Effect of temperature on mechanical response of austenitic materials

(1)

Effect of Temperature on

Mechanical Response of

Austenitic Materials

Mattias Calmunger

LIU-IEI-TEK-A–11/01236–SE

Department of Management and Engineering, Division of Engineering Materials Linköping University, 581 83, Linköping, Sweden

http://www.liu.se Linköping, December 2011

(2)

Distributed by: Linköping University

Department of Management and Engineering 581 83, Linköping, Sweden

(3)

Abstract

Global increase in energy consumption and global warming require more energy production but less CO2 emission. Increase in efficiency of energy

production is an effective way for this purpose. This can be reached by increasing boiler temperature and pressure in a biomass power plant. By increasing material temperature 50 ℃, the efficiency in biomass power plants can be increased significantly and the CO2 emission can be greatly reduced.

However, the materials used for future biomass power plants with higher temperature require improved properties. Austenitic stainless steels are used in most biomass power plants. In austenitic stainless steels a phenomenon called dynamic strain aging (DSA), can occur in the operating temperature range for biomass power plants. DSA is an effect of interaction between mov-ing dislocations and solute atoms and occurs durmov-ing deformation at certain temperatures. An investigation of DSA influences on ductility in austenitic stainless steels and nickel base alloys have been done. Tensile tests at room temperature up to 700 ℃ and scanning electron microscope investigations have been used. Tensile tests revealed that ductility increases with increased temperature for some materials when for others the ductility decreases. This is, probably due to formation of twins. Increased stacking fault energy (SFE) gives increased amount of twins and high nickel content gives a higher SFE. Deformation mechanisms observed in the microstructure are glide bands (or deformations band), twins, dislocation cells and shear bands. Damage due to DSA can probably be related to intersection between glide bands or twins, see figure 6 a). Broken particles and voids are damage mechanisms observed in the microstructure.

(4)

(5)

Nomenclature

ASS austenitic stainless steel BCC body centered cubic DC dislocation cell DSA dynamic strain aging

EBSD electron backscatter diffraction ECCI electron channeling contrast imaging FCC face centered cubic

PLC Portevin-LeChatelier RT room temperature

SEM scanning electron microscopy (or microscope) SFE stacking fault energy

SSA static strain aging

STEM transmission electron microscopy in a SEM TEM transmission electron microscopy (or microscope)

(8)

(9)

1

Introduction

1.1 Background

Global increase in energy consumption and global warming require more energy production but less CO2 emission. Increase in efficiency of energy

production is an effective way for this purpose. This can be reached by increasing boiler temperature and pressure in a biomass power plant. By increasing material temperature 50 ℃, the efficiency in biomass power plants can be increased significantly and the CO2 emission can be greatly reduced.

However, the materials used for future biomass power plants with higher temperature require improved properties. Austenitic stainless steels (ASS) are used in most biomass power plants. Strain aging is a phenomenon of the interaction between interstitial or substitutional atoms and dislocations in steels or other metals. There are two types of strain aging: static strain aging (SSA) and dynamic strain aging (DSA) [1–5]. For boiler tube material used in biomass power plant, the temperature range is in the region where DSA will occur in some austenitic steels and nickel base alloy [1, 6]. DSA may affect the high temperature behaviors.

1.2 Strain aging

The pinning of dislocations produced by straining and the associated return of discontinuous yielding, is referred to as strain aging [5].

1.2.1 Types of strain aging

As mentioned before usually strain aging are divided into two types, SSA and DSA, which can occur during or after deformation. The common features

(10)

CHAPTER 1. INTRODUCTION

are that they hardens the material and can also increase the ultimate ten-sile strength, raise the ductile-to-brittle transition temperature and decrease the ductility [1–4, 6]. For steels the responsible elements for strain aging in general are carbon and nitrogen. This is because they have relatively rapid diffusion in steels, even at fairly low temperatures, 200 ℃ to 400 ℃ [1–4]. The level of strengthening depends on the aging time and on the aging temper-ature, this because the diffusivities are a strong function of temperature [1]. It becomes more complex when the temperature reach above 800 ℃ because of phenomena such as recovery and recrystallization. This phenomena will change the mechanical behavior and decrease the occurrence of strain aging because the recovery and the recrystallization annihilate dislocations [3].

1.2.2 Static strain aging

SSA occurs after that a material has undergone plastic deformation and the material is aged for a period of time. The cause of SSA is the relocation of solute atoms to the stationary dislocations and their effective pinning of dislocations. The elements responsible for SSA are in general carbon and nitrogen for steels. It is possible for carbon and nitrogen to produce aging effects at relatively short holding times and low temperatures [1, 3]. This is because these interstitial elements have considerably higher diffusivities than the substitutional elements. Nitrogen has a higher solubility and diffusion coefficient in steels, which gives a more distinctive effect in strain aging [1]. During cold working the SSA can lead to flow localization [3].

1.2.3 Dynamic strain aging

DSA is an effect from interaction between the moving dislocations and solute atoms and occurs during deformation. The appearance of DSA are strongly depending on temperature and strain rate, if the temperature increase or the strain rate decrease the DSA intensifies. DSA occur when the solute atoms obtain enough mobility to keep up with the moving dislocations and form so-lute atmospheres [1–3, 5, 7]. According to Prasad Reddy et al. [7] DSA occur either during the viscous glide type dislocation motion or during the period of their temporary stay at local obstacles in the glide plane. This consequently increases the resistance to the dislocation motion, and contributes to matrix hardening. Therefore an increase in the flow stress is required to maintain the imposed strain rate, either to unlock the dislocations from the obstacles or to generate new dislocations. Both of these results in an increase of the mobile dislocation density [7]. A typical character of DSA is the formation of serrated yielding. This is often called the Portevin-LeChatelier (PLC) effect. 4

(11)

1.3. AUSTENITIC STAINLESS STEELS

DSA can also lead to an increase in flow stress, work hardening rate and most important a negative strain rate sensitivity [1–3, 5]. Y.B. Chun et. al. deduce that an increase in twinning activity with strain rate is responsible for the strain rate sensitivity for Mg alloy AZ31 [8]. DSA influence on ductility depends on the alloy composition [2, 6].

