Fracture toughness characterization in the lower transition of neutron irradiated Eurofer97 steel

Fracture toughness characterization in the lower transition of neutron irradiated Eurofer97 steel

Journal of Nuclear Materials 442 (2013) S58–S61 Contents lists available at SciVerse ScienceDirect Journal of Nuclear Materials journal homepage: ww...

947KB Sizes 1 Downloads 43 Views

Journal of Nuclear Materials 442 (2013) S58–S61

Contents lists available at SciVerse ScienceDirect

Journal of Nuclear Materials journal homepage: www.elsevier.com/locate/jnucmat

Fracture toughness characterization in the lower transition of neutron irradiated Eurofer97 steel N. Ilchuk a, P. Spätig a,⇑, G.R. Odette b a b

Fusion Technology-Materials, CRPP EPFL, Association EURATOM-Confédération Suisse, 5232 Villigen PSI, Switzerland Department of Mechanical and Environmental Engineering, University of California Santa Barbara, Santa Barbara, CA 93106-5070, USA

a r t i c l e

i n f o

Article history: Available online 11 January 2013

a b s t r a c t This research investigated the evolution of tensile, hardness, and fracture properties of Eurofer97 tempered martensitic steel following neutron irradiation. The irradiation-hardening was measured with Vickers hardness tests on broken parts of sub-sized compact tension specimens as well as with tensile tests deformed at room temperature. The fracture toughness was measured with pre-cracked sub-sized 0.18T compact tension specimens. Two specimen sets were irradiated up to a nominal dose of about 0.35 dpa at two different temperatures, 423 and 623 K, in the experimental reactor at AEKI-KFKI in Budapest. The median fracture toughness–temperature curve K(T) was characterized in the lower to middle transition region for each irradiation condition using the master-curve method. The irradiation-induced temperature shifts of K(T) were determined by calculating the reference temperature T0 at which the median toughness is 100 MPa m1/2. A significantly larger shift was determined for Eurofer97 irradiated at 423 K than at 623 K. Indeed, an upper shift of 98 K was found for the 423 K irradiation while only 50 K was measured for the 623 K. On the one hand, that observation reflects the difference in the irradiation-hardening following those two irradiation temperatures. On the other hand, when compared with other published data, the DT0 shift at 623 K irradiation was found to be greater than expected for the corresponding irradiation-hardening. Thus, it was suggested that non-hardening embrittlement mechanisms start to operate around 623 K. Ó 2013 Elsevier B.V. All rights reserved.

1. Introduction Reduced activation ferritic–martensitic (RAFM) steels are leading candidates as structural materials for the future fusion reactors due to their good physical and mechanical properties, swelling resistance in 573–823 K operating temperature window, and ability to meet low activation waste requirements [1]. Eurofer97 is a RAFM steel chosen as a reference structural material for the test blanket modules (TBMs) that will be tested in ITER [2]. A typical consequence of high energy neutron irradiations in fusion reactor is the degradation of mechanical properties such as irradiation-induced hardening leading to embrittlement and loss of fracture toughness [3,4]. Although the fracture toughness database for unirradiated Eurofer97 is quite large [5,6], the development of a high quality database on the effects of irradiation on the constitutive and fracture properties of irradiated material for different irradiation temperatures and doses remains an important objective. Thus, considerable efforts have been devoted on characterization of post-irradiation mechanical and micro-structural properties of Eurofer97. We emphasize that the embrittlement ef⇑ Corresponding author. Address: CRPP/EPFL, ODGA-C109A, 5232 Villigen PSI, Switzerland. Tel.: +41 56 310 29 34; fax: +41 56 310 45 29. E-mail address: [email protected] (P. Spätig). 0022-3115/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jnucmat.2013.01.002

fects observed at irradiation temperatures lower than 673–723 K stem from irradiation-induced hardening [7], whereas at higher irradiation temperatures where irradiation-induced softening occurs, the fracture properties degradations are associated with non-hardening embrittlement (NHE) contributions [8]. The goal of this paper is to present new results of mechanical characterization of Eurofer97 after two irradiations in experimental reactor at AEKI-KFKI in Budapest to 2.5  1020 n/cm2 (0.33 dpa) at 423 K and to 2.8  1020 n/cm2 (0.37 dpa) at 623 K. In particular, irradiation-induced hardening and embrittlement, measured by a shift of the reference temperature T0 of the master-curve method, are reported.

