1. Introduction
The increasing demand for energy efficiency in the transportation industry has accelerated the development of advanced, high-performance, and eco-friendly structural materials1-5). Among them, austenitic Fe-Mn-Al-C lightweight steels, which have a remarkable combination of high strength, ductility, and reduced density compared to conventional steels, have attracted substantial attention for next-generation automotive applications6-10). These superior mechanical properties of Fe-Mn-Al-C lightweight steels are primarily achieved through precise alloying and thermomechanical processing, by utilizing the unique precipitation behavior of κ-carbide within the austenite matrix, which has a crystal structure of (Fe,Mn)3AlC. Many researchers have studied κ-carbide precipitation and its influence on various Fe-Mn-Al-C lightweight steels. For instance, Chen et al. reported the precipitation behavior at different temperatures and its impact on Fe-30.5Mn-8Al-1.0C11). Xie et al. investigated the effect of Cu addition and heat treatment on κ-carbide precipitation and mechanical properties in Fe-25Mn-10Al-1.1C12). Furthermore, the application of these lightweight steels extends beyond the automotive industry, with potential uses in aerospace and defense industry, where weight reduction and strength are critical13,14). As such, the literature has presented a variety of studies on κ-carbide and its strengthening effect, highlighting its critical role in the development of high-performance lightweight steels for various applications.
Despite the demands and expectations, recent work has shown that κ-carbide in austenitic lightweight steels could negatively affect certain mechanical properties, such as ductility and toughness, depending on the thermal history. In particular, the welding process is known to make the heat affected zone (HAZ) of austenitic lightweight steels susceptible to metallurgical degradation. Jeong et al. confirmed the influence of intragranular and inter-granular κ-carbide precipitation on the microstructural and mechanical characteristics of the HAZ in various Fe-Mn-Al-C lightweight steels15). It was also reported that the welding thermal cycle can cause both unexpected hardening and softening effects in the HAZ depending on heat treatment conditions, peak temperature, and weld heat input in Fe-31.4Mn- 11.4Al-0.9C16). Considering the findings from the literature, the dual behavior of κ-carbide precipitation illustrates the complex effects of the welding thermal cycle on the austenitic lightweight steel system, emphasizing the necessity for a comprehensive understanding of welding metallurgy and recognizing the optimization of the welding process as critical research topics.
However, the lack of discussion regarding the welding of austenitic lightweight steels remains a significant hurdle to the broader application and practical implementation of Fe-Mn-Al-C alloys. Since welding is inevitable in most modern industrial fields, it is conceivable to say that understanding the precipitation behavior of κ-carbide as a function of alloy chemistry and the thermal cycle during welding is crucial. Accordingly, this study aims to investigate the influence of chromium (Cr) addition on the microstructural and mechanical characteristics of the weld and HAZ in a specific alloy. To this end, austenitic lightweight steels with and without Cr addition were fabricated and welded using an autogenous gas tungsten arc welding (GTAW) process, while Gleeble simulations were employed for manufacturing representative HAZ specimens. To achieve a comprehensive understanding of the effect of Cr addition on welding, a comparative analysis using micro hardness tests was carried out on the actual welded specimens, while tensile tests were performed on simulated HAZ, followed by microstructural analysis. The microstructural characteristics of each alloy were investigated using various techniques, including optical microscopy (OM), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). The results confirmed the significant effect of Cr addition on the weld and HAZ in austenitic Fe-Mn-Al-C lightweight steels.
2. Experimental Methods
To investigate the influence of Cr addition on the microstructural and mechanical behavior of austenitic lightweight steels during welding, a series of alloy fabrication, welding, and characterization procedures were conducted. Table 1 lists the chemical compositions of standard and Cr-added austenitic lightweight steels employed for the current study. Each ingot was fabricated from the original ingots, manufactured by vacuum in- duction melting. The alloys were fabricated from original ingots, manufactured by vacuum induction melting. Each ingot underwent homogenization at 1,200 °C for 2 hours, hot rolling into plates with a thickness of 13 mm, and solution treatment at 1,050 °C for 2 hours. Water quenching was performed after each process to minimize uncontrolled microstructural transitions.