1.2.4 Portevin-LeChatelier effect

The Portevin-LeChateliers (PLC) effect, also called the jerky flow occurs with DSA and is created by the pinning and unpinning of dislocations. PLC effect is recognized in stress and strain curves because of its characteristic serrated look [1, 6, 9, 10]. There are different types of PLC effects and they are divided from their look [1, 6, 9]. They are called type A, B, C, D and E serration and they can appear simultaneously. Type A appear from one end of the specimen gauge length to the other and arise from the propagation of Lüders bands [1, 11]. These are considered as locking serrations, they abrupt rise and then drop to a stress level below the general level. They appear at high temperatures or high strain rates in the DSA regime [11]. Type B occur when the Lüders band front has an irregular movement and is characterize by small oscillations about the general level of the curve [1, 11]. They can develop from A-type at higher strain values or start immediately with plastic flow at higher temperatures and lower strain rates then in the case of A-type [11]. Type C leads to unlocking serration which is when the curve abrupt drops below the general stress level. It appears when many Lüders bands form at various locations on the gauge length [1, 11]. They appear at higher temperatures and lower strain rates then A- and B-type [12]. Type D are characterized by plateaus on the curve, which appear due to shear band propagation with no evidence of work hardening [1, 11]. They can also appear mixed with the B-type [11]. Type E are less structured and occur often after type A at high strain values, therefore harder to recognize [1, 11]. The E-type do not result in work hardening [11]. Types A, B and E occur at lower temperatures and high strain rates, while types C and D generally appear at higher temperatures [1, 11].

1.3 Austenitic stainless steels

The most commonly used stainless steel is ASS. This is because it offers ex-cellent properties, like superior corrosion resistance compared to both ferritic and martensitic stainless steels [13].

(12)

1.3.1 What are austenitic stainless steels

The ASS offers excellent ductility, formability, corrosion and creep resistance [6, 13, 14]. They have greater heat capacity and thermal expansion, with lower thermal conductivity than other stainless steels [13]. These materials are strengthened by work-hardening and cannot be hardened by heat treat-ment [5, 13, 14]. ASS exhibit no yield point, unlike martensitic and ferritic stainless steels. At low temperatures they are not subjected to an ductile to brittle transition and they have high toughness down to cryogenic temper-atures [13]. Alloys containing typically 18 wt.% Cr and 8 wt.% Ni are fully austenitic from well below RT to melting temperature [5]. They are non-ferromagnetic [5, 13, 14], but after machining or that they been worked they may become slightly magnetic [13].

1.3.2 Main alloying elements

The main alloying elements in ASS are chromium, nickel, manganese, molyb-denum, titanium, niobium, carbon and nitrogen.

To obtain corrosion resistance and other characteristics typically for stain-less steels in iron alloys an amount of 12 wt.% or more of chromium is required [5, 13]. Chromium reacts rapidly with oxygen which creates a protective layer of chromium oxide on the surface. If the oxide layer get damaged it self-repairs because of the rapidly reaction between chromium and oxygen [13].

The alloy element nickel stabilizes the face centred cubic (FCC) structure in iron. Nickel increases the size of the austenitic field, while nearly elimi-nating body centred cubic (BCC) ferrite structure from the iron-chromium-carbon alloys. Together with chromium it produce high-temperature strength and scaling resistance [5].

Manganese form austenite and can be used to replace nickel. Manganese improves the solubility of nitrogen.

Molybdenum improves both the local and the general corrosion resistance [5, 13]. Molybdenum are a ferrite stabilizer and must therefore be balanced with austenitic stabilizers to maintain the austenitic structure [13].

Titanium are used to stabilize stainless steel against intergranular corro-sion, if the carbon content is at a high level. Titanium reacts more easily with carbon than chromium does. The titanium carbides are formed in preference to chromium carbides and thus localised reduction of chromium is prevented. Titanium is a ferrite stabilizer [5, 13].

Niobium have the similar affect as titanium, it is a ferrite stabilizer and it can be used for intergranular corrosion resistance. It also, similar to titanium, 6

(13)

1.4. STACKING FAULT ENERGY

creates carbides more easily than chromium. Because it have greater affinity with carbon than chromium [13]. Niobium can reduce the creep rate and shortening the tertiary creep stage [15].

Carbon is an austenitic stabilizer. But it have an negative affect on corrosion resistance, because of the formation of chromium carbides [13]. If the carbon content is below about 0.03 wt.%, the carbides do not form and the steel is virtually all austenitic at room temperature (RT) [5].

1.3.3 Dynamic stain aging in austenitic stainless steels

DSA causes strengthening in a specific temperature range which depends on the strain rate, if the strain rate increases the temperature range increases [1, 3]. At or near RT the DSA also can cause negative strain rate sensitivity. The strain rate sensitivity then decreases with increasing temperature until it reaches 950 ℃ where it sharply climbs until it reaches 1100 ℃. It never becomes negative as in the lower temperature range [3]. In ASS and Ni-base alloys a minor content of interstitial atoms in the solid solution, as carbon and nitrogen, results in DSA above 200 ℃ [11]. DSA causes an increased rate of dislocation multiplication and delay in recovery of dislocation structure. Further more DSA promotes an increased propensity towards uniform distri-bution of dislocations rather than cellular structures [16]. In type 316L ASS the dislocation distribution are uniform in the beginning of deformation and becomes cellular at higher strain values [11].

1.4 Stacking fault energy

Stacking fault energy (SFE) is the energy per unit area of the planar defect between the two partial dislocations [17, 18]. The magnitude of the SFE is an important parameter for high-temperature materials. This is because screw dislocations, dissociated into their partial dislocations, must constrict locally if they are to cross-slip. This is an important mechanism when obstacles as precipitates are circumvented [17, 18]. Materials with low SFE have widely separated partial dislocations, this type of dislocation movement is called planar slip. Materials with high SFE have less separated partial dislocations, which makes cross-slip more easy and this dislocation movement is called wavy glide [17].

