IM465 – Conformação Plástica dos Metais – 2º semestre de 2011 – Terceiro Artigo

Exercício de caráter estritamente individual. Envie-o para meu e-mail até 10/11/2011

O artigo a seguir apresenta um estudo sobre a influência da anisotropia sobre as propriedades

mecânicas do aço 14Cr ODS.

Responda as seguintes questões considerando o conteúdo do artigo e das aulas da disciplina:

1) Quais os objetivos dos autores e como são justificados?

2) Por que a anisotropia desse aço extrudado é útil para a aplicação em reatores?

3) Descreva o procedimento experimental empregado e justifique-o em função dos objetivos

apresentados.

4) Como se pode afirmar que o material extrudado está realmente texturizado e, portanto,

anisotrópico?

5) Como o Cr influencia a texturização da liga durante o processamento?

6) Como a dispersão de óxidos influencia a anisotropia da liga estudada?

7) Critique a frase que inicia o item 3.1: “The study of the microstructure of the 14Cr ODS by

means of light and electron microscopy reveals a more elongated microstructure along the

extruded direction than in the bar section, as shown in Figs. 1 and 2”

8) Como a anisotropia afetou as propriedades mecânicas?

9) Como os autores analisaram a anisotropia em função dos modos de fratura? Há como

associar essas conclusões à textura obtida?

Journal of Nuclear Materials xxx (2011) xxx–xxx

Contents lists available at SciVerse ScienceDirect

Journal of Nuclear Materials

journal homepage: www.elsevier .com/ locate / jnucmat

Microstructural anisotropy effect on the mechanical properties of a 14Cr ODS steel

Marta Serrano ⇑, Mercedes Hernández-Mayoral, Andrea García-JuncedaStructural Materials Division, Technology Department, CIEMAT, Avda de la Complutense 22, 28040 Madrid, Spain

a r t i c l e i n f o

Article history:Available online xxxx

0022-3115/$ - see front matter � 2011 Elsevier B.V. Adoi:10.1016/j.jnucmat.2011.08.016

⇑ Corresponding author.E-mail address: marta.serrano@ciemat.es (M. Serra

Please cite this article in press as: M. Serrano e

a b s t r a c t

During the last years in the field of materials for energy production, oxide dispersion strengthened (ODS)alloys have been reintroduced as candidate materials for high temperature and aggressive operating con-ditions. In particular, the Fe-ODS alloys developed in the 1970s for fast reactor cladding are beingimproved under various national and international projects. One of the main open issues in the use ofODS alloys as fuel cladding is their anisotropy, characteristic of extruded alloys. The aim of this workis to investigate the effect of the anisotropy of an extruded bar of a14Cr ODS on the tensile properties.The microstructure in longitudinal direction (L, parallel to the extrusion direction) is characterized bygrains elongated along the extrusion direction, while in the transverse direction (T, perpendicular tothe direction of extrusion) grains are smaller and more equiaxed. This microstructural anisotropy isreflected in a loss of ductility in the transverse direction. Furthermore, after tensile tests different fracturemode are observed depending on the orientation under study.

� 2011 Elsevier B.V. All rights reserved.

1. Introduction

Iron-based oxide dispersion strengthened alloys were consid-ered excellent candidates for fast reactor cladding in the 1970s be-cause of their better swelling resistance and creep behaviour thanferritic/martensitic steels [1–3]. The premier ODS ferritic alloydeveloped and patented by Fischer and fabricated by INCO wasdesignated MA957 (MA – mechanical alloyed), which is a high-chromium ferritic alloy that is mechanically alloyed with Y2O3

[2]. Similar ODS steels where produced by Dour Metal (Belgium)[1], Special Metals and Plansee. These entire suppliers no longerexists and a lack of commercial ODS steels, except small quantitiessupplied now by Dour Metal Sro. (Slovack republic) and KobeSteels (Japan), exists. The rising interest on ODS application inthe energy industry with the lack of industrial ODS steels, makesthe situation of ODS research quite complex. Several barriers tothe practical use of ODS alloys exists as difficulties in joining byconventional fusion techniques, the anisotropy on properties(important in extruded tubes/bar) and the cost.

