- Email: [email protected]
Investigation of beryllium pebbles produced by powder metallurgy for HCPB breeding blanket
Fusion Engineering and Design xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Investigation of beryllium pebbles produced by powder metallurgy for HCPB breeding blanket ⁎
I.B. Kupriyanova, , G.N. Nikolaeva, S.K. Zavjalova, L.A. Kurbatovaa, N.E. Zabirovaa, V.P. Chakinb a b
A.A. Bochvar High Technology Research Institute of Inorganic Materials, Moscow, Russia Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany
A R T I C LE I N FO
A B S T R A C T
Keywords: Beryllium pebbles Breeding blanket Neutron Multiplier Neutron irradiation Tritium release
This paper presents the results of investigation of three batches of beryllium pebbles with average pebble size of 1.2–1.3 mm and different average grain sizes (13–14 μm, ∼50 μm and ∼615 μm). Microstructure and chemical composition of produced beryllium pebbles are presented as well as packing density and pebble size distribution. The influence of grain size on tritium release and retention in Be pebbles during temperature programmed desorption (TPD) after high-temperature loading of tritium/hydrogen gas mixture are also described.
1. Introduction Beryllium is planned to be used as the neutron multiplier in helium cooled pebble-bed (HCPB) breeding blanket concept for DEMO power plants. Under neutron irradiation due to nuclear transmutation a large amount of helium and tritium is produced in beryllium. The key issues of neutron irradiation of beryllium are helium-induced swelling and tritium retention and release. The in-pile tritium release should be high enough to eliminate risk to personnel in case of a serious accident in a fusion power plant which can lead to abrupt release of all accumulated tritium. In addition, an increase in-pile tritium release from beryllium will help ensure the necessary efficiency in tritium breeding ratio of HCPB blanket. In the present European HCPB breeder blanket design, beryllium is used in the form of pebbles with diameter of ∼1 mm. Currently, the main producers of beryllium pebbles are NGK (Japan), which uses “Rotating Electrode Method” (REM,) and Materion Brush (USA), which uses “Fluoride Reduction Method " (FRM) [1–7]. A main feature of both methods is that the resulting pebbles are coarse-grained in the range of 500–1000 μm that makes it difficult to release tritium out of beryllium. It is assumed that helium and tritium release can be increased significantly if beryllium has fine grain structure with average grain size of a few microns [6,7]. In order to produce the pebbles with fine grain structure, an R&D project was performed at the Bochvar Institute. Several experimental batches of Be pebbles with average pebble size of 1.2–1.3 mm and different grain sizes have been fabricated by powder metallurgy and then characterized. This paper presents some results of investigations for three batches
⁎
of beryllium pebbles fabricated with average pebble size of 1.2–1.3 mm and different average grain sizes (13–14 μm, ∼50–55 μm and 615 μm). 2. Fabrication of beryllium pebbles by powder metallurgy The technology for powder metallurgy fabrication of Be pebbles was developed for the production of the experimental batches of pebbles. The process consists of the following key steps: 1. Fabrication of Be billet with the required grain size and chemical composition 2. Billet fragmentation (to ∼10 mm pieces) by hydraulic press; 3. Further fragmentation in jaw crusher; 4. Intermediate screen sizing (+1.0–1.6 mm); 5. Spherodizing by ball milling; 6. Cleaning in water/Etching in HNO3/Cleaning in water/Drying; 7. Final screen sizing (+0.8–1.25 mm). Manufacturing of Be pebbles is described in detail in Ref. [8]. The average grain size and chemical composition of Be pebbles mostly depend on average grain size and chemical composition of initial Be billets (Step 1), and shape and pebbles size distribution depend on milling parameters and screen sizing (Steps 2–7). The following initial materials were used for the fabrication of three types of beryllium pebbles: 1. For the Be pebbles with average grain size < 30 μm (Type 1) the
Corresponding author. E-mail addresses: [email protected], [email protected] (I.B. Kupriyanov).
https://doi.org/10.1016/j.fusengdes.2018.04.073 Received 28 September 2017; Received in revised form 10 April 2018; Accepted 19 April 2018 0920-3796/ © 2018 Elsevier B.V. All rights reserved.
Please cite this article as: Kupriyanov, I.B., Fusion Engineering and Design (2018), https://doi.org/10.1016/j.fusengdes.2018.04.073
Fusion Engineering and Design xxx (xxxx) xxx–xxx
I.B. Kupriyanov et al.
