Next Article in Journal
Study on the Application of the GTAW Process in Strengthening the Welding Quality of Short Duplex Stainless Pipe
Previous Article in Journal
Deep Learning to Predict Deterioration Region of Hot Ductility in High-Mn Steel by Using the Relationship between RA Behavior and Time-Temperature-Precipitation
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Metals
Volume 12
Issue 10
10.3390/met12101690
metals-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Microstructure Refinement of 301 Stainless Steel via Thermomechanical Processing

by
Khaled J. Al-Fadhalah
1,*,
Yousif Al-Attal
1 and
Muhammad A. Rafeeq
2
1
Department of Mechanical Engineering, College of Engineering and Petroleum, Kuwait University, P.O. Box 5969, Safat 13060, Kuwait
2
Nanotechnology Research Facility, College of Engineering and Petroleum, Kuwait University, P.O. Box 5969, Safat 13060, Kuwait
*
Author to whom correspondence should be addressed.
Metals 2022, 12(10), 1690; https://doi.org/10.3390/met12101690
Submission received: 25 August 2022 / Revised: 2 October 2022 / Accepted: 4 October 2022 / Published: 10 October 2022
Browse Figures
Review Reports Versions Notes

Abstract

:
The current study applied thermomechanical processing (TMP) on 301 austenitic stainless steel to produce an ultrafine-grained austenitic structure, examining the dual effects of deformation at subzero temperature and TMP cycles on the strain-induced α′-martensitic transformation and austenite reversion occurring upon subsequent annealing. Three TMP schemes were adopted: (1) one cycle using a strain of 0.30, (2) two cycles using a strain of 0.20, and (3) three cycles using a strain of 0.15. Each cycle consisted of tensile deformation at −50 °C followed by annealing at 850 °C for 5 min. Compared to other schemes, the use of three cycles of the 0.15 strain scheme resulted in a significant formation of the martensitic phase to about 99 vol.%. Consequently, the austenite reversion occurred strongly, providing a mixture of the austenitic structure of reverted ultra-fine grains and retained coarse grains with an average grain size of 1.9 µm. The development of a mixed austenitic structure was found to lower the austenite stability and thus enhance the α′-martensitic transformation upon deformation in subsequent cycles. Moderate growth of high-angle grain boundaries occurred in the austenitic phase for all schemes, reaching a maximum of 64% in cycle 3 of the 0.15 strain scheme. The tensile behavior during subzero deformation was generally characterized by an initial strain hardening by slip (stage I), followed by a remarkable increase in strain hardening rate due to the strain-induced α′-martensitic transformation (stage II). Further straining promoted breakage of the α′-martensite banded lath structure for forming dislocation cell-type martensite, which was marked by a decline in strain hardening rate (stage III). Accordingly, the latter hardening stage had a lesser hardness enhancement of deformed samples with an increasing number of cycles. Nevertheless, the yield strength for samples processed by the 0.15 strain scheme improved from 450 MPa in cycle 1 to 515 MPa in cycle 3.