Table 1
Chemical compositions of austenitic Fe-Mn-Al- C lightweight steels
| Alloy (wt.%) | C | Mn | Al | Cr | Fe |
|---|---|---|---|---|---|
| Standard | 0.9 | 29.8 | 10.4 | - | Bal. |
| Cr-added | 0.9 | 29.7 | 10.5 | 3.1 |
For autogenous GTAW, samples were prepared as rectangular specimens (130 mm × 25.4 mm × 3 mm) by machining and grinding. Argon shielding gas was used, and the welding direction was set parallel to the rolling direction of the specimen. The detailed welding parameters are listed in Table 2, and a schematic diagram along with a representative image of the welded plate is shown in Fig. 1(a). After welding, the fusion zone and HAZ can be observed, as indicated in Fig. 1(b). A separate specimen from the welded plate for hardness measurement was fabricated at an intermediate position, as indicated by the colored dashed line in Fig. 1(b).
Table 2
Welding parameters for autogenous GTAW
| Voltage | 12 V |
| Current | 100 A |
| Travel speed | 4 mm s-1 |
| Arc length | 2 mm |
| Temperature | RT |
| Shielding gas | Ar |
| Gas flow rate | 15 L/min |
Fig. 1
(a) An illustration of the rectangular-shaped specimen used for conducting autogenous GTAW, and (b) a photograph of the actual sample after welding, with indicating marks of fusion zone, HAZ, and the specimen for the hardness scan
Fig. 2 shows a schematic of the applied thermal cycle and the tensile specimen, which was prepared in accordance with ASTM standards. HAZ simulations were conducted at the center of each tensile specimen, using a Gleeble simulator (Gleeble 1500, Dynamic Systems Inc., USA). Based on a previous study, which confirmed the zero ductility temperature (ZDT) of Fe- 30Mn-10.5Al-0.9C lightweight steel as 1,210 °C, the peak temperature of the HAZ was set to 1,150 °C for both alloys with a heat input of 3 kJ/cm17). The actual thermal history for the simulated HAZ was calculated based on Rosenthal’s heat flow equation18).
Fig. 2
(a) A schematic of the thermal cycle for the HAZ simulation with a peak temperature of 1,150 °C with a heat input of 3 kJ/cm and (b) a diagram of the tensile test specimen with a red box indicating the location for HAZ simulation
For a discussion of the metallurgical characteristics of each alloy, micro hardness and tensile tests were employed. Vickers hardness was measured using a micro-indenter (HMV-2, Shimadzu, Japan) with a load of 1.96 N and a dwell time of 10 seconds for each actual welded specimens. Hardness measurements were taken 20 times at each point along the distance from the weld center. For clarity of discussion, only the average hardness values are reported in this paper. Tensile tests were performed at room temperature with a strain rate of 2.4 mm/min, using a commercial tensile testing machine (Z100, Zwick Roell Group, Germany). For comparison, tensile test specimen fabricated from each alloy was named as ‘Base sample’, while the tensile test specimen with HAZ simulation at its center was named as ‘HAZ sample’, as indicated in Fig. 2(b).
For microstructural analyses, a combination of OM, field emission scanning electron microscopy (FE-SEM, JSM-6700F, JEOL, Japan), and transmission electron microscopy (TEM, JEM 2100F, JEOL, Japan) was employed to investigate the morphological features and phase distributions in the welded samples. OM was primarily used for the initial examination of the macro- and microstructures across the weld cross-section, allowing identification of the fusion boundary, HAZ, and base metal regions. FE-SEM was utilized to observe the finer microstructural details, such as grain morphology, phase contrast, and precipitate distributions, especially in the HAZ and fusion zone. Specimens for SEM observation were prepared by standard metallographic procedures, including mechanical grinding with silicon carbide (SiC) abrasive papers of progressively finer grit sizes, followed by micro-polishing using a 1 μm diamond suspension to achieve a mirror-like surface finish suitable for high-resolution imaging. Prior to SEM observation, the polished samples were chemically etched by immersion in a 6% nitric acid solution diluted in ethanol. For nanoscale microstructure analysis, TEM was conducted to investigate fine precipitates and crystallographic features. Site-specific TEM specimens were prepared using a focused ion beam (FIB) milling technique. The FIB-prepared specimens were thinned to electron transparency, ensuring minimal damage for accurate imaging and diffraction analysis.
