Kumar2011.pdf

Materials and Design 32 (2011) 3617–3623

Contents lists available at ScienceDirect

Materials and Design journal homepage: www.elsevier.com/locate/matdes

Technical Report

Effect of heat input on the microstructure and mechanical properties of gas tungsten arc welded AISI 304 stainless steel joints Subodh Kumar, A.S. Shahi ⇑ Department of Mechanical Engineering, Sant Longowal Institute of Engineering & Technology, Longowal, Sangrur, Punjab 148 106, India

a r t i c l e

i n f o

Article history: Received 20 October 2010 Accepted 7 February 2011 Available online 3 March 2011

a b s t r a c t Influence of heat input on the microstructure and mechanical properties of gas tungsten arc welded 304 stainless steel (SS) joints was studied. Three heat input combinations designated as low heat (2.563 kJ/ mm), medium heat (2.784 kJ/mm) and high heat (3.017 kJ/mm) were selected from the operating window of the gas tungsten arc welding process (GTAW) and weld joints made using these combinations were subjected to microstructural evaluations and tensile testing so as to analyze the effect of thermal arc energy on the microstructure and mechanical properties of these joints. The results of this investigation indicate that the joints made using low heat input exhibited higher ultimate tensile strength (UTS) than those welded with medium and high heat input. Significant grain coarsening was observed in the heat affected zone (HAZ) of all the joints and it was found that the extent of grain coarsening in the heat affected zone increased with increase in the heat input. For the joints investigated in this study it was also found that average dendrite length and inter-dendritic spacing in the weld zone increases with increase in the heat input which is the main reason for the observable changes in the tensile properties of the weld joints welded with different arc energy inputs. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Austenitic stainless steels have been used widely by the fabrication industry owing to their excellent high temperature and corrosion resistance properties. Some of the typical applications of these steel include their use as nuclear structural material for reactor coolant piping, valve bodies, vessel internals, chemical and process industries, dairy industries, petrochemical industries etc. Out of 300 series grade of these steels type 304 SS is extensively used in industries due to its superior low temperature toughness and corrosion resistance. One of the typical applications of type 304 SS include storing and transportation of liquefied natural gas (LNG), whose boiling point is 162 °C under 1 atmosphere. A study on fatigue crack growth rate for type 304 SS over a temperature range from room to 162 °C has shown that base metal possesses superior resistance to crack growth relative to weld metals over the entire temperature range [1]. Another typical application of this material includes its use as bellows used as conduit for liquid fuel and oxidizer in propellant tank of satellite launch vehicle [2]. Chen et al. [3] found that when Cu–Si enriched type 304 SS (containing 2–2.5 wt.% copper and 1–1.5 wt.% silicon) and a conventional type 304 SS was welded using gas metal arc welding (GMAW), process ductility decreased and ferrite levels increased ⇑ Corresponding author. Tel.: +91 1672 253272; fax: +91 1672 280057. E-mail address: [email protected] (A.S. Shahi). 0261-3069/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.matdes.2011.02.017

in both weldments, as the heat input was increased. A comparative study by Yan et al. [4] on the microstructure and mechanical properties of 304 SS joints by tungsten inert gas (TIG) welding, laser welding and laser-TIG hybrid welding showed that laser welding could give highest tensile strength and smallest dendrite size in all joints whereas TIG welding gave lowest tensile strength and biggest dendrite size. Work reported by Muthupandi et al. [5] on the effect of weld chemistry and heat input on the structure and properties of duplex stainless steel welds using autogenous-TIG and electron beam welding process shows that chemical composition exerts a greater influence on the ferrite–austenite ratio than the cooling rate. Jana [6] has reported the effect of varying heat inputs on the properties of the HAZ of two different duplex steels and found that as arc energy increased hardness of both weld metal and the HAZ decreased, whereas width of the HAZ increased with increased arc energies. Study on the influence of welding heat input on submerged arc welding (SAW) welded duplex steel joints imperfections has been reported by Nowacki et al. [7] where heat input from 2.5 to 4.0 kJ/mm was used for plate thickness of 10–23 mm and it was concluded that usage of larger welding heat input provided the best joints quality. Zumelzu et al. [8] studied the mechanical behaviour of AISI 316L welded joints using shielded metal arc welding (SMAW) and GMAW process with different electrodes types. Their work concludes that a direct correlation exists between the thermal contribution and tensile strength for the materials studied. The effects of minor elements and shielding gas on the penetration of TIG

