Evolution of residual stress and microstructure of 2205 duplex stainless steel welded joints during di ﬀ erent post-weld heat treatment

: Duplex stainless steel (DSS) has been widely used in various applications due to the combination of excellent mechanical properties and corrosion resistance. However, shielded manual arc welding (SMAW) always deteriorates its phase balance and further changes its mechanical properties. Therefore, an appropriate post-weld heat treatment (PWHT) is of the essence to gain the superior performance of the DSS SMAW joint. In this article, the e ﬀ ects of PWHT temperature on the microstructure and residual stress of 2205 DSS SMAW joint were investigated by both experimental and simulation methods. The microstructural characteristics including phase ratio, morphology, grain misorientation, and boundary type were analyzed by the electron backscattered di ﬀ raction, while the evolution of residual stress was investigated by a thermal – mechanical coupled ﬁ nite-element simulation and hole drilling method. The results showed that the residual stress decreased signi ﬁ cantly after PWHT, particularly under the higher PWHT temperature. The maximum longitudinal residual stress had dropped by 20.4 and 66.8% at the PWHT temperatures 380 and 1,050°C, which were both far below the yield strength. However, the increase in PWHT temperature promoted the phase proportion imbalance due to the excessive precipitation of intragranular austenite and the formation of low-angle grain boundaries. The fraction of austenite had reached 75.5% when the PWHT temperature was 1,050°C. In order to obtain a reasonable distribution of residual stress and microstructure for the 2205 DSS SMAW joint, it is recommended to perform PWHT at 380°C.


Introduction
Duplex stainless steel (DSS) is characterized by its high strength and excellent corrosion resistance, making it a suitable choice for various applications, including filter presses, chemical cargo tanks, sulfuric acid evaporators, and heat exchangers handling chloride-containing media [1][2][3].Shielded manual arc welding (SMAW) is a common welding method for DSS equipment.However, during the welding process, the weld zone and the heat-affected zone (HAZ) undergo complex phase transformations due to the influence of thermal cycling, resulting in an unbalanced phase ratio, which would weaken the mechanical properties.Furthermore, the generation of welding residual stresses is inevitable, which adversely affects the performance of the joint, thereby reducing its service life.
Post-weld heat treatment (PWHT) is a crucial method for reducing residual stresses and modifying the microstructure in the welded joints.Numerous researchers have investigated the reduction of welding residual stress through PWHT.Jabran et al. [4] examined a 2.25Cr-1Mo circumferential butt-welded pipe fabricated using tungsten inert gas (TIG) welding and observed that PWHT could decrease the stress gradient, although stress peaks persisted at the edges of the heated region.Yu et al. [5] explored the impact of PWHT on residual stress relief in SMA490BW steel melt active gas arc welding joints, finding significant stress reduction during the heating stage, with the final residual stress distribution relatively unaffected by the cooling rate.Liu et al. [6] investigated laser beam welding of Ti-45Al-5Nb-0.2C-0.2Balloy and determined that PWHT at 1,260°C for 2 h completely alleviated longitudinal residual stresses.Chao et al. [7] conducted selective laser melting forming on 316L stainless steel powder, analyzing changes in residual stress and microstructure pre-and post-heat treatment.Their findings indicated a decrease in residual stress occurred after heat treatment, changing to a stable state with smaller compressive stresses.Despite numerous studies confirming the efficacy of PWHT in significantly reducing welding residual stress, limited attention has been directed toward the residual stress evolution in DSS SMAW joints due to PWHT.
Muthupandi et al. [8] investigated the optimal microstructure and mechanical properties of joints by controlling the heat input of DSS welds.They found that the proportion of dual phases in DSS often failed to meet the requirements of balance during rapid cooling.Ureña et al. [9] welded 2205 DSS using the plasma arc welding method and observed that at high energy inputs, the ferrite content in the weld and fusion zone exceeded 45%, significantly higher than that in the base material.Cui et al. [10] employed the K-TIG welding method for DSS welding and reported an austenite content of approximately 42% in the weld.These studies suggest that different welding methods employed for DSS welding result in varying phase contents and imbalanced phase proportions in the welds.In order to address the imbalance in phase proportions, Vijayalakshmi et al. [11] investigated the influence of annealing temperature on the microstructure of DSS, revealing that a temperature of 1,350°C results in coarsening of ferrite grains, while rapid quenching promotes the formation of austenite.Zhang et al. [12] studied the evolution of the weld microstructure of 2507 DSS under short-term PWHT at different temperatures.The results indicated a significant increase in the austenite content in the weld zone after a brief heat treatment.Singh et al. [13] examined the effect of isothermal treatment on the microstructure of 2205 DSS gas tungsten arc welded joints, finding a notable decrease in ferrite content after exposure to 850°C isothermal treatment for 2 h.Badji et al. [14] found that the HAZ of 2205 DSS gas tungsten arc welding joints had a higher ferrite content, while the weld center contained more austenite.The ferrite content increased with the increase of annealing temperature, and the rate of formation in the HAZ was higher than that in the base material and fusion zone.Zhang et al. [15] investigated the microstructural evolution of UNS S32750 super DSS TIG welded joints before and after heat treatment.They observed a significant increase in the volume fraction of austenite in the HAZ and weld metal after annealing at 1,080°C.Zhang et al. [16] studied the microstructural evolution of electron beam welded joints of DSS under different heat treatment temperatures.They found that PWHT effectively promoted the formation of austenite.Further investigations by Zhang et al. [17] revealed that heat treatment facilitated the formation of austenite and the dissolution of Cr 2 N, eliminating dendritic segregation.Shrikrishna and Sathiya [18] performed PWHT on friction-welded joints of DSS.The PWHT involved quenching and water cooling after annealing at 1,080°C, achieving a balanced volume fraction of ferrite and austenite without the precipitation of other phases.
Previous research studies have pointed out that there are significant differences in the austenite content of DSS prepared by different welding methods.However, the effects of PWHT on the formation of austenite in different welding methods are different; some would promote, while some would restrain.Furthermore, there has been no comprehensive study from the perspective of residual stress and microstructure on the effects of PWHT processes on SMAW joints.Therefore, the appropriate heat treatment process for SMAW joints remains unclear.This article focuses on 2205 DSS SMAW welded joints for investigation.The evolution of residual stress was investigated by a thermalmechanical coupled finite-element simulation and hole drilling method.The microstructural characteristics including phase ratio, morphology, grain misorientation, and boundary type were also analyzed.The optimal PWHT method was then proposed according to the distribution of residual stress and microstructure.