are presented in Fig. 1. The pebbles could be characterized as the irregularly shaped with rounded corners. To determine the pebbles’ dimensions, 100 pebbles were taken from the top and the bottom parts of a container with pebbles. Two dimensions were measured for each pebble: the biggest (L1) and the smallest (L2) which is perpendicular to the biggest one. With these measurements the average size Lave = L1 + L2/2 and the aspect ratio L1/L2 were calculated for every pebble. The measurements were carried out using a BMI-1 optical instrumental microscope at 20-x magnification. The results of calculations are summarized in Table 1. The average pebble size calculated on the basis of these measurements is 1.21–1.29 mm. The pebble size distributions are presented in Fig. 2. Attention is drawn to the fact that more than 35% of pebbles have larger average size than it expected from the size of the sieves used (0.8–1.25 mm). The probable reason for this is that ∼22–34% of pebbles have the aspect ratio L1/L2 > 1.5 and the pebbles, having average size > 1.25 mm but the L2 dimension less than 1.25 mm, can pass through the sieve with cell size of 1.25 mm. 3.2. Chemical composition Chemical composition of all pebbles' types investigated is shown in Table 2. The main purpose of this study is to define a method for production of beryllium pebbles with fine grain structure, so we did not prescribe the limits on chemical composition of initial materials. 3.3. Microstructure Typical microstructure of three types of beryllium pebbles is shown in Fig. 3. Average grain sizes are of 13–14 μm, 50–51 μm and 614–615 μm for pebbles of Type 1, Type 2 and Type 3, correspondingly (Table 3). The perceptible pebbles' surface microtoughness has been revealed that is a result of an intensive deformation of a near-surface layer during fabrication of the pebbles (Fig. 4). One can see that the surface microroughness of Type 2 pebbles is slightly higher than that of Type 1 and Type 3 pebbles. Fig. 1. Photo of beryllium pebbles.
3.4. Packing density vacuum hot pressed (VHP) Be billet with average grain size of < 30 μm was used; 2. For the Be pebbles with average grain size of 30–60 μm (Type 2) the hot isostatic pressed (HIP) Be billet was used with an average grain size of 50-60 μm, which was produced from spherical powder; 3. For the fabrication of Be pebbles with average grain size > 100 mm (Type 3) the Be ingot with an average grain size of > 100 mm was used.
Packing density of beryllium pebbles was defined at RT on 200 cm3 of pebbles, using a measuring cell with 70 mm in diameter and 200 mm in the height (no vibrations were applied). The packing density was 1.09, 1.05 and 1.11 g/cm3 or ∼59.0, 56.8 and 60.1% of theoretical density of beryllium for pebbles of Type 1, Type 2 and Type 3, correspondingly (Fig. 5). After vibration (50 Hz, 2 h) the packing density of pebbles increased up to 1.23, 1.14 and 1.18 g/cm3 or ∼66.6, ∼61.4 and ∼63.8% of theoretical density of beryllium for Type 1, Type 2 and Type 3, correspondingly. Although the dimensions of the pebbles are ∼20-30% higher than the reference pebble’s dimensions (1 mm) for the HCPB blanket, nevertheless the packing density of these pebbles after vibration has the necessary value (> 60%) for the Be multiplier material for HCPB. The differences in packing density could be associated with different pebbles size distribution (Fig. 2) and surface roughness (Fig. 4) in three types of pebbles. The higher packing density of Type 1 and Type 3 pebbles compared to Type 2 is likely due to the following factors: (1) a slightly smaller average pebble size of these
Three experimental batches of beryllium pebbles (∼0.9 kg total amount) with average pebble size of 1.2–1.3 mm and three average grain sizes were fabricated by this technology.
3. Characterization of beryllium pebbles 3.1. Shape and size of pebbles Typical view of beryllium pebbles from the experimental batches Table 1 Shape and size of beryllium pebbles. Type of pebbles
Average pebble size Σ (Lave), mm
Max dimension (L mm
Type 1 (< 30 μm) Type 2 (30–60 μm) Type 3 (> 100 μm)
1.27 1.29 1.21
1.98 1.65 1.82
ave) max,
Min dimension (L mm 0.83 0.94 0.18
2
ave) min,
Fraction with aspect ratio L1/ L2 ≤ 1.5 %
Fraction with aspect ratio L1/ L2 > 1.5 %
67 78 66
33 22 34
Fusion Engineering and Design xxx (xxxx) xxx–xxx
I.B. Kupriyanov et al.
