Exploring Nb-Stabilized Super Duplex Stainless Steel Welding Impact (PDF)

crystals Article Exploring the Impact of Cooling Rate on Microstructural Features, Mechanical Properties, and Corrosion Resistance of a Novel Nb-Stabilized Super Duplex Stainless Steel in Shielded Metal Arc Welding Ángelo Oñate 1,2 , Enrique Torres 1 , Diego Olave 1 , Jesús Ramírez 1 , Carlos Medina 3 , Juan Pablo Sanhueza 1 , Manuel Melendrez 1 , Víctor Tuninetti 4, * and David Rojas 1, * 1 Department of Materials Engineering (DIMAT), Faculty of Engineering, Universidad de Concepción, Edmundo Larenas 315, Concepción 4030000, Chile; [email protected] (Á.O.); [email protected] (J.P.S.); [email protected] (M.M.) 2 Department of Mechanical Engineering (DIMEC), Faculty of Engineering, Universidad del Bío-Bío, Av. Collao 1202, Concepción 4030000, Chile 3 Department of Mechanical Engineering (DIM), Faculty of Engineering, Universidad de Concepción, Edmundo Larenas 219, Concepción 4030000, Chile; [email protected] 4 Department of Mechanical Engineering, Universidad De La Frontera, Francisco Salazar 01145, Temuco 4780000, Chile * Correspondence: [email protected] (V.T.); [email protected] (D.R.) Abstract: The corrosion and mechanical response produced by quenching in the welded joint of a new Nb-doped stainless steel designed by the CALPHAD method and produced by open-atmosphere casting with recycled materials were investigated to contribute to the circular economy and to establish disruptive manufacturing criteria based on metallurgical principles. The steel was initially subjected to solubilization heat treatment and partial solubilization treatment at 1090 ◦ C to obtain Citation: Oñate, Á.; Torres, E.; Olave, an appropriate α/γ balance and carbide solubilization. It was then welded by the SMAW process, D.; Ramírez, J.; Medina, C.; Sanhueza, quenched, and tempered at three different cooling rates. As a result, a good fit between the phases J.P.; Melendrez, M.; Tuninetti, V.; predicted by the CALPHAD method and those observed by X-ray diffraction and scanning electron Rojas, D. Exploring the Impact of microscopy were obtained, with minor differences attributable to the precipitation and diffusion Cooling Rate on Microstructural kinetics required for dissolution or nucleation and growth of the phases in the system. The forced air Features, Mechanical Properties, and quenching mechanism was identified as providing an α/γ phase equilibrium equivalent to 62/38 as Corrosion Resistance of a Novel the most effective quenching method for achieving the optimum mechanical and corrosion response, Nb-Stabilized Super Duplex Stainless even with the post-weld σ phase and showing superior results to those of the base metal. The Steel in Shielded Metal Arc Welding. outstanding mechanical and corrosion responses resulted from a proper balance of the primary Crystals 2023, 13, 1192. https:// phases in the duplex steel with a precipitation-strengthening mechanism. The damage tolerance doi.org/10.3390/cryst13081192 obtained by forced air quenching was superior to that obtained by water and air quenching, with a Academic Editor: Sanbao Lin PSE of 24.71 GPa% post-welding. Received: 5 July 2023 Revised: 27 July 2023 Keywords: corrosion; sigma phase; super duplex stainless steel; mechanical response Accepted: 28 July 2023 Published: 31 July 2023 1. Introduction Duplex stainless steels (DSS) are key structural materials for highly demanding indus- Copyright: © 2023 by the authors. trial sectors such as chemical, petrochemical, desalinization, pulp and paper, and the power Licensee MDPI, Basel, Switzerland. industry [1–5]. These materials combine high mechanical response and excellent corrosion This article is an open access article resistance, mainly due to their biphasic microstructure of austenite (γ) and ferrite (δ) in distributed under the terms and conditions of the Creative Commons a proportion close to 50% of each phase [6,7]. However, fusion welding processes, often Attribution (CC BY) license (https:// involved in producing industrial equipment and its maintenance, promote microstructural creativecommons.org/licenses/by/ changes that may alter these alloys’ mechanical properties and corrosion behavior [8–10]. 4.0/). Crystals 2023, 13, 1192. https://doi.org/10.3390/cryst13081192 https://www.mdpi.com/journal/crystals Crystals 2023, 13, 1192 2 of 28 Moreover, the existence of non-metallic inclusions, Cr carbides (sensitization), and inter- metallic phases (σ, χ, and Laves phase), commonly obtained during the welding procedure, increases the driving force for pitting formation in combination with intergranular corro- sion [11,12]. In particular, the heat input of the welding process affects the balance between the austenite and the ferrite, producing the segregation of elements and precipitation of secondary phases. For instance, a low heat input during the welding process would pro- duce ferritization leading to the precipitation of Cr2 N due to the low miscibility of N in ferrite. On the other hand, a high heat input during the welding process would increase the diffusion kinetics promoting the stability of precipitates such as the σ phase, χ phase, Cr2 N, Cr23 C6, and Cr7 C3, which would decrease the localized corrosion resistance. Due to the sensitivity of the effect of heat input during the welding process, it should be limited to reduce the adverse effects in DSS materials between 0.5–2 kJ/mm and in SDSS between 0.5–1.5 kJ/mm. The corrosion effect arises from corrosive substances on the alloy surface. The contact area initiates an aggressive chemical reaction, leading to localized sensitization, which promotes metal dissolution. These effects result in significant industrial damage, especially when the microstructure is metallurgically sensitized due to processes like welding, which profoundly influence phase transformation. Additionally, welding induces damage, includ- ing the production of microcracks and residual stresses. A notable industrial example is the work of Tavares et al. , who analyzed ductile fissure failure in the welding metal zone. The damage was caused by an excess of ferrite with intense Cr2N precipitation, creating an imbalance and potential difference that enhanced corrosion and mechanical damage. Similarly, Yang et al. investigated a reactor failure due to intercolumnar cracks caused by solidification