Materials and Design 54 (2014) 331–341
Contents lists available at ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/matdes
The effect of repeated repair welding on mechanical and corrosion properties of stainless steel 316L Iman AghaAli a, Mansour Farzam a, , Mohammad Ali Golozar b, Iman Danaee a ⇑
a
Abadan Faculty of Petroleum Engineering, Petroleum University of Technology, Abadan, Iran Faculty of Materials Engineering, Isfahan University of Technology, Isfahan, Iran
b
a r t i c l e
i n f o
Article history: Received 6 June 2013 Accepted 15 August 2013 Available online 27 August 2013 Keywords: Austenitic stainless steel Weld repairs Heat affected zone Microstructure Mechanical characterization Corrosion
a b s t r a c t
The purpose of this study is to evaluate changes in the mechanical, micro structural and the corrosion properties of stainless steel 316L under repeated repair welding. The welding and the repair welding were conducted by shielded metal arc welding (SMAW). The SMAW welding process was performed usingE316L usingE316L filler filler metal metals. s. Specim Specimen en of the base base metaland metaland diffe differen rentt condit condition ionss of shield shielded ed metal metal arc weldwelding repairs were studied by looking in the micro structural changes, the chemical composition of the phases, the grain size (in the heat affected zone) and the effect on the mechanical and corrosion properties. The microstructure was investigated using optical microscopy (OM) and scanning electron microscopy copy (SEM) (SEM).. The The chemi chemical cal compo composit sition ion of the phas phases es was was deter determi mine ned d using using energ energy y disper dispersiv sive e spectrom spectrometry etry (EDS). (EDS). The corrosion corrosion behavior behavior in 1 M H 2SO4 + 3.5% NaCl solution solution was evalua evaluated ted using a potentiodynam potentiodynamic ic polarization polarization method. Tensile tests, Charpy-V Charpy-V impact resistance and Brinell hardness tests were conducted. Hardness of the heat affected zone decreased as the number of repairs increased. Generally an increase in the yield strength (YS) and the ultimate tensile strength (UTS) occurred with welding. After the first repair, a gradual decrease in YS and UTS occurred but the values of YS and UTS were not less than values of the base metal. Significant reduction in Charpy-V impact resistance with the number of weld repairs were observed when the notch location was in the HAZ. The HAZ of welding repair specimen is more sensitive to pitting corrosion. The sensitivity of HAZ to pitting corrosion was increased by increasing the number of welding repair. 2013 Elsevier Ltd. All rights reserved.
1. Introduction One of the importan importantt mainte maintenan nance ce and repair process processes es is repair repair weldin welding. g. In the metal metal indust industry, ry, the volume volume of repair repair and maintenance is far more than the manufacturing. There is no limitation in the number of repairs in the welding procedures such as API-1104 [1] [1] and ASME Section Section IX [2] [2].. The reference referencess found found in which which the number number of weld weld repairs repairs is lim limited ited are: DNV-OS DNV-OS-F10 -F101 1 [3],, IPS-C-PI-270(2) [3] IPS-C-PI-270(2) [4] [4] and GB50236-98 GB50236-98 [5,6] [5,6] standard standards. s. In the DNV-OS DNV-OS-F10 -F101 1 Append Appendix ix C, sub-sec sub-section tion G 300 it is express expressed ed ‘‘Weld ‘‘Weld seams may only be repaired twice in the same area’’. In the IPS-CPI-270(2), it is stated ‘‘a weld with unacceptable defects may be repaired once only’’. According to the GB50236-97 and GB5023698 standards, no more than two repair welds should be performed in the same area. Most Most of the investi investigat gation ionss about about stainle stainless ss steels steels 316L studied studied on the effects effects of alloying alloying element elements, s, various various heat heat treatme treatments nts and weldwelding techniques on micro structures, oxide film, creep and fatigue behaviors [7–18]. [7–18]. Many Many research researchers ers paid attention attention to weld weld and ⇑
Corresponding Corresponding author. Tel./fax: +98 6314429937. 6314429937. E-mail addresses: addresses:
[email protected],
[email protected], farzam@pu
[email protected] t.ac.ir (M. (M. Farzam).