Table 2 Chemical composition of beryllium pebbles. Element
Be Fe Al Si O2 C Mn Mg Ni Cr Ti Ca Mo Zn U Ag Co
wt% Type 1 (13–14 μm)
Type 2 (50–51 μm)
Type 3 (614–615 μm)
98.6 0.07 0.011 0.019 1.17 0.075 0.0023 0.0016 0.0051 0.025 0.0086 0.0047 0.0013 0.0014 0.0005 < 0.0001 < 0.0005
99.5 0.025 0.0051 0.013 0.28 0.076 0.0023 0.0023 < 0.001 0.011 0.001 0.0035 0.011 0.0013 0.00001 < 0.0001 < 0.0005
99.2 0.15 0.014 0.013 0.39 0.064 0.0069 0.0012 0.016 0.052 0.02 0.0037 0.003 0.0008 0.0002 < 0.0001 < 0.0005
The beryllium pebbles were saturated at a temperature 873 K for 15 h at a pressure of 4 bar in a gas mixture of 1H2 with 500 appm 3H2. The tritium release measurements were performed using a permanent heating rate ramp of 0.12 K/s up to 1373 K followed by 3 h exposure [9]. During tritium release experiments, the pebbles were aerated by a purge gas mixture of high-purity helium with a small addition of hydrogen (4He + 0.1 vol.% 1H2). The release activity was measured with a proportional counter (PC). Fig. 6a shows tritium release rates for beryllium pebbles with different grain sizes. The pebbles of grain sizes 50–51 mm (Type 2) and 615 mm (Type 3) have single peaks which are practically coincident with each other. The tritium release rate curve for the pebbles of grain size of 13–14 mm (Type 1) differs significantly from that for the pebbles with the larger grain size. The tritium release starts immediately from the beginning of the TPD test, i.e. from even just a small increment above room temperature after heating, and continues with increasing rate until it reaches its maximum peak. Despite the difference in the behavior of the release rate for pebbles with grain size of 13–14 mm, however, the peaks for all three pebble types are grouped closely at temperatures in the region of 1267–1300 K. The highest total tritium release (TTR) was detected for the pebbles with the grain size of 13–14 mm (Fig. 6b). With increase in grain size, the total tritium release decreases. For pebbles with av.g.s ∼13–14 μm the TTR is about 2 and 2.2 times higher than the TTRs from the pebbles with av.g.s. 50–51 μm and 615 μm correspondingly. For comparison, Fig. 7 presents data on tritium release from beryllium pebbles with a diameter of 1 mm produced by REM and FRM which were obtained under exactly the same conditions of tritium/ hydrogen saturation and TDP testing [10]. The major peak for the FRM pebbles is at the lower temperature (1228 K) compared to the REM pebbles (1308–1317 K). The TTR of pebbles with av.g.s ∼13–14 μm produced by powder metallurgy is significantly higher (∼in 7–12 times) than that for the pebbles produced by FRM and REM. TTR of these pebbles with av.g.s ∼615 μm is in ∼3–5 times higher, than that for the pebbles produced by FRM and REM.
Fig. 2. Pebble size distribution, where a) Type 1 (13–14 μm), b) Type 2 (50–51 μm), c) Type 3 (614–615 mm).
types (Table 1); (2) the higher fraction of fine (< 1.10 mm) pebbles (Fig. 4), which increases the ability to fill the space between larger pebbles and (3) the smoother surface of Type 1 pebbles in comparison with Type 2 pebbles, that improves the mutual movement of the pebbles. 4. Tritium release behavior
5. Discussion Investigation of tritium release behavior of beryllium pebbles were performed by temperature-programmed desorption (TPD) tests. Before tests beryllium pebbles were saturated in tritium/hydrogen gas mixture at high-temperature.
The results of this study show that the use of powder metallurgy makes it possible to manufacture beryllium pebbles with a wide range of grain sizes and, in particular, with a fine-grain structure (< 15 μm).
3
Fusion Engineering and Design xxx (xxxx) xxx–xxx
I.B. Kupriyanov et al.
Fig. 3. Microstructure of beryllium pebbles (metallography, polarized light), where a) Type 1 (13–14 μm), b) Type 2 (50–51 μm), c) Type 3 (614–615 mm).
these pebbles after vibration (50 Hz, 2 h) has the necessary value (> 60%) for the Be multiplier material for HCPB. The total tritium release decreases with increase of grain size. For the pebbles with av.g.s ∼13–14 μm the TTR is about 2 and 2.2 times higher than the TTRs for the pebbles with av.g.s. 50–51 μm and 615 μm. The TTR of pebbles with av.g.s ∼13–14 μm produced by powder metallurgy is significantly higher (∼in 7–12 times), than the TTRs for the pebbles produced by FRM and REM. Noteworthy is the fact that the TTR of coarse-grained (∼615 μm) pebbles produced by the powder metallurgy is ∼3–5 times higher than that for the coarse-grained pebbles produced by FRM and REM. This indicates that in addition to the grain size other structural factors also have a significant influence on tritium release, for example, deformation and the surface roughness of pebbles, the state and thickness of the
Table 3 Grain size in different Types of beryllium pebbles. Type of pebbles
Average grain size, μm
Max grain size, μm
Min grain size, μm
Type 1 (< 30 μm) Type 2 (30–60 μm) Type 3 (> 100 μm)
13–14 50–51 614–615
30 64.7 1163.5
5 34.6 216.7
In addition, this method makes it possible to manufacture beryllium pebbles of a variety different dimensions. Despite the fact that the dimensions of the pebbles are ∼20–30% higher than the reference pebble size (1 mm) for the HCPB blanket, nevertheless the packing density of
4
Fusion Engineering and Design xxx (xxxx) xxx–xxx
I.B. Kupriyanov et al.
Fig. 4. Surface morphology of the pebbles (SEM), where a) Type 1 (13–14 μm), b) Type 2 (50–51 μm), c) Type 3 (614–615 mm).
5
Fusion Engineering and Design xxx (xxxx) xxx–xxx
I.B. Kupriyanov et al.
