- Email: [email protected]
Effect of post weld heat treatment (PWHT) on the microstructure, mechanical properties, and corrosion resistance of dissimilar stainless steels
Materials Science & Engineering A 688 (2017) 470–479
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Effect of post weld heat treatment (PWHT) on the microstructure, mechanical properties, and corrosion resistance of dissimilar stainless steels ⁎
MARK
⁎
Saeid Ghorbania, , Reza Ghasemib,c, Reza Ebrahimi-Kahrizsangia, Akbar Hojjati-Najafabadid, a
Advanced Materials Research Center, Materials Engineering Department, Najafabad Branch, Islamic Azad University, Najafabad, Iran Iranian Academic Center for Education, Culture and Research (ACECR), Khorasan Razavi Branch, Azadi Square, P.O. Box 91775-1376, Mashhad, Iran c University of Applied Sciences and Technology (UAST), Khorasan Razavi Branch, Mashhad, Iran d Young Researchers and Elite Club, Najafabad Branch, Islamic Azad University, Najafabad, Iran b
A R T I C L E I N F O
A BS T RAC T
Keywords: Stainless steels Heat treatment Mechanical properties Corrosion properties
The dissimilar austenitic stainless steel (AISI304L) and ferritic stainless steel (AISI430) have been welded with two types of filler metals (316L and 2594L) by GTAW. Then, the effect of heat treatment on the microstructure, mechanical properties, and corrosion properties of welded joint was investigated. Due to remove chromium carbide created during welding process and homogenize the microstructure, heat treatment was carried out on the whole series of the samples at 860 °C and 960 °C. The microstructural, fracture cross-section and corrosive areas of the samples were investigated by SEM. Tensile, bending and potentiodynamic polarization tests were employed to characterize mechanical and corrosion performance of the samples. The results indicate that the welded joints had good mechanical properties after heat treatment so that the best tensile strength was obtained at 960 °C. Due to reduce grain size, the heat treatment samples show minimum corrosion resistance at 960 °C in comparison to 860 °C.
1. Introduction
welding, they are formed the chromium carbide, hence, it cause to decrease the corrosion properties lower than base metal [7]. There are various methods to remove chromium carbide like reduce volume of carbon in base metal, using laser welding and electron beam welding to reduce the heat input to the base metal during welding, consequently the base metal be less at the temperature range of formation chromium carbide. However, they are so expensive methods [8]. The addition to stainless steels of elements such as Nb, V and Ti which have a higher affinity of carbon in comparison to chrome and they prevent the formation of chromium carbide [6,9]. Heat treatment is another process to remove chromium carbide. In dissimilar joints, for decreasing metallurgical destructive effects on the base metal, it is essential to select a suitable temperature for removal of chromium carbide in heataffected zones (HAZ) and weld metal [10,11]. The previous work in this field have focused on modern and expensive welding methods such as laser welding and electron beam welding for reducing formation of chromium carbide in ferriticaustenitic stainless steels joint. The particular aims of present work were to investigate the effect of heat treatment on complete removal of chromium carbide, the microstructure and improvement of mechanical and corrosion properties of ferritic-austenitic stainless steels joint. After heat treatment the welding joins, they have been characterized
Due to high flexibility, low production costs, and joint ability of ferritic stainless steels (AISI400) have been used as the resistant steels against corrosion in different industries such as oil and gas, electricity, chemical, petrochemical industries and etc [1–4]. These steels are suitable replacement to austenitic steel in chloride environments and its microstructures undergo many changes like reduce of toughness and ductility properties during welding [1]. Today, in major industries are trying to optimize materials composition and properties for achieving to high quality products and low production costs. One of the ways to manufacture high quality products is the use of jointing of dissimilar metals [3]. In recent years, the joint of austenitic steels to ferritic steels have attracted more attention. One the most important issues must be considered to avoid chromium carbide formation in different areas of the junction [5]. In fact, the precipitates are developed into the intergranulars besides the insulating grain boundaries due to welding on these steels at jointing different areas. There are various carbide precipitates at different steels such as M23C6 type in unstabilized AISI 430 ferritic stainless steels and MC type in stabilized AISI 444 ferritic stainless steels [3,6]. When the stainless steels are subject to temperature range 550–850 °C during
⁎
Corresponding authors. E-mail addresses: [email protected] (S. Ghorbani), [email protected] (A. Hojjati-Najafabadi).
http://dx.doi.org/10.1016/j.msea.2017.02.020 Received 3 December 2016; Received in revised form 4 February 2017; Accepted 4 February 2017 Available online 05 February 2017 0921-5093/ © 2017 Elsevier B.V. All rights reserved.