3618

S. Kumar, A.S. Shahi / Materials and Design 32 (2011) 3617–3623

understanding about the metallurgical aspects that affect the service performance of these welded joints made using different heat input combinations.

welding in type 304 SS have been studied using bead on plate experimentation technique and it is concluded that minor elements such as oxygen, aluminium and sulfur have a significant effect on the weld depth to width ratio [9]. Experimental investigations on the effect of hydrogen in argon as a shielding gas in TIG welding of austenitic stainless steel show that mean grain size in the weld metal increases with increasing hydrogen content besides increasing the weld metal penetration depth and its width [10]. Lu et al. [11] have reported in their experimental results that small addition of oxygen content to the He–Ar mixed shielding can significantly change the weld shape from a wide shallow type to a narrow deep one. Lee et al. [12] have reported in their studies on effects of strain rate and failure behaviour of 304L SS SMAW weldments and find that as the strain rate increases, the flow stress increases and the fracture strain decreases. Korino et al. [13] have reviewed the considerations for weldability of 304L SS and recommend Creq to Nieq ratio of 1.52–1.9 to control the primary mode of solidification. Lee et al. [14] while investigating the pitting corrosion behaviour of welded joints of AISI 304L using flux cored arc welding (FCAW) process found that tensile and yield strengths were increased with increasing equivalent ratio of Creq/Nieq. Milad et al. [15] found that yield and tensile strengths of 304 SS increased gradually at the same rate with increasing degree of cold work. Shyu et al. [16] have investigated the effect of oxide fluxes on weld morphology, arc voltage, mechanical properties, angular distortion and hot cracking susceptibility of autogenous TIG bead on plate welds. Their results indicate that penetration is significantly increased which in turn increases depth to bead-width ratio and tends to reduce angular distortion. Other studies which show that 304 SS and 304L SS grade has been the topic of research of many researchers include various studies like experimental determination of grain boundary composition of 304 SS in low temperature sensitization condition using a scanning Auger microprobe [17], measuring chromium depletion after various thermal heat treatments [18], modelling of low temperature sensitization of austenitic stainless steel [19], studying sensitization behaviour of grain boundary engineered austenitic stainless steel [20], arresting weld decay in 304 SS by twin-induced grain boundary engineering [21] etc. From the literature reviewed on the material processing of 304 SS it is observed that no systematic work on the effect of heat input on microstructure and tensile properties of gas tungsten arc (GTA) welded has been reported. In view of the fact that arc welding processes like GTAW offer a wide spectrum of thermal energy for joining different thicknesses of steels it was considered important that undertaking the present study would be beneficial in gaining an

2. Experimental details 2.1. Base and filler material combination The base material used in the present investigation was in the form of AISI 304 SS plates of sizes 200 mm  100 mm  6 mm which were cut from a rolled sheet and the filler was 308 SS solid electrode of 3.15 mm diameter. Table 1 shows the chemical composition of the base and the filler used. 2.2. Welding procedure In the present work double V-groove design was used so that welding could be accomplished in two numbers of passes ensuring full penetration. Before welding all the edges were thoroughly cleaned mechanically and chemically in order to avoid any source of contamination like rust, scale, dust, oil, moisture etc. that could creep into the weld metal and later on, could result possibly into a weld defect. After tacking the plates together the first weld pass was given using GTAW process with welding conditions as mentioned in Table 2 and prior to giving of second pass an interpass temperature of around 150 °C was maintained. No preheat or post heat treatment was given to the specimens. Although GTAW process was used in the manual mode, still utmost care was taken during recording of the arc on time so as to facilitate calculations of welding speed for heat input calculations. It is worth mentioning here that the best welding practice available in the fabrication industry was used in the present work. It is a well established fact that among all the welding variables in arc welding processes welding current is the most influential variable since it affects the current density and thus the melting rate of the filler as well as the base material. So in accordance with this fundamental fact three different heat input combinations corresponding to different welding currents i.e. 120 A (low heat input), 150 A (medium heat input) and 180 A (high heat input) combinations were selected for the present study. The reason for using these specific welding current values was twofold firstly, this spectrum of heat input combinations results in arc energies which are sufficient to cause adequate fusion of the base and weld metal selected for the present study and secondly, a step increase of 30A was anticipated to be sufficient enough to cause a direct and significant influence on the microstructure and tensile properties of the

Table 1 Chemical composition (wt.%) of the base metal and filler used. Alloy element

C

Si

Mn

P

S

Cr

Ni

Fe

Base (304 SS) Filler (ER 308 SS)

0.06 0.08

0.42 1.0

1.89 1.59

0.032 0.045

0.014 0.03

18.67 18.15

8.53 10.02

Balance Balance

Table 2 Process parameters used for fabricating butt welded joints. Specimen no.