1. Introduction

Austenitic stainless steels have been widely used due to their excellent corrosion resistance, toughness, and weldability [1,2,3]. However, their strength is relatively low in the annealed state, limiting their use in many potential applications. Therefore, the strength of austenitic stainless steel is generally enhanced by cold working or solid solution strengthening. On the other hand, conventional thermomechanical processing (TMP) has been applied to austenitic stainless steels to reduce the grain size to the range of 10–30 μm [4]. Developing a strain-free and carbide-free austenitic microstructure typically requires cold working and subsequent solution annealing at temperatures ranging from 900 °C to 1200 °C. Nevertheless, conventional TMP has a negligible effect on the strength and hardness of austenitic stainless steels.
On the contrary, a special TMP developed for the formation of an ultra-fine grained (UFG) microstructure in metastable austenitic stainless steels has been proven to improve strength and hardness while maintaining ductility and toughness significantly [5,6,7,8]. It also improves the workability, allowing for a broader range of austenitic stainless steels for metal forming applications. The application of special TMP often necessitates heavy cold working to induce sufficient transformation of relatively soft austenite γ into strong α′-martensite in the metastable austenitic structure, displaying the so-called transformation-induced plasticity (TRIP) effect. Deformation by cold working generally elongates the austenite grains and forms a martensitic phase. It is expected that some deformed retained austenite remains in the deformed microstructure. Therefore, subsequent annealing might lead to the reversion of strain-induced martensite to UFG austenite and recrystallization and growth of the retained austenite to equiaxed grains of micron size, mainly if the amount of retained austenitic structure is relatively large. In general, reverse annealing in austenitic stainless steels is initiated at 600 °C and higher, depending on the alloy’s chemical composition, deformation condition, and duration and heating rate of the annealing. In the case of short-term heating (less than 10 min), reversion annealing is preferably employed in the range of 750–900 °C to produce complete austenite reversion with grain size in the submicron level [9]. To achieve remarkable grain refinement, it is essential to have a large amount of strain-induced martensite before reversion annealing takes place. Massive martensitic transformation is practically possible by cooling to subzero temperatures and/or increasing the amount of strain applied during cold working.
Moreover, the features of reversion transformation might differ from one austenitic steel to another, generally depending on the amount of cold working, reversion annealing parameters, and the type of reversion mechanism (diffusional type or martensitic-shear type). Tomimura et al. [8] reported that diffusional reversion involves the nucleation and growth of austenitic grains within a martensitic matrix with a low density of dislocations. On the contrary, the refinement of austenitic grains generated by shear reversion occurs by recovery and recrystallization, producing thin plates with morphological traits similar to lath martensite. The thin plates typically contain a high density of dislocations and traces of the microstructure from the earlier lath martensite.
The formation of the UFG microstructure in austenitic stainless steels has been typically obtained using single-step TMP [10,11,12,13,14,15,16,17,18,19]. Di Schino et al. [10,11] performed single-step TMP on sheets of 301 and 304 stainless steels and found that heavy cold rolling and subsequent reversion annealing produced a refined microstructure with an average grain size of 800 nm. As a result, the strength and hardness of the material increased significantly. Tao et al. [12] also reported that reversion annealing led to grain refinement of austenite to 350 nm, enhancing strain hardening and ductility and increasing yield strength. Furthermore, Misra et al. [16,18,19] demonstrated that increasing the degree of cold rolling caused the martensite structure to change from lath type to dislocation cell type. It was noticed that the generation of ultra-fine austenitic grains during reversion annealing required the transformation to dislocation-cell-type martensite, involving the refinement of lath size and breakup of the lath martensitic structure.
Other TMP works on austenitic steels have also focused on studying reversion annealing parameters (annealing time, annealing temperature, and repetitive annealing) for the enhancement of grain refinement [20,21,22,23]. Ravi Kumar and Sharma [20] examined the use of short-time repetitive annealing in cold-rolled 304 L stainless steel. They indicated that the reverted austenite was initially characterized by a fine polygonal-type structure of low-angle boundaries and high dislocation population, which gradually transformed into newly recrystallized grains of high-angle boundaries with increasing annealing iterations. On the other hand, the retained austenite grains underwent slow recrystallization, resulting in grain fragmentation through subgrain formation and subsequent formation of refined grains. Sharifian et al. [21] also utilized two-step annealing in cold-rolled 316 L stainless steel and reported that the first annealing step using low temperature (750 °C) only promoted austenite reversion without recrystallization and grain growth of the retained austenite. On the contrary, higher temperatures in the second annealing step (850–950 °C) enhanced recrystallization in the retained austenite without remarkable grain growth.