6. Summary The experimental technology of fabrication of beryllium pebbles with fine grain structure by powder metallurgy has been developed and successfully tested in Bochvar Institute. Three experimental batches of Be pebbles with average pebble size of 1.2–1.3 mm and different average grain sizes (∼ 13–14 μm, ∼50–51 μm and 614–615 μm) have been fabricated and characterized. The technology seems is scalable for further industrial production of beryllium pebbles. Microstructure examination shows that surface of the pebbles is characterized by a noticeable micro-toughness that is a result of the intensive deformation processes occurring in a near-surface layer of the pebbles during their fabrication. Packing density at free filling was ∼59, 56.8 and 60.1% of theoretical density of beryllium, for Type 1, Type 2 and Type 3 pebbles, correspondingly. After vibration (50 Hz, 2 h) packing density of pebbles increased up to ∼66.6%, ∼61.4 and 63.8% of theoretical density of beryllium for Type 1, Type 2 and Type 3 correspondingly. Total tritium release (TTR) decreases with increase of grain size. For pebbles with av.g.s ∼13–14 μm the TTR is about 2–2.2 times higher than that for the pebbles with av.g.s. 50–51 μm and 615 μm. The TTR of pebbles with av.g.s ∼13–14 μm produced by powder metallurgy is ∼7–12 times higher than that for the pebbles produced by FRM and REM.
Fig. 5. Packing density (PD) of beryllium pebbles before and after vibration (50 Hz, 2 h).
surface oxide film, internal defects of the pebbles, etc. The results obtained make it possible to conclude that from the point of view of the tritium release, the Be pebbles with smaller grain size (∼13–14 μm) are preferable for use in the ITER (and DEMO) blanket modules.
Fig. 6. Tritium release behavior for beryllium pebbles with different grain sizes, where: a) tritium release rate versus temperature, b) total tritium release.
Fig. 7. Tritium release behavior for 1 mm-diameter beryllium pebbles fabricated by REM and FRM: a) tritium release rate versus temperature, ‘REM (pores)’ indicates the presence of a higher relative density of pores compared to the “REM” batch with a significantly lower amount of pores; b) total tritium release. 6
Fusion Engineering and Design xxx (xxxx) xxx–xxx
I.B. Kupriyanov et al.
The TTR of coarse-grained (∼615 μm) pebbles produced by powder metallurgy is ∼3–5 times higher than that for the pebbles produced by FRM and REM.
[4]
Acknowledgements [5]
This work was funded by European Fusion for Energy (Contract EFDA 05/994).
[6]
References
[7]
[1] T. Iwadachi, N. Sakamoto, K. Nishida, H. Kawamura, Production of various sizes and some properties of beryllium pebbles by the rotating electrode method, Proc. Third IEA Int. Workshop on Beryllium Technology for Fusion, Mito, Japan, JAERI –Conf 98-001, 1997, pp. 33–38 Oct. 22–24. [2] M. Dalle-Donne, D.L. Baldwin, D.S. Gelles, L.R. Greenwood, et al., Behavior of beryllium pebbles under irradiation, Proc. Third IEA Int. Workshop on Beryllium Technology for Fusion, Mito, Japan, JAERI –Conf 98-001, 1997, pp. 296–306 Oct. 22–24. [3] M. Dalle-Donne, A. Goraieb, G. Piazza, G. Sordon, Measurement of thermal
[8]
[9]
[10]
7
conductivity and heat transfer coefficient of a binary bed of beryllium pebbles, Proc. Third IEA Int. Workshop on Beryllium Technology for Fusion, Mito, Japan, JAERI –Conf 98-001, 1997, pp. 17–24 Oct. 22–24. T. Iwadachi, D. Schmidt, H. Kawamura, The distribution of impurities in beryllium pebbles produced by rotating electrode method, Proc. Of the 4th IEA Int. Workshop on Beryllium Technology for Fusion, Karlsruhe, Germany, FZKA 6462, 2000, pp. 15–24 Sept. 15–17, 1999. E. Ishitsuka, H. Kawamura, T. Terai, S. Tanaka, Effect of oxide layer on tritium release from beryllium pebbles, Proc. of 18th SOFT, Fusion Technology 2 (1994) 1345–1348. E. Rabaglino, Helium and Tritium in Neutron Irradiated Beryllium, FZKA Report No.6939, (2006). F. Scaffidi-Argentina, et al., Tritium and helium release from neutron irradiated beryllium pebbeles from EXOTIC-8 irradiation, Fusion Eng. Des. 58–59 (2001) 641–645. I. Kupriyanov, et al., Production and investigation of beryllium pebbles with fine grain structure for the HCPB breeder blanket, Fusion Eng. Des. 87 (2012) 1338–1341. V. Chakin, R. Rolli, P. Vladimirov, A. Moeslang, Tritium release from highly neutron irradiated constrained and unconstrained beryllium pebbles, Fusion Eng. Des. 95 (2015) 59–66. V. Chakin, et al., Tritium release from advanced beryllium materials after loading by tritium/hydrogen gas mixture, Fusion Eng. Des. 107 (2016) 75–81.