0261-3069/$ - see front matter 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.matdes.2013.08.052
repair welding, focused on studying the effect or distribution of residual stress. Majority of these studies are based on simulation by finite element [19–25] element [19–25],, and so far very little research on the effects on mechanical properties and corrosion repair welding has been conducted. Jiang Jiang et al. al. [25] studied studied the effect effect of weldin welding g heat heat input input and layer number number on residual stress in repair welds for a stainless steel clad clad plate plate by finite finite eleme element nt meth method od.. Based Based on their their study study with with heat heat input input increas increase, e, transv transverse erse stress stress decrease decreasess while while longitu longitudin dinal al stress changes little and with the welding layer number increase, the residual stresses decrease. They concluded that using multiple-layer and high heat input weld can be useful to decrease the resid residua uall stress stress.. Vega Vega et al. [26] studied studied the effect effect of mu multip ltiple le repairs repairs in the same area in seamless API X52 micro alloyed steel pipe and obtained that a fourth weld repair is also possible. The mechanical properties properties satisfied the requirements requirements of the different different standards. standards. However their investigation did not refer to the corrosion properties. Lin et al. [27] [27] investigated investigated repeated repair welding effect on the the micro micro struc structu tura rall and and mech mechan anica icall prop proper ertie tiess of AISI AISI 304L 304L stainle stainless ss steel. steel. Accordin According g to their their investi investigat gation ion,, an increase increase in number of repair welding in AISI 304L caused uniform and pitting corrosion. The number of weld repair did not have any significant
I. AghaAli et al. / Materials and Design 54 (2014) 331–341
332 Table 1
Chemical composition of the AISI 316L austenitic stainless steel base metal (weight%). C
Si
Mn
P
S
Cr
Mo
Ni
0.03
0.06
1.05
0.022
0.004
16.5
2.1
10.4
Al
Co
Ti
Cu
Nb
V
W
Fe
0.005
0.15
0.019
0.31
0.014
0.034
0.06
Base
Table 2
Chemical composition of the AWS E316L-16 austenitic stainless steel weld metal (weight%). C
Mn
Si
Cr
Ni
Mo
Fe
0.025
0.8
0.9
18.5
12
2.7
Balance
effect on the impact strength but did affect the fracture characteristics. Silva et al. [28] evaluated the effect of welding heat input on the microstructure, hardness and corrosion resistance of AWS E309MoL-16 weld metal, with AISI 316L austenitic stainless steel plates. Their investigation revealed that as heat input increased, the corrosion rate reduced. Jiang et al. [29] studied of effect of multiple repairs welds on residual stress, microstructure and hardness for a stainless steel clad plate. According to their investigation as the repair times
increase, the content of short ferrite is increased, longitudinal and transverse residual stress decreased, the hardness in the diffusion layer is increased because more Fe and C are diffused to the diffusion layer. Therefore, the diffusion layer should be removed completely before re-repair, in order to decrease the risk of crack generation. Based on the considerations of microstructure, residual stress and hardness, it is proposed that this clad plate should not be repaired more than 2 times. Up to day a systematic investigation into the multiple weld repairs effect on the microstructure and mechanical properties of AISI 316L stainless steel has not been performed. Hence this study evaluates change in the mechanical, microstructural and corrosion properties of stainless steel 316L under the multiple weld-repairs effect.
2. Experimental details The material used was AISI 316L austenitic stainless steel, welded using AWS E316L-16 electrode. Tables 1 and 2 show chemical compositions of materials. V-shaped butt welds with dimension shown in Fig. 1a were prepared. The method of welding used was shielded metal arc welding (SMAW). Welding and removing of the weld bead was conducted using a qualified technician. Details of the welding procedure and parameters are shown
Fig. 1. (a) Dimensions of weldment specimens; (b) Schematic illustration showing regions of interest when evaluating microstructural characteristics and corrosion properties of various specimens; (c) Schematic illustrations of tensile test specimen; (d) schematic illustrations of impact test specimen. (Note that illustrations are not to scale, and dimensions are in mm.) Table 3
Welding parameter. Pass Welding process Filler metal
1 2 3 4 5 6 7 8
SMAW SMAW SMAW SMAW SMAW SMAW SMAW SMAW
Class
Diameter (mm)
E316L-16 E316L-16 E316L-16 E316L-16 E316L-16 E316L-16 E316L-16 E316L-16
3.25 3.25 3.25 3.25 3.25 3.25 3.25 3.25
Current (A) Voltage (V) Welding speed (Cm min 1) Welding heat input (kJCm 1) Inter-pass temperature (C)
120 120 120 120 120 120 120 120
20 24 24 25 25 25 25 24
20 20 20 20 20 20 20 20
7.2 8.6 8.6 9 9 9 9 8.6
Note: The inter-pass temperature was maintained at 140 C to avoid variations in the cooling rate among the passes.
