Scholarly article on topic 'Characterization of a Metastable Austenitic Stainless Steel with Severe Plastic Distortions'

Characterization of a Metastable Austenitic Stainless Steel with Severe Plastic Distortions Academic research paper on "Materials engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Procedia Engineering
OECD Field of science
Keywords
{"deformation induced phase transformation" / "plasticity mode" / "fatigue damage" / "fatigue life"}

Abstract of research paper on Materials engineering, author of scientific article — Zeng Wu, Yuan Huang

Abstract Surface treatment, e.g. shot peening and deep rolling, is popular in aero engine industry, which induces not only high residual stresses but also severe distortions into the subsurface of mechanical parts. In the present work a meta-stable austenitic steel is experimentally investigated by taking into account of martensitic transformations in plastic deformations and fatigue life assessment. The mechanical behavior of the deformed material with phase transformation is described by J2 plasticity combining with the Santacreu model. The plastic strain hardening is decomposed into plastic strain related part and martensitic phase part verified experimentally. It is found that the contribution of the martensitic phase transformation is limited only for large strain beyond 20% in the AISI 304 stainless steel. Experiments confirm, furthermore, that the martensitic phase transformation arises yield strength and improves fatigue limit of the material. However, the severe deformations diminish material's ductility. Further fatigue tests reveal the fatigue life increases significantly, should the applied load not be much higher than the fatigue limit. In this loading region the martensitic phase transformation improve material's fatigue resistance. With the increasing loading amplitude, however, the benefits from the martensitic phase transformation decrease and the plastic deformation may even reduce fatigue life of the material in the LCF region.

Academic research paper on topic "Characterization of a Metastable Austenitic Stainless Steel with Severe Plastic Distortions"

(I)

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Procedía Engineering 99 (2015) 1323 - 1329

Procedía Engineering

www.elsevier.com/locate/procedia

"APISAT2014", 2014 Asia-Pacific International Symposium on Aerospace Technology,

APISAT2014

Characterization of a Metastable Austenitic Stainless Steel with

Severe Plastic Distortions

ZENG Wua, YUAN Huanga,b*

aSchool of Mechanical Engineering, Beijing Institute of Technology, Beijing, 100081, China bSchool of Aerospace Engineering, Tsinghua University, Beijing, 100084, China

Abstract

Surface treatment, e.g. shot peening and deep rolling, is popular in aero engine industry, which induces not only high residual stresses but also severe distortions into the subsurface of mechanical parts. In the present work a meta-stable austenitic steel is experimentally investigated by taking into account of martensitic transformations in plastic deformations and fatigue life assessment. The mechanical behavior of the deformed material with phase transformation is described by J2 plasticity combining with the Santacreu model. The plastic strain hardening is decomposed into plastic strain related part and martensitic phase part verified experimentally. It is found that the contribution of the martensitic phase transformation is limited only for large strain beyond 20% in the AISI 304 stainless steel. Experiments confirm, furthermore, that the martensitic phase transformation arises yield strength and improves fatigue limit of the material. However, the severe deformations diminish material's ductility. Further fatigue tests reveal the fatigue life increases significantly, should the applied load not be much higher than the fatigue limit. In this loading region the martensitic phase transformation improve material's fatigue resistance. With the increasing loading amplitude, however, the benefits from the martensitic phase transformation decrease and the plastic deformation may even reduce fatigue life of the material in the LCF region.