Moreover, several studies have focused on developing bimodal grain distribution in thermomechanically processed austenitic stainless steels for enhancing the strain-hardening capability via minimizing strain localization in the fine grains during deformation [24,25,26,27,28,29]. Kisko et al. [24] examined the martensite to austenite reversion by varying the annealing time and temperature. A bimodal distribution of austenitic grains was produced in a temperature range of 700–800 °C via the formation of fine reverted austenitic grains and recrystallization of the retained austenitic structure into relatively coarse grains. The bimodal grain distribution was found to enhance strength and ductility. Additionally, Xu et al. [25] investigated the reversion process in 90% cold-rolled 316 LN stainless steel and showed that short annealing times (5–10 min) were sufficient to produce a bimodal distribution of austenitic grains. More recently, Niu and Wu [26] examined the development of bimodal distribution in TMP stainless steel and reported that reversion annealing at 800 °C produced ultra-fine austenite grains from the reverted structure via shearing reversion and subsequent recrystallization. At the same time, it formed micro-sized austenite grains from the retained austenite regions. The difference in grain size was attributed to the higher nucleation rate and stored energy in the reverted structure. Other studies examined the effect of bimodal-type grain size distribution on the tensile properties of TMP austenitic steels, indicating that the coarse grains accommodated more significant strains during tensile deformation, which allowed for more substantial strain hardening while restricting localized strain in the fine grains and thus enhancing ductility [28,29]. It was also identified that fine grains deformed at large strains via martensitic transformation due to local “grain-to-grain” interactions that resulted in high local stresses, providing a significant improvement in strength.
Repetitive TMP of austenitic stainless steels has been shown to enhance further the refinement of the austenitic microstructure and the mechanical properties via the production of a bimodal distribution of austenitic grains [30,31,32,33,34,35,36,37]. Naghizadeh and Mirzadeh [30] examined repetitive TMP of 304 L stainless steel for two cycles, using 75% cold roll reduction and annealing at 850 °C for 1 min, and showed that the average grain size of the austenitic phase was reduced from an initial value of 13 mm to 2.45 and 2.05 mm in cycle 1 and cycle 2, respectively. Additionally, Järvenpää et al. [31] examined the enhancement of grain refinement in 301 LN stainless steel processed by two cycles of TMP. The annealing treatment of the second cycle was performed in the range of 750–900 °C. The average grain size was found to be 0.48 and 0.75 mm for samples annealed at 750 °C and 900 °C, respectively. The use of two-cycle TMP was also shown to enhance strength and ductility compared to single TMP. Lee et al. [32] also utilized repetitive TMP and indicated that the simultaneous formation of reverted austenite and recrystallization of retained austenite produced a bimodal distribution of ultra-fine grains and coarser grains during reversion annealing. Additionally, Sun et al. [34] studied repetitive TMP in 304 stainless steel, focusing on the effect of annealing temperature (700–850 °C) on the reversion process during the second TMP cycle. It was shown that shear reversion and subsequent recrystallization took place at shorter times with increasing annealing temperature, reaching complete recrystallization within 1 min for the sample annealed at 850 °C. The development of a bimodal distribution of grains was also shown to occur for samples annealed above 800 °C, where coarse grains mainly formed due to the recrystallization of retained austenite. Al-Fadhalah and Aleem [37] applied repetitive TMP on 304 austenitic stainless steel using two schemes. The first scheme consisted of four cycles of tensile deformation at −100 °C to a moderate strain of 0.4, while the second applied two cycles of a large tensile strain of 0.6 at −100 °C. Both schemes utilized short-term annealing of 5 min at 800 °C. The volume fraction of α′-martensite increased with increasing TMP cycles, reaching a maximum of 98 vol.% using the second scheme. Besides, increasing the strain and/or the number of cycles resulted in a stronger reversion to fine austenitic grains of 1 mm, leading to an increase in yield strength and strain hardening rate as examined by tensile testing.
The above works demonstrate that an effective TMP requires a tremendous amount of α′-martensite to form during plastic deformation, which can be further induced by increased TMP cycles. Eventually, substantial grain refinement of the austenitic phase can be achieved upon reversion annealing. Therefore, the current study sought to investigate the application of repetitive TMP to produce a refined austenitic microstructure in 301 stainless steel, employing controlled deformation via tensile testing at subzero temperature and subsequent annealing. In particular, the study focused on the effect of applying small-to-moderate tensile strains on the α′-martensite formation during subzero deformation and on the austenite refinement and its bimodal grain distribution upon reversion annealing. The current study adopted the TMP from the work of Al-Fadhalah and Aleem [37] on 304 stainless steel with small changes in deformation and annealing conditions to allow for successful repetitive cycles. The changes are as follows: use of small-to-medium strains (0.15, 20, 0.30) during tensile deformation at −50 °C and reversion annealing at a higher temperature (850 °C) for 5 min. The changes were necessary, since 301 stainless steel has weaker austenite stability against strain-induced martensitic transformation than 304 stainless steel. Therefore, it is more susceptible to plastic instability and premature fracture during tensile deformation at subzero temperatures. At the same time, 850 °C was found to provide complete reversion upon short-term annealing. Accordingly, three TMP schemes were employed: (1) three cycles using a strain of 0.15, (2) two cycles using a strain of 0.20, and (3) one cycle using a strain of 0.30. In each cycle, the sample was deformed under tension at −50 °C and later annealed at 850 °C for 5 min. Microstructure refinement and grain boundary development were examined using scanning electron microscopy (SEM) and electron backscattered diffraction (EBSD). Tensile and Vickers microhardness tests were used to evaluate the mechanical behavior of the TMP samples.

