author. Tel.: +371 67033883. E-mail address: [email protected] (E. Pajuste). 0920-3796/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.fusengdes.2011.02.058
Three types of beryllium samples that have been irradiated in the experiments BERYLLIUM, EXOTIC 8/3-13 (Extraction of Tritium in Ceramics) and PBA (Pebble Bed Assembly) have been studied. All these irradiation programs have been performed in the High Flux Reactor (HFR) in Petten, the Netherlands. Description of the samples and irradiation conditions are summarized in Table 1.
2.2. Chemical scavenger and dissolution method In order to determine total tritium activity, bulk distribution and abundance ratios of chemical forms (T2 , T0 , T+ ) in irradiated beryllium pebbles, chemical scavenger and dissolution method has been used. Pebbles were dissolved in the solutions of 2 mol L−1 H2 SO4 and 2 mol L−1 H2 SO4 + 0.5 mol L−1 Na2 Cr2 07 in a special setup [2]. One hydrogen molecule evolved during beryllium dissolution corresponds to one beryllium atom. The dissolution rate of beryllium and hereby the thickness of dissolved layer can be calculated from the hydrogen measurements. The rate of hydrogen evolution was measured with a catarometer. In the solution of 2 mol L−1 H2 SO4 molecular and atomic tritium (T2 and T0 ) of the activities AT2 and AT0 respectively, present in a
2126
E. Pajuste et al. / Fusion Engineering and Design 86 (2011) 2125–2128
Table 1 Description of the samples.
Manufacturer Production process Diameter, mm Main impurities Irradiation time Neutron fluence (E > 0.1 MeV) Irradiation temp. 4 He content Year of end of irradiation
PBA
EXOTIC 8/3-13
BERYLLIUM
NGK Insulators Ltd., Handa City, Japan Rotating electrode process (REP) 0.9–1.1 2300 ppm BeO, 300 ppm Mg 294 days (3–4) × 1025 m−2 420–820 K 300–600 appm 2004
Brush Wellman Inc., Cleveland, OH, USA Inert gas atomization (IGA) 0.1–0.2 3400 ppm BeO, 100 ppm Mg 449.8 days 2.70 × 1025 m−2 800–900 K 285 appm 2000
Brush Wellman Inc., Cleveland, OH, USA Fluoride reduction process (FRP) Approximately 2 3125 ppm BeO, 1200 ppm Mg 97.4 days 1.24 × 1025 m−2 780 K 480 appm 1994
Be sample transfer as T2 + HT into a flow of carrier gas, where the tritium activity released (1) was measured with tritium monitor TEM 2100A with a proportional gas flow-through detector DDH 32. AT,
gas, acid
= AT2 + AT0
(1)
T+ localized in a Be layer remains in the solution and the tritium activity AT sol. acid was measured with liquid scintillation method. 90% of H0 (T0 ) reacts with the scavenger Na2 Cr2 O7 in the solution of 2 mol L−1 H2 SO4 and remains in the solution. Activity of the tritium released into a gas phase and retained in the solution are the respective sums: AT,
gas, Cr(VI)
AT,
sol., Cr(VI)
= AT2 + x · AT0
(2)
= AT+ + (1 − x) · AT0
(3)
where value of x was found to be 0.1 (10%). The contents of T0 , T2 , T+ (Bq g−1 ) in a sample were determined separately from the corresponding differences in the activities by following equations: AT0 =
AT, gas, acid − AT, gas, Cr(VI) 1−x
(4)
AT2 = AT, gas, acid − AT0
(5)
AT+ = AT,
(6)
sol., acid
2.3. Structure analysis The microstructure of the samples was evaluated by means of scanning electron microscopy (Hitachi S-4800, 2 kV, 15 A). Samples were prepared by polishing first with SiC paper and then with diamond paste of 1 m particle size. 2.4. Thermo-annealing To describe and predict the diffusion process of tritium during thermo-annealing the estimation also of the resulting abundance ratio of tritium chemical forms is needed. Pebbles were heated in the special setup with and without action of fast electron irradiation (14 MGy h−1 , 5 MeV) and high magnetic field (1.7 T). Experimental setup, conditions and tritium thermo-desorption curves have been described previously [3]. 3. Results and discussion 3.1. Structure Structure analysis has been performed for the samples irradiated in the PBA experiment. Pebbles from BERYLLIUM experiment have been described elsewhere by other authors [4], whereas pebbles from experiment EXOTC 8/3-13 have not sufficient diameter (0.1–0.2 mm) for the polishing procedure.
Fig. 1. SEM image of polished cross-section of PBA pebbles: (a) not treated and (b) after treatment in high temperature (1260 ◦ C).