27 ± 1 30 ± 1 140 ± 1 140 ± 1 140 ± 1 140 ± 1 140 ± 1 140 ± 1
I. AghaAli et al. / Materials and Design 54 (2014) 331–341
333
Fig. 2. Microstructure of AISI 316L stainless steel. Etch Solution: Glyceregia; etching time: 45–60 s. (a) BM; (b) 0R HAZ; (c) 1R HAZ; (d) 2R HAZ; (e) 3R HAZ; (f) 4R HAZ.
in Table 3. Removal of the weld bead was done by milling (Drilling and milling machine, Model ZX7032). The fusion zone and line removed completely before re-repair. Sample which was welded once is assigned 0R and sample which was repaired once assigned 1R and so on. Thus, five different samples: original and with different number of welding repairs: 0R, 1R, 2R, 3R, 4R were prepared. Detailed microstructural observations were carried out in the heat-affected zone (HAZ) and base metal (BM) region of the specimens are shown in Fig. 1b. The specimens were polished mechanically and then were etched chemically in a Glyceregia solution (3 parts glycerol, 5 parts HCl, 1 part HNO3). Glyceregia etchant attacks phases and outlines the carbides [30]. The surface of each specimen was examined using optical and scanning electron microscope (SEM, VEGA-TESCAN-XMU). The chemical composition and element distribution were determined using energy dispersive X-ray spectrometry (EDS). According to ASTM: E-3 and ASTM: E407, mechanical polishing with emery paper of grit Nos. 120, 240, 400, 600, 800, 1000, 1200, and 2000 was conducted before final polishing with 0.25 lm Al2O3 suspensions and then etched for 45–60 s in the Glyceregia solution. The SEM specimens were etched in the Glyceregia solution for 100–120 s. The assessment of the grain size of the HAZ was carried out according to ASTM: E-112 and ASTM: E-883 using an optical microscope coupled to a digital images analyzer. The hardness measurement was carried out using Brinell test in accordance the ASTM: E-10. Considering the limitations, tensile tests were performed at room temperature according the ISO 6892-1 standards [31]. The samples of the tensile test were prepared to sub size dimension. The location of the ten-
sile test sample has been shown in Fig. 1c. The extension rate was 0.05 mms 1. Several tests for each welding repair condition were carried out. The impact properties of the specimens were tested at a temperature 25 C using an impact test machine. The tests were performed using notch type A (V-notch by radius 0.25 mm) specimens with standard size dimensions (55 mm 10 mm 10 mm: ASTM: E-23 standard) (Fig. 1d). The electrochemical corrosion properties of the specimen immersed in a corrosive solution (chemical composition: 3.5% NaCl + 1 mol H2SO4) at 25 C were determined using a potentiodynamic method
Fig. 3. ASTM grain size number in HAZ as a function of the number of repairs for central area.
I. AghaAli et al. / Materials and Design 54 (2014) 331–341
334
(AUTOLAB; ASTM: G-59). Platinum and the silver/silver chloride (Ag/AgCl) were used as counter and referenced electrode. Each specimen was scanned potentiodynamically at scan rate 1 mV s 1 from an initial potential of 0.8 V to a final potential of 1.4 V.
3. Results and discussion 3.1. Microstructural properties Fig. 2 presents OM images of the original AISI 316L stainless steel BM, and the HAZ regions of the 0R, 1R, 2R, 3R and 4R specimens. The images show that the solidified microstructures of the BM specimen, and the HAZs of the five weld-repair speci-
mens, comprised of austenite matrix, ferrite precipitates and black carbide particles. BM had predominantly d -ferrite lathy morphology, characteristic in both ferrite–austenite (FA) and completely ferrite (F) (Fig. 2a). The morphology of d -ferrite was altered with number of welding. The evaluation of the microstructure, HAZ exposed to the lowest number of welding (R0) had predominantly dferrite lathy morphology similar to BM ( Fig. 2b), but less lathy d ferrite, and d -ferrite with vermicular morphology, fine short ferrite precipitates and black carbide particles were detected. 1R, 2R had a similar microstructure to the previous set of results, but less lathy d-ferrite and more d-ferrite with vermicular morphology were present (Fig. 2c and d). The evaluation of the microstructure HAZs of 3R and 4R show reduced percentage of d -ferrite lathy morphology and increased d-ferrite with vermicular morphology ( Fig. 2e
Fig. 4. Morphologies after Glyceregia chemical etching; etching time: 100–120 s. (a) BM; (b) 0R HAZ; (c) 1R HAZ; (d) 2R HAZ; (e) 3R HAZ; (f) 4R HAZ.