2. Materials and Methods

301 stainless steel sheets of 2 mm in thickness with the following chemical composition were received in a solution annealing condition: 18.03 Cr, 7.18 Ni, 1.08 Mn, 0.44 Si, 0.26 Mo, 0.34 Cu, 0.02 C, 0.01 N wt.% and the remainder of Fe. The chemical composition was determined using X-ray fluorescence (XRF) (ZSX Primus IV, Rigaku, Tokyo, Japan). Tensile specimens in the shape of a dog bone were produced with a gauge length of 25 mm and dimensions that met the ASTM-E8 standard. The samples were produced using a precision laser cutting machine and then polished to remove any burrs or irregularities. Three repetitive TMP schemes were chosen during tensile deformation to promote different amounts of α′-martensite formation and subsequent austenite reversion upon annealing. The first scheme involved three cycles of 0.15 tensile strain deformation at −50 °C and annealing at 850 °C for 5 min, once after each deformation. The second scheme used two cycles of tensile deformation at −50 °C of a strain of 0.20, followed by annealing at 850 °C for 5 min, once after each deformation. The third scheme applied one cycle of tensile deformation at −50 °C to a strain of 0.30 and subsequent annealing at 850 °C for 5 min. An electromechanical testing machine with an environmental chamber (Instron-5581, Instron, High Wycombe, UK) was used to perform the tensile test at a rate of 1 × 10−3 s−1. Stress–strain and strain hardening curves were plotted from the tensile tests to assess martensitic transformation qualitatively. The tensile behavior of TMP samples was compared to a sample deformed at room temperature to assess the effect of strain-induced martensitic transformation on strain hardening. The annealing was carried out using an environmental chamber furnace (AWF 12/14, Lenton, Hope Valley, UK), and the samples were heated up to 850 °C at a heating rate of 22 °C/min and remained at this temperature for 5 min to allow phase reversion and recrystallization. Samples were then water quenched to maintain the resultant microstructure. Details of the thermomechanical processing of the three schemes are summarized in Table 1.
Coupons were cut for microstructural examination, undergoing standard metallographic preparation before being polished with colloidal silica. Electropolishing was next applied to prepare the samples for microstructure analysis using SEM and EBSD. Electropolishing was performed at room temperature using an electropolishing machine (LectroPol-5, Struers, Ballerup, Denmark) in a standard A2 solution at 30 V. The EBSD measurements were taken using an EBSD detector (Aztec, Oxford, High Wycombe, UK), which was coupled to a field-emission scanning electron microscope (7001F-JSM, JEOL, Tokyo, Japan) operated at 20 kV. The EBSD maps were generated with a 0.10–0.25 μm step size. The EBSD data were employed to evaluate several microstructure characteristics, including phase volume fraction, grain size and distribution, and grain boundary characters. Phase identification in EBSD is based on the crystallographic differences between the phases present in the material. For the current material, the crystal structure databases available in the EBSD software were used to identify the crystal crystallographic information of each phase (fcc structure for austenite, bcc structure for α′-martensite, and hcp structure for ε-martensite). The emphasis was put on analyzing austenitic grains with coincidence site lattice (CSL) boundaries, specifically the Σ3, Σ9, and Σ27 boundaries. Identifying misorientation angles was set using a critical misorientation angle of 15° to distinguish low-angle grain boundaries (LABs) from high-angle grain boundaries (HABs). Grain size measurements were determined by excluding LABs and using the linear intercept method in the EBSD post-processing software.
The Vickers microhardness test was used to determine the hardness of TMP samples. Using a Buehler hardness tester (Falcon 500, Innovatest, Maastricht, The Netherlands), microhardness indentations were carried out at room temperature on metallographically prepared coupons from tensile samples under a load of 500 gf and a dwell time of 15 s. Ten measurements were collected at different positions for each coupon, and the average value was determined.