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Investigation of beryllium pebbles produced by powder metallurgy for HCPB breeding blanket ⁎
I.B. Kupriyanova, , G.N. Nikolaeva, S.K. Zavjalova, L.A. Kurbatovaa, N.E. Zabirovaa, V.P. Chakinb a b
A.A. Bochvar High Technology Research Institute of Inorganic Materials, Moscow, Russia Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany
A R T I C LE I N FO
A B S T R A C T
Keywords: Beryllium pebbles Breeding blanket Neutron Multiplier Neutron irradiation Tritium release
This paper presents the results of investigation of three batches of beryllium pebbles with average pebble size of 1.2–1.3 mm and different average grain sizes (13–14 μm, ∼50 μm and ∼615 μm). Microstructure and chemical composition of produced beryllium pebbles are presented as well as packing density and pebble size distribution. The influence of grain size on tritium release and retention in Be pebbles during temperature programmed desorption (TPD) after high-temperature loading of tritium/hydrogen gas mixture are also described.
1. Introduction Beryllium is planned to be used as the neutron multiplier in helium cooled pebble-bed (HCPB) breeding blanket concept for DEMO power plants. Under neutron irradiation due to nuclear transmutation a large amount of helium and tritium is produced in beryllium. The key issues of neutron irradiation of beryllium are helium-induced swelling and tritium retention and release. The in-pile tritium release should be high enough to eliminate risk to personnel in case of a serious accident in a fusion power plant which can lead to abrupt release of all accumulated tritium. In addition, an increase in-pile tritium release from beryllium will help ensure the necessary efficiency in tritium breeding ratio of HCPB blanket. In the present European HCPB breeder blanket design, beryllium is used in the form of pebbles with diameter of ∼1 mm. Currently, the main producers of beryllium pebbles are NGK (Japan), which uses “Rotating Electrode Method” (REM,) and Materion Brush (USA), which uses “Fluoride Reduction Method " (FRM) [1–7]. A main feature of both methods is that the resulting pebbles are coarse-grained in the range of 500–1000 μm that makes it difficult to release tritium out of beryllium. It is assumed that helium and tritium release can be increased significantly if beryllium has fine grain structure with average grain size of a few microns [6,7]. In order to produce the pebbles with fine grain structure, an R&D project was performed at the Bochvar Institute. Several experimental batches of Be pebbles with average pebble size of 1.2–1.3 mm and different grain sizes have been fabricated by powder metallurgy and then characterized. This paper presents some results of investigations for three batches
⁎
of beryllium pebbles fabricated with average pebble size of 1.2–1.3 mm and different average grain sizes (13–14 μm, ∼50–55 μm and 615 μm). 2. Fabrication of beryllium pebbles by powder metallurgy The technology for powder metallurgy fabrication of Be pebbles was developed for the production of the experimental batches of pebbles. The process consists of the following key steps: 1. Fabrication of Be billet with the required grain size and chemical composition 2. Billet fragmentation (to ∼10 mm pieces) by hydraulic press; 3. Further fragmentation in jaw crusher; 4. Intermediate screen sizing (+1.0–1.6 mm); 5. Spherodizing by ball milling; 6. Cleaning in water/Etching in HNO3/Cleaning in water/Drying; 7. Final screen sizing (+0.8–1.25 mm). Manufacturing of Be pebbles is described in detail in Ref. [8]. The average grain size and chemical composition of Be pebbles mostly depend on average grain size and chemical composition of initial Be billets (Step 1), and shape and pebbles size distribution depend on milling parameters and screen sizing (Steps 2–7). The following initial materials were used for the fabrication of three types of beryllium pebbles: 1. For the Be pebbles with average grain size < 30 μm (Type 1) the
Corresponding author. E-mail addresses: [email protected], [email protected] (I.B. Kupriyanov).
https://doi.org/10.1016/j.fusengdes.2018.04.073 Received 28 September 2017; Received in revised form 10 April 2018; Accepted 19 April 2018 0920-3796/ © 2018 Elsevier B.V. All rights reserved.
Please cite this article as: Kupriyanov, I.B., Fusion Engineering and Design (2018), https://doi.org/10.1016/j.fusengdes.2018.04.073
Fusion Engineering and Design xxx (xxxx) xxx–xxx
I.B. Kupriyanov et al.