during shielded metal arc welding (SMAW) on duplex stainless steel (DSS). This failure was attributed to cooling, which facilitated an 85% ferrite formation in the metal. Several authors indicate that the best DSS performance occurs when the volume frac- tion of significant phases (δ/γ) is in a 50/50 ratio [2,5,6]. Nevertheless, such a narrow phase balance may not achieve the best mechanical and corrosion behavior for heavy alloyed super duplex stainless steels (SDSS) and hyper duplex stainless steels (HDSS), because secondary phases play a fundamental role in the mechanical and corrosion behavior. This statement is consistent with the findings obtained by Ha et al. , who found that in a duplex stainless steel, UNS S32205, the highest corrosion resistance occurred for a 57% ferrite fraction approaching a main phase fraction δ/γ 60/40. Therefore, duplex stainless steel’s optimum mechanical and corrosion characteristics are not always obtained for the δ/γ 50/50 phase fraction. The use of scrap from the steel industry for steel production has increased during the last few years due to the high demand for steel worldwide [16–18]. An example is the sustained increase in stainless steel, with a growth rate of 5.33% per year. However, its use has not been deeply explored in the fabrication of stainless steels, and its microstructural and functional study in the structural application is reduced. Exam- ples of the production process of stainless steels based on scrap are detailed by Holappa et al. , where it is detailed that ordinary recycled steel and additions of ferroalloys such as FeCr, FeMo, and FeNi are used. However, the production process is carried out under controlled atmosphere processes or undergoes decarburization processes. These processes make the steel more expensive and make production logistics more challenging. Welding parameters such as welding speed, temperature, and cooling rate are crucial in the joining processes of stainless materials [20–23]. If these parameters are not ade- quately controlled, the welding process can induce ferritization and intermetallic phase transformations, such as the sigma phase, leading to material sensitization due to poten- tial differences, thereby increasing susceptibility to corrosive damage. Among the parameters mentioned, the cooling rate is one of the most significant factors in achieving the desired microstructure for optimal functional performance during service. Rapid cool- ing rates are known to increase the likelihood of generating Cr2N, while slower cooling rates can trigger phase transformations, such as the sigma or chi phases, through diffu- Crystals 2023, 13, 1192 3 of 28 sional mechanisms [20,21]. The study of the effects of welding on duplex stainless steels is extremely important for the industry due to the sensitization it causes in the material, promoting corrosion damage [24–27]. One of the parameters for assessing sigma phase precipitation is microhardness. This is demonstrated by Putz et al. obtained that the hardness in UNS S32205 duplex steel by four-layer deposition welding was 300 HV. They analyzed the change in hardness upon electric arc heat treatment and observed that harnesses close to 230 HV were obtained in nitrogen-depleted zones. However, the highest hardness values were found next to the depleted zone, reaching 471 HV for a heat treatment duration of 60 min. The temperature reached by the heat treatment in the contact zone was higher than 1400 ◦ C, decreasing as the distance increased from the contact center towards the periphery, reaching 300 ◦ C for a 7 mm radius. The observed increase in hardness was attributed to the sigma phase and ferrite fraction, which was consistent with N impoverishment of the surrounding zones, providing sigma precipitation by CrN and Cr2 N. Souza et al. evaluated the mechanical behavior of a UNS S31803 stainless steel welded by MIG/MAG finding hardness values between 235 and 295 HV. These results indicated a higher homogeneity of the phases with controlled metallurgical distortion. Hardness in duplex steels is restricted to limit values after welding to avoid embrittlement. ANSI/NACE MR0103-2012 states that the maximum allowable hardness should be 35 HRC (345.45 HV). On the other hand, ANSI/NACE MR0175 indicates that the maximum allowable hardness should be 28 HRC (286.43 HV) in base metal. The novelty of the present work is to deepen the understanding of the kinetic mecha- nisms after the hardening process of welded samples and how these impact the microstruc- ture, altering its corrosion and mechanical response. In the same way, the research work is developed on a new SDSS alloy fabricated in an open atmosphere. Consequently, it is necessary to deepen the feasibility of production processes of stainless steels based on scrap with functional characteristics and rigorous characterization of their corrosion and mechanical response, enhancing the circular economy at the industrial level. Further studies are required to determine the main factors governing the corrosion response and mechanical properties in DSS and SDSS. In addition, suitable cooling rates for welded joints by the SMAW method in structural components are unknown, which significantly influences the mechanical response and corrosion behavior due to changes in phase volume fraction, precipitation of secondary phases, and segregation, among others. Based on the previous background, this research aims to estimate the influence of cooling rates on the corrosion resistance and mechanical performance after the SMAW process on a high-carbon super duplex stainless steel stabilized by Nb. To achieve the proposed objective, a microstructural analysis was performed using CALPHAD computational simulation and correlated by a complete microstructural characterization of the welding joint. Furthermore, the corrosion response was studied by cyclic potentiodynamic polarization in a 3.5% NaCl solution at 40 and 70 ◦ C, respectively. Finally, the mechanical behavior was investigated by microhardness and quasi-static uniaxial tensile test. 2. Methodology The alloy was designed based on physical metallurgical principles supported by Thermo-Calc modeling using the TCFe-8 database to obtain a super duplex stainless steel. The alloy was produced by induction melting in the open air, based on a scrap shaft of 329 DSS as base material aiming for metallic waste valorization, plus the addition of ferroalloy elements, such as FeCr, FeNb, Mn, Ni, FeMo, and FeSi. The mass balance resulting from the Thermo-Calc design is shown in Table 1. The raw materials for the melting were dried at 70 ◦ C for 24 h to remove moisture before casting. The chemical composition of the resulting alloy is detailed in Table 2, which was obtained by optical emission spectroscopy using SPECTROMAXX equipment. Crystals 2023, 13, 1192 4 of 28 Table 1. Mass balance for free atmosphere melting. Base Material Material Cu Ni FeCr FeNb Mn FeMo FeSi 329 Load (g) 5300 24 112 130 25 117 193 1 Table 2. Chemical composition wt.