(14)

1.5 Slip and twinning

Austenitic materials have face centered cubic (FCC) crystals, which has twelve slip systems, three slip directions at each of the four slip planes. These slips most often occur on the {111} octahedral planes and in the h110i di-rections that are parallel to cubic face diagonals [14, 17]. Of the twelve slip systems there are five that are independent. Independent slip system is de-fined as one producing a crystal shape change that cannot be reproduced by any combination of other slip systems. Materials ability to withstand a gen-eral homogeneous strain involving an arbitrary shape change of the crystal affects the ductility. Plastic deformation in a single crystal begins when the shear stress acting on the incipient slip plane and in the slip direction reaches a critical values. Planar slip and wavy slip are two different dislocation move-ments and are depending on the movement of dislocation. Planar slip can be seen on a polished surface when the slip offset take on a straight pattern. This is because the cross-slip of an extended screw dislocation around obsta-cles needs to be thermally activated. When cross-slip is easy the slip offset on a polished surface take on a wavy pattern, this is called wavy slip [17].

Twin modes can be activated to bring an arbitrary shape change, because the shape change needs some additional deformation mechanism. The shape change caused by twins results from atom movements taking place on all planes. The twinning displacement in a cubic lattice is a rotation of the lat-tice such that the atom position in the twin represent a mirror image of those in the untwined material. There are different types of twins, deformation and annealing twins. Deformation twins are generated as a result of plastic de-formation. Annealing twins are formed during prior heat treatment. The formation of annealing twins could be a result of the interaction between packing sequence defects in the original grains and the emerging grains. An-nealing twins appear more often in materials with low stacking fault energy, because of the higher probability of prior defects in the original grains. The stress needed to initiate a twin is greater than the stress to propagate a pre-existent twin. The growth of a twin exhibits smooth loading behavior, while the nucleation of a twin is associated with a sudden load drop [14, 17].

In FCC metals slip are more favored than twin induced deformation. This is due to the many close-packed planes that the partial dislocations can move in [17].

(15)

1.6. PURPOSES OF THE STUDY

1.6 Purposes of the study

To increase the efficiency in energy production the material temperature in boilers for biomass power plants are proposed to be increased. This creates a need for materials with good high temperature properties. Today ASS are used in most biomass power plants. The phenomenon DSA can appear in ASS and change the mechanical responses of the material. DSA is intensified with increased temperature in the operating temperature range for biomass power plants [1–3, 5, 7]. The purpose of this study is to evaluate the effect of increased temperature on mechanical response of austenitic materials.

(16)

(17)

2

Experimental

2.1 Materials

The materials that are used in this study are Sanicro 25, Sanicro 28, AISI 310, AISI 310M, AISI 316L, Alloy 800HT and Alloy 617. They are austenitic stainless steels and nickel base alloys and their composition are listed in table 1. The materials have similar alloy content but the amount of the alloying elements are different. They are all relatively stable because they all contain a great amount of austenitic stabilizers as nickel. All materials have been heat treated by solution treatment and then water quenched, see table 2.

2.1.1 Sample design

The tensile specimens were manufactured according to the standard SS-112113 (tensile test pieces for screwed holders - round test pieces) with a diameter of 5mm and gauge length of 50mm.

2.2 Experimental methods

To be able to find and understand the mechanisms of DSA and its influence on ductility of austenitic materials. The experimental method is to find the typically DSA sign, the PLC effect, which can be done with tensile testing. With the data from the tensile tests the engineering stress and engineering strain curve can be plotted where the PLC effect can be discovered. Ev-ery stress-strain curve throughout this thesis would be an engineering stress and engineering strain curve. After discovering DSA the microstructure can

(18)

CHAPTER 2. EXPERIMENTAL Tabl e 1: Comp osition of austenitic materials Name C Si Mn P S Cr Ni Mo Cu W Co Nb N Ti A l V F e Sanicro 25 0.067 0.25 0.47 0.018 0.006 22.33 24,91 0.24 2.9 5 3.37 1.44 0.52 0.236 0.005 0.031 0.046 b Sanicro 28 0.019 0.43 1.83 0.018 0. 0005 27.02 30.76 3.39 0.9 0.02 0.088 -0.047 0.003 -0.054 b AISI 310 0.046 0.55 0. 84 0.015 0.001 25.43 19.21 0.11 0.08 -0.04 0.001 -b AISI 310 M 0.052 0.15 1.29 0.018 0.01 25.3 19.73 0.18 0.09 -0.27 0.51 0.24 0.002 -b AISI 316 L 0.04 0.4 1.7 0.03 0.015 17.0 12.0 2.6 -b Allo y 800 HT 0.063 0.71 0.5 0.009 0.001 20.32 30.06 0.005 0.053 0.01 0.031 0.01 0.013 0. 52 0.47 0.048 b Allo y 617 0.061 0.04 0.02 0.003 0.001 22.53 53.8 9.0 0.011 0.02 12.0 0.02 0.005 0.46 0.0094 0.017 1.1 b balance Ta ble 2: Solution treatment Name T emp er atu r e [_◦ C ] Time [min] Sanicro 25 1250 10 Sanicro 28 1150 15 AISI 310 1050 10 AISI 310 M 1250 10 AISI 316 L 1050 10 Allo y 800 HT 1200 15 Allo y 617 1175 20 12

(19)

2.2. EXPERIMENTAL METHODS

be explored by electron backscatter diffraction (EBSD), electron channeling contrast imaging (ECCI) and transmission electron microscopy (TEM) in a scanning electron microscope (SEM). Transmission electron microscopy pre-formed in a SEM is called STEM. The materials were tensile tested at several temperatures, from RT up to 700 ℃.

2.2.1 Tensile tests

All the tensile test were made according to the standard SS-EN10 002-1. The tensile tests were carried out using a Roell-Korthaus tensile testing machine equipped with a MTS 653 furnace and Magtec PMA-12/2/V7-1 extensometer in air environment. The temperatures that was used were RT, 400 ℃, 500 ℃, 600 ℃, 650 ℃ and 700 ℃. After the sample were broken, the specimens were cooled down in the furnace without forced cooling.