2. Material and experimental procedures 2.1. Material The material used in this research was the reduced activation Eurofer97 steel, heat E83697, 25 mm-thick plate, produced by Böhler AG. The chemical composition (wt%) was Fe–8.9Cr–0.12C– 0.46Mn–1.07W–0.2V–0.15Ta. The heat-treatment was 0.5 h at 1253 K for 0.5 h and tempering at 1033 K for 1.5 h. This steel was fully martensitic after quenching.

N. Ilchuk et al. / Journal of Nuclear Materials 442 (2013) S58–S61

the hot cells of the Paul Scherrer Institute hot-laboratory for performing mechanical tests.

2.2. Mechanical testing The static fracture toughness data reported here were obtained with fatigue pre-cracked compact tension C(T) specimens. Subsized 0.18T C(T) specimens were used with thickness (crack front) (B) equal to 4.5 mm with width (W) equal to 2B and the crack length a to specimen width W ratio (a/W) was about 0.5. The fracture toughness tests were performed over the temperature range (153–273 K). Temperature control was provided by a PID controller equipped with a regulated N2 gas flow. The results of the fracture toughness tests were evaluated using the ASTM E1921 standard in terms of KJc, an elastic–plastic equivalent stress intensity factor derived from the value of the Jc integral at the onset of cleavage fracture, Jc [9].

K Jc ¼

qffiffiffiffiffiffiffiffi J c E0

ð1Þ

where E0 is the plane strain Young’s modulus. The reference temperature (T0) was determined in accordance with ASTM E1921-03 standard [9] (master-curve method). However, slight modifications of the master-curve shape were considered in this work as proposed by Mueller et al. [5]. As demonstrated in [5], the modified-master curve for Eurofer97 reads:

K med;Jc ¼ 12 þ 88 expð0:019ðT  T o ÞÞ

ð2Þ

This modified master curve allows determining To (at which the median toughness of 1T C(T) specimens is 100 MPa m1/2) with data obtained in the very low part of the transition and with sub-sized specimens. The crack front length adjustments, from 0.18T to 1T size, were done as per ASTM standard as:

K B2 ¼ K min þ ½K B1  K min 

 1=4 B1 B2

ð3Þ

Note that in this study we considered a KJc,limit calculated with an M factor equal to 80 instead of 30 as recommended in the ASTM-E1921 standard. The justification for this modification is given in [5].

K Jc

lim it

¼

S59

rffiffiffiffiffiffiffiffiffiffiffiffi E 0 b ry M

ð4Þ

Tensile tests have been conducted at room temperature on irradiated small flat specimens. The tensile tests were carried out with a screw-driven Zwick 010 machine at an imposed nominal strainrate of 1.5  103 s1. The elongation of specimens was deduced from the displacement of the machine crosshead, measured using a linear variable differential transformer with compliance correction. The stresses and strains reported hereafter are expressed in engineering units. Hardness tests up to HV10 were performed on both as-received and irradiated Eurofer97 samples using a hardness tester equipped with a Vickers indenter tip. The indentations on the irradiated specimens were done on the 0.18T C(T) actual specimens.

2.4. Finite element (FE) model Significant loss of the uniform tensile elongation after irradiation at 423 K did not allow direct evaluation of true stress true strain curve beyond 0.5% plastic strain. Finite element (FE) modeling was performed, using ABAQUS 6.10-3, in order to estimate the average flow stress of Eurofer97 irradiated at 423 K to 0.33 dpa. In general, for the stress analysis problems, ABAQUS uses incremental J2 plastic flow theory and the Newton’s method as a numerical technique for solving the nonlinear equilibrium equations. Required inputs to the ABAQUS code include the Young’s modulus (E), Poisson’s ratio (v), and the flow stress as a function of the effective plastic strain after the yield. Fig. 1 shows a three dimensional (3D) elastic–plastic FE model of small flat tensile specimen used in this study. The specimen instance was meshed with 26796 linear hexahedral elements of the type 8-node linear brick, reduced integration. The contact between the specimen and the two pins simulated as analytical rigid bodies was assumed frictionless. The required r(e) curve of irradiated material was obtained by iteratively modifying trial r(e) input function until the model output reproduced the experimental engineering stress–strain s(e) curve (or in other words the load–displacement curve); this work followed previous tests conducted by Yamamoto et al. [10].