© 2015TheAuthors. Published byElsevierLtd.Thisis an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of Chinese Society of Aeronautics and Astronautics (CSAA) Keywords: deformation induced phase transformation, plasticity mode, fatigue damage, fatigue life

* Corresponding author. Tel.: +86-10-68918528; fax: +86-10-68918528. E-mail address: yuan.huang@tsinghua.edu.cn

1877-7058 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of Chinese Society of Aeronautics and Astronautics (CSAA) doi: 10.1016/j.proeng.2014.12.666

1. Introduction

Surface treatment of critical mechanical parts is an important step manufacturing of life limited parts of aero engine. The improvement of fatigue performance of the mechanical parts is realized due to compressive residual stresses and distortions of the surface material [1]. Quantifying effects of surface treatment in fatigue life assessment requires detailed understanding of the mechanical behavior of the material with severe distortions and variations of the residual stresses in the subsurface material. Many stainless steels, e.g. AISI304, are meta-stable austenitic. The crystallographic structures can transform to martensitic phase under plastic deformations [2-4]. The deformation-induced martensitic transformations have been studied for many years, especially in material science communities on kinetics of transformation [5-7]. Due to martensitic phase the stainless steel behaviors significantly different from the austenitic steel, both in plastic deformation and failure behavior [8-11]. Quantifying effects of surface treatment need detailed understanding of evolution of microstructure and meso-mechanical behavior of the distorted material [5-7]. Especially, phase transformation under severe compressive deformation is less investigated.

The present work dedicates to identify characterization of AISI304 with phase transformation under severe plastic deformations and to clarify effects of the pre-strains to fatigue failure. Based on extensive experiments. The plasticity model taking into account martensitic transformation should be applied to predict fatigue life of the compressive deformed specimens and to quantify effects of the compressive strains to material failure evolution.

2. Representation of material behavior

2.1. Martensitic phase transformation

It is known that austenitic phase of the stainless steel AISI304 can transfer into martensitic phase during severe deformations [12-14]. Experiments demonstrated clearly that the martensitic phase content increases with plastic strain (Fig. 1). The phase transformation changes the microstructure of grains and raises resistance of the dislocation motions. The material with martensitic phase obtains higher strength and less ductility.

TrueStrain (mm/mm)

Fig 1. Evolution of the martensitic phase in AISI304 under both compression and tension. Symbols denote experimental data measured by the ferritescope and the solid lines are predictions from the Santacreu model, Eq. (2).

In the strain-induced phase transformation, the martensitic phase forms only when the material is plastically deformed. The content of the martensitic phase can be expressed as a function of plastic strain. Olson and Cohen [5] proposed a martensitic kinetics law that the increment of martensitic fraction is proportional to the plastic strain. Further experiments reveal that the phase transfor- mation depends additionally on the stress triaxiality, n = oh lae. For introducing a stress-state dependent maximum fraction, Santacreu et al. [6] suggested martensitic evolution equations:

(* _ \m-1

De" ) e

where the maximum fraction of the martensitic content, Xmax, is introduced as the upper limit of the phase transformation and depends on the stress triaxiality. For Xmax = 1 the martensitic phase reaches 100%. D is introduced to consider influence of the stress triaxiality. m should be identified following experimental curves. Since only austenite-to-martensitic transformation is considered, and the reverse transformation is restricted, the total form of the martensitic phase fraction under monotonic loading condition is given as follows:

In the integration the stress triaxiality is assumed to be constant.

The ferritescope [15-18] was fixed on the specimen surfaces and the martensitic phase was measured continuously from its magnetism. The experimental data are shown in Figure 1, in which the martensitic content under compressive and tensile uniaxial loading is plotted as a function of the equivalent true plastic strain.

The experimental data reveal that the evolution of the martensitic phase in the stainless steel AISI304 is determined by the plastic deformation and affected by the stress triaxiality as shown Figure 1. Effects of stress triaxiality in the present work are limited in tension and compression with n=1/3 and -1/3, respectively. Dependence on the stress triaxiality is expressed by the parameter D in the Santacreu model, which reads,

The solid lines in Fig. 1 denote the prediction from the Santacreu model, Eq. (2), with %m!x=1, m=2.16, D0=1.04452 and D1=0.3408. The model agrees with the experimental data, as shown in the figure.