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Table 1 Chemical composition of base and filler metals. Chemical compounds Element
Fe
C
Si
Mn
P
S
Cr
Mo
Ni
Co
Cu
W
AISI304L AISI430 ER316L ER2594L
71.8 82.4 BAL BAL
0.03 0.18 0.03 0.02
0.5 0.42 0.5 0.4
0.96 0.42 1.8 0.4
0.03 0.02 – –
0.01 0.006 – –
18.3 16.1 19 25
0.27 0.03 2.8 4
8.24 0.17 11.5 9.8
0.1 0.02 – –
0.14 0.14 0.3 ≥0.3
0.02 0.02 – ≥1
tungsten arc weld (GTAW) process using a 316L and 2594L filler metal. The chemical composition of materials used in this study are shown in Table 1. The corresponding welding parameters are shown in Table 2. Samples with the dimensions of 3 mm×90 mm×300 mm welded by GTAW process in a single pass. V-shaped butt welds with the dimensions shown in Fig. 1. Furthermore, the schematically of different zone of specimens after welding was shown in Fig. 2. The mechanical evaluation was also done through tensile mechanical test (INSTRON, 4486, USA) according to AWS instructions. Transverse tensile specimens with a gage length of 24 mm and a width of 6 mm (overall length: 100 mm) were prepared from the weld coupons in as-welded condition. Room-temperature tensile tests were conducted on three samples as per ASTM E8 on a universal tensile testing machine. To minimize the machining error (noise) three specimens were prepared at each levels of the designed matrix. The dimensions of tensile specimen are shown in Fig. 3. The prepared tensile specimens were subjected to tensile test and their ultimate tensile strengths were evaluated. Bending test was also according to ASME Sec IX standard. Metallographic and microstructural studies
Table 2 Heat treatment parameters. Base metal
Filler metal
Heat treatment temperature (°C)
304L 304L 304L 304L
2594 2594 316L 316L
860 960 860 960
& & & &
430 430 430 430
using scanning electron microscopy (SEM). Potentiodynamic polarization tests were used to study the corrosion behavior of the welding joints, and mechanical properties of them was investigated using tensile and bending tests. 2. Experimental method The tests were performed using plates of austenitic stainless steel (AISI304L) and ferritic stainless steel (AISI430) welded by the gas
Fig. 1. Dimension of welding plate of 304/430 dissimilar metals.
471
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Fig. 2. The schematically of different zone of specimens after welding.
Fig. 3. Dimensions of tensile specimen.
Fig. 4. SEM micrograph of HAZ area of both stainless steels (A) 430 and (B) 304 before heat treatment process.
472
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Fig. 5. Heat treatment cycle of samples at (a) 860 °C and (b) 960 °C for 2 h.
Fig. 6. The SEM micrograph of heat-affected zone after heat treatment process at 860 °C and 960 °C, (a) stainless steel 430 at 860 °C, (b) Stainless steel 430 at 960 °C, (c) Stainless steel 304 at 860 °C and (d) Stainless Steel 304 at 960 °C.
473
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Table 3 The result of the tensile test. Type Filler Filler Filler Filler Filler Filler
metal metal metal metal metal metal
2594L 2594L 2594L 316L 316L 316L
Heat treatment temperature (°C)
Percent of increasing length (%)
Tensile strength (MPa)
Yield strength (Mpa)
– 860 960 – 860 960
10 ± 2 28 ± 3 35 ± 1 15 ± 2 24 ± 3 40 ± 2
389 ± 5 433 ± 4 624 ± 5 409 ± 4 419 ± 3 615 ± 4
230 ± 4 265 ± 5 440 ± 4 222 ± 5 252 ± 6 410 ± 3
Fig. 7. SEM micrographs of fracture cross-section of the ferritic steel 430 samples in HAZ area (a) unheated treatment, (b) heat treatment at 860 °C and (c) heat treatment at 960 °C.
Ag/AgCl/Cl- reference electrode. All tests were repeated at least three times in order to assess reproducibility and were performed in an aqueous 3.5 wt% NaCl+0.1% FeCl3 solution at room temperature.
after etching samples with oxalic acid were carried out by optical microscope and scanning electrone microscope (LEO-435VP, Germany) in order to identify chromium carbide. Samples heat treatment was carried out in induction furnace (EXITON, EX.14506LA, Iran) at 860 °C and 960 °C. In this study, the whole series of samples were kept under two temperatures (860 °C and 960 °C) for 120 min; then, in order to prevent carbide precipitation, creation of high-temperature brittleness, and precipitation of sigma phase, the samples were evicted from the furnace and cooled in water. Corrosion properties of base metals and weld different areas and corrosion comparison of the two weld metals were carried out by two TAFLE methods (Princeton Applied Research, 2273 EG & G PARSTAT, USA). A three-electrode cell equipped with a platinum counter electrode, an
3. Results and discussion 3.1. Microstructure properties Fig. 4 shows the SEM micrograph of HAZ area of both stainless steels (430 & 304L) before heat treatment process. As it can be seen in HAZ areas, due to exposure to temperatures above 450 °C, both of steels are sensitive to grain boundary corrosion and chromium carbide has been formed in grain boundary. It can be attributed to combination 474
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Fig. 8. The typical SEM micrograph of fracture cross-section sample of bending test before heat treatment.
Fig. 9. The images of bending samples (A) before and (B & C) after heat treatment process.
because of rapid cooling. The heat treatment of both base metals implies the decrease of δ-ferrite in the base metal structure at 960 °C compared to 860 °C, as shown in Fig. 3c and d. The decrease of δferrite lead to improvement of mechanical properties [13].
of carbon with chromium around the grain boundary areas that causing the lack of chromium around the grain boundary areas and these areas are prone to corrosion [12]. In order to evaluation the effect of temperature on the microstructure, mechanical and corrosion properties as well as removal chromium carbide, the heat treatment was performed at two different temperatures. The heat treatment cycle was performed in according to Table 2 and Fig. 5. The SEM micrograph of heat-affected zone after heat treatment process at 860 °C and 960 °C have been shown in Fig. 6. As it observed in Fig. 3a and b, at both temperatures, the chromium carbide have been completely dissolved and not again precipitated in grain boundary
3.2. Tensile properties The results of the tensile test for different samples are reported in Table 3. As it can be seen that the heat treatment samples at 960 °C have highest strength. In according to SEM micrograph in Fig. 6, with increasing temperature from 860 °C to 960 °C, the chromium carbides 475
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Fig. 10. The potentiodynamic curves of the different areas of samples include base metals, heat-affected zone and welded metal at temperatures 860 °C and 960 °C with different filler metals, (a) base metal 430 (b) base metal 304 (c) HAZ of steel 430 (d) HAZ of steel 304 (e) weld metal 316L (f) weld metal 2594L.