% of the alloy obtained. C Cr Ni Mo N Mn Cu Si W Nb P S Fe 0.067 26.5 7.31 3.46 0.14 2.70 0.84 0.545 0.034 0.24 0.011 0.018 Bal. The ingot obtained from casting with a mass of about 6 kg was cut into 2 cm pieces to be further processed by thermomechanical treatment. The thermomechanical treatment was carried out in a Nabertherm electric muffle furnace at a solubilization temperature of 1200 ◦ C for 60 min. Subsequently, the samples were hot rolled in a JOLIOT symmetrical rolling mill until a 60% section reduction was obtained, followed by water quenching. Finally, a solution annealing heat treatment was performed at 1090 ◦ C for 60 min on the rolled samples and then quenched in water at 15 ◦ C. The temperature of 1090 ◦ C in the heat treatment was obtained by CALPHAD simulation in Thermo-Calc. The computational simulations were performed in Thermo-Calc using the chemical composition presented in Table 2, considering atmospheric pressure. Thermo-Calc sim- ulations considered a phase diagram calculation, volume fractions of different phases at interest temperatures, and equilibrium point temperatures. Subsequently, to comple- ment the computational simulation considering the thermal cycling caused by the welding process, simulations were performed in JMatPro V7.0 using the duplex stainless-steel database. The simulations performed in JMatPro consisted of Time-Temperature Transfor- mation (TTT) and Continuous Cooling Transformation (CCT) diagrams, which were used to analyze the phase transformation during heat input in welding and subsequent phase transformation during quenching. For the welding process, 3.5 cm × 3.5 cm × 1 cm plates were fabricated, which were joined by the SMAW method with E-2209-16 electrode due to the similar characteristics concerning the alloy, ensuring dilution during the process, and obtaining a homogeneous microstructure. The chemical composition of the electrode used can be seen in Table 3. A 65-degree bevel with a root of 2 mm and a separation of 3 mm was used to guarantee a good weld joint (see Figure 1). The current intensity in the welding process was 70 (A) with a voltage of 24 (V) for an electrode diameter of 2.5 mm. The welding feed rate used was 9 cm/min with an efficiency of 80%. The welding temperature reached 1100 ◦ C measured by thermographic camera. The number of steps considered was 2 due to the thickness of the sample. The behavior of the microstructure sensitized by SMAW was analyzed according to AWS D1.1-6 for stainless steels with electrode E2209. The welded samples were processed by three types of quenching: water quenching, air quenching, and forced air quenching. Table 3. Chemical composition of consumable E2209-16 wt.%. %C %Cr %Ni %Mo %N %Mn %Si 0.02 23.32 8.15 3.15 0.15 1.33 0.46 Sample preparation for microstructural characterization consisted of grinding using SiC sandpaper with mesh sizes from 240 to 1200. The samples were then polished in a Mecatech 264 metallographic polisher using diamond paste from 6 µm to 1 µm at a speed of 350 RPM for 4000 s. The samples were further polished in an alumina suspension with a ness of the sample. The behavior of the microstructure sensitized by SMAW was analyzed according to AWS D1.1-6 for stainless steels with electrode E2209. The welded samples were processed by three types of quenching: water quenching, air quenching, and forced air quenching. Crystals 2023, 13, 1192 5 of 28 Table 3. Chemical composition of consumable E2209-16 wt.%. %C %Cr %Ni %Mo %N %Mn %Si particle 0.02size of 0.05 23.32 µm for 2000 s 8.15at 250 RPM. Finally, 3.15 the samples 0.15 were 1.33 microstructurally 0.46 developed using electrolytic etching using KOH—30% at a potential of 9V for 10 s. Figure 1. Scheme of welded joint used in the investigated samples. The post-welding microstructure was analyzed by optical microscopy (OM) and scanning electron microscopy with an energy-dispersive X-ray spectrometer (SEM-EDS). Characterization by optical microscopy was performed on a Leica DMi8-M microscope us- ing magnification between 100× and 1000×. Scanning electron microscopy was performed on a TESCAN VEGA 3 EASYPROBE SBU microscope with an accelerating voltage of 20 kV and a 10 mm working distance. The energy dispersive X-ray spectrometer was used to control the number of counts over 25,000 cps to maintain the representativeness of the results. The characterization of the phases present in the welded samples was performed by X-ray diffraction using a Cu Kα filter with a wavelength, λ = 1.5406 Å. The operating parameters were an accelerating voltage of 40 kV and a current of 20 mA. The diffraction angles were between 20◦ and 90◦ with analysis steps of 0.02◦ with a permanence of 1 s. The corrosion tests were performed using a cell with a saturated Calomel reference electrode with 244 mV to a hydrogen electrode, a Pt-alloyed stainless steel auxiliary elec- trode, and the obtained alloy samples as working electrodes. The measurements were performed on a VersaStat three potentiostat with a PolyScience thermostat. The time for each point is 1 s with an open circuit duration of 1000 s. Pitting corrosion was evaluated by a cyclic potentiodynamic polarization test in an aggressive 3.5% NaCl medium using ASTM G-61. The mechanical characterization considered microhardness measurements and uniax- ial tensile tests. The Vickers microhardness tests were performed with a load of 300 g in the base metal area, heat-affected zone, and melting zone. ASTM E8 performed the quasi-static tensile test on specimens with a cross-section in the calibrated section of 7 mm × 4 mm using a 1 mm/min speed in an Instron 8801 universal testing machine with a 100 kN load cell. 3. Results and