I. AghaAli et al. / Materials and Design 54 (2014) 331–341
335
Table 4
Major alloying-element concentration in BM and HAZ regions of the 0R, 1R, 2R, 3R and 4R specimens. Specimen
Phases
Cr (wt.%)
Ni (wt.%)
Mo (wt.%)
Mn (wt.%)
BM
d-ferrite
23.91 14.97 16.5
4.64 10.52 10.04
4.39 2 2.1
0.15 0.2 1.05
22.08 17.15 16.7
5.46 9.27 10.18
3.67 2.53 2.2
0.76 0.84 1.03
22.28 16.72 16.76
4.91 9.38 10.24
3.69 2.19 2.23
0.72 0.84 1.03
22.59 16.24 16.81
4.74 9.72 10.29
4.39 2.18 2.25
0.7 0.85 1.04
22.83 15.71 16.85
4.7 10.53 10.33
4.86 2.17 2.28
0.68 0.86 1.04
23.95 15.48 16.89
4.5 11.71 10.41
4.93 2.16 2.3
0.6 0.87 1.05
Austenite Total HAZ 0R
d-ferrite
Austenite Total HAZ 1R
d-ferrite
Austenite Total HAZ 2R
d-ferrite
Austenite Total HAZ 3R
d-ferrite
Austenite Total HAZ 4R
d-ferrite
Austenite Total
Table 6
Result of the tensile tests for the different repair conditions.
Fig. 5. d-ferrite as a function of the number of repairs.
Yield strength (offset 0.2%) (Mpa)
Tensile strength (Mpa)
L f (mm)
Elongation (%) Lo = 40 mm
Failure zone
Base
270
554.4
64.15
60.37
R0
340.2
580.5
55.50
38.75
R1
392.4
585
58.00
45
R2
355.5
577.35
57.30
43.35
R3
327.6
576
57.05
42.62
R4
343
566.55
55.20
38
Base metal Base metal Base metal Base metal Base metal Base metal
Table 5
Brinell hardness values for each one of the repair condition and for different locations.
HBW BM HBW 0R HBW 1R HBW 2R HBW 3R HBW 4R
Weld
HAZ
2 cm
3 cm
4 cm
5 cm
– 187 187 187 187 187
146 170 172 167 163 160
146 152 152 152 152 152
146 149 149 149 149 149
146 147 147 147 147 147
146 146 146 146 146 146
Fig. 6. Longitudinal section of specimens from tensile tests.
and f). Fig. 2 shows that the addition of d-ferrite, carbide black deposits in the area is distributed. However, these carbides are more apparent in the austenitic grain boundaries and especially at the border of d -ferrite and austenite phases. The assessment of the grain size in the HAZ was carried out according to ASTM: E-112. It should be noticed that each specimen was analyzedwith 10 fields of view per data with magnifications of 100 X and 200 X, in the central area. Fig. 3 shows ASTM grain size
Fig. 7. Stress vs. elongation profile.
336
I. AghaAli et al. / Materials and Design 54 (2014) 331–341
Fig. 8. Average value of absorbed energy for each one of the repair condition.
number (G) in HAZ for central area is largest for 1R. The grain size number is a function of the number of repairs. At first, the heat of the welding cause formation of new grains and the grain size number is increased. After 1R, by repeating repair welding, the grains grow and get bigger and the grain size number is decreased. Fig. 4 presents scanning electron microscopy (SEM) images of the original AISI 316L stainless steel BM, and the HAZ regions of the 0R, 1R, 2R, 3R and 4R specimens, respectively. According to Table 4, the ferrite phase has a higher percentage of chromium compared to the austenite phase, so ferrite passive layer is more corrosion resistant than that of austenite. Increase or decrease of alloying element results in a concentration gradient of alloying elements (between base and the filler metal). However, significant change at percent of the element has not been observed due to the
Fig. 9. Fracture surface of specimens from impact test for each one of the repair condition; (a) base metal, (b) as-welded, (c) first, (d) second, (e) third and (f) fourth repairs.