476
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
be more dissolved and also the microstructure is more homogeneous and more fine-grained. The whole of series of AISI 430 ferritic stainless steels heat treatment samples at 860 °C and 960 °C have been fractured from HAZ area and base metal, respectively. The results indicate that dissolving precipitates and decreasing the ferrite matrix grains leading to increase the toughness of the welding joint. Actually, with decreasing the ferrite grain size, due to improvement of size distribution of martensitic areas, the mechanical properties of the welding joint increased [14,15]. Fig. 7a shows the fracture cross-section of the unheated treatment of sample after welding. It is revealed that the fracture occurred intergranulars, which is represented the brittle fracture. It may be attributed to presence of coarse and continuous martensitic areas and precipitated of hard and brittle chromium carbide in grain boundaries [16,17]. The fracture cross-section of heat treatment samples at 860 °C and 960 °C are shown in Fig. 4b and c, respectively. It was found that both of samples show almost the ductile fracture in these areas. The fracture cross-sections have many big and small dimples. This indicate that the samples have high tensile strength and good ductility [8]. Due to different temperatures of heat treatment process, the size and depth of fracture cross-section and dimples are different. However, the microstructure of ferritic-martensitic in stainless steels have good mechanical properties in compared to ferritic and martensitic steels, which is in agreement with the results of this study and previous literature in this field [18,19].
acquired for samples are listed in Table 4. The results show that the corrosion current density (ICorr) values reduced at 860 °C in comparison to 960 °C for all of the samples. Due to higher temperature at 960 °C and the presence of carbon in 430 ferritic stainless steel base metal, the austenite phase is stabilized. Because of quenching in water, the austenite phase is transformed to martensitic phase. The martensitic phase is an energetic and unstable phase, hence it acts as a sacrificed anode and lead to increase of corrosion rate at corrosive environment (Fig. 10a) [13]. For austenite stainless steel base metal in Fig. 10b, the residual stress increased in grain boundaries after quenching in water. On the other hand, it can be possible, due to more temperature for diffusion at 960 °C in comparison to 860 °C, the formation and precipitation of impurities in grain boundaries increased. Furthermore, these factors will be increased the energy in grain boundaries and cause to decrease of corrosion resistance. For example, Mn forms the manganese sulfide inclusions in the grain boundary and increases the corrosion rate [10,20]. Similar with 430 base metal, the formation of austenite phase and transform to martensitic phase as well as reduction of chromium in HAZ of 430 ferritic steel in Fig. 10c, the corrosion resistance decrease at 960 °C [9]. The corrosion results acquired for welded metal with 316L filler metal in Fig. 10d show that the corrosion resistance decrease at 960 °C due to presence of Mn in 316L filler metal. In fact, the Mn forms MnS in grain boundaries more comfortable at 960 °C in compared to 860 °C and lead to decrease corrosion properties. In according to increase Cr and Mo in 2594 filler metal in comparison to other samples, it is possible to form the σ-phase at 960 °C in welded metal [21]. So, it can be decreased the corrosion resistance at welded joint at 960 °C. The SEM micrographs were used to investigate the surface of whole series of samples after corrosion test. As it can be observed in Fig. 11, the corrosion severity of samples at 960 °C is more than 860 °C and the pores created on the metal surface at 860 °C are smaller than 960 °C. Also, due to the size and distribution of precipitations is non-uniform, the corrosion severity of whole series of the samples is not the same in all areas and the areas devoid of precipitations have been protected from corrosion. The precipitations adjacent areas and energetic grain boundaries are the sites for beginning of localized corrosion at Cr and Fe-rich matrix. These areas are shown the cathodic behavior in comparison to matrix and the potential difference between the two areas is causing a small galvanic cell. Therefore, the localized corrosion increased.
3.3. Bending test
4. Conclusion
To determine the qualification of welding process and improvement of welded joints using 316L and 2594L filler metal, the bending test was performed. The results show that the whole series of unheated treatment of 430 ferritic stainless steel samples were fractured from heat-affected zones (HAZ). The intergranulars cleavage fracture were seen in all of the samples. Fig. 8 show the typical SEM micrograph of fracture cross-section sample of bending test before heat treatment. The cause of the brittle fracture in these areas are likely to coarse grain, intra and intergranulars martensitic, chromium carbide precipitation in grain boundaries and among martensitic boundary and ferritic matrix [4]. After heat treatment process, the whole series of samples were passed bending test at both temperatures (860 °C and 960 °C). It can be attributed to remove the chromium carbide and decrease in grain size [13]. Fig. 9 show the images of bending samples before and after heat treatment process.