Sample preparation
Two SAF2205 DSS plates were butt-welded by SMAW, and the welding consumable is ER2209.The chemical composition of the base metal and welding material is provided in Table 1 [19].To ensure sufficient penetration, a 60°bevel angle was adopted.In order to match welding performance and avoid welding defects, welding parameters, as shown in Table 2, were selected.As illustrated in Figure 1, the welding joint has a dimension of 400 mm (X) × 10 mm (Y) × 300 mm (Z).The welding joint consists of four welding passes and one root pass.
The sample was subjected to bulk heat treatment after welding, and the heat treatment temperature profile is shown in Figure 2. The heating time (t 1 ) is 6 h, the holding time (t 2 ) is 6 h, and the sample is cooled to room temperature in the furnace.In order to investigate the influence of heat treatment temperatures on residual stress and microstructure, three heat treatment processes were designed: low temperature, medium temperature, and high temperature, in conjunction with the phase transformation of DSS [20,21], as illustrated in Table 3.

Residual stress measurement by the hole drilling method
The principle of the hole drilling method is to paste strain gauges on the surface of the workpiece to be measured and punch holes to relax the stress around the holes and form a new stress/strain field distribution.By calibrating strain release coefficients A and B, the original residual stress and strain of the workpiece can be calculated based on the principle of elastic mechanics.The strain generated on the three strain gauges by drilling is related to the residual principal stress as follows: (1) (2) where σ 1 and σ 2 represent the principal stresses (MPa); ε 1 , ε 2 , and ε 3 represent strains in the 0°, 45°, and 90°directions,    respectively; A and B are the strain release coefficients; and φ is the angle between σ 1 and the 0°strain gauge [22,23].
In this hole drilling method measurement, an HT101C strain and stress measurement instrument was used with a drill hole diameter of 1 mm and a depth of 2 mm.BX120-3CA three-directional strain gauges were used.The measurement positions are shown in Figure 1.

Microstructure characterization
For welding and PHWT samples, microstructure analysis was performed at the locations of the first and fourth weld seams.The microstructure of the as-welded sample has been observed in our previous work [24].The sample size is 15 mm (X) × 10 mm (Y) × 3 mm (Z).The samples were polished with 240#, 400#, 600#, 800#, 1,000#, 1,500#, and 2,000# sandpapers, followed by mechanical polishing with 5, 2.5, and 1 μm diamond polishing agents.After electrolytic corrosion with NaOH solution (40 g NaOH + 200 g H 2 O), the microstructure observation can be carried out under the optical microscope.The samples were then conducted for the electron backscattered diffraction (EBSD) test after vibratory polishing.