Discussion 3.1. Thermochemical Simulation Using the CALPHAD Method For better and simplified naming, the produced alloy was identified as SDSS-Nb (see chemical composition in Table 2). The thermodynamic equilibrium calculations of the phases present in the SDSS-Nb alloy were modeled using Thermo-Calc with the TCFE-8 module, obtaining the phases diagram (T◦ vs. %C). Figure 2 shows the alloy present below 1000 ◦ C, the M23 C6 , σ, and Z phases. The M23 C6 plays a crucial role in the material’s mechanical strength and in controlling the subgrain’s coarsening, even more so than the MX precipitates. Due to their reduced Gibbs, the free energy is highly stable, generating a high contribution to dislocation locking and creep resistance. The σ phase in welded super duplex steels has a higher precipitation potential in the thermally affected zone for temperatures between 900 ◦ C and 1000 ◦ C due to the decrease in interfacial energy favoring nucleation in this zone, producing a reduction in energy absorption and embrittlement. Figure 3a shows the volume fraction of the phases present in the SDSS-Nb alloy. The approximate fraction α/γ (50/50) is obtained for a temperature of 1040 ◦ C, at which secondary phases are also present. Figure 3b shows the magnification of the volume fraction in the SDSS-Nb alloy. It can be observed that for a temperature of 1040 ◦ C, the stable secondary phases are Cr23 C6 carbides and the Z phase. It Crystals 2023, 13, 1192 6 of 28 is also possible to see that for a temperature of 1090 ◦ C, complete dissolution of the Cr23 C6 carbides occurs, having its highest energetic stability at temperatures below 1070 ◦ C. Table 4 shows the composition of each phase obtained from Figure 3 for a temperature of 1090 ◦ C, Crystals 2023, 13, x FOR PEER REVIEW resulting in increased stability between ferrite and austenite with Nb-MX precipitates 6 of 30 at 0.223%. Figure 2. Phase diagram of SDSS-Nb steel. The σ phase in welded super duplex steels has a higher precipitation potential in the thermally aﬀected zone for temperatures between 900 °C and 1000 °C due to the decrease in interfacial energy favoring nucleation in this zone, producing a reduction in energy absorption and embrittlement. Figure 3a shows the volume fraction of the phases pre- sent in the SDSS-Nb alloy. The approximate fraction α/γ (50/50) is obtained for a temper- ature of 1040 °C, at which secondary phases are also present. Figure 3b shows the magni- ﬁcation of the volume fraction in the SDSS-Nb alloy. It can be observed that for a temper- ature of 1040 °C, the stable secondary phases are Cr23C6 carbides and the Z phase. It is also possible to see that for a temperature of 1090 °C, complete dissolution of the Cr23C6 car- bides occurs, having its highest energetic stability at temperatures below 1070 °C. Table 4 shows the composition of each phase obtained from Figure 3 for a temperature of 1090 °C, resulting in increased stability between ferrite and austenite with Nb-MX precipitates at Figure2.2.Phase 0.223%. Figure Phasediagram diagramofofSDSS-Nb SDSS-Nbsteel. steel. The σ phase in welded super duplex steels has a higher precipitation potential in the thermally aﬀected zone for temperatures between 900 °C and 1000 °C due to the decrease in interfacial energy favoring nucleation in this zone, producing a reduction in energy absorption and embrittlement. Figure 3a shows the volume fraction of the phases pre- sent in the SDSS-Nb alloy. The approximate fraction α/γ (50/50) is obtained for a temper- ature of 1040 °C, at which secondary phases are also present. Figure 3b shows the magni- ﬁcation of the volume fraction in the SDSS-Nb alloy. It can be observed that for a temper- ature of 1040 °C, the stable secondary phases are Cr23C6 carbides and the Z phase. It is also possible to see that for a temperature of 1090 °C, complete dissolution of the Cr23C6 car- bides occurs, having its highest energetic stability at temperatures below 1070 °C. Table 4 shows the composition of each phase obtained from Figure 3 for a temperature of 1090 °C, resulting in increased stability between ferrite and austenite with Nb-MX precipitates at Figure 3. Volume fraction plots of SDSS-Nb, (a) total volume fraction, and (b) volume fraction mag- 0.223%. Figure 3. Volume fraction plots of SDSS-Nb, (a) total volume fraction, and (b) volume fraction niﬁcation in the M23C6 carbide dissolution zone. magnification in the M23 C6 carbide dissolution zone. Table 4. ThermoCalc phase fraction simulation at temperature 1090 °C with chemical composition in 4. ThermoCalc phase fraction simulation at temperature 1090 ◦ C with chemical composition wt.%. Table in wt.%. Phases Vol. %Fe %Cr %Ni %Mo %Mn %Si %Cu %N %Nb %C δ Phases 53.29 Vol. 57.14 %Fe 30.16 %Cr 5.13 %Ni 4.1 %Mo 2.25 %Mn 0.585 %Si 0.513 %Cu 0.036 %N 0.032 %Nb 0.021 %C δ γ 46.41 53.29 59.83 57.14 22.39 30.16 9.83 5.13 2.73 4.1 3.17 2.25 0.5 0.585 1.21 0.513 0.188 0.036 0.097 0.032 0.017 0.021 γ 46.41 59.83 22.39 9.83 2.73 3.17 0.5 1.21 0.188 0.097 0.017 z 0.078 5.99 27.03 - 6.36 - - - 8.52 52.09 - Nb-MX 0.223 0.009 11.08 - 0.389 0.005 - - 8.786 75.14 4.57 The achieved results by thermochemical simulation in Thermo-Calc are complemented Figure with the3.TTT Volume andfraction plots transformation CCT phase of SDSS-Nb, (a) total volume results fraction, obtained byand (b) volume JMatPro fraction on the SDSS-Nbmag- niﬁcation alloy. in the Figure M23C6 carbide 4 shows dissolution no σ phase zone. at 1090 ◦ C temperature. However, a stable formation Table 4. ThermoCalc phase fraction simulation at temperature 1090 °C with chemical composition in wt.%. Phases Vol. %Fe %Cr %Ni %Mo %Mn %Si %Cu %N %Nb %C z 0.078 5.99 27.03 - 6.36 - - - 8.52 52.09 - Nb-MX 0.223 0.009 11.08 - 0.389 0.005 - - 8.786 75.14 4.57 The achieved results by thermochemical simulation in Thermo-Calc are comple Crystals 2023, 13, 1192 mented with the TTT and CCT phase transformation results obtained by JMatPro 7 of 28 on th SDSS-Nb alloy. Figure 4 shows no σ phase formation at 1090 °C temperature. However, stable volume fraction of M23C6 is equivalent to 0.13 wt.% is predicted. It can be deter mined fraction volume from theofTTT diagram M23 C (see Figure 6 is equivalent 4a) to 0.13 thatisthe wt.