Despite this limitations, there is currently a resurgence ofindustrial interest in ODS alloys for fossil fuel energy applicationsas high temperature heat exchanger, fuel nozzles and combustorscan for turbines fired with H2 as well as in the ‘‘reborn’’ nuclearindustry as cladding for Generation IV reactors and several struc-tural components in DEMO Fusion reactor [4–7].

In this context, the European Commission has launched a 7thFramework Programme Research Project, named ‘Generation IV

ll rights reserved.

no).

t al., J. Nucl. Mater. (2011), doi

and Transmutation Materials’ (GETMAT) [8]. One of the objectivesof GETMAT project is to develop and characterize Fe–Cr ODS alloys.One of the materials under study is a 14Cr ODS alloy provided byCEA (France).

A microstructural characterisation and tensile tests in longitu-dinal and transverse orientation of this 14Cr ODS have been under-taken at CIEMAT (Spain) in order to assess the effect of theanisotropy.

2. Experimental

2.1. Material and specimens

The material tested is a 14Cr ODS alloy supplied by CEA in theframework of GETMAT project. The CEA material code is J27-M2.The master alloy of composition (wt%) 13.98Cr, 1.03W, 0.39Ti,0.29Mn, 0.32Si, 0.17Ni was mechanically alloyed with 0.3 wt%Y2O3 under hydrogen atmosphere within a vertical attritor. Subse-quently, the material was hot extruded in the form of bars at1100 �C and finally annealed for 1.5 h at 1050 �C.

Microstructural characterisation has been carried out in the as-received condition by optical microscopy (OM), scanning electronmicroscopy (SEM) and transmission electron microscopy (TEM).Specimens for OM and SEM, parallel and transverse to the hot ex-truded direction were ground and polished by means of standard-ised techniques for metallographic examination. A Hitachi SU6600field emission gun scanning electron microscope (FEG-SEM)equipped with an Oxford Instruments HKL NordlysF detector wasused to obtain the electron micrographs and the electron backscat-ter diffraction (EBSD) patterns. All the EBSD acquisitions werecarried out with a step size of 0.05 lm.

:10.1016/j.jnucmat.2011.08.016

Fig. 1. Optical micrographs of the extruded samples (a) longitudinal and (b) transverse.

Fig. 2. Inverse pole figure maps obtained by EBSD on the (a) longitudinal, (b) transverse samples and (c) colour key for the crystallographic orientation. (For interpretation ofthe references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3. Phi(2) = 45� sections of the Euler space obtained for the (a) longitudinal and (b) transverse sample.

2 M. Serrano et al. / Journal of Nuclear Materials xxx (2011) xxx–xxx

Please cite this article in press as: M. Serrano et al., J. Nucl. Mater. (2011), doi:10.1016/j.jnucmat.2011.08.016

Fig. 4. TEM image of the 14Cr ODS alloy (a) in the longitudinal direction, (b) in the transverse direction and (c) EDX analysis of the Cr-rich phase decorating the grainboundary.

M. Serrano et al. / Journal of Nuclear Materials xxx (2011) xxx–xxx 3

TEM thin foils in the form of discs of 3 mm in diameter werealso punched out parallel and transverse to the hot extruded direc-tion. Electrochemical thinning until electron transparency was per-formed in a Tenupol 5 machine. The solution employed was 5%perchloric acid in methanol at�60 �C. The microstructural examin-ations were performed in a JEOL JEM-2010 microscope operating at200 keV. EDX analyses were also carried out by a Link Oxfordmicroanalysis system coupled to the TEM.

Vickers hardness tests were performed on longitudinal andtransverse orientation. The results correspond to the average ofat least five tests.

Tensile specimens were mechanized in longitudinal and trans-verse orientations. The specimens have a dog-bone geometry witha gauge section of 15 � 3 � 2 mm. Tensile tests were performed at22, 400, 600 and 700 �C. Above room temperature, the time ofholding at temperature prior to the start of the test is minimum30 min and the temperature is maintained before and during thetests ±2 �C. The tests were performed under stroke control at a dis-placement rate of 0.1 mm/min (corresponding to a strain rate of1 � 10�4 /s). No extensometer is used due to the small size of thespecimens. The crosshead displacement values were recordedand used to obtain both uniform elongation and strain. Total elon-gation and reduction in area measurements were performed on thebroken specimens. No machine compliance correction has beenmade and thus strain values are only indicative.