Finite-element model (FEM)
In this study, a three-dimensional FEM was established based on the actual welding sample dimension, as shown in Figure 3.The model consists of 67,200 elements and 74,186 nodes.During the simulation of the temperature field, hexahedral heat transfer elements (DC3D8) were used.For the analysis of the welding stress field, eight-node reduced integration three-dimensional stress elements (C3D8R) were used.

Temperature field simulation
A double ellipsoid heat source was adopted to model the heat generation from the moving welding arc [25].The heat flow density distributions of the front and rear ellipsoids in the heat source are shown in equations ( 3) and (4), respectively where Q is the heat input of the arc; a, b, c f , and c r are the shape parameters of the ellipsoid; and m and n represent the energy fractions of the front and rear hemispheres (m + n = 2).In general, m = 0.4, and n = 1.6.

Stress field simulation
The sequential coupling method was then carried out, where the results of the temperature field were applied to the stress field as a predefined field.In order to avoid rigid displacement of the specimen, displacement constraints are carried out at the endpoints on the bottom surface.The total strain is decomposed into three components as follows: where ε e , ε p , and ε th denote elastic strain, plastic strain, and thermal strain, respectively.The elastic strain was calculated using isotropic Hooke's law with temperature- dependent Young's modulus and Poisson's ratio.The plastic strain was calculated using the von Mises yield criterion, temperature-dependent mechanical properties, and isotropic hardening model of the plastic model.Thermal strains were calculated using temperature-dependent thermal expansion coefficients.
The mechanical properties of DSS are calculated by the parameter calculation software JMatPro, including the modulus of elasticity E, the coefficient of linear expansion, Poisson's ratio μ, and so on, which are shown in Table 5.

Heat treatment simulation
Heat treatment simulation is also a thermal-mechanical sequential coupled process, which is similar to the welding simulation.The heat treatment temperature simulation would be carried out through the actual process with three PWHT temperatures.The material parameters required for simulation are the same as those in Tables 4 and 5.

Effect of PWHT temperature on residual stresses
Figure 4 shows the residual stress distribution in the welded component without undergoing heat treatment.Initially, the residual stress is symmetrically distributed, with higher residual stress near the weld seam and the HAZ, approaching the material's yield strength.In the longitudinal direction, the residual stress exhibits tensile stress in the middle of the weld seam, reaching a maximum value of 530 MPa.To ensure overall stress balance, compressive stress is observed at both ends, with a maximum value of 216 MPa.As further away from the weld seam, the stress values decrease.The transverse residual stress is significantly lower than the longitudinal residual stress.Its highest tensile stress, reaching 262 MPa, is observed at the upper and lower surfaces of the weld toes, and the stress decreases as we move away from the weld seam.Two paths, P1 and P2, were selected to analyze the distribution and magnitude of welded residual stresses, as shown in Figure 1.P1 is perpendicular to the weld seam, while P2 is in the thickness direction at the center of the weld seam.The transverse and longitudinal stress distributions along these two paths are shown in Figures 5  and 6.It is evident that the maximum tensile stress is observed in the HAZ, with a maximum longitudinal residual stress value of 488 MPa.Transverse residual stress is lower than the longitudinal stress, with a maximum value of 169 MPa.As away from the weld seam, residual stresses gradually decrease and approach 0 MPa.Comparing the data obtained from the hole drilling method experiments with the numerical simulation results, as shown in Figure 6, it is found that, within an acceptable range, there is some deviation, but the distribution is consistent.This, to some extent, validates the accuracy and reliability of the finite- Residual stress and microstructure of 2205 DSS welded joints  5 element simulation method.The experimental results show lower values compared to the finite-element simulation.
There are several possible reasons for this difference: the hole drilling method involves polishing the specimen surface during the initial preparation, releasing some of the residual stresses, resulting in lower residual stress values during testing compared to the initial state.Additionally, the assumption is made that all residual stresses are released after drilling.Inevitably, there may be some remaining stresses that are not completely released, leading to slightly lower measurement results.Figure 4(c)-(h) shows the residual stress after heat treatment.The maximum residual stresses in the welded components are reduced after heat treatment, with varying degrees of reduction at different heat treatment temperatures.As shown in Figure 4, after PWHT at 1,050°C, the stress distribution trend is generally the same as that of the aswelded sample.However, the residual stresses are greatly released and the reduction degree is significantly increased with the increase of temperature.
Residual stress distribution curves after heat treatment were plotted perpendicular to the weld path, as shown in Figure 5.It can be observed that at PHWT temperatures ranging from 380 to 1,050°C, the transverse residual stress in the weld seam decreases from 75 to 42 MPa, and the longitudinal residual stress decreases from 345 to 133 MPa.Residual stress peaks still exist in the HAZ after heat treatment.Combining with Figure 4, it can be seen that treatment option A (380°C) has a weaker effect, but the maximum stress is also reduced.After PWHT, the maximum transverse and longitudinal residual stresses are 223.7 and 422.6 MPa, with reductions of 14.8 and 20.4%, respectively.This indicates that although residual stresses are reduced after heat treatment, but for better residual stress distribution, the heat treatment temperature should be increased.Treatment option B (750°C) results in a more effective reduction of residual stresses compared to option A. The transverse and longitudinal residual stress values after heat treatment are 198.0 and 368.0 MPa, respectively, with reductions of 24.6 and 30.7%, making it a more Along the through-thickness direction of the weld seam, stress distribution curves before and after heat treatment are plotted, as shown in Figure 6.The transverse residual stresses gradually decrease from the upper surface to the lower surface, with a slight increase near the lower surface.The stress is lowest at the location of the first weld bead, and due to the tempering effect of subsequent weld passes on the preceding weld bead, the stress is higher at the positions of subsequent weld passes [26].It can be observed that with higher heat treatment temperatures, the stress curves become smoother and closer to the 0 stress axis.Furthermore, the longitudinal stress does not have an obvious fluctuation along the thickness; its maximum ranges from 465 to 130 MPa from the as-welded condition to the 1,050°C PWHT condition.