I. AghaAli et al. / Materials and Design 54 (2014) 331–341
337
Fig. 10. SEM micrographs of impact surface in beginning of fracture: (a) base metal, (b) as-welded, (c) first, (d) second, (e) third and (f) fourth repairs.
use of the similar filler with base metal. Considerable differences between the percent of manganese in the ferrite and the austenite and total were observed. This is due to the lack of manganese a solid solution. Nevertheless, increased number of repair does not alter the solubility of manganese. Table 4 shows as the number of repair is increased, the solubility of chromium and molybdenum in the ferrite phase is increased. The solubility of nickel and manganese at ferrite phase is decreased. Chromium and molybdenum solubility in the austenite phase is also decreased and the solubility of nickel and manganese in the austenite phase is increased. Fig. 5 shows the amount of d -ferrite tends to decrease with an increase in number of repairs. Silva et al. [28] have concluded that reductions in the level of d-ferrite have been attributed to a slower cooling rate when the welding heat input is increased. This assumption is based on the theory that the cooling rate has a sig-
nificant influence on solidification and solid state transformations of stainless steel weld metals, especially for levels of d-ferrite above 14% [32]. Slower cooling rates would result in lengthy d to c transformation, thus causing a greater percentage of d-ferrite transformation into austenite. Assuming that the heat welding does not melt the HAZ; increase in repair number promotes conditions for the d to c transformation. 3.2. Hardness evaluation Brinell hardness results are reported in Table 5. According to ASTM: E-10, the test force was 29.42 KN (3000 Kgf), ball diameter 10 mm, and test duration was 10–15 s. Weld hardness value is more than the other regions because of differences in percent of alloying elements. Due to the use of same filler in all welding pro-
I. AghaAli et al. / Materials and Design 54 (2014) 331–341
338
cesses, hardness remained constant. Table 5 shows the hardness value of HAZ as a function of weld repair. The Brinell hardness of HAZ has a tendency to decrease with increase in the number of repair. 3.3. Tensile tests The tensile samples were prepared and tested as shown in Fig. 6. The results of the tensile tests are presented in Table 6 and Fig. 7. This behavior indicates a gradual increase in yield strength (YS) and ultimate tensile strength (UTS) reaching maximum value in the 1R, followed by a slight decrease in the second, third and fourth repair. The variation in the YS and UTS can be attributed to the contribution of the HAZ grain size and grain
refinement (Fig. 3). The change in elongation may be looked at the similar manner. 3.4. Impact properties analysis Impact test was conducted using ASTM: E-23. Fig. 8 shows the impact strength of BM was higher than that of the HAZ, but falls as the number of weld repairs increased. Fig. 9 shows the fractographs of the fracture surfaces of different weld repair. The fracture surfaces were dull and fibrous. Fig. 10 shows details of the fractured surface morphologies in beginning of fracture. Increased number of weld-repairs leads to change of the fracture from planar fracture to appearance of ridges. This behavior can be attributed to changing fracture toughness value with the number of weld re-
Fig. 11. SEM micrographs of impact surface in the center or the end of fracture: (a) base metal, (b) as-welded, (c) first, (d) second, (e) third and (f) fourth repairs.
I. AghaAli et al. / Materials and Design 54 (2014) 331–341
pairs. Fig. 11 shows the fractured surface morphologies in the center (or the end) of fracture. The fractured surface of specimens contains planar fracture, in-depth ridges and shear dimples. 3.5. Corrosion properties analysis Fig. 12 shows the polarization curves and the measured data of the electrochemical parameters are presented in Table 7. The results show that the corrosion potentials for the BM specimen, the welded specimen and repaired specimens varied from 0.337 to 0.298 V. The important parameter in pitting corrosion and crevice corrosion is difference between breakdown potential (E B, the lowest potential at which pitting occurs) and protection potential (E P , repassivation potential). When a pit has been initiated, the potential must be decreased down below the protection potential to stop the pit from growing and repassivate the surface. The BM specimen has the highest corrosion and breakdown potentials. On the other hand corrosion potential and breakdown potential and its relevant current density of weld are decreased. With the first repair welding, corrosion current density decreased and difference between breakdown potential and protection potential ( E B – E P ) increased. With each repair, corrosion current decreased and ‘‘E B – E P ’’ increased. Such behavior has been reported for AISI 304L but their corrosion current density did not reduce [27]. 3.6. Discussion The multiple repair welds has great effect on the mechanical properties and corrosion. It is induced by the heat input. When solidification occurs in the primary ferrite phase (FA) or completely ferrite (F), the decrease in the cooling rate leads to a decrease in the d-ferrite level from the solidified liquid metal, and an increase in the ferrite–austenite solid state transformation. Slower cooling rates would result in lengthy d to c transformation. Increase in repair number promotes conditions for the d to c transformation. The ferrite phase has a higher percentage of chromium compared to the austenite phase, so ferrite passive layer is more corrosion resistant than that of austenite. Increase in ’’E B – E P ’’ indicated a reduced pitting and crevice corrosion resistance and decreased corrosion current density indicated an increased uniform corrosion resistance. Such behavior is due to the changing morphology and volume of d -ferrite. By performing a repair welding, volume of d -ferrite is reduced. Uniform corrosion is decreased with the decreasing percent of d -ferrite as demonstrated [28]. The effect of d-ferrite on corrosion resistance can also be attributed to the difference in the chemical composition of d -ferrite and austenite. The presence of two phases can form an active–passive region, accelerating the attack on the
Fig. 12. Potentiodynamic curves of AISI 316L stainless steel, base metal and HAZ of the different welding repair in 1 M H2SO4 + 3.5% NaCl solution at 25 C.