2.2.2 Scanning electron microscopy

To evaluate the microstructure of the materials after tensile testing a SEM HITACHI SU-70 were employed. SEM examinations were preformed for AISI 310, AISI 310M, Sanicro 25 and Alloy 617 for the temperatures, RT and 650 ℃. Also samples not effected by deformation and temperature were examined. The samples were taken from the fracture of the tensile test specimen and 15mm down in the longitudinal cross-section. The specimen preparation were carefully done to avoid surface deformation. Then three locations were used to be able to compare the different specimens, see figure 1. The locations were examined systematically at each specimen. The distances

X1, X2 and X3 are 0.6mm, 6.4mm and 12.0mm ± 0.02mm respectively. The

distance were taken from the lowest point at the fracture surface, see figure 1. In this way it is ensured that approximately the same region is compared in all specimens.

The microscope technique ECCI were used in the SEM. The technique is based on that the backscattered electrons intensity are strongly dependent on the orientation of the crystal lattice planes with respect to the incident electron beam. Due to dislocations small local distortions in the crystal lattice causes a modulation of the backscattered electron intensity, allowing the defect to be imaged [19].

EBSD were also used to evaluate the microstructure. This microscope technique in the SEM made it possible to observe the very deformed grains and grain boundaries. It also made it possible to verify if there were any twins in the microstructure.

(20)

CHAPTER 2. EXPERIMENTAL

1

2

3

Lowest point at the fracutre surface

X1 X2

X 3

Figure 1: Schematic drawing of the three different locations on the specimens for examination of microstructure in SEM.

STEM have been used to observe the microstructure at a higher mag-nification. STEM was used to examine AISI 316L, after tensile testing at RT.

(21)

3

Results

3.1 Mechanical signs of dynamic strain aging

Signs of DSA by the PLC effect for the studied materials are shown in fig-ures 2 and 3. The different types of serration for the materials are listed in table 3 [11, 12, 20].

The tensile tests showed that DSA appear at 500 ℃ and are present up to 700 ℃ in Sanicro 25. Sanicro 25 have A and C types of serration, were A occur at 500 ℃ and 600 ℃ while C occur at 650 ℃ and 700 ℃. The tensile tests revealed that in Sanicro 28 the PLC effect is well defined at testing temperatures from 400 ℃ to 700 ℃. At 400 ℃ the serrations are a combina-tion of A and B (called A+B in table 3), near the highest strain values the serrations transforms into E. From 500 ℃ and up to middle strain values at 600 ℃ the serrations are A-type. Then it transforms to B up to 650 ℃ and then at 700 ℃ C serrations occur. AISI 310M showed PLC effects at 600 ℃ and up to 700 ℃. At 600 ℃ and 650 ℃ A serrations occur which transform into C-type at 700 ℃. For AISI 316L DSA appear with A serrations at 400 ℃ and then at 600 ℃ up to 650 ℃ at low strain values. Intermediate values of strain gives B-type and high strain values gives C serrations at 650 ℃. 700 ℃ showed both B and C serrations, were B serrations appear at low values of strain while C-type occur at high values of strain. AISI 310 showed a PLC effect at 600 ℃ and at lower values of strain, a combination of D and B (called D+B in table 3) serrations were noticed while at higher values of strain A serrations were seen. At 650 ℃ B serrations occur and at 700 ℃ A+B serra-tions at lower strain values appear which transforms to B serraserra-tions at higher strain values. Alloy 800HT showed PLC effects of A-type at 400 ℃ for low values of strain and B serrations for high values of strain. For 500 ℃ the PLC effect showed a combination of A and B serrations. From 600 ℃ up to 650 ℃

(22)

CHAPTER 3. RESULTS

B serrations appear, also for low values of strain at 700 ℃ B serrations occur and transform then to C serrations at higher strain values. For Alloy 617 DSA occur at test temperature 400 ℃ with A-type for low strain values and A+B serrations for high values of strain. At 500 ℃ and up to 650 ℃ the PLC effect showed B serrations. For 700 ℃ C serrations appear [11, 12, 20]. The stress and strain graphs from the tensile tests showing the PLC effect for each material at the temperatures DSA appears are showed in Appendix A. The stress and strain curves revealed that increasing temperature influ-ence the ductility differently in the seven different materials. Figures 2 and 3 shows the stress and strain curves for each material and test temperature, there are three different ductility responses on the increased temperature. At the first response the ductility decreases with increased temperature, see a) and b) in figure 2 and c) in figure 3. AISI 316L, AISI 310 and AISI 310M showed such response. AISI 310 hade decreased ductility with increased tem-perature up to 650 ℃ and then the ductility increased when the temtem-perature reached 700 ℃. Sanicro 25, Sanicro 28 and Alloy 800HT showed the second response were the ductility first increases with increased temperature and then decreases over a specific temperature, see c) and d) in figure 2 and a) in figure 3. Alloy 617 showed a third response where the ductility increases with increased temperature up to the highest test temperature where ductil-ity decreases, see b) in figure 3.

Table 3: Types of serration

Name RT 400 ℃ 500 ℃ 600 ℃ 650 ℃ 700 ℃ Sanicro 25 - - A A C C Sanicro 28 - A+B, E A A, B B C AISI 316L - A - A A, B, C B, C AISI 310 - - - D+B, A B A+B, B AISI 310M - - - A A C Alloy 800HT - A, B A+B B B B, C Alloy 617 - A, A+B B B B B, C 16

(23)

3.1. MECHANICAL SIGNS OF DYNAMIC STRAIN AGING −10 0 10 20 30 40 50 60 −100 0 100 200 300 400 500 600

AISI 316L − stress−strain curve

strain [%] st ress [MP a] RT 400°C 500°C 600°C 650°C 700°C

a)

b)

−10 0 10 20 30 40 50 60 −100 0 100 200 300 400 500 600 700

800 Sanicro 25 − stress−strain curve

strain [%] st ress [MP a] RT 400°C500°C 600°C 650°C 700°C

c)

−10 0 10 20 30 40 50 60 70 −100 0 100 200 300 400 500

600 Sanicro 28 − stress−strain curve

strain [%] st ress [MP a] RT 400°C 500°C 600°C 650°C 700°C

d)

−10 0 10 20 30 40 50 0 100 200 300 400 500 600 700

AISI 310M − stress−strain curve

strain [%] stress [MPa] RT _400°C 500°C 600°C 650°C 700°C

Figure 2: Stress-strain curves for room temperature to 700 ℃ a) AISI 316L, b) AISI 310M, c) Sanicro 25 and d) Sanicro 28.