Effect of heat treatment temperature on microstructure 4.2.1 Microstructure morphology
Figure 7 displays the morphology of the weld root region before and after heat treatment at different temperatures.
After PWHT at 380°C, the weld's microstructure remains largely unchanged, resembling the microstructure of the as-welded joint, with minimal changes in phase content.As the temperature is relatively low, which does not reach the temperature of recrystallization.Following PWHT at 750°C, the austenite phase content has a slight increase.Some grains become larger, causing a slight reduction in the ferrite phase content.However, after heat treatment at 1,050°C, the weld's microstructure appears more uniform, primarily existing in the form of Intragranular austenite, with smaller and evenly distributed grains.Figure 8 illustrates the microstructure morphology on the surface of the weld seam before and after different heat treatment temperatures.As shown in Figure 8(b), after heat treatment at 380°C, there is also no significant change in the microstructure morphology.It remains similar to the aswelded condition, with little difference in the content of austenite and ferrite phases.After heat treatment at 750°C (Figure 8(c)), a noticeable growth in the austenite grain size, an increase in the austenite phase content, a more pronounced dendritic structure, and the formation of a dark precipitate phase between the austenite grains can be observed.As a result, the content of the ferrite phase decreases.Following heat treatment at 1,050°C (Figure 8(d)), the weld's microstructure becomes more uniform.The coarse dendritic structure disappears, and the higher temperature causes a refinement of the austenite grain size.Austenite continues to form, and the content of the ferrite phase significantly decreases, while the austenite content increases.Observations under an optical microscope indicate that higher-temperature heat treatment promotes further transformation of ferrite into austenite.With increasing heat treatment temperature, the extent of this transformation process becomes more pronounced.To further quantify the impact of heat treatment on microstructure evolution, EBSD experiments were conducted to determine specific phase content and structure types.
After observations using an optical microscope, it was found that the changes in microstructure morphology were more pronounced near the weld seam surface (at the fourth weld pass).Therefore, EBSD tests were conducted at the weld seam surface to analyze and discuss the specific effects.The EBSD results of as-welded states are shown in Figure 9(a).Figure 9 presents the distribution and content of the two phases near the weld seam surface after heat treatments at different temperatures.Following heat treatment at 380°C, the microstructure of the weld seam remains largely unchanged, mainly due to the relatively low temperature.However, some austenite grains show slight growth, resulting in an increase in the austenite content to 67.2%, with the ferrite content at 32.8%.During the holding process at 750°C, intragranular austenite continues to grow within ferrite grains.The ferrite grains become more fragmented, and a significant amount of grain boundary austenite surrounds the ferrite grains.As a result, the austenite content increases significantly to 71.8%, while the ferrite content decreases to 28.2%.When the heat treatment temperature reaches 1,050°C, the transformation of ferrite to austenite becomes more thorough.Austenite grains further refine and exhibit a "wheat-like" shape, distributing more uniformly within the ferrite matrix.At this point, the austenite content reaches 75.5%.Simultaneously, small austenite phases melt away, and alloy elements redissolve into both the austenite and ferrite phases [27,28].
Based on the above analysis, PWHT induces the formation of more austenite in the weld seam, primarily classified into primary austenite and secondary austenite [29].Primary austenite solidifies directly from the molten metal, while secondary austenite is generated during the solid- Residual stress and microstructure of 2205 DSS welded joints  9 phase transformation from ferrite to austenite, including grain boundary austenite, Widmanstätten austenite, and intragranular austenite.When the post-weld microstructure is reheated, secondary austenite forms within ferrite and at the interface between ferrite and primary austenite [30].Studies have found that after heat treatment in the range of 1,000-1,100°C, secondary austenite readily precipitates [27].At 1,050°C, the predominant forms of austenite are primary and secondary austenite [31].Therefore, the increase in heat treatment temperature promotes the growth of austenite and the generation of intragranular austenite.The increased content of austenite in the weld seam is ultimately attributed to the non-equilibrium nature of the postweld microstructure [32].