339
Table 7
Electrochemical properties of AISI 316L stainless steel BM and weld-repair specimens when immersed in 1 M H 2SO4 + 3.5% NaCl solution at 25 C.
Base metal 0R 1R 2R 3R 4R
E corr (V)
icorr (Acm
2)
E B–E P (V)
0.298 0.318 0.333 0.337 0.320 0.329
3.07 10 2.01 10 1.69 10 1.61 10 1.01 10 9.47 10
6
0.786 0.791 0.833 0.910 0.922 0.928
6
6
6
6
7
E corr : corrosion potential, i corr : corrosion current density, E B–E P : difference between breakdown potential and protection potential.
austenite matrix [28,33]. Increase in ‘‘E B – E P ’’ implies the enhancement to chloride sensitivity of the HAZ of repeated welding. Its result is the transformation of the original lathy ferrite phases to a fine distributed short ferrite precipitates in the austenite. The repeated weld repair process prompts a greater transformation of the lathy ferrite phase to short ferrite precipitates, and therefore increasing pitting corrosion [15,27]. The greater sensitivity of the 4R is the result of the higher grain boundary energy induced by the repeated welding repair process. This is due to more transformation of the lathy d-ferrite phase to d-ferrite of vermicular morphology, therefore increasing corrosion attack. However, the uniform corrosion resistance increased due to the reducing d -ferrite phase. The grain size number is a function of the number of repairs. After the first repair, increasing the number of weld repairs promotes grain growth in the coarse grained heat affected zone (CGHAZ). The bead technique generates overlap beads producing grain refinement in the CGHAZ of the previous bead and decreasing the residual stresses due to the input of additional thermal energy [34]. This is the reason that the one repair presents the maximum value of YS and UTS. Generally, the Brinell hardness of HAZ has a tendency to decrease with increase in the number of repair. This behavior can be attributed to the bigger grain size, the reduction in the level of d-ferrite and the morphology of d-ferrite. In 0R and 1R, the reduction in grain size is the dominant factor, whereas in 2R, 3R, and 4R all the factors are responsible for reduction in hardness. Hardness increases in the first repair due to the generation of new grain. Hardness decrease in the subsequent repairs is due to grain growth, reduction of ferrite and the change in morphology of d -ferrite. The increase in hardness after the first repair and a decrease in the second and third repairs, because of an apparent grain refinement generated in the HAZ and grains grow during second and third repair as demonstrated [26,35,36]. Silva et al. [28] showed the average hardness of lathy d-ferrite morphology was larger than the average value of vermicular d -ferrite morphology. Padilha and Guedes [37] showed that the increase in the volumetric fraction of d-ferrite leads to a higher value of hardness. Vega et al. [26] studied the effect of multiple repairs in the same area in seamless API X52 micro alloyed steel pipe and reported a similar results. The impact strength of BM was higher than that of the HAZ, but falls as the number of weld repairs increased. Repeating repair welding changes the fracture mechanism from a planar fracture to ridges present. The change of morphology of d-ferrite from lathy to vermicular (and d -ferrite dispersion) may be the reason for the changes mentioned. The micro voids presence may be related to d-ferrite. Another reason for decrease in the absorbed energy may be related to repeat welding which reduced percentage of d ferrite. According to research conducted by Yılmaz and Tümer [38], toughness is directly related to the amount of ferrite.