3. Results and Discussion
The representative microstructures of each alloy are shown in Fig. 3. As shown in the figure, both alloys exhibit an austenitic microstructure with angled grains and twin boundaries, along with a small fraction of elongated phase, indicated by yellow arrows. This elongated phase was identified as δ-ferrite according to the TEM observation with selected area diffraction (SAD) patterns. The average grain size and ferrite fraction in the standard specimen were measured to be 85.3 μm and 0.3%, respectively, whereas those in the Cr-added specimen were 47.2 μm and 2.3%. These microstructures indicate that both alloys exhibit an F-A solidification mode, which is reasonable as a previous study confirmed that the formation of ferrite is inevitable when the Al content exceeds 9.7% in the Fe-(25-31)Mn- (0.7-1)C-xAl alloy system19). In addition, the ferrite content increment in the Cr-added specimen compared to the standard sample is also expected, as Cr is a typical ferrite stabilizer20). The presence of the δ-ferrite phase in each alloy is a positive sign from the viewpoint of welding metallurgy, as it has adequate solubility for harmful elements that may cause hot cracking behavior, such as sulfur and phosphorus. However, it can also be a negative sign in the context of the Fe- Mn-Al-C system, since it stimulates the formation of an ordered DO3 phase with a composition of (Fe,Mn)3Al instead of κ-carbide, because of the low solubility of carbon. The formation of the DO3 phase can negatively affect the ductility of the material due to its hard and brittle characteristics21,22).
Fig. 3
OM images of the base metal microstructure in (a) the standard and (b) the Cr-added alloy exhibit a typical austenitic matrix with angled grains and prominent annealing twins, with elongated phases indicated by yellow arrows. In addition, (c) TEM micrograph along with selected area diffraction (SAD) patterns confirms that the elongated phases are δ-ferrite
Fig. 4 presents the results of Vickers hardness measurements taken from the base metal, across the fusion zone, to the other side of the welded samples. As shown in the figure, the addition of Cr resulted in a different transition behavior of mechanical properties in both the fusion zone and the HAZ compared to the standard specimen. In the standard specimen, a complex behavior was observed in micro hardness, where an increase in micro hardness was measured in the HAZ relative to the base steel. In contrast, for the Cr-added specimen, an increase in hardness was noted only in the fusion zone, while no noticeable transition was detected in the HAZ. It is reasonable to say that the HAZ microstructures between the standard and Cr-added specimens differ because of κ-carbide precipitation behavior, since it is well established that an increase in hardness is a representative indication of κ-carbide formation when Fe-Mn-Al-C alloys are thermally exposed at around 500 °C23). On the other hand, the hardness trend in the fusion zone shows opposite behaviors between the standard and Cr-added alloys. This suggests that different factors may be influencing the fusion zone characteristics, which are not related to κ-carbide precipitation.
Fig. 4
Vickers hardness distribution profiles across the welded joint, ranging from the base metal to the opposite side of the weld, in both standard and Cr-added specimens
To understand the different hardness transitions observed in the fusion zones of the samples, Fig. 5 presents representative high-magnification SEM micrographs of the fusion zone for the Cr-added sample. As shown in the figure, numerous sub-micrometer scale precipitates, estimated to be DO3 or carbides such as M3C, M7C3, M6C, and M23C6, are observed along the grain boundaries in the fusion zone24-26). These precipitates are known for their inherent high hardness, which significantly increases the hardness of the alloy27). In addition, it was observed that the weld of austenitic lightweight steel with Cr addition exhibited substantial formation of M3C carbide in the fusion zone, leading to an increased susceptibility to solidification cracking28). Therefore, it is believed that the segregation of alloying elements and formation of the carbides in the Cr-added alloy, during the solidification process after welding, lead to a relatively high micro hardness compared to the base metal. Although further research is required to clarify details about these precipitates, such as their crystal structures, formation mechanisms, fraction, and distribution, it is believed that welding processes utilizing filler materials, such as flux cored arc welding (FCAW), shielded metal arc welding (SMAW), or submerged arc welding (SAW), rather than autogenous welding may mitigate potential microstructural risks in specific applications29). Fig. 6 presents the stress-strain curves for the ‘Base sample’ and ‘HAZ sample’ of each alloy, along with the macro images after the tensile fracture for each case (Fig. 6(b-e)). As shown in Fig. 6(a), both the standard and Cr-added alloys exhibit high strength and ductility, with the Cr-added alloy demonstrating slightly higher strength and lower elongation compared to the standard alloy. The slight difference in tensile characteristics between both alloys seems reasonable since the Cr-added alloy has a relative advantage in solid solution hardening effect compared to the standard alloy due to the addition