Grain-type distribution
The content of grain types in the weld seam before and after heat treatment is illustrated in Figure 10.After undergoing heat treatment at different temperatures, the predominant grain types at the weld seam surface are recrystallized grains and substructured grains, with a minimal presence of deformed grains.Substructured grains result from lattice distortions, while recrystallized grains require significant stress to induce the recrystallization phenomenon [33].As the heat treatment temperature increases, the content of deformed grains at the weld seam decreases.The content of deformed grain in the austenite decreases from 12.1 to 0%, while that in the ferrite changes from 21.9 to 2%.This is because there is a large deformation after welding.However, after heat treatment, the plastic deformation is eliminated due to the dislocation rearrangement caused by thermal recovery.Besides, new grains tend to form at pre-existing high-angle boundaries that have a significant density of dislocations.These new grains then grow by absorbing the adjacent deformed grains.Due to the effect of PWHT temperature increase, the driving force of recrystallization also increases, and recrystallization is more prone to occur.Without heat treatment, the recrystallized grain content is 20.30%.However, after heat treatment at 380, 750, and 1,050°C, the recrystallized grain content near the weld seam surface in the two-phase region is 33.25, 32.70, and 46.50%, respectively.The higher the heat treatment temperature, the greater the plastic deformation inside the material, which is closely related to the deformation of the material and the dynamic recrystallization process [34].When the material deforms, dislocations are generated inside the grains, and a recrystallization process occurs.When this process is not sufficient to offset the dislocations generated, the irregular dislocations on the slip plane redistribute, and the dislocations align vertically to form walls.This distribution releases a large amount of strain energy, and this reorganization of dislocations also leads to the formation of substructures in the microstructure [35].The formation of these substructures consumes the surrounding sub-grains, ultimately forming substructured grains with relatively low dislocation densities.When recrystallized grains nucleate, these substructured grains are further consumed.In addition, the higher the heat treatment temperature, the more thorough the conversion of elastic deformation to plastic deformation, leading to higher energy storage and the formation of more nucleation sites, effectively reducing the average grain size.In summary, heat treatment increases the content of recrystallized grains and substructured grains.As the heat treatment temperature increases, the weld microstructure still consists mainly of recrystallized grains and  Residual stress and microstructure of 2205 DSS welded joints  11 substructured grains, with an increase in the content of recrystallized grains and a decrease in the content of substructured grains.