Pass

Current (A)

Voltage (V)

Average welding speed (mm/s)

Average heat input per unit length per pass (kJ/mm)

Total heat input per unit length of the weld (kJ/mm)

A (low heat)

First Second First Second First Second

120 120 150 150 180 180

30 30 35 35 40 40

2.252 2.243 3.030 3.003 3.846 3.787

1.280 1.283 1.386 1.398 1.497 1.520

2.563

B (medium heat) C (high heat)

2.784 3.017

S. Kumar, A.S. Shahi / Materials and Design 32 (2011) 3617–3623

3619

tested on a servo hydraulically controlled digital tensile testing machine of 400 kN capacity. 2.5. Metallography

Fig. 1. Photograph showing the base plates in the as welded condition at different heat inputs.

welded joints. During and after welding the joints were visually inspected for their quality and it was ensured that all weld beads possessed good geometrical consistency and were free from visible defects like surface porosity, blow holes etc. Fig. 1 shows the plates in the as welded condition using different heat inputs. Other details related to the process and procedures used in the present work include:Type and size of the non-consumable for the joints investigated in this study tungsten electrode = EW-Th-2 (Thoriated tungsten) of 3 mm diameter, Shielding gas flow rate of industrially pure Argon = 15 L/min, Electrode to work angle = 45°, Polarity = DC electrode positive. 2.3. Specimen sampling

In order to observe the microstructural changes that take place during welding, corresponding to each heat input combination, specimens were machined out from the weld pads as shown in Fig. 2. After polishing and macroetching the cross sections of the joints were captured with the help of Image analysis software coupled with a stereozoom microscope at a magnification of 10 to facilitate measuring of the details like cross sectional areas of the fusion zone and HAZ. Standard polishing procedures were used for general microstructural observations [23]. An electrolytic oxalic acid etch was used with the conditions (Electrolyte used: Oxalic acid (10 g) + distilled water (100 mL), Cell voltage: 6 V, Etching time: 1 min). Microstructures of different zones of interest like weld metal, HAZ and fusion boundary under different heat input combinations were viewed and captured with an optical microscope coupled with an image analyzing software. Microhardness of different zones of the weldments was measured using Vickers’s micro hardness testing machine with a load of 0.5 kg. Fractured ends of the tensile tested specimens were analyzed using Scanning electron microscopy (SEM) to assess the nature of the fracture mode. 3. Results and discussion

The specimens for tensile testing, micro hardness testing and microstructural studies were taken from the weld pads as schematically illustrated in Fig. 2. 2.4. Tensile test Three specimens per heat input combinations, were machined out from the weld pads as mentioned in Fig. 2. Each tensile specimen size was prepared in accordance with ASTM E08 standards [22] as illustrated schematically in Fig. 3. The specimens were

3.1. Metallographic studies Full penetration welds were obtained in all the three combinations of heat input as shown in Fig. 4. Measured areas of fusion zone and HAZ of different weldments are shown in Table 3. As indicated by these values it is found that as heat input increases the fusion areas of the joints also increase proportionately. The same trend is followed for the HAZ area associated with each of these joints. Yan [4] and Jana [6] have reported similar trends while studying TIG welded 304 SS and SMAW welded duplex SS respec-

Fig. 2. Schematic illustration of the specimen sampling from the weld pads.

Fig. 3. Specifications of the tensile specimen used in the present work.

3620

S. Kumar, A.S. Shahi / Materials and Design 32 (2011) 3617–3623

Fig. 4. Stereozoom images showing the cross sections of the weld joints at different heat inputs (a) low heat (b) medium heat (c) high heat (10).