are presented in Fig. 1. The pebbles could be characterized as the irregularly shaped with rounded corners. To determine the pebbles’ dimensions, 100 pebbles were taken from the top and the bottom parts of a container with pebbles. Two dimensions were measured for each pebble: the biggest (L1) and the smallest (L2) which is perpendicular to the biggest one. With these measurements the average size Lave = L1 + L2/2 and the aspect ratio L1/L2 were calculated for every pebble. The measurements were carried out using a BMI-1 optical instrumental microscope at 20-x magnification. The results of calculations are summarized in Table 1. The average pebble size calculated on the basis of these measurements is 1.21–1.29 mm. The pebble size distributions are presented in Fig. 2. Attention is drawn to the fact that more than 35% of pebbles have larger average size than it expected from the size of the sieves used (0.8–1.25 mm). The probable reason for this is that ∼22–34% of pebbles have the aspect ratio L1/L2 > 1.5 and the pebbles, having average size > 1.25 mm but the L2 dimension less than 1.25 mm, can pass through the sieve with cell size of 1.25 mm. 3.2. Chemical composition Chemical composition of all pebbles' types investigated is shown in Table 2. The main purpose of this study is to define a method for production of beryllium pebbles with fine grain structure, so we did not prescribe the limits on chemical composition of initial materials. 3.3. Microstructure Typical microstructure of three types of beryllium pebbles is shown in Fig. 3. Average grain sizes are of 13–14 μm, 50–51 μm and 614–615 μm for pebbles of Type 1, Type 2 and Type 3, correspondingly (Table 3). The perceptible pebbles' surface microtoughness has been revealed that is a result of an intensive deformation of a near-surface layer during fabrication of the pebbles (Fig. 4). One can see that the surface microroughness of Type 2 pebbles is slightly higher than that of Type 1 and Type 3 pebbles. Fig. 1. Photo of beryllium pebbles.
3.4. Packing density vacuum hot pressed (VHP) Be billet with average grain size of < 30 μm was used; 2. For the Be pebbles with average grain size of 30–60 μm (Type 2) the hot isostatic pressed (HIP) Be billet was used with an average grain size of 50-60 μm, which was produced from spherical powder; 3. For the fabrication of Be pebbles with average grain size > 100 mm (Type 3) the Be ingot with an average grain size of > 100 mm was used.
Packing density of beryllium pebbles was defined at RT on 200 cm3 of pebbles, using a measuring cell with 70 mm in diameter and 200 mm in the height (no vibrations were applied). The packing density was 1.09, 1.05 and 1.11 g/cm3 or ∼59.0, 56.8 and 60.1% of theoretical density of beryllium for pebbles of Type 1, Type 2 and Type 3, correspondingly (Fig. 5). After vibration (50 Hz, 2 h) the packing density of pebbles increased up to 1.23, 1.14 and 1.18 g/cm3 or ∼66.6, ∼61.4 and ∼63.8% of theoretical density of beryllium for Type 1, Type 2 and Type 3, correspondingly. Although the dimensions of the pebbles are ∼20-30% higher than the reference pebble’s dimensions (1 mm) for the HCPB blanket, nevertheless the packing density of these pebbles after vibration has the necessary value (> 60%) for the Be multiplier material for HCPB. The differences in packing density could be associated with different pebbles size distribution (Fig. 2) and surface roughness (Fig. 4) in three types of pebbles. The higher packing density of Type 1 and Type 3 pebbles compared to Type 2 is likely due to the following factors: (1) a slightly smaller average pebble size of these
Three experimental batches of beryllium pebbles (∼0.9 kg total amount) with average pebble size of 1.2–1.3 mm and three average grain sizes were fabricated by this technology.
3. Characterization of beryllium pebbles 3.1. Shape and size of pebbles Typical view of beryllium pebbles from the experimental batches Table 1 Shape and size of beryllium pebbles. Type of pebbles
Average pebble size Σ (Lave), mm
Max dimension (L mm
Type 1 (< 30 μm) Type 2 (30–60 μm) Type 3 (> 100 μm)
1.27 1.29 1.21
1.98 1.65 1.82
ave) max,
Min dimension (L mm 0.83 0.94 0.18
2
ave) min,
Fraction with aspect ratio L1/ L2 ≤ 1.5 %
Fraction with aspect ratio L1/ L2 > 1.5 %
67 78 66
33 22 34
Fusion Engineering and Design xxx (xxxx) xxx–xxx
I.B. Kupriyanov et al.