(24)

CHAPTER 3. RESULTS −10 0 10 20 30 40 50 60 −100 0 100 200 300 400 500

600 Sanicro 31HT − stress−strain curve

strain [%] st ress [MP a] RT 400°C 500°C 600°C 650°C 700°C −10 0 10 20 30 40 50 60 70 80 0 100 200 300 400 500 600 700

Alloy 617 − stress−strain curve

strain [%] st ress [MP a] Room temperature 500°C 400°C 600°C 650°C 700°C

a)

b)

−10 0 10 20 30 40 50 0 100 200 300 400 500 600

AISI 310 − stress−strain curve

strain [%] st ress [MP a] RT 400°C 500°C _600°C 650°C 700°C

c)

Figure 3: Stress-strain curves for room temperature to 700 ℃ a) Alloy 800HT (Sanicro 31HT), b) Alloy 617 and c) AISI 310.

(25)

3.2. MICROSTRUCTURAL INVESTIGATION

3.2 Microstructural investigation

The microstructural investigation revealed that the material have undergone high plastic deformation. The plastic deformation are higher at RT com-pared to the elevated temperatures. Deformation mechanisms as glide band or deformation bands, twins, dislocation cells (DC) and shear bands have been observed in ECCI investigations [21, 22]. Dislocation structures such as forests and cells are present in the microstructure [21]. DC are displayed in figure 6 b) and dislocation forest are displayed in figure 6 d). Twins have been identified by EBSD for Sanicro 25 and Alloy 617. Figure 4 a) and b), shows glide bands in multi-directions in Alloy 617 at RT and 650 ℃. The pictures are taken at location 1, see figure 1. There are differences in the size of glide bands, at a higher temperature the bands are faded compared to glide bands in RT. Similar feathers are observed in Sanicro 25, in figure 4 c) and b). The appearance of large glide bands are higher in the material with high nickel content. Figure 4 and figure 5 display that Alloy 617 have more and larger glide bands in the microstructure than Sanicro 25, AISI 310M and AISI 310. Sanicro 25 have higher nickel content and more and larger sized glide bands than AISI 310M and AISI 310. The occurrence of multi-direction glide bands increases with increased nickel content. Interaction between glide bands or twins causing a damage have been observed in the microstructure, as seen in figure 6 a). Damage mechanisms observed in the microstructure are broken particles and voids. These mechanisms has been observed at every location (figure 1) in AISI 310M at both temperatures. In the microstructure of Al-loy 617 broken particles has been observed in location 2 at RT and location 2 and 3 at 650 ℃. Voids has been observed at every location and both tem-perature in Alloy 617. In Sanicro 25 broken particles and voids are observed at every location and both temperatures. In the microstructure of AISI 310 voids has been observed in every location at both temperatures. Broken particles are shown in figure 4 c), displaying Sanicro 25 at RT. More ECCI pictures displaying the damages mechanisms at the three different locations and the test temperatures RT and 650 ℃ are shown in Appendix B. ECCI images displaying microstructure of material not affected by deformation or temperature are shown in figure 7. Alloy 617 (figure 7 a)) have bigger grains than Sanicro 25 (figure 7 b)) and AISI 310M (figure 7 c)). Sanicro 25 and AISI 310M have more particles than Alloy 617.

(26)

CHAPTER 3. RESULTS

a) b)

c) d)

Figure 4: ECCI image of Alloy 617 and Sanicro 25. a) Alloy 617 at RT, b) Alloy 617 at 650 ℃, c) Sanicro 25 at RT and d) Sanicro 25 at 650 ℃. Displaying differences in the microstructure due to a higher temperature and alloy content. All images are taken at location 1, see figure 1. The loading directions are displayed with the arrows.

(27)

a) b)

c) d)

Figure 5: ECCI image of AISI 310M and AISI 310. a) AISI 310M at RT, b) AISI 310M at 650 ℃, c) AISI 310 at RT and d) AISI 310 at 650 ℃. Displaying differences in the microstructure due to a higher temperature and alloy content. All images are taken at location 1, see figure 1. Observe that c) and d) have different magnification. The loading directions are displayed with the arrows.

(28)

CHAPTER 3. RESULTS a) b) DC c) Ti-nitride d)

Figure 6: a) shows an ECCI image of Alloy 617 at 650 ℃. Displaying interaction between glide bands or twins, causing a damage. The loading directions are displayed with the arrows. b) shows an STEM image of AISI 316L at RT, displaying DC and an interaction between glide bands. c) shows Alloy 617 at 650 ℃, displaying Ti-nitride particle. d) displaying dislocation forest in Alloy 617 at 650 ℃.

(29)

a) b)

c)

Figure 7: ECCI images of microstructure not affected by deformation and tem-perature. a) Alloy 617, b) Sanicro 25 and c) AISI 310M. Observe that image c) have a lower magnification than a) and b).

(30)

CHAPTER 3. RESULTS

3.3 Stacking fault energy calculations

The SFE is an important material property. Jinkyung Kim et al.[23] writes that the deformation mechanisms changes as function of SFE. High SFE values leads to formation of deformation twins [23]. The SFE was calculated using the following equation [24]:

SFE=25.7+2*(%Ni)+410*(%C)-0.9*(%Cr)-77(%N)-13*(%Si)-1.2*(%Mn) Table 4 shows the SFE values for the different materials, all the values are considered to be high according to Jinkyung Kim et al. [23].