Table 3 Macro and microstructural details of the weld joints. Heat input

Low Medium High Base metal

Tensile properties

Macrostructural details (Cross sectional area)

Microstructural details

Ultimate tensile strength (MPa)

Percentage elongation (%)

Fusion zone with reinforcement (mm2)

Fusion zone without reinforcement (mm2)

HAZ area (mm2)

Dendrite length in the weld zone (lm)

Interdendrite spacing (lm)

657.32 639.45 622.8 610.8

24.28 22.85 21.42 38.57

36.74 38.86 43.02 –

21.68 23.57 26.29 –

12.83 14.79 16.24 –

111.10 151.75 201.14 –

10.29 15.42 22.87 –

tively, that fusion zone and HAZ area increase with increase in heat input. Optical micrographs showing the microstructures of weld zone, fusion boundary and HAZ for different heat input combinations are presented from Figs. 5–7. The measured values of dendrite lengths and inter-dendritic spacings for these joints are mentioned in Table

a

Location of fracture

Joint efficiency (%)

Base metal Base metal Base metal –

107.61 104.69 101.96 –

3. It is observed from these optical micrographs that as heat input increases the dendrite size and inter-dendritic spacing in the weld metal also increase. This dendrite size variation can be attributed to the fact that at low heat input, cooling rate is relatively higher due to which steep thermal gradients are established in the weld metal, which in turn allow lesser time for the dendrites to grow,

b

FB

HAZ

Fig. 5. Optical micrograph showing the microstructure of (a) weld metal (b) fusion boundary and HAZ (low heat, at 100).

a

b

FB HAZ

Fig. 6. Optical micrograph showing the microstructure of (a) weld metal (b) fusion boundary and HAZ (medium heat, at 100).

3621

S. Kumar, A.S. Shahi / Materials and Design 32 (2011) 3617–3623

b

a

HAZ FB

Fig. 7. Optical micrograph showing the microstructure of (a) weld metal (b) fusion boundary and HAZ (high heat, at 100).

whereas at high heat input, cooling rate is slow which provides ample time for the dendrites to grow farther into the fusion zone.

Low heat Medium heat High heat

280 270

3.2. Microhardness

Low heat Medium heat High heat

250

Vicker hardness (HV0.5)

240 230 220 210 200 190 180 170 160

-4

-3

-2

-1

0

1

2

3

4

Distance from the weld centre (mm) Fig. 8. Microhardness profile showing micro hardness at different points in the weld metal at different heat inputs.

Vicker hardness (HV0.5 )

260

Microhardness measurements were taken in two directions firstly in the transverse direction i.e. perpendicular to the base plate surface and secondly, in the longitudinal direction i.e. parallel to the base plate surface and the same are shown in Figs. 8 and 9 respectively. Fig. 8 shows that the micro hardness near the top of the weld bead surface is high and as the centre of the fusion/weld zone is approached by the indentor it gradually reduces, which is due to the fact that cooling rate is relatively higher at the top of the weld bead surface than at the centre of the weld metal. From Fig. 9 it is observed that as the indentor traverses outwards (parallel to the base plate surface) from the centre of the weld/fusion zone towards the fusion boundary, micro hardness increases from 205.5 to 228.8 VHN for low heat input, 194.0–210.2 VHN for medium heat and 181.1–197.4 VHN for high heat input welded joint. Fusion boundary or transition zone encountered while traversing in this direction is indicated by a steep rise in the micro hardness with value of 272.4 VHN, 262.6 VHN and 251.6 VHN respectively for low, medium and high heat input respectively. High hardness as possessed by the fusion boundary zone (FBZ) in all the joints can be attributed to the presence of partially unmelted grains at the fusion boundary which are partially adopted as nuclei by the new precipitating phase of the weld metal during the solidification

250 240 230 220 210 200 190 180 170 160 150

-10

-8

-6

-4

-2

0

2

4

6

8

10

Distance from the weld centre (mm) Fig. 9. Microhardness profile showing micro hardness of different zones of the weldments at different heat inputs.