Grain boundaries character distribution
Quantitative analysis of the grain boundary characteristics near the upper surface region of the weld seam after different temperature heat treatments is presented, as shown in Figure 11.The distribution of the misorientation angle in austenite exhibits a bimodal distribution, while the misorientation angle in the ferrite phase is predominantly concentrated at low angles, displaying a unimodal distribution.The misorientation angle is further calculated to obtain the grain boundary distribution, as shown in Figure 12.The heat treatment temperature of 380°C results in a high content of low-angle grain boundaries (LAGBs) in the ferrite phase, reaching 89.94%, with only 2.04% being high-angle grain boundaries (HAGBs).The medium-angle grain boundaries (MAGBs) have a lower content of about 8.02%.In the austenite phase, both the LAGBs and HAGBs are relatively dominant, with contents of 49.02 and 32.23%, respectively.In the austenite phase, both the LAGBs and HAGBs are relatively dominant, with contents of 49.02 and 32.23%, respectively.When the heat treatment temperature is increased to 750°C, the dominance of LAGBs continues in the ferrite phase, with a content of 93.85%.HAGBs account for only 3.14%.The content of LAGBs in the austenite phase increases compared to the heat treatment at 380°C, reaching approximately 61.00%, while MAGBs and HAGBs are found to be around 9.41 and 29.59%.When the heat treatment temperature reaches 1,050°C, the distribution of austenite phase grain boundaries resembles that at 750°C, with LAGB, MAGB, and HAGB contents of 64.81, 7.57, and 27.62% respectively.Additionally, the presence of HAGBs (5.14%) is observed in the ferrite phase while the content of LAGBs remains as high as 94.50%.Comparing the grain boundary characteristics after PWHT with those of the as-welded state, it is found that heat treatment is conducive to the transformation of LAGBs in two phases.This is due to the large welding residual deformation and a large number of sub-structures generated in the welding process.The sub-grains rotate and the misorientation angle changes during the PHWT process, resulting in the transformation of HABGs to LAGBs.Second, the LAGBs can absorb the motion dislocation and reduce the residual deformation of the weld [36].If the LAGBs are changed to HAGBs, more energy should be provided by PHWT for the transformation.In addition, due to the reduction of the  residual stress in the PHWT process, the grain changes to be in the LAGBs, which is more stable.Therefore, the content of LAGBS increases while that of HAGBs decreases.
From the above results of microstructure and residual stress, it can be found that PHWT at lower temperatures would not have a significant influence on the microstructure morphology and phase ratio.However, the PWHT at high temperatures would promote the growth of austenite and result in a severely unbalanced phase ratio, which would weaken the strength of the welded joint [37].Besides, the content of LAGBs also increases, which would decrease the yield strength because dislocations rearrange themselves to reduce their energy through the rotation and merging of subgrains, as well as the migration of subgrain boundaries [38].Although the increase of PWHT temperature can lead to an increase in the degree of stress release, the maximum longitudinal stress at PHWT temperature of 380°C has reached only 422.6 MPa, below the yield strength.Therefore, it is recommended to carry out PWHT at a relatively low temperature for the DSS SMAW joint.It then can promote stress relief and avoid the severely unbalanced phase ratio between two phases, which would guarantee excellent mechanical properties [39,40].

Conclusion
This study investigated the impact of PWHT temperature on residual stresses and microstructure in DSS SMAW joint, which was expected to provide an appropriate PHWT process to gain excellent mechanical properties.The conclusions are summarized as follows: (1) In the as-welded joint, large tensile longitudinal residual stress was found in the weld seam and HAZ, which was close to the yield strength of the material.Along the through-thickness direction, the longitudinal residual stress decreases from 465 to 400 MPa.The transverse stress ranges from −48.9 to 126 MPa along the thickness.(2) PWHT significantly reduced the residual stresses.When the PWHT reached 380°C, the longitudinal and the transverse residual stresses were reduced by 20.4 and 14.8% compared to those of the as-welded joint, respectively.When the PWHT reached 1,050°C, the longitudinal and the transverse residual stresses were reduced by 66.8 and 62.2%, respectively.(3) After welding, the fraction of austenite was 66.4%.
PWHT at 380°C has a slight influence on the phase ratio.However, PWHT at 1,050°C resulted in the fraction of austenite reaching 75.5%.Furthermore, the increase in PWHT temperature would promote the growth of recrystallized grains and LAGBs.
(4) It is recommended to perform PWHT at relatively low temperatures around 380°C to achieve good combination performance of DSS manual arc welding joints.

Figure 1 :
Figure 1: Geometric dimensions of welding specimens and testing points using drilling method.

Figure 3 :
Figure 3: FEM model of the weld joint.

Figure 5 :Figure 6 :
Figure 5: Effect of heating temperature on residual stress perpendicular to the weld seam: (a) transverse stress and (b) longitudinal stress.

Figure 10 :
Figure 10: Statistical graph of weld grain-type content before and after heat treatment.

Table 1 :
Chemical composition of the SAF2205 and ER2209 (in wt%)

Table 3 :
Heat treatment scheme

Table 4 :
Thermophysical properties of SAF2205 steel

Table 5 :
Mechanical properties of SAF2205 steel