E. Pajuste et al. / Fusion Engineering and Design 86 (2011) 2125–2128
2127
Fig. 2. Abundance ratio of tritium chemical forms in the beryllium pebbles irradiated in experiments BERYLLIUM, EXOTIC 8/3-13 and PBA.
Large voids became visible in the bulk of most of the examined PBA pebbles after polishing (Fig. 1). These voids are generated during the cooling phase of the fabrication process and according to the production method patent have been developed to prevent beryllium swelling and failure when irradiated with neutrons (helium produced in the nuclear reactions is stored in this void instead of beryllium lattice) [5]. Nevertheless, existence of such technical void increases tritium retention since it may act as a pressurized sink for tritium trapping [6]. Cross-section diameter of these voids ranges from 100 to 250 m that corresponds to approximately 0.5% of the pebble volume. Porosity of the PBA pebbles before thermal treatment was not observed since resolution of the method for these particular samples allows seeing pores larger than 500 nm, but according to literature [7,8] at the temperatures below ∼400 ◦ C size of helium inclusions are less than several tens of nm. Thermo-annealing experiment at the temperature of 720 ◦ C has been performed and at this temperature visible cracks are formed and the porosity starts to appear. After heating of the sample at high temperature (up to 1260 ◦ C) structure changes drastically – large radial cracks and high porosity appear (Fig. 1(b)). This is a result both of thermal stresses (cracks) and the coalescence of small helium inclusions into the larger ones. 3.2. Tritium chemical forms and distribution in pebbles irradiated in different experiments Tritium produced in the neutron induced transmutation of beryllium can diffuse into the lattice or can be trapped by structure traps (such as intragranular He bubbles, closed porosity and grain boundaries) or it may react with BeO to form Be(OT)2 with the formation energy of −0.7 eV (at the standard temperature, pressure) [9]. The abundance ratio of tritium chemical forms in the beryllium pebbles was measured for samples from all three irradiation experiments (Fig. 2). Large fraction (up to 96%) of the tritium accumulated in irradiated beryllium pebbles was found to be in the molecular form. We assume that it coexists with the helium in the gas inclusions. Large fraction of gaseous species might be trapped also in the technical void in the bulk of the pebbles that might explain the highest T2 content in the pebbles from the PBA experiment (fabricated by the REP method). Presence of molecular tritium in the pebbles is undesirable. In order to extract T2 from the metallic matrix it is required T2 to dissociate into T0 (dissociation energy for hydrogen −4.52 eV). In the investigated samples fraction of atomic tritium was in range from 1 to 32%. Atomic tritium can either exist as interstitial or can be trapped in the vacancy-based defect of the beryllium [10]. Atomic tritium can diffuse without any transformation – energy is required only for passing through the diffusion energy barrier. The role of molecular and atomic tritium ratio is obvious if one
Fig. 3. Distribution of the chemical forms of tritium in beryllium pebbles after thermo-annealing experiments, where R – radiation of accelerated electrons 5 MeV, 14 MGy h−1 , MF – magnetic field 1.7 T, and T – temperature: (a) 280 ◦ C and (b) 500 ◦ C.
compares the tritium release from EXOTIC 8/3-13 and PBA pebbles. From the EXOTIC 8/3-13 pebbles tritium release is starting at much lower temperatures and full detritiation is reached much faster (the diameter difference definitely also has an impact). Chemically bonded tritium T+ was found to be in range from 3 to 12%. Highest concentration of T+ was in the pebbles from the experiment EXOTIC 8/3-13. According to the chemical analysis these pebbles have the highest beryllium oxide content. Oxide could be formed as a result of the chemical interaction with lithium ceramics during the irradiation (signs of Si were found on the surface of these pebbles). Due to large specific surface oxide content could be increasing also during the storage in air. Tritium retained in the form of beryllium hydroxide stays immobile unless temperature required for dehydration of Be(OT)2 is reached (complete dehydration is reached at temperatures above 950 ◦ C) [11]. Bulk distribution of tritium in the pebbles has been determined for the pebbles of BERYLLIUM and PBA experiments. Distribution was calculated by measuring tritium activity in the gas phase released during the dissolution, therefore distribution of chemically bonded tritium T+ was not determined (more likely most of it is in the surface oxide layer and in the oxide inclusions in the bulk). For both types of samples distribution of tritium was similar–low concentration in the first micrometers from the surface, more or less uniform distribution in the bulk and in some pebbles a sharp peak somewhere in the bulk. The peak might indicate the position of the large void that was found during the structure investigations. 3.3. Evolution of tritium chemical forms at the thermo-annealing Changes of the abundance ratio of tritium chemical forms during the thermal treatment of sample at different conditions, such as fast electron irradiation and high magnetic field have been done for the samples from EXOTIC 8/3-13 experiment. Treatment has