Table 2 Chemical composition of beryllium pebbles. Element
Be Fe Al Si O2 C Mn Mg Ni Cr Ti Ca Mo Zn U Ag Co
wt% Type 1 (13–14 μm)
Type 2 (50–51 μm)
Type 3 (614–615 μm)
98.6 0.07 0.011 0.019 1.17 0.075 0.0023 0.0016 0.0051 0.025 0.0086 0.0047 0.0013 0.0014 0.0005 < 0.0001 < 0.0005
99.5 0.025 0.0051 0.013 0.28 0.076 0.0023 0.0023 < 0.001 0.011 0.001 0.0035 0.011 0.0013 0.00001 < 0.0001 < 0.0005
99.2 0.15 0.014 0.013 0.39 0.064 0.0069 0.0012 0.016 0.052 0.02 0.0037 0.003 0.0008 0.0002 < 0.0001 < 0.0005
The beryllium pebbles were saturated at a temperature 873 K for 15 h at a pressure of 4 bar in a gas mixture of 1H2 with 500 appm 3H2. The tritium release measurements were performed using a permanent heating rate ramp of 0.12 K/s up to 1373 K followed by 3 h exposure [9]. During tritium release experiments, the pebbles were aerated by a purge gas mixture of high-purity helium with a small addition of hydrogen (4He + 0.1 vol.% 1H2). The release activity was measured with a proportional counter (PC). Fig. 6a shows tritium release rates for beryllium pebbles with different grain sizes. The pebbles of grain sizes 50–51 mm (Type 2) and 615 mm (Type 3) have single peaks which are practically coincident with each other. The tritium release rate curve for the pebbles of grain size of 13–14 mm (Type 1) differs significantly from that for the pebbles with the larger grain size. The tritium release starts immediately from the beginning of the TPD test, i.e. from even just a small increment above room temperature after heating, and continues with increasing rate until it reaches its maximum peak. Despite the difference in the behavior of the release rate for pebbles with grain size of 13–14 mm, however, the peaks for all three pebble types are grouped closely at temperatures in the region of 1267–1300 K. The highest total tritium release (TTR) was detected for the pebbles with the grain size of 13–14 mm (Fig. 6b). With increase in grain size, the total tritium release decreases. For pebbles with av.g.s ∼13–14 μm the TTR is about 2 and 2.2 times higher than the TTRs from the pebbles with av.g.s. 50–51 μm and 615 μm correspondingly. For comparison, Fig. 7 presents data on tritium release from beryllium pebbles with a diameter of 1 mm produced by REM and FRM which were obtained under exactly the same conditions of tritium/ hydrogen saturation and TDP testing [10]. The major peak for the FRM pebbles is at the lower temperature (1228 K) compared to the REM pebbles (1308–1317 K). The TTR of pebbles with av.g.s ∼13–14 μm produced by powder metallurgy is significantly higher (∼in 7–12 times) than that for the pebbles produced by FRM and REM. TTR of these pebbles with av.g.s ∼615 μm is in ∼3–5 times higher, than that for the pebbles produced by FRM and REM.
Fig. 2. Pebble size distribution, where a) Type 1 (13–14 μm), b) Type 2 (50–51 μm), c) Type 3 (614–615 mm).
types (Table 1); (2) the higher fraction of fine (< 1.10 mm) pebbles (Fig. 4), which increases the ability to fill the space between larger pebbles and (3) the smoother surface of Type 1 pebbles in comparison with Type 2 pebbles, that improves the mutual movement of the pebbles. 4. Tritium release behavior
5. Discussion Investigation of tritium release behavior of beryllium pebbles were performed by temperature-programmed desorption (TPD) tests. Before tests beryllium pebbles were saturated in tritium/hydrogen gas mixture at high-temperature.
The results of this study show that the use of powder metallurgy makes it possible to manufacture beryllium pebbles with a wide range of grain sizes and, in particular, with a fine-grain structure (< 15 μm).
3
Fusion Engineering and Design xxx (xxxx) xxx–xxx
I.B. Kupriyanov et al.
Fig. 3. Microstructure of beryllium pebbles (metallography, polarized light), where a) Type 1 (13–14 μm), b) Type 2 (50–51 μm), c) Type 3 (614–615 mm).
these pebbles after vibration (50 Hz, 2 h) has the necessary value (> 60%) for the Be multiplier material for HCPB. The total tritium release decreases with increase of grain size. For the pebbles with av.g.s ∼13–14 μm the TTR is about 2 and 2.2 times higher than the TTRs for the pebbles with av.g.s. 50–51 μm and 615 μm. The TTR of pebbles with av.g.s ∼13–14 μm produced by powder metallurgy is significantly higher (∼in 7–12 times), than the TTRs for the pebbles produced by FRM and REM. Noteworthy is the fact that the TTR of coarse-grained (∼615 μm) pebbles produced by the powder metallurgy is ∼3–5 times higher than that for the coarse-grained pebbles produced by FRM and REM. This indicates that in addition to the grain size other structural factors also have a significant influence on tritium release, for example, deformation and the surface roughness of pebbles, the state and thickness of the
Table 3 Grain size in different Types of beryllium pebbles. Type of pebbles
Average grain size, μm
Max grain size, μm
Min grain size, μm
Type 1 (< 30 μm) Type 2 (30–60 μm) Type 3 (> 100 μm)
13–14 50–51 614–615
30 64.7 1163.5
5 34.6 216.7
In addition, this method makes it possible to manufacture beryllium pebbles of a variety different dimensions. Despite the fact that the dimensions of the pebbles are ∼20–30% higher than the reference pebble size (1 mm) for the HCPB blanket, nevertheless the packing density of
4
Fusion Engineering and Design xxx (xxxx) xxx–xxx
I.B. Kupriyanov et al.
Fig. 4. Surface morphology of the pebbles (SEM), where a) Type 1 (13–14 μm), b) Type 2 (50–51 μm), c) Type 3 (614–615 mm).
5
Fusion Engineering and Design xxx (xxxx) xxx–xxx
I.B. Kupriyanov et al.