% weldingItprocess predicted. would not produc can be determined σ phase precipitation because the welding time is less than 1 min in the samples. from the TTT diagram (see Figure 4a) that the welding process would not produce σHoweve phase precipitation because the welding time is less than 1 min in the samples. However, there is σ phase precipitation at approximately 1 wt.%. From Figure 4b, it is possible t there is σ phase precipitation at approximately 1 wt.%. From Figure 4b, it is possible to determine the secondary phase precipitation during the cooling cycle for the diﬀeren determine the secondary phase precipitation during the cooling cycle for the different quenchingmechanisms quenching mechanisms through through thethe CCT CCT diagram. diagram. It isIt possible is possible to observe to observe fromfrom the the CC CCT diagram that quenching in water would not produce secondary phase precipitation. How diagram that quenching in water would not produce secondary phase precipitation. ever, quenching However, in forced-air quenching in forced-air andandnatural-air natural-airwill willmainly mainly produce produce σσ andandchi chiphase phase precip itation. The The precipitation. precipitation precipitationofofthe thesigma phasewill sigma phase willundoubtedly undoubtedly generate generate an alteration i an alteration inthe themechanical characteristics mechanical characteristics of of thethe material. material. However, However, this alteration this alteration can be or can be positive positive o negative depending on the fraction of the sigma phase present and its size. negative depending on the fraction of the sigma phase present and its size. Figure 4. CALPHAD simulation of the soldering process in SDSS-Nb; (a) TTT diagram for a tempera- ture of 1090 ◦ C; (b) CCT diagram for a temperature of 1090 ◦ C. Crystals 2023, 13, 1192 8 of 28 To evaluate the formation of oxides during the welding process, an atmosphere of 0.03 wt.% oxygen was considered during the welding process to analyze the activation energy and temperature of oxide formation. The oxygen content used for the simulation is based on the typical ranges of oxygen in welding processes proposed by Dellam et al. not to affect the welded joint’s toughness. These values are also considered in previous studies by Coetsee et al.. The CALPHAD simulation developed in JMatPro resulted in the predominant formation of SiO2. It was found by the simulation that the onset of forma- tion is at a temperature of 1200 ◦ C with an activation energy of −399.677 (kJ/Mol). The SiO2 fraction increases dramatically until it reaches 940 ◦ C and remains stable. Secondary oxides are formed at 370 ◦ C and are mainly composed of Cr and Fe with an activation energy of −318.913 (kJ/mol). These results obtained will be corroborated by scanning electron microscopy with energy-dispersive X-ray spectroscopy to analyze the formation of oxides on the welded samples. 3.2. Optical Microscopy Figure 5 shows that the optical microscopy (OM) images revealed a microstructure formed by a ferritic matrix with columnar austenite branching in the center of the sample (Figure 5a). These columnar branchings are produced due to the cooling rates obtained in the center of the sample. The microstructure obtained is characteristic of solidification in duplex alloys, starting from the primary ferrite, with nucleation of austenite from the solid in the ferritic matrix preferentially at the grain boundaries. This process occurs via a diffusion-controlled state, where Ni is expelled from the ferrite into the austenite, and Cr is rejected by the advancing interface, producing a Ni enrichment and Cr deficit in the austenite compared to the ferrite. Figure 5b shows the microstructure of the eutectoid decomposition resulting from slow cooling. Figure 5c,d show that the microstructure can be observed with a ferritic-austenitic microstructure after the thermomechanical treatment. It can be observed that the austenite is the elongated product of the rolling direction in an equal proportion with the ferrite. The phase ratio was analyzed using ImageJ software by the fraction of area used by each phase in the micrograph. The assay results indicated that the phases are in a proportion of 45% austenite and 55% ferrite, respectively. These results were further corroborated by thermodynamic simulation by CALPHAD in Thermo-Calc software, obtaining a volume fraction of 46.1% for austenite and 53.29% for ferrite at the temperature of 1090 ◦ C, respectively (see Table 4). Figure 5b, a transition zone δ→γ2 + σ can be seen. The transition zone noted above is due to the enrichment of the Z phase with Cr, Si, and Mo from the ferrite, thus producing a decomposition of the matrix and the σ and γ2 phase is formed. Additionally, the formation of M23 C6 or MX carbides is possible due to the high segregation that can occur during the welding process and the activation of carbon diffusion to carbide stabilizing elements such as Cr and Nb. These are consistent with those obtained previously by CALPHAD simulation. Figure 6 shows the microstructure in the HAZ zone–root weld bead of SDSS-Nb steel with different cooling conditions. Figure 6a shows the microstructure of SDSS-Nb steel cooled in air, Figure 6b shows the microstructure of forced air cooling, and Figure 6c shows the microstructure of water cooling. The change in the microstructure is evident, and its relation is direct to the type of cooling. Due to localized heat input, this change in the microstructure is because the welding process generates a phase transformation in the duplex steels, leading them to a ferritic state when reaching temperatures above 1250 ◦ C. Consequently, the austenite will reform again upon cooling, but its size and quantity depend mainly on the cooling rate. Therefore, a higher fraction of austenite can be observed in Figure 6a as it has been subjected to cooling in air. From Figure 6b,c, it can be deduced that the variation of the austenite fraction is due to the cooling process, highlighting, at first sight, the higher fraction of austenite in the sample cooled in forced air concerning the sample cooled in water. Crystals 2023, Crystals 13,13, 2023, x FOR 1192 PEER REVIEW 9 28 9 of of 30 Figure 5. Results obtained from optical microscopy on SDSS-Nb As-Cast and thermomechanical treatment; (a) SDSS-Nb As-Cast optical microscopy; (b) SDSS-Nb As-Cast optical microscopy; (c) SDSS-Nb optical microscopy with thermomechanical treatment; (d) SDSS-Nb