stage. After reaching this peak value micro hardness shows a decreasing trend in the HAZ. In all the joints, HAZ area adjacent to the fusion boundary was coarse grained HAZ (CGHAZ) which possessed low hardness whereas the HAZ area adjacent to the base metal was fine grained HAZ (FGHAZ) which possessed high hardness. The reason for this trend of micro hardness in the HAZ of all the joints is that the area adjacent to the weld/fusion zone experiences relatively slow cooling rate and hence has coarse grained microstructure, whereas the area adjoining the base metal undergoes high cooling rate due to steeper thermal gradients and consequently has fine grained microstructure. This is evident from the trend depicted by the micro hardness profile within the HAZ of each of these joints. In general it is observed from these micro hardness studies that hardness follows an increasing trend in the order of weld metal, HAZ, unaffected base metal and fusion boundary for all the joints made at different heat inputs. It is also observed that there is significant grain coarsening in the HAZs of all the joints. Further it is observed from the optical micrographs shown from Fig. 5b–7b that the extent of grain coarsening in the HAZ increases with increase in heat input. 3.3. Tensile properties The transverse tensile strength of all the joints made using different heat input conditions has been evaluated. In each condition

3622

S. Kumar, A.S. Shahi / Materials and Design 32 (2011) 3617–3623

three specimens were tested and the average tensile strength of three specimens per heat input and their corresponding percentage elongations thus obtained is mentioned in Table 3. The tensile results so obtained show that maximum tensile strength of 657.32 MPa is possessed by the specimens made using low heat input combination followed by 639.45 MPa using medium heat input and 622.8 MPa using high heat input combination. Table 3 shows the microstructural details of the weld metal in terms of dendrite size and cell spacing, which indicates that high tensile strength and ductility is possessed by the joints at low heat input, which can be attributed to smaller dendrite sizes and lesser inter-dendritic spacing in the fusion zone. Relatively lower tensile strength and ductility is possessed by the joints with long dendrite sizes and large inter-dendritic spacing in the fusion zone of the joint welded using high heat input. Further it is found that all the tensile

specimens fractured in the base metal as shown in Fig. 10 which indicates that weld metal in all the joints possessed higher tensile strength than the base metal and thus joint efficiencies [defined as (UTSweldjoint)/(UTSbasemetal)  100] of 107.61%, 104.69% and 101.96% were achieved for low, medium and high heat input combination respectively. The fractured surfaces of the tensile specimens were analyzed using SEM. Figs. 11–13 show the SEM fractographs of all the joints tensile tested. Dimples of varying size and shape were observed in all the fractured surfaces which indicate that major fracturing mechanism was ductile. From Fig. 11 it is observed that fractured surface of the specimen at low heat input contains a large population of small and shallow dimples which is indicative of its relatively high tensile strength and ductility. From Figs. 12 and 13 it is observed that as heat input increases coarse and elongated dim-

Fig. 10. Photograph of the tensile tested specimens showing the location of fracture in the base metal (a) low heat (b) medium heat (c) high heat input.

Fig. 11. SEM fractograph of the tensile specimen welded at low heat input (a) at 1000 (b) at 2000.

Fig. 12. SEM fractograph of the tensile specimen welded at medium heat input (a) at 1000 (b) at 2000.

S. Kumar, A.S. Shahi / Materials and Design 32 (2011) 3617–3623

3623

Fig. 13. SEM fractograph of the tensile specimen welded at high heat input (a) at 1000 (b) at 2000.

ples are observed. It is also observed that small dimples are surrounded by the large ones in all the specimens and a small quantity of tearing ridge is also present. A similar fractograph observation has been reported for 3 mm thick TIG welded 304 SS where relatively minor size dimples surround coarse dimples besides the presence of small quantity of tearing ridge [4]. 4. Conclusions The following conclusions can be drawn from the present work: Good joint strength is exhibited by all the joints which show that for welding 6 mm thick AISI 304 SS the operating envelope of GTAW process offers a wide range of parameters to the fabricator.  As the dendrite size in the fusion zone is smaller in low heat input joints than the dendrites in medium and high heat input joints, it is found that maximum tensile strength and ductility is possessed by the weld joints made using low heat input.  As heat input increases, the fusion zone and HAZ area also increase. Significant grain coarsening is found in the HAZs of all the joints. It is also observed that the extent of grain coarsening increases with increasing heat input.  Near to the fusion boundary the size of the grains in the HAZ of the joints is found to be relatively coarser at high heat input and finer at low heat input. Based upon the present study it is recommended that low heat input should be preferred when welding AISI 304SS using GTAW process because of the reason that besides giving good tensile strength and ductility, the size of the HAZ and the extent of grain coarsening obtained in these weld joints is less. References [1] Baek Jong-Hyun, Kim Young-Pyo, Kim Woo-Sik, Young-Tai Kho. Fracture toughness and fatigue crack growth properties of the base metal and weld metal of a type 304 stainless steel pipeline for LNG transmission. Int J Press Vessels Pip 2001;78:351–7. [2] Jha Abhay K, Diwaker V, Sreekumar K. Metallurgical investigation on stainless steel bellows used in satellite launch vehicle. Eng Fail Anal 2006;13:1437–47.