Please cite this article in press as: M. Serrano et al., J. Nucl. Mater. (2011), doi

3. Results and discussion

3.1. Microstructure characterisation

The study of the microstructure of the 14Cr ODS by means oflight and electron microscopy reveals a more elongated micro-structure along the extruded direction than in the bar section, asshown in Figs. 1 and 2. Thus, the inverse pole figure maps obtainedfor both sections (Fig. 2) demonstrate that the longitudinal sampleexhibits a great quantity of elongated grains (mean length =855 nm and mean width = 533 nm), whereas the transverse sec-tion presents a relatively high number of equiaxed grains (meanlength = 417 nm and mean width = 394 nm). Vickers hardness isslightly higher in the T orientation (HV0.1 = 392 ± 6) than in the Lorientation (HV0.1 = 362 ± 6), as it could be expected because ofits smaller grain size.

The black lines drawn in Fig. 2 represent grain boundarieswhich rotation angles are above 5�. This figure highlights the exis-tence of a preferential crystallographic orientation of the grainsalong h110i in the extrusion direction. This assumption is con-firmed analysing the orientation distribution function (ODF).Hence, Fig. 3 shows the measured phi (2) = 45� sections of the Eulerspace, using Bunge’s angular system, for both sections to obtain adeeper detail of the microtexture of the material. The results of thisODF study support the preferential orientation of the grains along

:10.1016/j.jnucmat.2011.08.016

Fig. 5. Oxide particles of the 14Cr ODS (a) TEM image and EDX, (b) from a particle containing Y–Ti–O, (c) from a Ti-rich particle containing Al–Y–Si–O and (d) from a particlecontaining Y–Ti–Si–Al–O (Fe and Cr signals are the contribution from the matrix).

4 M. Serrano et al. / Journal of Nuclear Materials xxx (2011) xxx–xxx

h110i. The h110i fibre texture along the extrusion direction is re-ported by other authors [9,10] and it is considered as a main char-acteristic for ODS ferritic steels by Ratti [11].

TEM examinations reveal the microstructural characteristics al-ready observed by OM and SEM. In the longitudinal sample, grainsare elongated along the h110i direction (Fig. 4a), whereas in thetransverse sample, the grains are smaller and they do not exhibitelongated shape (Fig. 4b). Decorating some grain boundaries, aCr-rich phase is observed in the transverse sample, which can bechromium carbides. In addition, this phase contains Ti, Y and W,as can be seen in the EDX analysis of Fig. 4c. Inside the grains, aquite homogeneous distribution of oxide particles is observed, withdifferent sizes and compositions (Fig. 5a). Precipitates containingY–Ti–O are found together with Al-rich and Ti-rich precipitates.

Please cite this article in press as: M. Serrano et al., J. Nucl. Mater. (2011), doi

Additionally, oxides with a mixed composition of Y–Ti–Al–Si arefound. An example of these analyses is shown in Fig. 5b–d. Alumin-ium is not reported in the chemical analysis of the master alloy andthus it is believed that it is an impurity coming from the fabricationprocess. It is worth pointing out that EDX studies showing pureY2O3 particles have not been obtained in the analyses performedso far.

3.2. Tensile tests

Results of tensile tests are shown in Fig. 6, where YS is the 0.2%offset yield strength, UTS is the ultimate tensile strength, UE is theuniform elongation, TE is the total elongation and RA is the reduc-tion in area. All the specimens were broken within the calibrated

:10.1016/j.jnucmat.2011.08.016

Fig. 6. Tensile tests results (a) 0.2% offset yield strength (YS) and ultimate tensilestrength (UTS) for L and T orientation, (b) reduction in area (RA) for L and Torientation and (c) uniform elongation (UE) and total elongation (TE) for L and Torientation.