340
I. AghaAli et al. / Materials and Design 54 (2014) 331–341
Based on the Jiang’s [25] study, using multiple-layer and high heat input weld can be useful to decrease the residual stress. Although the residual stress is not studied in this paper independently but the results of tensile and hardness tests represent the residual stress decreased with increasing frequency of repair. The harness has a close relationwith the strength [39,40], whichmeans the local strength in HAZ has been degraded because of the multiple heating. 4. Conclusions This paper investigated the effect of repeating repair welding on microstructure, hardness, tensile strength, impact strength and corrosion resistance for a stainless steel 316L by experimental, the following conclusions are obtained: (1) The microstructures of the BM specimen, and the HAZs of the weld-repair specimens, comprised of austenite matrix, ferrite precipitates and black carbide particles. Heat induced of welding repair changes the structure and amount of d-ferrite. Repeating repair welding will transform d -ferrite morphology to fine short ferrite precipitates and will reduce the amount of ferrite. (2) Brinell hardness of HAZ decreased with increasing number of repair welding. The analysis showed reduction of the percentage of d -ferrite is the main reason for such behavior. (3) The tensile test results showed that repeating repair welding did not have much adverse effect on yield and ultimate tensile strength. The yield strength and the ultimate tensile initially increased and then decreased, this was due to grain size and grain refinement effect. The grain size decreased during the first repair and later increased with increasing number of repairs. (4) The impact test showed a significant fall in the absorbed energy values as the number of weld repairs increased. This was due to the change in the fracture mechanism because of the transformed d-ferrite morphology and the increased grain boundary energy and decreased d -ferrite phase. (5) The HAZ of weld-repair specimen is sensitive to pitting corrosion when immersed in 1 M H2SO4 + 3.5% NaCl solution. Repeating repair welding causes the sensitivity of HAZ to pitting corrosion and crevice corrosion increased. This is due to more transformation of the lathy d-ferrite phase to d-ferrite of vermicular morphology, therefore increasing corrosion attack. However, the uniform corrosion resistance increased due to the reducing d -ferrite phase. (6) Based on the compressive considerations of residual stress, microstructure, hardness, tensile strength, impact strength and corrosion resistance, it concludes that the stainless steel 316L can be repaired for 4 times in chloride-free environment. But in chloride environment, because of damaging effects of chlorine and formation of stress corrosion cracking (SCC), repairing more than 2 is not suggested. According to the results in chlorine-free environment, there is no limit to numbers of repairs.
References [1] API STANDARD 1104: Welding of pipeline and related facilities. 20 ed.; 2005. [2] ASME SECTION IX: Qualification standard for welding and brazing procedures, welders, brazers, and welding and brazing operators; 2002. [3] DNV-OS-F101: Submarine pipeline systems; 2000. [4] IPS-C-PI-270(2): Construction standard for welding of transportation pipeline, 2009. [5] GB Standard 50236–97, GB, China; 1999.
[6] GB Standard 50236–98. GB, China; 1999. [7] Zareie Rajani HR, Torkamani H, Sharbati M, Raygan Sh. Corrosion resistance improvement in gas tungsten arc welded 316L stainless steel joints through controlled preheat treatment. Mater Des 2012;34:51–7. [8] Katherasan D, Sathiya P, Raja A. Shielding gas effects on flux cored arc welding of AISI 316L (N) austenitic stainless steel joints. Mater Des 2013;45:43–51 . [9] Cui Y, Lundin Carl D. Evaluation of initial corrosionlocationin E316L austenitic stainless steel weld metals. Mater Lett 2005;59:1542–6 . [10] Mathew MD, Sasikala G, Bhanu Sankara Rao K, Mannan SL. Influence of carbon and nitrogen on the creep properties of type 316 stainlesssteel at 873 K. Mater Sci Eng A Struct Mater 1991;148:253–60 . [11] Sasikala G, Ray SK, Mannan SL. Kinetics of transformation of delta ferrite during creep in a type 316(N) stainlesssteel weld metal. MaterSci Eng A Struct Mater 2003;359:86–90. [12] Robertson C, Fivel MC, Fissolo A. Dislocation substructure in 316L stainless steel under thermal fatigue up to 650 K. Mater Sci Eng A Struct Mater 2001;315:47–57. [13] Ganesh P, Moitra A, Tiwari P, Sathyanarayanan S, Kumar H, Rai SK, et al. Fracture behavior of laser-clad joint of Satellite 21 on AISI 316L stainless steel. Mater Sci Eng A Struct Mater 2010;527:3748–56 . [14] Liqing H, Guobiao