6. Summary The experimental technology of fabrication of beryllium pebbles with fine grain structure by powder metallurgy has been developed and successfully tested in Bochvar Institute. Three experimental batches of Be pebbles with average pebble size of 1.2–1.3 mm and different average grain sizes (∼ 13–14 μm, ∼50–51 μm and 614–615 μm) have been fabricated and characterized. The technology seems is scalable for further industrial production of beryllium pebbles. Microstructure examination shows that surface of the pebbles is characterized by a noticeable micro-toughness that is a result of the intensive deformation processes occurring in a near-surface layer of the pebbles during their fabrication. Packing density at free filling was ∼59, 56.8 and 60.1% of theoretical density of beryllium, for Type 1, Type 2 and Type 3 pebbles, correspondingly. After vibration (50 Hz, 2 h) packing density of pebbles increased up to ∼66.6%, ∼61.4 and 63.8% of theoretical density of beryllium for Type 1, Type 2 and Type 3 correspondingly. Total tritium release (TTR) decreases with increase of grain size. For pebbles with av.g.s ∼13–14 μm the TTR is about 2–2.2 times higher than that for the pebbles with av.g.s. 50–51 μm and 615 μm. The TTR of pebbles with av.g.s ∼13–14 μm produced by powder metallurgy is ∼7–12 times higher than that for the pebbles produced by FRM and REM.
Fig. 5. Packing density (PD) of beryllium pebbles before and after vibration (50 Hz, 2 h).
surface oxide film, internal defects of the pebbles, etc. The results obtained make it possible to conclude that from the point of view of the tritium release, the Be pebbles with smaller grain size (∼13–14 μm) are preferable for use in the ITER (and DEMO) blanket modules.
Fig. 6. Tritium release behavior for beryllium pebbles with different grain sizes, where: a) tritium release rate versus temperature, b) total tritium release.
Fig. 7. Tritium release behavior for 1 mm-diameter beryllium pebbles fabricated by REM and FRM: a) tritium release rate versus temperature, ‘REM (pores)’ indicates the presence of a higher relative density of pores compared to the “REM” batch with a significantly lower amount of pores; b) total tritium release. 6
Fusion Engineering and Design xxx (xxxx) xxx–xxx
I.B. Kupriyanov et al.
The TTR of coarse-grained (∼615 μm) pebbles produced by powder metallurgy is ∼3–5 times higher than that for the pebbles produced by FRM and REM.
[4]
Acknowledgements [5]
This work was funded by European Fusion for Energy (Contract EFDA 05/994).
[6]
References
[7]
[1] T. Iwadachi, N. Sakamoto, K. Nishida, H. Kawamura, Production of various sizes and some properties of beryllium pebbles by the rotating electrode method, Proc. Third IEA Int. Workshop on Beryllium Technology for Fusion, Mito, Japan, JAERI –Conf 98-001, 1997, pp. 33–38 Oct. 22–24. [2] M. Dalle-Donne, D.L. Baldwin, D.S. Gelles, L.R. Greenwood, et al., Behavior of beryllium pebbles under irradiation, Proc. Third IEA Int. Workshop on Beryllium Technology for Fusion, Mito, Japan, JAERI –Conf 98-001, 1997, pp. 296–306 Oct. 22–24. [3] M. Dalle-Donne, A. Goraieb, G. Piazza, G. Sordon, Measurement of thermal
[8]
[9]
[10]
7
conductivity and heat transfer coefficient of a binary bed of beryllium pebbles, Proc. Third IEA Int. Workshop on Beryllium Technology for Fusion, Mito, Japan, JAERI –Conf 98-001, 1997, pp. 17–24 Oct. 22–24. T. Iwadachi, D. Schmidt, H. Kawamura, The distribution of impurities in beryllium pebbles produced by rotating electrode method, Proc. Of the 4th IEA Int. Workshop on Beryllium Technology for Fusion, Karlsruhe, Germany, FZKA 6462, 2000, pp. 15–24 Sept. 15–17, 1999. E. Ishitsuka, H. Kawamura, T. Terai, S. Tanaka, Effect of oxide layer on tritium release from beryllium pebbles, Proc. of 18th SOFT, Fusion Technology 2 (1994) 1345–1348. E. Rabaglino, Helium and Tritium in Neutron Irradiated Beryllium, FZKA Report No.6939, (2006). F. Scaffidi-Argentina, et al., Tritium and helium release from neutron irradiated beryllium pebbeles from EXOTIC-8 irradiation, Fusion Eng. Des. 58–59 (2001) 641–645. I. Kupriyanov, et al., Production and investigation of beryllium pebbles with fine grain structure for the HCPB breeder blanket, Fusion Eng. Des. 87 (2012) 1338–1341. V. Chakin, R. Rolli, P. Vladimirov, A. Moeslang, Tritium release from highly neutron irradiated constrained and unconstrained beryllium pebbles, Fusion Eng. Des. 95 (2015) 59–66. V. Chakin, et al., Tritium release from advanced beryllium materials after loading by tritium/hydrogen gas mixture, Fusion Eng. Des. 107 (2016) 75–81.