[3] Chen TF, Chen YR, Wu W. Properties of Cu–Si enriched type 304 stainless steel welds. Sci Technol Weld Joining 1998;3:75–9. [4] Yan Jun, Gao Ming, Zeng Xiaoyan. Study on microstructure and mechanical properties of 304 stainless steel joints by TIG, laser and laser-TIG hybrid welding. Opt Lasers Eng 2010;48:512–7. [5] Muthupandi V, Srinivasan P Bala, Seshadri SK, Sundaresan S. Effect of weld metal chemistry and heat input on the structure and properties of duplex stainless steel welds. Mater Sci Eng, A 2003;358:9–16. [6] Jana S. Effect of heat input on the HAZ properties of two duplex stainless steels. J Mater Process Technol 1992;33:247–61. [7] Nowacki Jerzy, Rybicki Pawel. The influence of welding heat input on submerged arc welded duplex steel joints imperfections. J Mater Process Technol 2005;164–165:1082–8. [8] Zumelzu E, Sepulveda J, Ibarra M. Influence of microstructure on the mechanical behaviour of welded 316 L SS joints. J Mater Process Technol 1999;94:36–40. [9] Hsieh R-I, Pan Y-T, Liou H-Y. The study of minor elements and shielding gas on penetration in TIG 304 stainless steel. J Mater Eng Perform 1999;8:68–74. [10] Durgutlu Ahmet. Experimental investigation of the effect of hydrogen in argon as a shielding gas on TIG welding of austenitic stainless steel. Mater Des 2004;25:19–23. [11] Lu Shanping, Fujii Hidetoshi, Nogi Kiyoshi. Arc ignitability, bead protection and weld shape variations for He–Ar–O2 shielded GTA welding on SUS304 stainless steel. J Mater Process Technol 2009;209:1231–9. [12] Lee Woei-Shyan, Tzeng Fan-Tzung, Lin Chi-Feng. Mechanical properties of 304L stainless steel SMAW joints under dynamic impact loading. J Mater Sci 2005;40:4839–47. [13] Korinko PS, Malene SH. Considerations for the weldability of types 304L and 316L stainless steel. Pract Fail Anal 2001;4:61–8. [14] Lee DJ, Jung KH, Sung JH, Kim YH, Lee KH, Park JU, et al. Pitting corrosion behavior on crack property in AISI 304L weld metals with varying Cr/Ni equivalent ratio. Mater Des 2009;30:3269–73. [15] Milad M, Zreiba N, Elhalouani F, Baradai C. The effect of cold work on structure and properties of AISI 304 stainless steel. J Mater Process Technol 2008;203:80–5. [16] Shyu SW, Huang HY, Tseng KH, Chou CP. Study of the performance of stainless steel A-TIG welds. J Mater Eng Perform 2008;17:193–201. [17] Lumsden JB, Stocker PJ. Composition of grain boundaries of type 304 stainless steel in low temperature sensitization condition. Scr Metall 1981;15:1295–8. [18] Bruemmer SM, Charlot LA. Development of grain boundary chromium depletion in type 304 and 316 stainless steels. Scr Metall 1986;20:1019–24. [19] Mozhi TA, Juhas MC, Wilde BE. Modeling low temperature sensitization of austenitic stainless steels. Scr Metall 1987;21:1547–52. [20] Jones Richard, Randle Valerie. Sensitization behaviour of grain boundary engineered austenitic stainless steel. Mater Sci Eng A 2010;527:4275–80. [21] Kokawa H, Shimada M, Michiuchi M, Wang ZJ, Sato YS. Arrest of weld-decay in 304 austenitic stainless steel by twin-induced grain boundary engineering. Acta Mater 2007;55:5401–7. [22] ASTM International E08-03. Standard test methods for tension testing of metallic materials. [23] ASTM International E407-07. Standard practices for microetching metal and alloys.