Table 4: Stacking fault energies

Name SFE [mJm−2_] Sanicro 25 60.9 Sanicro 28 59.3 AISI 310 48.7 AISI 310M 41.7 AISI 316L 43.6 Alloy 800HT 82.5 Alloy 617 137.1 24

(31)

4

Discussion

4.1 Choice of experimental methods

As mentioned in chapter 2.2 Experimental methods, methods were chosen to show as much as possible of DSA influence on the mechanical response and microstructure. Tensile testing gives information that can effectively detect the PLC effect by studying the stress and strain curves. SEM in-vestigations gives more and better information of microstructure than light microscope investigations do. To be able to see any microstructure in a light microscope the sample must be etched and that does not give enough information about the effects caused by plastic deformation. SEM have a higher magnification capability than light microscope and specimens can be studied without etching. In this study a method based on an interaction be-tween primary electrons and the crystal called electron channeling contrast imaging was chosen. This method can image plastic deformation down to a dislocation level using an annular solid state detector. The channeling ef-fect gives important information about deformation and can therefore reveal more information about the deformation and damage mechanisms. By using an SEM in standard secondary electron mode very little information about the plastic deformation would have been obtained.

4.2 Evolution of tensile test curves

There seems to be a relationship between the nickel content and the different ductility responses. Three different responses have been observed. The first gives decreased ductility with increased temperature, observed in AISI 310, AISI 310M and AISI 316L. These three materials also have the lowest nickel

(32)

CHAPTER 4. DISCUSSION

contents, 20wt%, 19wt% and 12wt% respectively. The second ductility re-sponse first gives increased ductility with increased temperature and then above a specific temperature the ductility decreases. These response have been observed in Sanicro 25 and Sanicro 28 and Alloy 800HT, with 25wt%, 31wt% and 30wt% nickel content respectively. The third ductility response shows similar character as the second but the ductility decrease comes at a higher temperature. The third response have been observed in Alloy 617, with an nickel content of 53.8wt%. Figure 8 shows a comparison of the maximum strain values for each material at each test temperature. The maximum stresses decrease with increased temperature for all seven materi-als as expected, see figure 2 and 3 [14]. The stress and strain curves revemateri-als that alloy content and temperature affect the PLC effect. At the two lowest test temperatures the most common serration type were A and four of the tested materials showed PLC effects at these temperatures. Some what re-markable is that AISI 316L showed PLC effects at 400 ℃ but not at 500 ℃. For 600 ℃ the most common serration types were A and B and at this tem-perature all the tested materials showed PLC effects. For the two highest test temperatures the most common serration types were B at 650 ℃ and C at 700 ℃. Table 3 shows the serration responses for the tested materials at each test temperature. The basis for this evaluation are reports from I. Rodriguez[20], L. H. de Almeida et al.[12] and M. Ivanchenko[11] which illustrated and described the various serration types.

(33)

4.2. EVOLUTION OF TENSILE TEST CURVES 0 100 200 300 400 500 600 700 30 35 40 45 50 55 60 65 70 75 80 85 Strain−temperature curve temp. [°C] Strain [%] Alloy 617

Sanicro 28 Alloy 800HT Sanicro 25 AISI 310M

AISI 310 AISI 316L

Figure 8: Display the maximum strain value for each material at the test tem-peratures.

(34)

CHAPTER 4. DISCUSSION

4.3 Evolution of microstructure and deformation

mechanisms

There are a higher degree of plastic deformation in the specimens deformed at RT than at higher temperatures. The plastic deformation seems to fade with increased temperature, which was expected [14, 17, 22]. The deforma-tion mechanisms that are dominant at RT are glide band and twinning in all five tested materials. Were the glide bands mostly showed planar slip mode [17]. At higher temperatures the deformation mechanism that are dominant changes to only glide bands with decreasing amount of twins with increased temperature. Decreased density of twins with increased temperature for a modified 316L austenitic steel has been reported by S. Scheriau et al.[22]. Dislocation cells has also been seen in the microstructure of AISI 316L at room temperature, see figure 6 b). I. Gutierrez-Urrutia et al. reports of dislo-cation cells at room temperature in TWIP steels during tensile deformation. Damage mechanisms such as broken particles and voids are observed in the microstructure in most of the investigated materials. AISI 310 seems not to display any broken particles. In Alloy 617 broken particles seems not to display in the location 1 and 3 at RT and location 1 at 650 ℃. ECCI images displaying the microstructure of the virgin materials shows that Alloy 617 has less particles and lager sized grains than Sanicro 25 and AISI 310M, see figure 7). Which probably is the reason for less broken particles in the mi-crostructure of Alloy 617 at RT and 650 ℃. The increased amount of broken particles in 650 ℃ than can be related to precipitation of titanium-nitride, observed in the microstructure shown i figure 6 c).

(35)

5

Conclusions

This research includes studies of the high-temperature influence on mechani-cal responses and how the alloy content influence the high-temperature prop-erties. Mechanical responses such as ductility changes due to increased tem-perature in austenitic materials and the phenomenon DSA response due to increased temperature. Also couplings between the ductility changes, DSA and alloy content is observed. The presented results also reveal the deforma-tion and damage mechanisms in austenitic materials. The presented results will contribute to the continuing research of high-temperature influence on austenitic materials. The influence of high alloy content of nickel has been shown to increase the SFE. High SFE seems to increase the amount of twins which has been found in the microstructure investigation. The higher amount of twins gives probably higher ductility. The increase of ductility during DSA is probably due to formation of twins. Damage in the microstructure due to DSA can probably be related to intersection between glide bands or twins. Other damage mechanisms has been shown as broken particles and voids. The deformation mechanisms in the microstructure during tensile loading has been found to be glide bands, twins, dislocation cells and shear bands.

(36)

(37)

6

Future work

The present research has been as a smaller part of a PhD-project, called Long

Term High Temperature Behavior of Advanced Heat Resistant Materials. The

presented results will be used in the PhD-project together with the questions that this master thesis didn’t answered. The evaluation of ductility changes will continue. The microstructure evaluation will continue looking for signs of DSA, this is suggested to be carried out with use of in-situ EBSD tensile test. The PhD-project will also include long time aging tests, up to 30000 hours for temperatures up to 700 ℃. This is to evaluate the influence of long time aging and rough environment on structure integrity and safety at high temperatures. Stress relaxation testing will be carried out to evaluate the degradation mechanism responsible for stress relaxation cracking. The tensile deformation and cracking behavior will be evaluated with very slow strain rate testing at high temperatures.

(38)

(39)

Acknowledgement

This research has been funded by the Swedish Energy Agency and AB Sand-vik Material Technology through the Swedish research program KME, the support of which is gratefully acknowledged.