3. Results and discussion 3.1. Tensile properties Fig. 2 represents tensile data of Eurofer97 irradiated at 423 K and 623 K to 0.33 dpa and 0.37 dpa, respectively. As a consequence of neutron irradiation, a strong increase of yield stress (Dry = 235 MPa) and a loss of the uniform elongation (Deu = 5.9%) were observed for irradiation at 423 K. Less pronounced effect is shown for Tirr = 623 K, where the engineering stress–strain curve is very close to the unirradiated one, with Dry = 35 MPa and Deu = 0.8%.

2.3. Neutron irradiation conditions Two different irradiations of subsized 0.18T C(T) fatigue precracked fracture toughness and small flat tensile specimens were performed in the BAGIRA rig of the experimental reactor at AEKIKFKI in Budapest, Hungary. The first irradiation, up to a fluence of 2.5  1020 n/cm2 (E > 1.0 MeV; about 0.33 ± 0.05 dpa), was carried out at a temperature of about 423 K. The second irradiation, up to a similar fluence of 2.8  1020 n/cm2 (E > 1.0 MeV; about 0.37 ± 0.04 dpa), was performed at a higher temperature of 623 K. After the irradiations, the specimens were transported to

Fig. 1. 3D finite element model of the flat tensile specimen.

S60

N. Ilchuk et al. / Journal of Nuclear Materials 442 (2013) S58–S61

Fig. 2. Tensile properties of Eurofer97 measured in the unirradiated and irradiated conditions.

Fig. 4. Fracture toughness results and corresponding master-curves in the unirradiated and irradiated conditions.

3.3. DT0 shift and irradiation correlations

Fig. 3. Vickers hardness distribution of Eurofer97 measured in the unirradiated and irradiated conditions.

3.2. Hardness and fracture toughness data Fig. 3 shows Vickers hardness values for Eurofer97 after the two irradiations as well as for the unirradiated material obtained with an applied load of 10 kgf. The increase of the Vickers hardness value, DHV10 for 423 K the irradiation, was 12% and no observable change in hardness was observed for the 623 K irradiation. The evolution of the hardness values appears qualitatively consistent with that of the flow stress. The increase of the average flow stress (over 10% of plastic strain) following 623 K irradiation is very small, less than 3% of the average flow stress of the unirradiated material, which, taking into account the error bars on the data, makes it difficult to detect with Vickers hardness tests. On the contrary, the marked increase of the flow stress after the 423 K is reflected by a similar increase of the hardness. Fig. 4 shows the fracture toughness data of Eurofer97 steel obtained with unirradiated and irradiated 0.18T C(T) specimens after the two irradiation conditions. The reference temperature T0 was determined following the general procedure of the ASTM-E1921 standard but with the master-curve shape and M factor modifications discussed in Section 2.2. The irradiation-induced DT0 shifts of median fracture toughness–temperature curves were equal to 98 K and 50 K for the 423 K and 623 K irradiation conditions respectively. Qualitatively, the difference in DT0 shifts between the two irradiations is consistent with the measured difference in the irradiation-hardening (determined either by the flow stress or the Vickers hardness).

In a study on the relation between irradiation-induced reference temperature shift DT0 of the master-curve and increase of the flow stress of reactor pressure vessel and tempered martensitic steels, He et al. [11] suggested that a universal relation exists between DT0 and Drfl averaged over a certain range of e, . A good correlation was found for e = 10%, yielding the following relation: C 00 = DT0/ 10%  0.7 K/MPa, which is consistent with the relation obtained for reactor pressure vessel steels. In this work, the coefficient C 00 , as defined above, was estimated for the two irradiations. For the 623 K irradiation, the determination of the average flow stress was straightforward, the tensile exhibits a large amount of plasticity. On the contrary, the abrupt loss of the uniform tensile elongation after irradiation at 423 K did not allow direct evaluation of true stress true strain curve beyond 0.5% plastic strain. Thus, FE modeling was employed for the estimation of . The trial r(e) input function was modified until the model output reproduced the experimental s(e) curve. Fig. 5 shows the experimental engineering stress–strain curve, the fitted true stress–strain curve used to reconstruct the experimental engineering, and the calculated engineering stress–strain curve. A very good agreement between the experimental and calculated curves was obtained. Next was determined from the fitted true stress–strain curve. The results of the correlation between DT0,

Fig. 5. Engineering stress–strain curves (experiment and model) and fitted true stress–strain curve (model input), Tirr = 423 K.