optical microscopy with thermomechanical treatment. Figure 5b, a transition zone δ→𝛾𝛾2 + σ can be seen. The transition zone noted above is due to the enrichment of the Z phase with Cr, Si, and Mo from the ferrite, thus produc- ing a decomposition of the matrix and the σ and 𝛾𝛾2 phase is formed. Additionally, the formation of M23C6 or MX carbides is possible due to the high segregation that can occur during the welding process and the activation of carbon diﬀusion to carbide stabi- lizing elements such as Cr and Nb. These are consistent with those obtained previously by CALPHAD simulation. Figure 6 shows the microstructure in the HAZ zone–root weld bead of SDSS-Nb steel with diﬀerent cooling conditions. Figure 6a shows the microstructure of SDSS-Nb steel cooled in air, Figure 6b shows the microstructure of forced air cooling, and Figure 6c shows the microstructure of water cooling. The change in the microstructure is evident, Figure 5. Results obtained from optical microscopy on SDSS-Nb As-Cast and thermomechanical and its5.relation Figure Resultsisobtained direct tofrom the type of microscopy optical cooling. DueontoSDSS-Nb localized heat input, As-Cast this change in and thermomechani- treatment; (a) SDSS-Nb As-Cast optical microscopy; (b) SDSS-Nb As-Cast optical microscopy; (c) the cal microstructure treatment; is because (a) SDSS-Nb the As-Cast welding optical process generates microscopy; a (b) SDSS-Nb phase transformation As-Cast in the optical microscopy; SDSS-Nb optical microscopy with thermomechanical treatment; (d) SDSS-Nb optical microscopy duplex (c) steels, SDSS-Nb leading optical them microscopy with thermomechanical treatment. to a withferritic state when thermomechanical reaching treatment; temperatures (d) SDSS-Nb above optical 1250 °C microscopy.thermomechanical treatment. with Figure 5b, a transition zone δ→𝛾𝛾2 + σ can be seen. The transition zone noted above is due to the enrichment of the Z phase with Cr, Si, and Mo from the ferrite, thus produc- ing a decomposition of the matrix and the σ and 𝛾𝛾2 phase is formed. Additionally, the formation of M23C6 or MX carbides is possible due to the high segregation that can occur during the welding process and the activation of carbon diﬀusion to carbide stabi- lizing elements such as Cr and Nb. These are consistent with those obtained previously by CALPHAD simulation. Figure 6 shows the microstructure in the HAZ zone–root weld bead of SDSS-Nb steel with diﬀerent cooling conditions. Figure 6a shows the microstructure of SDSS-Nb steel cooled in air, Figure 6b shows the microstructure of forced air cooling, and Figure 6c Figure 6. Results obtained from optical microscopy on SDSS-Nb with SMAW process in the HAZ shows the microstructure of water cooling. The change in the microstructure is evident, zone–root weld; (a) SDSS-Nb optical microscopy with air cooling; (b) SDSS-Nb optical microscopy and its relation is direct to the type of cooling. Due to localized heat input, this change in with forced air cooling; (c) SDSS-Nb optical microscopy with water cooling. the microstructure is because the welding process generates a phase transformation in the duplexThe steels, leading cooling them rate also to a ferritic affected state whenofreaching the stabilization secondarytemperatures above phases. The high 1250 °C cooling. rate of water had a remarkable ability to suppress the formation of secondary phases at lower temperatures than that produced by the thermal effect of the solder joint, such as the σ and γ2 phases. In contrast, air cooling generated a kinetic state suitable for long-range diffusion and mobility of chemical elements to zones of lower chemical potential , as is the mobility of Cr and Mo towards the χ or Z phase to subsequently form σ [10,36,39]. The above is visualized in Figure 7, corresponding to the air-cooled sample, where a structure of primary austenite, Widmanstatten austenite, and secondary austenite is found. The latter is due to the decomposition reaction of ferrite (δ→γ2 + σ) or austenite (γ→γ2 + σ). Therefore, it range diﬀusion and mobility of chemical elements to zones of lower chemical potentia , as is the mobility of Cr and Mo towards the χ or Z phase to subsequently form [10,36,39]. The above is visualized in Figure 7, corresponding to the air-cooled sampl where a structure of primary austenite, Widmanstatten austenite, and secondary austenit Crystals 2023, 13, 1192 10 of 28 is found. The latter is due to the decomposition reaction of ferrite (δ→γ2 + σ) or austenit (γ→γ2 + σ). Therefore, it indicates sigma phase formation due to precipitation kinetic granted by natural cooling. These have two eﬀects. Firstly, the controlled form precipita indicates sigma phase formation due to precipitation kinetics granted by natural cooling. tion of secondary phases during cooling will generate attractive mechanical reinforcemen These have two effects. Firstly, the controlled form precipitation of secondary phases during by increasing cooling the yield will generate stress attractive and ultimate mechanical tensile strength. reinforcement Secondly, by increasing thestress the yield generation o Cr-ultimate and and Mo-depleted zones tensile strength. will reduce Secondly, corrosion the generation of resistance. The proportion Cr- and Mo-depleted zones will of primar phasescorrosion reduce is also relevant since resistance. Thethe mechanical proportion and corrosion of primary phases isresistance depends also relevant since thedirectly o them, determined by the type of cooling used after welding. The corrosion resistance an mechanical and corrosion resistance depends directly on them, determined by the type ofmechanical cooling used response to theThe after welding. proposed cooling corrosion conditions resistance will be response and mechanical evaluated later in th to the research. proposed cooling conditions will be evaluated later in this research. Figure Figure 7. 