In addition, I would like to thank the group of proficient researchers that I have had the pleasure of working with during these month: Sten Johansson, Guocai Chai and Johan Moverare.

I also would like to thank for the technical support from Annethe Billenius at Linköping University, the help with tensile testing from Håkan Nylén at AB Sandvik Material Technology and my colleagues at the division for fruitful discussions. A special thanks to my wife and family who have supported me all the time.

(40)

(41)

Bibliography

[1] Sandra Cunningham. Effect of substitutional elements on dynamic strain aging in steels. Master’s thesis, Department of Mining and Metallurgical Engineering McGill Univeristy, 1999.

[2] Guocai Chai. IAS 2008. In Influence of nitrogen content on the static

and dynamic strain ageing in super duplex stainless steels, 2008.

[3] G. R. Stewart and J. J. Jonas. Static and dynamic strain aging at high temperatures in 304 stainless steel. ISIJ International, 44:1263–1272, 2004.

[4] Soon-Bok Lee Seong-Gu Hong, Keum-Oh Lee. Dynamic strain aging effect on the fatigue resistance of type 316L stainless steel. International

Journal of Fatigue, 27:1420–1424, 2005.

[5] George Krauss. STEELS Processing, Structure, and Performance. ASM International, 2005.

[6] R. Kaibyshev I. Nikulin. Deformation behavior and the portevin-le chatelier effect in a modified 18Cr-8N stainless steel. Materials Science

and Engineering A, 528:1340–1347, 2011.

[7] M. Valsan K. Bhanu Sankara Rao G. V. Prasad Reddy, R. Sandhya. Temperature dependence of low cycle fatigue of 316(N) weld metals and 316L(N)/316(N) weld joints. Materials science and technology, 26: 1384 –1392, 2010.

[8] C.H.J. Davies Y.B. Chun. Twinning-induced negative strain rate sen-sitivity in wrought Mg alloy AZ31. Materials Science and Engineering

A, 528:5713–5722, 2011.

[9] Ulla Ehrnstén Yuriy Yagodzinskyy Hannu Hänninen Wade Karlsen, Mykola Ivanchenko. Microstructural manifestation of dynamic strain

(42)

BIBLIOGRAPHY

aging in AISI 316 stainless steel. Journal of Nuclear Materials, 395: 156–161, 2009.

[10] T. Yu H.J. Shi X.F. Ma M.C. Cai, L.S. Niu. Strain rate and temperature effects on the critical strain for portevin-le chatelier effect. Materials

Science and Engineering A, 527:5175–5180, 2010.

[11] Mykola Ivanchenko. Dynamic strain aging of austenitic stainless steels

and Ni-base alloys. PhD thesis, Aalto University, 2010.

[12] I. Le May L. H. de Almeida and P. R. O. Emygdio. Mechanistic mod-eling of dynamic strain aging in austenitic stainless steels. Materials

Characterization (USA), 41:4:137–150, 1998.

[13] International stainless steel forum: The stainless steel family.

[14] Wendelin J. Wright Donald R. Askeland, Pradeep P. Fulay. The

Sci-ence and Engineering of Materials, Sixth Edition. Global Engineering,

Christopher M. Shortt, 2011.

[15] Vlastimil Vodárek. Creep behavior and microstructural evolution in AISI 316LN + Nb steels at 650 ℃. Materials Science and Engineering

A, 528:4232–4238, 2011.

[16] R. N. Singh B. P. Kashyap, R. Kishore and T. K. Sinha. Effect of dynamic strain ageing on the tensile properties of a modified 9Cr-1Mo steel. Journal of Materials Science, 32:437–442, 1997.

[17] Richard W. Hertzberg. Deformation and Fracture Mechanics of

Engi-neering Materials. John Wiley & sons, 1995.

[18] Roger C. Reed. The Superalloys Fundamentals and Applications. Cam-bridge University Press, 2008.

[19] D. Raabe I. Gutierrez-Urrutia, S. Zaefferer. Electron channeling con-trast imaging of twins and dislocations in twinning-induced plasticity steels under controlled diffraction conditions in a scanning electron mi-croscope. Scripta MATERIALIA, 61:737–740, 2009.

[20] P. Rodriguez. Serrated plastic flow. Bulletin of Materials Science, 6:4: 653–663, 1984.

[21] D. Raabe I. Gutierrez-Urrutia. Dislocation and twin substructure evolu-tion during strain hardening of an Fe-22wt.% Mn-0.6wt.% C twip steel observed by electron channeling contrast imaging. Acta Materialia, 59: 6449–6462, 2011.

(43)

BIBLIOGRAPHY

[22] S. Kleber R.Pippan S. Scheriau, Z. Zhang. Deformation mechanisms of a modified 316L austenitic steel subjected to high pressure torsion.

Materials Science and Engineering A, 528:2776–2786, 2011.

[23] Jinkyung Kim and B.C. De Cooman. On the stacking fault energy of Fe-18 pct Mn-0.6 pct C-1.5 pct Al twinning-induced plasticity steel.

Metallurgical and Materials Transactions A, 42A:932–936, 2011.

[24] F.B. Pickering. Physical mettallurgical developments of stainless steels.

(44)

APPENDIX

A

Stress and strain curves

a)

30 35 40 45 50 55 500 510 520 530 540 550 560 570 580 590 600 san25 − 500 C strain [%] stress [MPa] A 25 30 35 40 45 50 440 460 480 500 520 540 560 san25 − 600 C strain [%] stress [MPa] A

b)

5 10 15 20 25 30 35 40 45 50 250 300 350 400 450 500 san25 − 650 C strain [%] stress [MPa] C

c)

6 8 10 12 14 16 18 20 22 24 26 28 250 300 350 400 450 san25 − 700 C strain [%] stress [MPa] C

d)

Figure 9: Stress-strain curves for Sanicro 25 at a) 500 ℃, b) 600 ℃, c) 650 ℃ and d) 700 ℃.