of Cr. In the standard alloy, the base sample exhibited a tensile strength of 810 MPa and an elongation of 52%, while the HAZ sample recorded a tensile strength of 812 MPa and an elongation of 51%; in the Cr-added alloy, the base sample showed a tensile strength of 867 MPa and an elongation of 50%, whereas the HAZ sample confirmed a tensile strength of 870 MPa and an elongation of 49%. While variations in fracturing were observed in samples during tensile testing, it is noticeable that only a negligible difference in tensile properties was observed between the base and HAZ samples for both alloys. On the other hand, the macro images after the tensile testing reveal a significant distinction between the standard and Cr-added alloys. In the case of the standard alloy, it is clear that the fracture location is different between the base sample and the HAZ sample. The fracture of the HAZ sample occurred in the region where it was not influenced by the welding thermal cycle, as indicated by black arrow in Fig. 6(c). Considering that the tensile stress applied to the specimen tends to concentrate in more vulnerable areas, it can be inferred that the HAZ of the standard alloy has relatively high strength compared to its base metal. Consequently, it seems that the tensile test result of the HAZ sample was predominantly governed by its base metal. This behavior aligns with the results of the Vickers hardness profile, resulting from the precipitation of κ-carbide. In contrast, the Cr-added alloy consistently fractured at the center of the tensile specimens in both the base sample and HAZ sample. From the viewpoint of microstructure, it might be considered that the fracture location shifted from the base metal to the HAZ after welding thermal cycle. However, given that the stress-strain curves for the Cr-added alloy remain consistent regardless of the HAZ simulation, this fracturing behavior indicates that the tensile test results for the Cr-added alloy are independent of the welding thermal cycle, suggesting that the HAZ has no contribution to changes in tensile properties induced by precipitation of κ-carbide, which again aligns with the results of the Vickers hardness profile.
Fig. 5
Representative microstructural images of the fusion zone in the Cr-added alloy, with (a) low magnification and a marking at a random location, and (b) high magnification with yellow arrows indicating precipitates along the grain boundaries
Fig. 6
(a) The stress-strain curves of the base and HAZ simulated specimens in standard and Cr-added alloys along with (b-e) the macro images of all specimens after tensile fracturing
To support the influence of Cr addition on HAZ characteristics in detail, microstructural investigation through TEM was carried out. Fig. 7 exhibits dark-field TEM images of the HAZ for the standard (Fig. 7(a)) and the Cr-added (Fig. 7(b)) samples, along with representative SAD patterns obtained by FIB extraction from HAZ simulated specimens. As shown in the figures, both samples contain nano-sized κ-carbide in their matrix, according to the dark-field TEM images. The κ-carbide, indicated by colored arrows, is well distributed in the standard specimen, exhibiting a more significant presence when compared to the Cr-added sample. In addition, it was confirmed that both alloys had a crystallographic relationship of [011]γ // [011]κ between κ-carbide and the austenite matrix30). Similar to the dark-field TEM observations, the superlattice peaks are more clearly defined in the HAZ of standard sample, indicating a stronger formation behavior of κ-carbide in the standard alloy, which results in an increase of hardness and strength. The difference in the distribution of κ- carbide on the HAZ between the standard and Cr-added alloys can be attributed to the influence of Cr. When certain alloying elements are added in the Fe-Mn-Al-C system, such as Mo, Si, Cr, substitution occurs at the atom sites within the κ-carbide31). In the substituted form, the interfacial energy and elastic strain energy can be changed according to thermodynamic calculations, resulting in a transition of the formation energy of κ-carbide precipitation. In the case of Cr addition, it has been reported that Cr prefers to substitute the site of Al in the κ-carbide structure, resulting in suppressed nucleation and growth of κ-carbide during isothermal heat treatment, due to the increased formation energy32). While the relatively high amount of κ-carbide in the HAZ of the standard specimen compared to the Cr-added specimen is expected, as reported in previous work, the current result is notable because Cr effectively retards the precipitation of κ-carbide not only in the steel making and heat treatment processes but also in actual welding processes33). Although κ-carbide formation in the HAZ can enhance mechanical properties such as strength and hardness, the literature consistently regards the hardening effect of the HAZ as a potential hazard. Previous study have reported that κ- carbide precipitates in the HAZ may cause a drastic reduction in cryogenic impact toughness by providing an easy path for crack propagation34). In other words, the difficulty in enhancing the usability of austenitic lightweight steels due to metallurgical degradation in the weld HAZ, induced by uncontrolled κ-carbide precipitation during welding, may be addressed by appropriate control of alloying elements and heat treatment processes.