L, Zidong W, Hong Z, Feng L, Long Y. Study on corrosion resistance of 316L stainless steel welded Joint. Rare Metal Mat Eng 2010;39:393–6. [15] Garcia C, Martin F, de Tiedra P, Blanco Y, Lopez M. Pitting corrosion of welded joints of austenitic stainless steels studied by using an electrochemical minicell. Corros Sci 2008;50:1184–94. [16] Garcia C, de Tiedra MP, Blanco Y, Martin O, Martin F. Intergranular corrosionof welded joints of austenitic stainless steels studied by using an electrochemical minicell. Corros Sci 2008;50:2390–7. [17] Li SX, He YN, Yu SR, Zhang PY. Evaluation of the effect of grain size on chromium carbide precipitation and intergranular corrosion of 316L stainless steel. Corros Sci 2013;66:211–6. [18] Kocabekir B, Kaçar R, Gündüz S, Hayat F. An effect of heat input, weld atmosphere and weld cooling conditions on the resistance spot weldability of 316L austenitic stainless steel. J Mater Process Technol 2008;195:327–35 . [19] Lahti KE, Hänninen H, Niemi E. Nominal stress range fatigue of stainless steel fillet welds — the effect of weld size. J Constr Steel Res 2000;54:161–72 . [20] Bouchard PJ, George D, Santisteban JR, Bruno G, Dutta M, Edwards L, et al. Measurement of the residual stresses in a stainless steel pipe girth weld containing long and short repairs. Int J Pres Ves Pip 2005;82(4):299–310 . [21] George D, Smith DJ. Through thickness measurement of residual stresses in a stainless steel cylindercontainingshallowand deep weld repairs. IntJ Pres Ves Pip 2005;82:279–87. [22] Mcdonald EJ, Hallam KR, Flewitt PEJ. A strategy for accommodating residual stresses in the assessment of repair weldments based upon measurement of near surface stresses. Int J Pres Ves Pip 2005;82:339–46 . [23] Xu S, Wang W. Numerical investigation on weld residual stresses in tube to tube sheet joint of a heat exchanger. Int J Pres Ves Pip 2013;101:37–44 . [24] Li Yongkui, Kaji Yoshiyuki, Igarashi Takahiro. Effects of thermal load and cooling condition on weld residual stress in a core shroud with numerical simulation. Nucl Eng Des 2012;242:100–7 . [25] Jiang WC, Wang BY, Gong JM, Tu ST. Finite element analysis of the effect of welding heat input and layer number on residual stress in repair welds for a stainless steel clad plate. Mater Des 2011;32:2851–7 . [26] Vega OE, Hallen JM, Villagomez A, Contreras A. Effect of multiple repairs in girth welds of pipelines on the mechanical properties. Mater Charact 2008;59:1498–507. [27] Lin CM, Tsai HL, Cheng CD, Yang C. Effect of repeated weld-repairs on microstructure, texture, impact properties and corrosion properties of AISI 304L stainless steel. Eng Fail Anal 2012;21:9–20 . [28] Silva CC, de Miranda HC, de Sant’Ana HB, Farias JP. Microstructure, hardness and petroleum corrosion evaluation of 316L/AWS E309MoL-16 weld metal. Mater Charact 2009;60:346–52. [29] Jiang W, Luo Y, Zhang G, Woo W, Tu ST. Experimental to study the effect of multiple weld-repairs on microstructure, hardness and residual stress for a stainless steel clad plate. Mater Des 2013;51:1052–9. [30] ASM Handbook Volume 09: Metallography and Microstructures, 1992. [31] ISO 6892–1: Metallic materials – Tensile testing – Part 1: Method of test at room temperature; 2009. [32] Folkhard E. Welding metallurgy of stainless steels. New York: Springer-Verlag Wien; 1998. [33] Cui Y, Lundin CD. Austenite-preferential corrosion attack in 316 austenitic stainless steel weld metals. Mater Des 2007;28:324–8 . [34] Lant T, Robinson DL, Spafford B, Storesund J. Review of weld repair procedures for low alloy steels designed to minimize the risk of future cracking. Int J Pres Ves Pip 2001;78:813–8. [35] T. McGaughy, The influence of weld repairs on changes in residual stress and fracture toughness in pipeline girth welds. In: Recent Advances in Structural Mechanics, ASME; 1992, PVP 248, p. 81–6. [36] McGaughy T. Significance of changes in residual stresses and mechanical properties due to SMAW repair girth welds in line pipe. Edison Welding Institute and Pipeline Research Council International, Inc.; 1990. p. 1–18 . [37] Padilha AF, Guedes, LC. In: Hemus editor. Austenitic stainless steels, São Paulo Brazil; 1994 [in Portuguese].
I. AghaAli et al. / Materials and Design 54 (2014) 331–341 [38] Y ılmaz R, Tümer M. Microstructural studies and impact toughness of dissimilar weldments between AISI 316 L and AH36 steels by FCAW. Int J Adv Manuf Technol 2012;29:262–8 . [39] Zhang P, Li SX, Zhang ZF. General relationship between strength and hardness. Mater Sci Eng A 2011;529:62–73.
341
[40] Zhu Ming-Liang, Xuan Fu-Zhen. Correlation between microstructure, hardness and strength in HAZ of dissimilar welds of rotor steels. Mater Sci Eng A 2010;527:4035–42.