2.2. Plasticity for the austenitic-martensitic steel

The formation of the martensitic phase changes the macroscopic material behavior. To study influence of severe plastic deformations as well as the martensitic phase individually, three kinds of specimens are considered in the present work: The base material, the pre-strained specimen with martensitic phase transformation and the pre-strained specimens without martensitic phase transformation. The last kind of specimens was pre-strained at elevated temperature to avoid martensitic phase. All specimens underwent solution treatment (annealing) before testing.

The static material behavior of the pre-strained specimens is summarized in Fig. 2(a), with different pre-strain levels finished at room temperature. The reduction in the specimen thickness, i.e. pre-strain, varies from 0% (Base material without initial pressing) to 38.6%. The thickness reduction of 38.6% means -48.8% of the logarithmic strain. For the pre-strained specimens, the total equivalent plastic strain sums the initial compressive plastic strain and plastic strain from tests. The true stress versus total equivalent true plastic strain is shown in Fig. 2(a). It is clear that the stress-strain curves depend on the initial pre-strains. Compressive pre-strains show more hardening than the tension loading. Due to cold work the pre-strained specimens were broken with smaller strains.

D = Do + Dn

Equvalent Strain (mm/mm) EquivalentStrain (mm/mm)

(a) Results of pre-strained specimens (b) Results of pre-strained specimens at elevated temperature

Fig 2. Stress-strain curves from tensile tests. The specimens were prepared very differently. The legends show the reduction of the specimen thickness. BM stands for the base material of AISI304. (a) The specimens were pressed in the thickness direction at room temperature. The material deformation is accompanied by martensitic phase transformation. (b) The specimens were compressed at elevated temperature 50°C-100°C, so that the specimen did not contain martensitic phase into the specimen before testing

As known, martensitic phase transformation depends on temperature. With elevating temperature of pre-straining, the phase transformation may be prevented. In Figure 2(b) the tensile tests of compressively pre-strained specimens without martensitic phase transformation are illustrated. In the figure the percentage denotes the pre-strains before testing. The different stress-strain curves represent the hardening purely from initial plastic strain. The difference to the base material curves increases with pre-strain amplitude, which implies effects of the martensitic phase. One may use such experiments to quantify influence of the martensitic phase transformation. The martensitic phase and material distortion contribute the increment of the material resistance.

Plastic Strain

Fig 3. Variations of yield stress as a function of the plastic strain of the base material as well as the pre-strained material without phase transformation. The contribution of martensitic phase is defined as the difference between the two curves based on the Santacreu model, Eq.(2).

The plasticity model for the meta-stable austenitic steel containing martensitic phase transformation has to contain both plastic strain hardening and phase hardening. For instance, the total hardening can be expressed as

k = Hip + H vy, (4)

that is, the deformation resistance is linear super-position of the plastic strain hardening and the martensite transformation hardening. Both plastic strain hardening modulus, H, and phase transformation hardening modulus,

H„ are functions of plastic strain, ep, and martensitic phase fraction, %. For the simplicity, the plastic strain hardening modulus can be represented as classical plasticity,

He = pH о exp (в),

with H0 as plastic hardening factor and в hardening exponent.

The additional contribution from the phase transformation hardening modulus, H„ has to be determined based on experimental observation. In the present work the power-law is assumed as

HX = hXX • (6)

Combining with the Santacreu's model the plastic strain hardening term is determined by different model parameters. Identifying the two sets of material parameters needs material tests under considering influence of martensitic phase and stress triailaity. Note the formulations above are only valid for monotonic loading.

2.3. Identification of the plasticity model

To decompose the hardening of plastic strain from that of martensitic phase, the stress-strain curves in Fig. 2(b) are manipulated and illustrated in Fig. 3. In the figure the yield stress of the material is expressed as function of plastic strain. The base material curve denotes the basic material behavior of AISI304, whereas the curve of the pure plastic strain stands for the stress-plastic strain relation without phase transformation. The discrepancy between the two curves is the contributions of the martensitic phase transformation. The curves in the figure are fittings of the three different stress-plastic strain curves, with identified model parameters in Table 1. In the table k0 denotes the initial yield stress of AISI304.