In this study, the effect of heat treatment on corrosion and mechanical properties and the dissimilar joint of stainless steel austenitic 304L and ferritic 430 was investigated at temperatures 860 °C and 960 °C. The main findings of this research can be summarized as follows:
Table 4 The results of corrosion testing. Sample
Ecorr (mV)
Icorr (µA/cm2)
Base304L – 860 °C Base304L – 960 °C Base 430 – 860 °C Base 430 – 960 °C HAZ 304L – 860 °C HAZ 304L – 960 °C HAZ 430 – 860 °C HAZ 430 – 960 °C FZ ER316L – 860 °C FZ ER316L – 960 °C FZ ER2594L – 860 °C FZ ER2594L – 960 °C
−91.41 −202.40 −230.07 −395.01 −109.71 −288.30 −318.41 −544.52 −135.54 −158.73 −60.42 −147.74
4.85 ± 0.15 5.12 ± 0.09 11.14 ± 0.41 11.81 ± 0.37 5.43 ± 0.33 4.37 ± 0.24 7.97 ± 0.51 11.70 ± 0.61 5.23 ± 0.08 7.54 ± 0.45 4.88 ± 0.44 5.12 ± 0.29
1. With increasing the heat treatment temperature from 860 °C to 960 °C, the more austenite phase stabilized at HAZ area of 430 ferritic steel. Due to quenching, the austenite phase was transformed to martensitic phase. 2. Due to brittle chromium carbide and coarse and continues martensitic phase in grain boundaries at non-heated treatment samples, the HAZ area was fractured during bending test at low stresses. The heat treatment changed the microstructure and improved the toughness and strength of HAZ area. 3. Because of removal chromium carbides, reduce of ferrite grains size and decrease of martensitic grains size, the heat treatment samples had good mechanical properties at 960 °C in comparison to 860 °C. 4. The corrosion current density values reduced at 860 °C in comparison to 960 °C for all of the samples. The main factors for increasing corrosion resistance were residual stress and presence of precipitation in grain boundaries due to quenching.
3.4. Corrosion resistance Fig. 10 shows the potentiodynamic curves of the different areas of samples include base metals, heat-affected zone and welded metal at temperatures 860 °C and 960 °C with different filler metals. Electrochemical characteristics derived from the polarization curves 477
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Fig. 11. SEM micrographs of samples after corrosion test (a) base metal 430 at 860 °C, (b) base metal 430 at 960 °C, (c) base metal 304L at 860 °C, (d) base metal 304L at 960 °C, (e) HAZ area of ferritic steel 430 at 860 °C, (f) HAZ area of ferritic steel at 960 °C, (g) HAZ area of austenitic steel at 860 °C, (h) HAZ area of austenitic steel at 960 °C, (i) weld metal 316L at 860 °C, (j) weld metal 316L at 960 °C, (k) weld metal 2594L at 860 °C, (l) weld metal 2594L at 960 °C.
2393–2399. [6] M. Vach, T. Kuníková, M. Dománková, P. Ševc, L. Čaplovič, P. Gogola, et al., Mater. Charact. 59 (2008) 1792–1798. [7] M. Greeff, M. Toit, Weld. J. 85 (2006) 243–251. [8] S. Kumar, S. Shahi, Mater. Des. 32 (2011) 3617–3623. [9] M. Van Warmelo, D. Nolan, J. Norrish, Mater. Sci. Eng. A 464 (2007) 157–169. [10] M. Dadfar, M.H. Fathi, F. Karimzadeh, M.R. Dadfar, A. Saatchi, Mater. Lett. 61 (2007) 2343–2346. [11] P. Sathiya, S. Aravindan, A.N. Haq, Mater. Des. 29 (2008) 1099–1109.
References [1] M.O.H. Amuda, S. Mridha, Int. J. Mech. Mater. Eng. 4 (2009) 160–166. [2] M.O.H. Amuda, S. Mridha, Adv. Mater. Res. 83–86 (2009) 1165–1172. [3] M.M.A. Khan, L. Romoli, M. Fiaschi, G. Dini, F. Sarri, J. Mater. Proc. Technol. 212 (2012) 856–867. [4] E. Taban, E. Kaluc, A. Dhooge, Mater. Des. 30 (2009) 4236–4242. [5] F.F. Curiel, R. García, V.H. López, J. González-Sánchez, Corros. Sci. 53 (2011)
478
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
[17] M.M.A. Khan, L. Romoli, M. Fiaschi, G. Dini, F. Sarri, J. Mater. Process. Technol. 212 (2012) 856–867. [18] H.W. Hayden, S. Floreen, Metall. Trans. 1 (1970) 1955–1959. [19] R. Wright, J.R. Wood, Metall. Trans. A 8 (1977) 2007–2011. [20] Y. Cui, C.D. Lundin, Mater. Des. 28 (2007) 324–328. [21] S. Kožuh, M. Gojić, L. Kosec, Mater. Geoenviron. 54 (2007) 331–344.
[12] C. Garcia, F. Martin, P. De Tiedra, Y. Blanco, M. Lopez, Corros. Sci. 50 (2008) 1184–1194. [13] J. Lippold, D. Kotecki, Welding Metallurgy and Weldability of Stainless Steels, John Wily & Sons, 2005. [14] K.M. Lee, H.C. Lee, J. Mater. Process. Technol. 210 (2010) 1574–1579. [15] D.K. Mondal, R.M. Dey, Sci. Eng. A 49 (1992) 173–181. [16] C.M. Lin, H.L. Tsai, C.D. Cheng, C. Yang, Eng. Fail. Anal. 21 (2012) 9–20.