Fig. 7. Photographs of the broken tensile specimens mechanised

M. Serrano et al. / Journal of Nuclear Materials xxx (2011) xxx–xxx 5

Please cite this article in press as: M. Serrano et al., J. Nucl. Mater. (2011), doi

gauge length. According to Klueh [12] variation in YS and UTS mea-sured by small specimens, was usually well below 10%. So no cor-rections on YS nor in UTS has been made in the present work.

The values of YS and UTS obtained from the tests are similarin both orientations. These results are in very good agreementwith those published by Steckmeyer et al. for the CEA J0514Cr–1W–0.26Ti ODS alloy. However, they are slightly lower thatthose reported by Oksiuta et al. on a 14Cr–2W–0.3Ti [13] andconsiderably lower than values reported by Kim et al. on the14YWT-SM10 alloy [14]. This lower strength could be explainedby an effect of the grain size, especially for the 14YWT wherethe reported grain size was between 100 and 300 nm in the lon-gitudinal orientation, smaller than the grain size of the 14Cr ODSpresented in this paper.

The effect of the orientation on the ductility is larger and morepronounced for the total elongation and reduction in area values,where the loss of reduction in area for the T orientation in respectof the L is around 70% at room temperature and around 50% at700 �C. Nevertheless, the values of uniform elongation are quitehigh in both orientations and the microstructure anisotropy effectis not as marked as for the total elongation values.

It is well known that the elongation values depend on the spec-imen geometry, and thus for comparison from different specimengeometries it is necessary to apply a correction factor due to thedifferent slimness ratios (K = L0/

pS0, where L0 is the gage length

and S0 is the cross-sectional area) [15]. The correction factor usedcomes from the Bertella–Oliver equation, assuming a b value of0.4:

e2

e1¼ K2

K1

� ��b

ð1Þ

where e2 and K2 and e1 and K1 are the respective elongation andslimness ratio values of the different specimens. The slimness ratioof our specimens is 6.12

Tensile properties of similar 14Cr extruded ODS alloys hasbeen reported by Oksiuta et al. [13] and by Kim et al. [14] withspecimens slimness ratios of 9.24 and 7.09 respectively. Applyingthe Bertella–Oliver equation, even for the less ductile orientation(the transverse one), the values of uniform elongation presentedhere are higher that those reported by Oksiuta et al. and Kim etal. [13,14].

The anisotropy observed on grain morphology has a noticeableeffect on the fracture mode. In Fig. 7, the photographs of the brokentensile specimens are shown. A typical flat-face tensile fracture isseen in the longitudinal specimens. On the other hand, a shear-facetensile fracture (45� to the surface) is found in the transverse ori-entation, which seems to indicate a lower ductile behaviour inthe T orientation.

SEM analysis of fracture surface reveals that in the L speci-mens tested at room temperature, secondary cracks propagateparallel to the tensile axis (along the extrusion direction), asshown in Fig. 8. This kind of fracture was also observed in highstrength and ferritic steels extruded or laminated in tensile barstaken parallel to the hot-working direction and was named like‘‘splitting’’ or ‘‘rosette star-type fracture’’ [16,17]. A similar typeof fracture was also reported on ODS alloys [9,14,18] and couldbe attributed to a grain boundary decohesion. By contrast, in

(a) parallel and (b) perpendicular to the extrusion direction.

:10.1016/j.jnucmat.2011.08.016

Fig. 8. Fracture surface of the longitudinal specimen tested at 22 �C, where some secondary cracks parallel to the extrusion direction are seen.

Fig. 9. Fracture surfaces of the tensile specimens tested at different temperatures in both orientations.

6 M. Serrano et al. / Journal of Nuclear Materials xxx (2011) xxx–xxx

the transverse section these cracks have not been observed in thespecimens under study or in the longitudinal section at tempera-tures higher than 22 �C.

In all the temperatures tested, transverse specimens fracturesurfaces are characterized by an elongated structure, while in thelongitudinal specimens a more uniform dimple formation is seen,as presented in Fig. 9. At 700 �C, the fracture surface of the L spec-imens seems to indicate some brittle features and less dimples

Please cite this article in press as: M. Serrano et al., J. Nucl. Mater. (2011), doi

than for lower temperatures. A deeper fractographic examinationof the specimens tested at 700 �C is on-going.