Fig. 7
Dark-field TEM images of the HAZ located 2 mm away from the fusion line for the (a) standard and the (b) Cr-added samples, along with SAD patterns showing κ-carbide superlattice peaks and austenite matrix
Based on the experimental data, it was found that even a small addition of Cr to austenitic Fe-Mn-Al-C lightweight steel significantly influences the metallurgical and mechanical characteristics of the welded structure. The summary of microstructural transitions occurring in welding for both standard and Cr-added alloys is exhibited in Fig. 8. In austenitic Fe-Mn-Al-C lightweight steels, the precipitation behavior of κ-carbide dominates the mechanical characteristics of the HAZ. In the standard alloy, nucleation and growth mechanisms of κ- carbide occur rapidly during the cooling thermal cycle of welding, resulting in an evident increment of the volume of κ-carbide distributed in the austenitic matrix and a notable enhancemnet in mechanical properties compared to its base metal. The mechanical characteristics of the HAZ in the Cr-added alloy, on the other hand, showed no meaningful change after the welding process. While it is well understood that κ-carbide has a coherent interface with the austenitic matrix, the substitutional occupation of Al sites by Cr atoms in the κ- carbide crystal structure increases the thermodynamic energy for κ-carbide precipitation, thereby suppressing the nucleation and growth associated with strengthening and hardening effects35). In the fusion zone, a different behavior is observed. Both alloys employed for the current study have F-A solidification mode. While the DO3 phase within the ferritic regions is possible, it is believed that the relatively decreased hardness in the fusion zone of the standard alloy originated from κ- carbide precipitation within the austenite phase, which was delayed due to the segregation of alloying elements accompanying phase separation in the fusion zone. In the Cr-added alloy, the relatively high mico hardness within the fusion zone is attributed to the precipitation of carbides along fusion zone grain boundaries and the increased potential for DO3 phase formation. These findings collectively demonstrate that the microstructural and mechanical properties of welded austenitic lightweight steels are the complex results of the chemical composition of the alloy and thermal history of the welding process, leading to various phases such as κ-carbide, DO3, and carbides. In other words, since the mechanical and microstructural evolutions confirmed that the precipitation of nano-sized κ-carbide was clearly suppressed by Cr addition, it is suggested that appropriate alloy design including Cr would be effective to improve the feasibility of austenitic lightweight steels for practical use, and further research is recommended to understand the effects of various alloying elements, particularly during the welding process in austenitic lightweight steel systems.
4. Summary
This study focused on the effects of Cr addition on the metallurgical and mechanical characteristics of welded austenitic Fe-Mn-Al-C lightweight steels. The key findings obtained from this study for optimization and further application are summarized as follows:
1) Both the standard and Cr-added alloys exhibited similar austenitic microstructures; however, the ferrite content in the Cr-added specimen was relatively higher due to the increased presence of ferrite stabilizers.
2) Cr addition resulted in different hardness behavior in the weld, with the Cr-added specimen showing uniform hardness in the HAZ and increased hardness in the fusion zone, while the standard specimen showed increased hardness in the HAZ but decreased hardness in the fusion zone.
3) The experimental results confirmed that Cr addition to austenitic lightweight steel effectively suppressed the precipitation of κ-carbide during welding, resulting in stable mechanical properties in the HAZ.
4) Although the fusion zone of the Cr-added alloy exhibited increased hardness due to carbide formation along the grain boundaries, welding methods utilizing filler materials are expected to effectively mitigate potential microstructural risks.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