479
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Effect of post weld heat treatment (PWHT) on the microstructure, mechanical properties, and corrosion resistance of dissimilar stainless steels ⁎
MARK
⁎
Saeid Ghorbania, , Reza Ghasemib,c, Reza Ebrahimi-Kahrizsangia, Akbar Hojjati-Najafabadid, a
Advanced Materials Research Center, Materials Engineering Department, Najafabad Branch, Islamic Azad University, Najafabad, Iran Iranian Academic Center for Education, Culture and Research (ACECR), Khorasan Razavi Branch, Azadi Square, P.O. Box 91775-1376, Mashhad, Iran c University of Applied Sciences and Technology (UAST), Khorasan Razavi Branch, Mashhad, Iran d Young Researchers and Elite Club, Najafabad Branch, Islamic Azad University, Najafabad, Iran b
A R T I C L E I N F O
A BS T RAC T
Keywords: Stainless steels Heat treatment Mechanical properties Corrosion properties
The dissimilar austenitic stainless steel (AISI304L) and ferritic stainless steel (AISI430) have been welded with two types of filler metals (316L and 2594L) by GTAW. Then, the effect of heat treatment on the microstructure, mechanical properties, and corrosion properties of welded joint was investigated. Due to remove chromium carbide created during welding process and homogenize the microstructure, heat treatment was carried out on the whole series of the samples at 860 °C and 960 °C. The microstructural, fracture cross-section and corrosive areas of the samples were investigated by SEM. Tensile, bending and potentiodynamic polarization tests were employed to characterize mechanical and corrosion performance of the samples. The results indicate that the welded joints had good mechanical properties after heat treatment so that the best tensile strength was obtained at 960 °C. Due to reduce grain size, the heat treatment samples show minimum corrosion resistance at 960 °C in comparison to 860 °C.
1. Introduction
welding, they are formed the chromium carbide, hence, it cause to decrease the corrosion properties lower than base metal [7]. There are various methods to remove chromium carbide like reduce volume of carbon in base metal, using laser welding and electron beam welding to reduce the heat input to the base metal during welding, consequently the base metal be less at the temperature range of formation chromium carbide. However, they are so expensive methods [8]. The addition to stainless steels of elements such as Nb, V and Ti which have a higher affinity of carbon in comparison to chrome and they prevent the formation of chromium carbide [6,9]. Heat treatment is another process to remove chromium carbide. In dissimilar joints, for decreasing metallurgical destructive effects on the base metal, it is essential to select a suitable temperature for removal of chromium carbide in heataffected zones (HAZ) and weld metal [10,11]. The previous work in this field have focused on modern and expensive welding methods such as laser welding and electron beam welding for reducing formation of chromium carbide in ferriticaustenitic stainless steels joint. The particular aims of present work were to investigate the effect of heat treatment on complete removal of chromium carbide, the microstructure and improvement of mechanical and corrosion properties of ferritic-austenitic stainless steels joint. After heat treatment the welding joins, they have been characterized
Due to high flexibility, low production costs, and joint ability of ferritic stainless steels (AISI400) have been used as the resistant steels against corrosion in different industries such as oil and gas, electricity, chemical, petrochemical industries and etc [1–4]. These steels are suitable replacement to austenitic steel in chloride environments and its microstructures undergo many changes like reduce of toughness and ductility properties during welding [1]. Today, in major industries are trying to optimize materials composition and properties for achieving to high quality products and low production costs. One of the ways to manufacture high quality products is the use of jointing of dissimilar metals [3]. In recent years, the joint of austenitic steels to ferritic steels have attracted more attention. One the most important issues must be considered to avoid chromium carbide formation in different areas of the junction [5]. In fact, the precipitates are developed into the intergranulars besides the insulating grain boundaries due to welding on these steels at jointing different areas. There are various carbide precipitates at different steels such as M23C6 type in unstabilized AISI 430 ferritic stainless steels and MC type in stabilized AISI 444 ferritic stainless steels [3,6]. When the stainless steels are subject to temperature range 550–850 °C during
⁎
Corresponding authors. E-mail addresses: [email protected] (S. Ghorbani), [email protected] (A. Hojjati-Najafabadi).
http://dx.doi.org/10.1016/j.msea.2017.02.020 Received 3 December 2016; Received in revised form 4 February 2017; Accepted 4 February 2017 Available online 05 February 2017 0921-5093/ © 2017 Elsevier B.V. All rights reserved.