N. Ilchuk et al. / Journal of Nuclear Materials 442 (2013) S58–S61 Table 1 Summary of the irradiation-hardening and DT0 shift of the mastercurve. Tirr = 423 K, 0.33 dpa

Tirr = 623 K, 0.37 dpa

DT0 = 98 K DT0 = 0.72 D10% DT0 = 0.42 Dry DHV10 = 12%

DT0 = 50 K DT0 = 2.5 D10% DT0 = 1.43 Dry No difference

Dry, 10%, and DHV10 for the two irradiations are summarized in Table 1. C 00 for the 423 K irradiation is fully consistent with previous published data on tempered martensitic steels irradiated in the irradiation-hardening regime. However, C 00 for the 623 K irradiation appears much larger than expected, because C 00 is equal to 2.5. Considering C 00 = 0.7 K/MPa as a reference value. The expected DT0 shift after the 623 K irradiation should be about 15 K, rather than the 50 K determined, meaning that the difference in about 35 K results from non-hardening embrittlement. This indicates that those mechanisms come into play already at 623 K and that they act synergistically with hardening mechanisms. Several factors can be used to explain non-hardening embrittlement (e.g. segregation of impurities on grain boundaries, irradiation-accelerated precipitate/particle coarsening, laves phase precipitation at grain boundaries [7]). Works is in progress to gain insight into the modifications of the microstructures following irradiation at 623 K. 4. Conclusions Following neutron irradiations on the Eurofer97 steel at a nominal dose of 0.35 dpa at 423 K and 623 K, the irradiation hardening and shift of the reference temperature T0 of the toughness temper-

S61

ature master-curve were determined. The irradiation-hardening was quantified by the increase of the yield stress, the increase of the average flow stress (over 10% of strain) and Vickers hardness. Owing to the very low uniform elongation, FE simulations were run to determine the average flow stress of the tensile specimen irradiated at 423 K. The degradation of the fracture properties were quantified by the DT0 shift of the reference temperature of the master-curve. After the 423 K irradiation, the ratio C 00 = DT0/10% was found to be in good agreement (C 00 = 0.72) with the existing data on similar steels. However, the coefficient C 00 measured for the 623 K irradiation (C 00 = 2.5) was much greater. This result suggests that nonhardening embrittlement processes are already operating at 623 K. References [1] H. Tanigawa, K. Shiba, A. Möslang, R.E. Stoller, R. Lindau, M.A. Sokolov, G.R. Odette, R.J. Kurtz, S. Jitsukawa, J. Nucl. Mater. 417 (2011) 9. [2] L.V. Boccaccini, A. Aiello, O. Bede, F. Cismondi, L. Kosek, T. Ilkei, J.-F. Salavy, P. Sardain, L. Sedano, Fusion Eng. Des. 86 (2011) 478–483. [3] G.R. Odette, T. Yamamoto, H.J. Rathbun, M.Y. He, M.L. Hribernik, J.W. Rensman, J. Nucl. Mater. 323 (2003) 313. [4] E. Lucon, J. Nucl. Mater. 376–370 (2007) 575–580. [5] P. Mueller, P. Spätig, R. Bonadé, G.R. Odette, D. Gragg, J. Nucl. Mater. 386–388 (2009) 323. [6] P. Mueller, P. Spätig, J. Nucl. Mater. 389 (2009) 374. [7] T. Yamamoto, G.R. Odette, H. Kishimoto, J.-W. Rensman, P. Miao, J. Nucl. Mater. 356 (2006) 27. [8] R.L. Klueh, K. Shiba, M.A. Sokolov, J. Nucl. Mater. 377 (2008) 427. [9] Standard Test Method for Determination of Reference Temperature, T0, for Ferritic Steels in the Transition Range, E1921, Annual Book of ASTM Standards 2004, vol. 03.01, ASTM International, 2004. [10] T. Yamamoto, G.R. Odette, M.A. Sokolov, J. Nucl. Mater. 417 (2011) 115. [11] M.Y. He, G.R. Odette, Y. Yamamoto, D. Klingensmith, J. Nucl. Mater. 367–370 (2007) 556.