4. Conclusions

This paper presents an study of the anisotropy of an extruded14Cr ODS steel bar, which includes the microstructural character-isation and tensile tests of specimens obtained in the longitudinal

:10.1016/j.jnucmat.2011.08.016

M. Serrano et al. / Journal of Nuclear Materials xxx (2011) xxx–xxx 7

(parallel to the extrusion direction) and transverse (perpendicularto the direction of extrusion) orientations. The main conclusions ofthis study are:

1. Elongated grains along the extrusion direction and small areaswith equiaxed grains were observed in the longitudinal direc-tion. Nevertheless, the grains were more equiaxed and with asmaller mean size in the transverse orientation.

2. A preferential crystallographic orientation of the grains alongh110i in the extrusion direction was seen by EBSD and TEM.

3. The observed microstructural anisotropy seems to exhibit adirect effect on the ductility of the ODS material studied. Itwas found that the transverse orientation was around 50% lessductile than the extruded direction, while yield stress and ulti-mate tensile strength were slightly affected by this anisotropy.

4. The differences on the grain morphology of each orientation alsoappear to affect to the fracture mode. The fracture surface of Tspecimens presented an elongated structure, while in the longi-tudinal specimens a more uniform dimple formation is seen.

5. A rosette star-type fracture was observed in longitudinal spec-imens tested at room temperature, probably due to a grainboundary decohesion.

Acknowledgements

This work is supported by the European Commission under theGrant Agreement ‘Generation IV and Transmutation Materials’(GETMAT) FP7-212175.

Please cite this article in press as: M. Serrano et al., J. Nucl. Mater. (2011), doi

References

[1] J.J. Huet, Powder Metall. 10 (1967) 208.[2] J.J. Fischer, US Patent 4075;010, February 21, 1978.[3] S. Ukai et al., J. Nucl. Mater. 204 (1993) 65–73.[4] B. Kang et al., in: 23rd Annual Conference on Fossil Energy Materials,

Pittsburgh, May 12–14, 2009.[5] S.J. Zinkle et al., Mater. Today 12 (2009) 12–19.[6] Y. de Carlan et al., J. Nucl. Mater. 386–388 (2009) 430–432.[7] A. Jones, in: ODS 2010 Materials Workshop, November 17–18, 2010.[8] Generation IV and Transmutation Materials (GETMAT) – Collaborative Project,

FP7-212175, December 2007.[9] A. Steckmeyer, M. Praud, B. Fournier, J. Malaplate, J. Garnier, J.L. Béchade, I.

Tournié, A. Tancray, A. Bougault, P. Bonnaillie, J. Nucl. Mater. 405 (2010) 95–100.

[10] R. Kasada, S.G. Lee, J. Isselin, J.H. Lee, T. Omura, A. Kimura, T. Okuda, M. Inoue,S. Ukai, S. Ohnuki, T. Fujisawa, F. Abe, J. Nucl. Mater., in press.

[11] M. Ratti, Thèse de doctorat, Institut Polytechnique de Grenoble, 2009.[12] R.L. Klueh, Fusion 2 (3) (1985) 407–416.[13] Z. Oksiuta, P. Olier, Y de Carlan, N. Baluc, J. Nucl. Mater. 393 (2009) 114–119.[14] J.H. Kim, T.S. Byun, D.T. Hoelzer, J. Nucl. Mater. 407 (2010) 143–150.[15] J. Chao, C. Capdevila-Montes, J.L. González-Carrasco, Mater. Sci. Eng. A 515

(2009) 190–198.[16] ASM Handbook, vol. 12, Fractography, ASM International, 1987, pp. 517, ISBN:

978-0-87170-018-6.[17] J. Blach, L. Falat, P. Sevc, Eng. Fail. Anal. 16 (2009) 1397–1403.[18] T.S. Byun, J.H. Kim, J.H. Yoon, D.T. Hoelzer, J. Nucl. Mater. 407 (2010) 78–82.

:10.1016/j.jnucmat.2011.08.016