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Table 1 Chemical composition of base and filler metals. Chemical compounds Element
Fe
C
Si
Mn
P
S
Cr
Mo
Ni
Co
Cu
W
AISI304L AISI430 ER316L ER2594L
71.8 82.4 BAL BAL
0.03 0.18 0.03 0.02
0.5 0.42 0.5 0.4
0.96 0.42 1.8 0.4
0.03 0.02 – –
0.01 0.006 – –
18.3 16.1 19 25
0.27 0.03 2.8 4
8.24 0.17 11.5 9.8
0.1 0.02 – –
0.14 0.14 0.3 ≥0.3
0.02 0.02 – ≥1
tungsten arc weld (GTAW) process using a 316L and 2594L filler metal. The chemical composition of materials used in this study are shown in Table 1. The corresponding welding parameters are shown in Table 2. Samples with the dimensions of 3 mm×90 mm×300 mm welded by GTAW process in a single pass. V-shaped butt welds with the dimensions shown in Fig. 1. Furthermore, the schematically of different zone of specimens after welding was shown in Fig. 2. The mechanical evaluation was also done through tensile mechanical test (INSTRON, 4486, USA) according to AWS instructions. Transverse tensile specimens with a gage length of 24 mm and a width of 6 mm (overall length: 100 mm) were prepared from the weld coupons in as-welded condition. Room-temperature tensile tests were conducted on three samples as per ASTM E8 on a universal tensile testing machine. To minimize the machining error (noise) three specimens were prepared at each levels of the designed matrix. The dimensions of tensile specimen are shown in Fig. 3. The prepared tensile specimens were subjected to tensile test and their ultimate tensile strengths were evaluated. Bending test was also according to ASME Sec IX standard. Metallographic and microstructural studies
Table 2 Heat treatment parameters. Base metal
Filler metal
Heat treatment temperature (°C)
304L 304L 304L 304L
2594 2594 316L 316L
860 960 860 960
& & & &
430 430 430 430
using scanning electron microscopy (SEM). Potentiodynamic polarization tests were used to study the corrosion behavior of the welding joints, and mechanical properties of them was investigated using tensile and bending tests. 2. Experimental method The tests were performed using plates of austenitic stainless steel (AISI304L) and ferritic stainless steel (AISI430) welded by the gas
Fig. 1. Dimension of welding plate of 304/430 dissimilar metals.
471
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Fig. 2. The schematically of different zone of specimens after welding.
Fig. 3. Dimensions of tensile specimen.
Fig. 4. SEM micrograph of HAZ area of both stainless steels (A) 430 and (B) 304 before heat treatment process.
472
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Fig. 5. Heat treatment cycle of samples at (a) 860 °C and (b) 960 °C for 2 h.
Fig. 6. The SEM micrograph of heat-affected zone after heat treatment process at 860 °C and 960 °C, (a) stainless steel 430 at 860 °C, (b) Stainless steel 430 at 960 °C, (c) Stainless steel 304 at 860 °C and (d) Stainless Steel 304 at 960 °C.
473
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Table 3 The result of the tensile test. Type Filler Filler Filler Filler Filler Filler
metal metal metal metal metal metal
2594L 2594L 2594L 316L 316L 316L
Heat treatment temperature (°C)
Percent of increasing length (%)
Tensile strength (MPa)
Yield strength (Mpa)
– 860 960 – 860 960
10 ± 2 28 ± 3 35 ± 1 15 ± 2 24 ± 3 40 ± 2
389 ± 5 433 ± 4 624 ± 5 409 ± 4 419 ± 3 615 ± 4
230 ± 4 265 ± 5 440 ± 4 222 ± 5 252 ± 6 410 ± 3
Fig. 7. SEM micrographs of fracture cross-section of the ferritic steel 430 samples in HAZ area (a) unheated treatment, (b) heat treatment at 860 °C and (c) heat treatment at 960 °C.
Ag/AgCl/Cl- reference electrode. All tests were repeated at least three times in order to assess reproducibility and were performed in an aqueous 3.5 wt% NaCl+0.1% FeCl3 solution at room temperature.
after etching samples with oxalic acid were carried out by optical microscope and scanning electrone microscope (LEO-435VP, Germany) in order to identify chromium carbide. Samples heat treatment was carried out in induction furnace (EXITON, EX.14506LA, Iran) at 860 °C and 960 °C. In this study, the whole series of samples were kept under two temperatures (860 °C and 960 °C) for 120 min; then, in order to prevent carbide precipitation, creation of high-temperature brittleness, and precipitation of sigma phase, the samples were evicted from the furnace and cooled in water. Corrosion properties of base metals and weld different areas and corrosion comparison of the two weld metals were carried out by two TAFLE methods (Princeton Applied Research, 2273 EG & G PARSTAT, USA). A three-electrode cell equipped with a platinum counter electrode, an
3. Results and discussion 3.1. Microstructure properties Fig. 4 shows the SEM micrograph of HAZ area of both stainless steels (430 & 304L) before heat treatment process. As it can be seen in HAZ areas, due to exposure to temperatures above 450 °C, both of steels are sensitive to grain boundary corrosion and chromium carbide has been formed in grain boundary. It can be attributed to combination 474
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Fig. 8. The typical SEM micrograph of fracture cross-section sample of bending test before heat treatment.
Fig. 9. The images of bending samples (A) before and (B & C) after heat treatment process.
because of rapid cooling. The heat treatment of both base metals implies the decrease of δ-ferrite in the base metal structure at 960 °C compared to 860 °C, as shown in Fig. 3c and d. The decrease of δferrite lead to improvement of mechanical properties [13].
of carbon with chromium around the grain boundary areas that causing the lack of chromium around the grain boundary areas and these areas are prone to corrosion [12]. In order to evaluation the effect of temperature on the microstructure, mechanical and corrosion properties as well as removal chromium carbide, the heat treatment was performed at two different temperatures. The heat treatment cycle was performed in according to Table 2 and Fig. 5. The SEM micrograph of heat-affected zone after heat treatment process at 860 °C and 960 °C have been shown in Fig. 6. As it observed in Fig. 3a and b, at both temperatures, the chromium carbide have been completely dissolved and not again precipitated in grain boundary
3.2. Tensile properties The results of the tensile test for different samples are reported in Table 3. As it can be seen that the heat treatment samples at 960 °C have highest strength. In according to SEM micrograph in Fig. 6, with increasing temperature from 860 °C to 960 °C, the chromium carbides 475
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Fig. 10. The potentiodynamic curves of the different areas of samples include base metals, heat-affected zone and welded metal at temperatures 860 °C and 960 °C with different filler metals, (a) base metal 430 (b) base metal 304 (c) HAZ of steel 430 (d) HAZ of steel 304 (e) weld metal 316L (f) weld metal 2594L.