Table l.Parameters for constitutive model of the stainless steel 304

Martensite transformation parameters Plastic hardening parameters

Xmax=1, m=2.16, D0=1.0445, D1=0.3408, #=1061.3MPa, n=0.9742 ¿0=221.8MPa, в=2.55, #0=907.2MPa,

3. Fatigue assessment of severe distorted materials

Martensitic phase transformation affects plastic behavior of the stainless steel AISI304. Since H in Eq. (4) is positive, the martensitic phase will generally increase the hardening of the material. Due to embrittlement through martensitic phase, however, fracture toughness of AISI304 may decrease with growth of the martensitic content. Effects of plasticstrainandthe martensitic phase to material fatigue behavior have to be considered carefully.

Effects of the pre-strain to fatigue life are studied systematically on the three differently treated materials, as mentioned in the previous section. Both strain and stress-controlled fatigue tests are performed to examine influence of the material cyclic behavior. It is confirmed that the severe plastic strains hardly further change the fatigue life if ep >20%. It implies that one may consider effects of the severe distortion uniformly without taking care of the pre-strain amplitudes.

As known, the base material AISI 304 demonstrated strong cyclic softening, so that the fatigue life from stress-controlled tests is obviously shorter than that of the strain-controlled (Fig. 4(b)). The difference increases with loading amplitude. The results become less sensitive to the test control only for high cycles >105. The fatigue life can essentially be expressed by the Basquin model.

Fig 4. Fatigue life curves of AISI304 material with different pre-strained states and under both stress- and strain-controlled loading conditions.Base material denotes the conventional fatigue test. The tests Pre-strained with martensitetook specimens with martensite, whereas tests Pre-strained without martensite stands for specimens pre-strained without phase transformation. Solid symbols represent tests under stress trolled tests and open symbols for strain controlled. (a) Stress vs. fatigue life; (b) Strain vs. fatigue life.

The pre-strain without phase transformation induces plastic hardening and increases fatigue limit of the material, as shown in Fig. 4(a), however, its effect in finite life region is not obvious. Whereas the S-N curve for the stress-controlled tests provides more or less the same results as the base material, the strain controlled results reveal worse life. In summary the pure severe material distortion would diminish material ductility and so reduce the finite fatigue life.

As shown in the previous section, the martensitic phase will increases material strength significantly, so the fatigue behavior of the material will further be changed if the pre-strained material contains phase transformation. Fig. 4 confirms that the fatigue life curves are hardly influenced by stress or strain controlling. Since the material including high fraction of the martensitic phase becomes little plastified, the material becomes cyclically more stable. The whole load amplitude versus fatigue life curves from both stress and strain-controlled tests can be approached by the Basquin equation since the material becomes brittle. However, the s-N life seems shorter than the base material, whereas S-N life shows significant improvement to the base material. Such difference is simply caused by the high strain hardening of the material after phase transformation.

Due to difference in micro-structures of the material after pre-straining, such non-uniform changes can induce different effects in mechanical design. If the stress of the mechanical part is related to mechanical load directly, the fatigue life of the part will be improved significantly due to the martensitic phase transformation. This is often the case at the high cycle fatigue. Should the local stress be related to the local strain, the improvement from the phase transformation may diminish due to behavior in the s-N curve.

4. Conclusions

In the present work the martensitic phase transformation in stainless steel AISI304 and its effects to material damage are presented. The experiments confirm that martensitic phase transformation in AISI304 can be described by the Santacreu model and shows significant dependence on stress triaxiality. The plasticity model with the martensite transformation is established based on J2 and Santacreu model and identified based on special prepared material tests. The model agrees with experiments well.

It is further shown that the severe material distortions can arise the fatigue limit, but reduce fatigue life in finite fatigue cycle region. The martensitic phase makes material cyclically stable and decreases cyclic softening of the steel AISI304. The tests under stress-controlling display significant enhancement, whereas the strain versus life curve gives even worse results than the base material. This non-uniform result implies complex material behavior under different loading condition. Generally the severe plastic strain can diminish fatigue performance of the material if the load is high.