7. Optical Optical microscopy microscopy ofstructure of the the structure formed formed due to due to the thermal the thermal eﬀect of soldering effect of soldering on the on th air-cooled air-cooled SDSS-Nb SDSS-Nb sample. sample. The The size size of the of the HAZHAZ zone zone was also was also quantiﬁed quantified using theusing the three three types typesThe of cooling. of size cooling. Th Crystals 2023, 13, x FOR PEER REVIEW 11 of 30 size was quantiﬁed using ImageJ software in Figure 8a–c, respectively. The results ind wasquantified using ImageJ software in Figure 8a–c, respectively. The results indicated that the samples cooled in the air show a larger size of the thermally affected zone, reaching cated that the samples cooled in the air show a larger size of the thermally aﬀected zon 820 microns. On the other hand, as the cooling rate increases, the thermally affected zone reaching 820 microns. On the other hand, as the cooling rate increases, the thermally a fected zone becomes becomes smaller and smaller smaller.andThis smaller. result isThis result is demonstrated demonstrated by obtained by the values the valuesfrom ob- tained the from cooled samples the samples cooled in forced air, in forced only reaching air, reaching onlyin589 589 microns themicrons in affected thermally the thermally zone, aﬀected while zone,inwhile cooling watercooling reachedina water reached thermally a thermally affected aﬀected zone of 271 zone microns. of phenomenon This 271 microns. This phenomenon is due to the remaining diﬀusive eﬀect during the cooling cycle,for is due to the remaining diffusive effect during the cooling cycle, which is higher which the is higher for the naturally cooled samples. The above results are presented in Table 5. naturally cooled samples. The above results are presented in Table 5. Figure 8. Eﬀect of cooling kinetics on the size of the thermally aﬀected zone: (a) sample cooled in Figure 8. Effect of cooling kinetics on the size of the thermally affected zone: (a) sample cooled in the the air; (b) sample cooled in forced air; (c) sample cooled in water. air; (b) sample cooled in forced air; (c) sample cooled in water. Table 5. HAZ zone sizes were obtained in SDSS-Nb by optical microscopy in diﬀerent cooling me- dia. Samples Heat-Affected Zone (HAZ) Water 271 microns Forced Air 589 microns Crystals 2023, 13, 1192 11 of 28 Table 5. HAZ zone sizes were obtained in SDSS-Nb by optical microscopy in different cooling media. Samples Heat-Affected Zone (HAZ) Water 271 microns Forced Air 589 microns Air 820 microns The outcome obtained for each thermally affected zone further indicates a change in the mechanical and corrosion response of the alloy. This change in material response is due to the change in grain sizes, fraction of major phases, and content of secondary phases precipitating during the heating and cooling cycle. The detrimental phase transfor- mations during these processes mainly produce mobility of Cr and Mo from the ferritic phase. Therefore, the ferritic phase content must be controlled to optimize the material’s mechanical and corrosion resistance. Our alloy’s main phase balance was quantified to indirectly analyze each cooling type’s effect on the material’s performance using ImageJ software. These results are presented in Table 6. Table 6. Fraction of ferrite obtained in the SDSS-Nb alloy according to its post-weld cooling method. Samples Ferrite Austenite Base Metal 60% 40% Water 71% 29% Forced Air 62% 38% Air 53% 47% 3.3. Scanning Electron Microscopy and Energy Dispersive X-ray Spectroscopy Figure 9 corresponds to the SEM-EDS analysis of the air-cooled samples. Figure 9a shows the existence of precipitates mainly in the ferritic matrix produced by the thermal effect of welding. Zone (1) and zone (2) represent the EDS profiles corresponding to the austenite and ferrite (see Figure 9a,b), observing the presence of Cu and higher intensity of Ni in the austenite compared to the ferrite, while in the ferrite the intensity of Cr and Mo is increased. Zones (3) and (4) correspond to Mn oxide in the austenitic phase (see Crystals 2023, 13, x FOR PEER REVIEW 12 of 30 Figure 9d) and Mn sulfide (see Figure 9e). The summary of the chemical composition obtained for each analysis zone can be seen in Table 7. It can be seen that the oxides obtained are predominantly particle oxides. However, the oxides in zones 3 and 4 are film Film oxides oxides. Filmare not recommended oxides in the welded are not recommended joint because in the welded these these joint because discontinuities would discontinuities promote the formation of cracks. would promote the formation of cracks. Figure Figure9.9.SEM-EDS analysisofofSDSS-Nb, SEM-EDS analysis SDSS-Nb, (a)(a) SEMSEM image image of SDSS-Nb of SDSS-Nb air-cooled air-cooled sample;sample; (b) EDS (b) zoneEDS zone 1 austenite; (c) EDS zone 2 ferrite; (d) EDS zone 3 Mn oxide; (e) EDS zone 4 1 austenite; (c) EDS zone 2 ferrite; (d) EDS zone 3 Mn oxide; (e) EDS zone 4 Mn sulfide. Mn sulﬁde. Table 7. EDS analysis of the chemical composition of each phase in SDSS-Nb wt.% in the air-cooled sample. Analysis %Cr %Ni %Mn %Cu %Mo %Si %Nb %S %Fe Zone 1 21.33 8.80 3.23 1.19 2.49 0.55 --- --- 62.41 Zone 2 25.59 5.52 2.72 --- 4.18 0.69 --- --- 61.30 Crystals 2023, 13, 1192 12 of 28 Figure 9. SEM-EDS analysis of SDSS-Nb, (a) SEM image of SDSS-Nb air-cooled sample; (b) EDS zone 1 austenite; (c) EDS zone 2 ferrite; (d) EDS zone 3 Mn oxide; (e) EDS zone 4 Mn sulﬁde. Table 7. EDS analysis of the chemical composition of each phase in SDSS-Nb wt.% in the Table 7. EDS analysis of the chemical composition of each phase in SDSS-Nb wt.% in the air-cooled air-cooled sample. sample. Analysis Analysis %Cr %Cr %Ni %Ni %Mn %Mn %Cu %Cu %Mo %Mo %Si %Si %Nb %Nb %S %S %Fe %Fe Zone1 1 Zone 21.33 21.33 8.80 8.80 3.23 3.23 1.19 1.19 2.49 2.49 0.55 0.55 ------ ------ 62.41 62.41 Zone2 2 Zone 25.59 25.59 5.52 5.52 2.72 2.72 ------ 4.18 4.18 0.69 0.69 ------ ------ 61.30 61.30 Zone 3 26.35 4.98 7.08 --- 1.73 0.54 --- --- 59.38 Zone 3 26.35 4.98 7.08 --- 1.73 0.54 --- --- 59.38 Zone 4 20.16 5.97 15.16 --- 1.56 0.44 --- 7.93 48.78 Zone 4 20.16 5.97 15.16 --- 1.56 0.44 --- 7.93 48.78 Figure 10 shows the outcomes of the SEM-EDS characterization of the sample cooled in forced in forced air. air. An An austenitic–ferritic austenitic–ferritic structure structure with with the the presence presence of of oxides oxides and and precipitates precipitates can be can be seen. seen. The The EDS EDS results results indicate indicate that the oxides that the oxides are are mainly mainly silicon silicon and and chromium chromium oxides (see Figure 10d,e). The detail of the chemical composition obtained by EDS of oxides (see Figure 10d,e). The detail of the chemical composition obtained by EDS of the the zones analyzed in this sample can be seen in Table 8. zones analyzed in this sample can be seen in Table 8. Figure 10. SEM-EDS analysis of SDSS-Nb, (a) SEM image of SDSS-Nb forced-air-cooled sample; (b) Figure 10. SEM-EDS analysis of SDSS-Nb, (a) SEM image of SDSS-Nb forced-air-cooled sample; (b) EDS zone 1 austenite; (c) EDS zone 2 ferrite; (d) EDS zone 3 Si oxide; (e) EDS zone 4 Cr oxide. EDS zone 1 austenite; (c) EDS zone 2 ferrite; (d) EDS zone 3 Si oxide; (e) EDS zone 4 Cr oxide. Table 8. Table 8. EDS EDS analysis analysis of of the the chemical chemical composition composition of of each each phase phase on on SDSS-Nb SDSS-Nb wt.