476
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
be more dissolved and also the microstructure is more homogeneous and more fine-grained. The whole of series of AISI 430 ferritic stainless steels heat treatment samples at 860 °C and 960 °C have been fractured from HAZ area and base metal, respectively. The results indicate that dissolving precipitates and decreasing the ferrite matrix grains leading to increase the toughness of the welding joint. Actually, with decreasing the ferrite grain size, due to improvement of size distribution of martensitic areas, the mechanical properties of the welding joint increased [14,15]. Fig. 7a shows the fracture cross-section of the unheated treatment of sample after welding. It is revealed that the fracture occurred intergranulars, which is represented the brittle fracture. It may be attributed to presence of coarse and continuous martensitic areas and precipitated of hard and brittle chromium carbide in grain boundaries [16,17]. The fracture cross-section of heat treatment samples at 860 °C and 960 °C are shown in Fig. 4b and c, respectively. It was found that both of samples show almost the ductile fracture in these areas. The fracture cross-sections have many big and small dimples. This indicate that the samples have high tensile strength and good ductility [8]. Due to different temperatures of heat treatment process, the size and depth of fracture cross-section and dimples are different. However, the microstructure of ferritic-martensitic in stainless steels have good mechanical properties in compared to ferritic and martensitic steels, which is in agreement with the results of this study and previous literature in this field [18,19].
acquired for samples are listed in Table 4. The results show that the corrosion current density (ICorr) values reduced at 860 °C in comparison to 960 °C for all of the samples. Due to higher temperature at 960 °C and the presence of carbon in 430 ferritic stainless steel base metal, the austenite phase is stabilized. Because of quenching in water, the austenite phase is transformed to martensitic phase. The martensitic phase is an energetic and unstable phase, hence it acts as a sacrificed anode and lead to increase of corrosion rate at corrosive environment (Fig. 10a) [13]. For austenite stainless steel base metal in Fig. 10b, the residual stress increased in grain boundaries after quenching in water. On the other hand, it can be possible, due to more temperature for diffusion at 960 °C in comparison to 860 °C, the formation and precipitation of impurities in grain boundaries increased. Furthermore, these factors will be increased the energy in grain boundaries and cause to decrease of corrosion resistance. For example, Mn forms the manganese sulfide inclusions in the grain boundary and increases the corrosion rate [10,20]. Similar with 430 base metal, the formation of austenite phase and transform to martensitic phase as well as reduction of chromium in HAZ of 430 ferritic steel in Fig. 10c, the corrosion resistance decrease at 960 °C [9]. The corrosion results acquired for welded metal with 316L filler metal in Fig. 10d show that the corrosion resistance decrease at 960 °C due to presence of Mn in 316L filler metal. In fact, the Mn forms MnS in grain boundaries more comfortable at 960 °C in compared to 860 °C and lead to decrease corrosion properties. In according to increase Cr and Mo in 2594 filler metal in comparison to other samples, it is possible to form the σ-phase at 960 °C in welded metal [21]. So, it can be decreased the corrosion resistance at welded joint at 960 °C. The SEM micrographs were used to investigate the surface of whole series of samples after corrosion test. As it can be observed in Fig. 11, the corrosion severity of samples at 960 °C is more than 860 °C and the pores created on the metal surface at 860 °C are smaller than 960 °C. Also, due to the size and distribution of precipitations is non-uniform, the corrosion severity of whole series of the samples is not the same in all areas and the areas devoid of precipitations have been protected from corrosion. The precipitations adjacent areas and energetic grain boundaries are the sites for beginning of localized corrosion at Cr and Fe-rich matrix. These areas are shown the cathodic behavior in comparison to matrix and the potential difference between the two areas is causing a small galvanic cell. Therefore, the localized corrosion increased.
3.3. Bending test
4. Conclusion
To determine the qualification of welding process and improvement of welded joints using 316L and 2594L filler metal, the bending test was performed. The results show that the whole series of unheated treatment of 430 ferritic stainless steel samples were fractured from heat-affected zones (HAZ). The intergranulars cleavage fracture were seen in all of the samples. Fig. 8 show the typical SEM micrograph of fracture cross-section sample of bending test before heat treatment. The cause of the brittle fracture in these areas are likely to coarse grain, intra and intergranulars martensitic, chromium carbide precipitation in grain boundaries and among martensitic boundary and ferritic matrix [4]. After heat treatment process, the whole series of samples were passed bending test at both temperatures (860 °C and 960 °C). It can be attributed to remove the chromium carbide and decrease in grain size [13]. Fig. 9 show the images of bending samples before and after heat treatment process.