2128
E. Pajuste et al. / Fusion Engineering and Design 86 (2011) 2125–2128
been performed at temperatures 280 ◦ C for 3 h and 500 ◦ C for 1 h (Fig. 3). The results show that the amount of chemically bonded tritium changes insignificantly after the thermal treatment of pebbles. At low temperature (280 ◦ C) only content of atomic tritium has decreased, other forms stayed immobile. Heating at higher temperature decreases the T2 content because of the molecule dissociation into the mobile T0 and also the formation of cracks and open porosity where molecular tritium may migrate. Dissociation of molecular tritium is also the reason the atomic form content virtually stays at the same level. Under action of fast electrons the radiolysis of T2 takes place and therefore the increase of atomic tritium content. The highest decrease of molecular tritium and tritium total content in samples is observed under the simultaneous action of temperature, irradiation and magnetic field. This phenomenon could be explained by the spin-spin transformations (radiolysis in high magnetic field). In the dissociation (radiolysis) process correlated radical pair is formed with anti-parallel electron spins, whereas magnetic field changes them to parallel and probability of recombination of the radicals decreases [12]. Therefore more tritium is in a mobile T0 form. 4. Conclusions Structure analysis revealed large technical voids in the bulk of the beryllium pebbles produced by REP method. Existence of the void may play crucial role in the tritium retention. The main chemical form of the tritium localized in the irradiated beryllium pebbles is T2 . High content of this immobile form might significantly extend the time and therefore increase the cost of detritiation method based on thermo-annealing process. Determination of the chemical forms of tritium before and after thermal treatment in different condition gives the possibility to estimate the processes occurring during the thermoannealing.
Acknowledgments This work was performed within the EFDA Goal Oriented Training Programme “Breeding Blanket Development for Fusion Reactors – EUROBREED”. The support of L’Oréal Latvian Fellowship for Women in Science 2010 with the support of the Latvian National Commission for UNESCO and the Latvian Academy of Sciences is also gratefully acknowledged. References [1] M. Gasparotto, L.V. Boccaccini, L. Giancarli, S. Malang, Y. Poitevin, Demo blanket technology R&D results in EU, Fusion Engineering and Design 61–62 (2002) 263–271. [2] J. Tiliks, G. Kizane, A. Vitins, E. Kolodinska, E. Rabaglino, Magnetic field effects on tritium release from neutron-irradiated beryllium pebbles, Nuclear Technology 159 (2007) 245–249. [3] J. Tiliks, A. Vitins, G. Kizane, V. Tilika, E. Kolodinska, S. Kaleja, B. Lescinskis, Effects of external energetic factors on tritium release from the EXOTIC 83/13 neutron-irradiated beryllium pebbles, Fusion Engineering and Design 84 (2009) 1842–1846. [4] A. Möslang, R.A. Pieritz, E. Boller, C. Ferrero, Gas bubble network formation in irradiated beryllium pebbles monitored by X-ray microtomography, Journal of Nuclear Materials 386–388 (2009) 1052–1055. [5] E. Ishitsuka, H. Kawamura, N. Sakamoto, K. Nishida, Process for preparing metallic beryllium pebbles, in: U.S. Patent (Ed.), Japan Atomic Energy Research Institute (JP) NGK Insulators Ltd. (JP) United States, 1999, p. 10. [6] F. Scaffidi-Argentina, M. Dalle Donne, H. Werle, Critical assessment of beryllium pebbles response under neutron irradiation: mechanical performance and tritium release, Journal of Nuclear Materials 258–263 (1998) 595–600. [7] L.L. Snead, Low-temperature low-dose neutron irradiation effects on beryllium, Journal of Nuclear Materials 326 (2004) 114–124. [8] I.B. Kupriyanov, R.R. Melder, V.A. Gorokhov, The effect of neutron irradiation on beryllium performance, Fusion Engineering and Design 51–52 (2000) 135–143. [9] F. Scaffidi-Argentina, Tritium and helium release from neutron irradiated beryllium pebbles from the EXOTIC-8 irradiation, Fusion Engineering and Design 58–59 (2001) 641–645. [10] M.G. Ganchenkova, P.V. Vladimirov, V.A. Borodin, Vacancies, interstitials and gas atoms in beryllium, Journal of Nuclear Materials 386–388 (2009) 79–81. [11] K.A. Walsh, Beryllium Chemistry and Processing, ASM International, Ohio, 2009. [12] H. Hayashi, Introduction to Dynamic Spin Chemistry: Magnetic Field Effects upon Chemical and Biochemical Reactions (World Scientific Lecture and Course Notes in Chemistry, vol. 8), World Scientific Publishing Company, 2004.