(45)

A. STRESS AND STRAIN CURVES

a)

b)

c)

d)

30 35 40 45 50 55 360 370 380 390 400 410 420 430 440 450 460 san28 − 400 C strain [%] stress [MPa] A+B E 15 20 25 30 35 40 45 50 55 300 350 400 450 san28 − 500 C strain [%] stress [MPa] A 10 15 20 25 30 35 40 45 50 55 220 240 260 280 300 320 340 360 380 400 san28 − 600 C strain [%] stress [MPa] A B 5 10 15 20 25 30 35 40 45 50 55 180 200 220 240 260 280 300 320 340 360 380 san28 − 650 C strain [%] stress [MPa] B

Figure 10: Stress-strain curves for Sanicro 28 at a) 400 ℃, b) 500 ℃, c) 600 ℃ and d) 650 ℃.

(46)

APPENDIX

a)

b)

c)

d)

6 8 10 12 14 16 18 20 22 24 26 28 170 180 190 200 210 220 230 240 250 260 270 280 san28 − 700 C strain [%] st ress [MP a] C 20 25 30 35 40 420 440 460 480 500 520 540 AISI 310M − 600 C strian [%] stress [MPa] A 5 10 15 20 25 30 35 40 250 300 350 400 450 500 AISI 310M − 650 C strian [%] stress [MPa] A 5 10 15 20 25 30 35 260 280 300 320 340 360 380 400 420 440 460 480 AISI 310M − 700 C strian [%] stress [MPa] C

Figure 11: Stress-strain curves for Sanicro 28 at a) 700 ℃ and AISI 310M at b) 600 ℃, c) 650 ℃ and d) 700 ℃.

(47)

a)

b)

c)

d)

25 26 27 28 29 30 31 32 33 34 35 380 385 390 395 400 405 AISI 316L − 400 C strain [%] stress [MPa] A 10 15 20 25 30 260 270 280 290 300 310 320 330 340 350 AISI 316L − 600 C strain [%] stress [MPa] A 5 10 15 20 25 30 160 180 200 220 240 260 280 300 320 AISI 316L − 650 C strain [%] stress [MPa] A B C 6 8 10 12 14 16 18 190 200 210 220 230 240 250 260 270 280 AISI 316L − 700 C strain [%] stress [MPa] B C

Figure 12: Stress-strain curves for AISI 316L at a) 400 ℃, b) 600 ℃, c) 650 ℃ and d) 700 ℃.

(48)

APPENDIX

a)

b)

c)

d)

5 10 15 20 25 30 240 260 280 300 320 340 360 380 400 420 AISI 310 − 600°C strain [%] stress [MPa] D+B A 5 10 15 20 25 30 220 240 260 280 300 320 340 360 380 AISI 310 − 650°C strain [%] stress [MPa] B 5 10 15 20 25 30 35 220 240 260 280 300 320 AISI 310 − 700°C strain [%]

stress [MPa] A+B

B 10 15 20 25 30 35 40 45 50 55 280 300 320 340 360 380 400 420 440 460 san31HT − 400 C strain [%] stress [MPa] A B A

Figure 13: Stress-strain curves for AISI 310 at a) 600 ℃, b) 650 ℃ and c) 700 ℃ and Alloy 800HT at d) 400 ℃.

(49)

a)

b)

c)

d)

10 15 20 25 30 35 40 45 50 55 280 300 320 340 360 380 400 420 440 460 san31HT − 500 C strain [%]

stress [MPa] A+B

5 10 15 20 25 30 35 40 45 50 200 250 300 350 400 450 san31HT − 600 C strain [%] stress [MPa] B 5 10 15 20 25 30 35 40 180 200 220 240 260 280 300 320 340 360 380 400 san31HT − 650 C strain [%] stress [MPa] B 5 10 15 20 25 30 35 180 200 220 240 260 280 300 320 340 360 san31HT − 700 C strain [%] stress [MPa] C B

Figure 14: Stress-strain curves for Alloy 800HT at a) 500 ℃, b) 600 ℃, c) 650 ℃ and d) 700 ℃.

(50)

APPENDIX

a)

b)

c)

10 20 30 40 50 60 300 350 400 450 500 550 alloy617 − 400 C strain [%] stress [MPa] A A+B 10 20 30 40 50 60 70 300 350 400 450 500 550 alloy617 − 500 C strain [%] stress [MPa] B 10 20 30 40 50 60 70 300 350 400 450 500 alloy617 − 600 C strain [%] stress [MPa] B

Figure 15: Stress-strain curves for Alloy 617 at a) 400 ℃, b) 500 ℃ and c) 600 ℃.

(51)

a)

b)

10 20 30 40 50 60 70 300 350 400 450 500 alloy617 − 650 C strain [%] stress [MPa] B 10 20 30 40 50 60 70 250 300 350 400 450 alloy617 − 700 C strain [%] stress [MPa] C B

(52)

APPENDIX

B

ECCI images of microstructure

a) b)

c)

Figure 17: ECCI pictures from the three different locations of Sanicro 25 mi-crostructure at RT, a) location 1, b) location 2 and c) location 3. The loading directions are displayed with the arrows.

(53)

B. ECCI IMAGES OF MICROSTRUCTURE

a) b)

c)

Figure 18: ECCI pictures from the three different locations of Sanicro 25 mi-crostructure at 650 ℃, a) location 1, b) location 2 and c) location 3. The loading directions are displayed with the arrows.

(54)

APPENDIX

a) b)

c)

Figure 19: ECCI pictures from the three different locations of AISI 310M mi-crostructure at RT, a) location 1, b) location 2 and c) location 3. The loading directions are displayed with the arrows.

(55)

a) b)

c)

Figure 20: ECCI pictures from the three different locations of AISI 310M mi-crostructure at 650 ℃, a) location 1, b) location 2 and c) location 3. The loading directions are displayed with the arrows.

(56)

APPENDIX

a) b)

c)

Figure 21: ECCI pictures from the three different locations of Alloy 617 mi-crostructure at 650 ℃, a) location 1, b) location 2 and c) location 3. The loading directions are displayed with the arrows.

(57)

a) b)

c)

Figure 22: ECCI pictures from the three different locations of Alloy 617 mi-crostructure at 650 ℃, a) location 1, b) location 2 and c) location 3. The loading directions are displayed with the arrows.

Effect of temperature on mechanical response of austenitic materials