In this study, the effect of heat treatment on corrosion and mechanical properties and the dissimilar joint of stainless steel austenitic 304L and ferritic 430 was investigated at temperatures 860 °C and 960 °C. The main findings of this research can be summarized as follows:
Table 4 The results of corrosion testing. Sample
Ecorr (mV)
Icorr (µA/cm2)
Base304L – 860 °C Base304L – 960 °C Base 430 – 860 °C Base 430 – 960 °C HAZ 304L – 860 °C HAZ 304L – 960 °C HAZ 430 – 860 °C HAZ 430 – 960 °C FZ ER316L – 860 °C FZ ER316L – 960 °C FZ ER2594L – 860 °C FZ ER2594L – 960 °C
−91.41 −202.40 −230.07 −395.01 −109.71 −288.30 −318.41 −544.52 −135.54 −158.73 −60.42 −147.74
4.85 ± 0.15 5.12 ± 0.09 11.14 ± 0.41 11.81 ± 0.37 5.43 ± 0.33 4.37 ± 0.24 7.97 ± 0.51 11.70 ± 0.61 5.23 ± 0.08 7.54 ± 0.45 4.88 ± 0.44 5.12 ± 0.29
1. With increasing the heat treatment temperature from 860 °C to 960 °C, the more austenite phase stabilized at HAZ area of 430 ferritic steel. Due to quenching, the austenite phase was transformed to martensitic phase. 2. Due to brittle chromium carbide and coarse and continues martensitic phase in grain boundaries at non-heated treatment samples, the HAZ area was fractured during bending test at low stresses. The heat treatment changed the microstructure and improved the toughness and strength of HAZ area. 3. Because of removal chromium carbides, reduce of ferrite grains size and decrease of martensitic grains size, the heat treatment samples had good mechanical properties at 960 °C in comparison to 860 °C. 4. The corrosion current density values reduced at 860 °C in comparison to 960 °C for all of the samples. The main factors for increasing corrosion resistance were residual stress and presence of precipitation in grain boundaries due to quenching.
3.4. Corrosion resistance Fig. 10 shows the potentiodynamic curves of the different areas of samples include base metals, heat-affected zone and welded metal at temperatures 860 °C and 960 °C with different filler metals. Electrochemical characteristics derived from the polarization curves 477
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
Fig. 11. SEM micrographs of samples after corrosion test (a) base metal 430 at 860 °C, (b) base metal 430 at 960 °C, (c) base metal 304L at 860 °C, (d) base metal 304L at 960 °C, (e) HAZ area of ferritic steel 430 at 860 °C, (f) HAZ area of ferritic steel at 960 °C, (g) HAZ area of austenitic steel at 860 °C, (h) HAZ area of austenitic steel at 960 °C, (i) weld metal 316L at 860 °C, (j) weld metal 316L at 960 °C, (k) weld metal 2594L at 860 °C, (l) weld metal 2594L at 960 °C.
2393–2399. [6] M. Vach, T. Kuníková, M. Dománková, P. Ševc, L. Čaplovič, P. Gogola, et al., Mater. Charact. 59 (2008) 1792–1798. [7] M. Greeff, M. Toit, Weld. J. 85 (2006) 243–251. [8] S. Kumar, S. Shahi, Mater. Des. 32 (2011) 3617–3623. [9] M. Van Warmelo, D. Nolan, J. Norrish, Mater. Sci. Eng. A 464 (2007) 157–169. [10] M. Dadfar, M.H. Fathi, F. Karimzadeh, M.R. Dadfar, A. Saatchi, Mater. Lett. 61 (2007) 2343–2346. [11] P. Sathiya, S. Aravindan, A.N. Haq, Mater. Des. 29 (2008) 1099–1109.
References [1] M.O.H. Amuda, S. Mridha, Int. J. Mech. Mater. Eng. 4 (2009) 160–166. [2] M.O.H. Amuda, S. Mridha, Adv. Mater. Res. 83–86 (2009) 1165–1172. [3] M.M.A. Khan, L. Romoli, M. Fiaschi, G. Dini, F. Sarri, J. Mater. Proc. Technol. 212 (2012) 856–867. [4] E. Taban, E. Kaluc, A. Dhooge, Mater. Des. 30 (2009) 4236–4242. [5] F.F. Curiel, R. García, V.H. López, J. González-Sánchez, Corros. Sci. 53 (2011)
478
Materials Science & Engineering A 688 (2017) 470–479
S. Ghorbani et al.
[17] M.M.A. Khan, L. Romoli, M. Fiaschi, G. Dini, F. Sarri, J. Mater. Process. Technol. 212 (2012) 856–867. [18] H.W. Hayden, S. Floreen, Metall. Trans. 1 (1970) 1955–1959. [19] R. Wright, J.R. Wood, Metall. Trans. A 8 (1977) 2007–2011. [20] Y. Cui, C.D. Lundin, Mater. Des. 28 (2007) 324–328. [21] S. Kožuh, M. Gojić, L. Kosec, Mater. Geoenviron. 54 (2007) 331–344.
[12] C. Garcia, F. Martin, P. De Tiedra, Y. Blanco, M. Lopez, Corros. Sci. 50 (2008) 1184–1194. [13] J. Lippold, D. Kotecki, Welding Metallurgy and Weldability of Stainless Steels, John Wily & Sons, 2005. [14] K.M. Lee, H.C. Lee, J. Mater. Process. Technol. 210 (2010) 1574–1579. [15] D.K. Mondal, R.M. Dey, Sci. Eng. A 49 (1992) 173–181. [16] C.M. Lin, H.L. Tsai, C.D. Cheng, C. Yang, Eng. Fail. Anal. 21 (2012) 9–20.
479