% wt.% in in forced-air- forced-air- cooled sample. cooled sample. Analysis %Cr %Ni %Mn %Cu %Mo %Si %Nb %S %Fe Analysis %Cr %Ni %Mn %Cu %Mo %Si %Nb %S %Fe Zone 1 23.16 10.02 1.63 --- 3.59 0.71 --- --- 60.89 Zone 1 23.16 10.02 1.63 --- 3.59 0.71 --- --- 60.89 Zone 2 Zone 2 28.65 28.65 6.59 6.59 1.28 1.28 --- --- 4.40 4.40 0.69 0.69 --- --- --- --- 58.39 58.39 Zone3 3 Zone 20.42 20.42 5.56 5.56 4.82 4.82 --- --- 3.04 3.04 3.73 3.73 --- --- ------ 62.43 62.43 Zone4 4 Zone 25.35 25.35 4.77 4.77 2.55 2.55 --- --- 2.88 2.88 0.40 0.40 --- --- ------ 64.05 64.05 Figure 11 shows the SEM-EDS results obtained for the welded and water-cooled samples. The results indicate the presence of chromium and molybdenum oxides (see Figure 11d,e). The findings from the SEM-EDS characterization in Figure 11 are shown in Table 9. The oxides found in all the samples are particle-type with isolated film-type oxides. Therefore, the welded joint would not be significantly affected mechanically. The results presented above indicate a large distribution of oxide precipitates in the welded speci- mens, which increase in size as the cooling process provides a higher diffusion rate. It is important to highlight that as the size of the oxide particles increases, the resistance to fatigue conditions will be negatively influenced. However, the mechanical response of the welded alloy can benefit from the formation of oxides by size and distribution. The above is based on mechanical reinforcement by precipitation. Therefore, if the metal ox- ides have a homogeneous distribution or adequate dispersion and reduced size, the shear stresses required to mobilize a dislocation anchored in the precipitate increase, favoring the mechanical response of the welded joint. The possible existence of this reinforcement generated by oxides is validated by Sun et al.. They quantified the size of the oxides in Crystals 2023, 13, x FOR PEER REVIEW 13 of 30 Crystals 2023, 13, 1192 13 of 28 Figure 11 shows the SEM-EDS results obtained for the welded and water-cooled sam- ples. The 17-4PH results indicate stainless the presence steel, which had a sizeofrange chromium and10 between molybdenum nm and 300 oxides (see Figure nm. Additionally, 11d,e). Theet Terashima ﬁndings al. from the SEM-EDS indicated characterization that the size in Figure 11 should be minimized to are shown delay in Table the possible 9. failure events that the oxides could cause. Figure 11. SEM-EDS analysis of SDSS-Nb, (a) SEM image of SDSS-Nb water-cooled sample; (b) EDS Figure 11. SEM-EDS analysis of SDSS-Nb, (a) SEM image of SDSS-Nb water-cooled sample; (b) EDS zone 1 austenite; (c) EDS zone 2 ferrite; (d) EDS zone 3 Cr oxide; (e) EDS zone 4 Mo oxide. zone 1 austenite; (c) EDS zone 2 ferrite; (d) EDS zone 3 Cr oxide; (e) EDS zone 4 Mo oxide. Table 9. Table 9. EDS EDSanalysis of of analysis thethe chemical composition chemical of each composition phasephase of each in SDSS-Nb wt.% in in SDSS-Nb the in wt.% water- the cooled sample. water-cooled sample. Analysis %Cr %Ni %Mn %Cu %Mo %Si %Nb %S %Fe Analysis %Cr %Ni %Mn %Cu %Mo %Si %Nb %S %Fe Zona 1 27.05 6.04 3.51 --- 4.94 0.52 --- --- 57.94 Zona 1 27.05 6.04 3.51 --- 4.94 0.52 --- --- 57.94 Zona 2 22.14 9.41 3.45 1.17 2.42 0.62 --- --- 60.79 Zona 2 22.14 9.41 3.45 1.17 2.42 0.62 --- --- 60.79 Zona3 3 Zona 26.63 26.63 5.85 5.85 2.98 2.98 ------ ------ ------ ------ 1.21 1.21 63.33 63.33 Zona4 4 Zona 22.35 22.35 4.77 4.77 2.55 2.55 ------ 6.75 6.75 1.25 1.25 ------ ------ 62.33 62.33 The oxides found in all the samples are particle-type with isolated ﬁlm-type oxides. Figure 12 shows the metal base structure after the cooling process using different Therefore, the welded joint would not be signiﬁcantly aﬀected mechanically. The results mechanisms investigated in this study. In Figure 12a, the sample quenched in water presented above indicate a large distribution of oxide precipitates in the welded speci- displays a clean microstructure with no apparent precipitation of the sigma phase. However, mens, which increase in size as the cooling process provides a higher diﬀusion rate. It is there are observed precipitates that are morphologically consistent with chromium nitrides important to highlight that as the size of the oxide particles increases, the resistance to (Cr2N) (see Figure 12a,d). This is reasonable since it is a rapid partitioning process in fatigue conditions will be negatively inﬂuenced. However, the mechanical response ferrite, leading to the instant formation of Cr2N. Similarly, this nitride will enable sigma of the welded alloy can beneﬁt from the formation of oxides by size and distribution. The phase formation over time, as it is a preferential site for nucleation or diffusion-induced above is based on mechanical reinforcement by precipitation. Therefore, if the metal ox- formation during slow cooling. This can be observed in Figure 12b, corresponding to the ides have a homogeneous distribution or adequate dispersion and reduced size, the shear sample quenched in forced air. The precipitates in this case change, preferentially locating stresses at the α/γrequired interface to mobilize (see Figure a dislocation 12e). Theseanchored in the precipitates areprecipitate morphologicallyincrease,consistent favoring the mechanical response of the welded joint. The possible existence with the sigma phase [28,44], and the precipitation mechanism remains the same, as it of this reinforcement generated occurs by region in the oxideswith is validated by Sun the highest et al.. diffusion rate They (grainquantiﬁed boundary)the size ofThe

Exploring Nb-Stabilized Super Duplex Stainless Steel Welding Impact (PDF)

Document Details

Tags

Related

Summary

Full Transcript

Upgrade to continue