References

[1] Schulze V, Modern Mechanical Surface Treatment-States, Stability, Effects. Wiley-VCH Verlag GmbH&Co, KGaA. 2006.

[2] V. Mertinger, E. Nagy, F. Tranta, J. Solyom, Strain-induced martensitic transformation in textured austeniticstainless steels. Materials Science and Engineering: A, 481-482 (2008) 718-722.

[3] S.S.M. Tavares, D. Gunderov, V. Stolyarov, J.M. Neto, Phase transformation induced by severe plasticdeformation in the AISI 304L stainless steel. Materials Science and Engineering: A, 358 (2003) 32-36.

[4] S.K. Ghosh, P. Mallick, P.P. Chattopadhyay, Effect of cold deformation on phase evolution and mechanicalproperties in an austenitic stainless steel for structural and safety applications. Journal of Iron and SteelResearch, International, 19 (2012) 63-68.

[5] G.B. Olson, M. Cohen. Kinetics of strain-induced martensitic nucleation. Metallurgical Transactions A, 6(1975) 791-795.

[6] PO Santacreu, JC Glez, G Chinouilh, T Frölich. Behaviour model of austenitic stainless steels for automotive structural parts. Steel Research International, 77(2006) 686-691.

[7] A.M. Beese, D. Mohr. Effects of stress triaxiality and Lode angle on the kinetics of strain induced austenite to martensite transformation. Acta Materialia, 59 (2011) 2589-2600.

[8] J.L. Chaboche, A review of some plasticity and viscoplasticity constitutive theories. International Journal of Plasticity, 24 (2008) 1642-1693.

[9] J.Lubin. Plasticity theory. Pearson Educational Inc., 2006.

[10] C.J. Miao et. al. Investigation of low-cycle fatigue behavior of austenitic stainless steel for cold-stretched pressure vessels. Journal of Zhejiang University- Science A (Applied Physics and Engineering), 14(2013):31-37.

[11] Z. Mei, J.W. Morris, Influence of deformation-induced martensite on fatigue crack propagation in 304-typesteels, Metallurgical Transactions

A, 21 (1990) 3137-3152.

[12] A. Das, S. Sivaprasad, M. Ghosh, P.C. Chakraborti, S. Tarafder, Morphologies and characteristics of deformation induced martensite during tensile deformation of 304LN stainless steel. Materials Science andEngineering: A, 486 (2008) 283-286.

[13] K. Mumtaz, S. Takahashi, J. Echigoya, Y. Kamada, L. F. Zhang, H. Kikuchi, K. Ara, M. Sato, Magneticmeasurements of martensitic transformation in austenitic stainless steel after room temperature rolling. Journal of Materials Science, 39 (2004) 85-97.

[14] W.S. Lee, C.F. Lin, The morphologies and characteristics of impact-induced martensite in 304L stainless steel, Scripta Materialia, 43 (2000) 777-782.

[15] G. Hu, C.C. Xu, J.G.Yuan, Deformation induced martensite transformation and its magnetic memory effectof austenitic 304 stainless steel. Nondestructive Testing, 30(2008) 216-219.

[16] Y.X. Ren, The comparison test of a-phases of austenite and ferrite austenite double phases of stainlesssteel,Physical Testing and Chemical Analysis Part A:Physical Testing, 41(2005) 184-185.

[17] C.C. Xu, X.S.Zhang, G.Hu, Microstructure change of AISI304 stainless steel in the course of cold working.Journal of Beijing University of Chemical Technology (Natural Science Edition), 29(2002) 27-31.

[18] J. Talonen, P. Aspegren, H. Hanninen, Comparison of different methods for measuring strain induced alpha'-martensite content in austenitic steels, Materials Science and Technology, 20 (2004) 1506-1512.