3. Results and Discussion

3.1. Phase Transformation

Figure 1 depicts the EBSD phase maps for TMP samples with a 0.3 strain. The HABs are depicted in black, austenite in blue, ε-martensite in yellow, and α′-martensite in red. Along with the shear bands in the deformed microstructure, extensive formation of α′-martensite laths occurred. Initially formed in the annealed structure, some pancake-shaped austenitic grains transformed into α′-martensite upon deformation. Nevertheless, the remnants of the pancaked γ-grains were still present (Figure 1a). Fine ε-martensite was also present, but its content was significantly lower than that of α′-martensite. Many ε-martensite grains formed in the shear band and near the retained austenitic structure, while others formed at grain boundaries. The austenite phase formed extensively during reversion annealing, and many α′-martensite laths took the shape of isolated islands, as shown in Figure 1b. Compared to the sample deformed by a strain of 0.30, Figure 2a indicates less α′-martensite transformation occurred in cycle 1 for the sample deformed by a strain of 0.2. At the same time, more ε-martensite was present, as demonstrated by the increase in regions in yellow. The deformed microstructure in cycle 1 did not fully revert to austenite after annealing. On the contrary, the deformed microstructure in cycle 2 was almost martensitic, containing small islands of retained austenite, whereas subsequent annealing resulted in a near complete reversion to austenite, with only a few islands of α′-martensite (Figure 2c,d). In addition, Figure 3 shows the phase maps for samples processed with a strain of 0.15. Compared to deforming by a strain of 0.20, the deformed microstructure by a strain of 0.15 in cycle 1 reveals less formation of the α′-martensitic phase (Figure 3a). Yet, the ε-martensite formed almost at the same volume observed for the sample deformed by a strain of 0.20 in cycle 1. The reversion to austenite upon annealing also occurred much less, as a large amount of α′-martensite was still preserved (Figure 3b). In cycle 2, martensitic transformation became much more robust, and a few regions of retained austenite are present in the deformed structure (Figure 3c). Additionally, the ε-martensite is at a much lower fraction than in cycle 1, randomly dispersed in the deformed structure. Annealing resulted in a near complete austenitic structure, with only a few isolated regions of α′-martensite, as presented in Figure 3d. Compared to cycle 2, deforming the sample in cycle 3 enhanced the α′-martensitic transformation, as much less ε-martensite and retained austenite were present in the deformed structure, and stronger reversion annealing occurred (Figure 3e,f).
The phase volume fraction for TMP samples is shown in Table 2. The martensitic transformation is enormous at strain 0.3, with a volume fraction of 97.4% for ε- and α′-martensite. However, the reversion annealing is incomplete, since a large volume fraction of α′-martensite (26.3%) is preserved. For the 0.20 strain scheme, martensitic transformation increases to a maximum volume fraction of 99.7% in cycle 2. Subsequent annealing resulted in a successful reversion to austenite, with a volume fraction of 98%. The use of the 0.15 strain resulted in progressive improvement in the volume fraction of α′-martensite as the number of cycles increased, resulting in a maximum martensitic transformation of 98.4% and reversion to austenite of 99.2% in cycle 3. Accordingly, one can suggest that using a lower strain in combination with increasing the number of TMP cycles improves martensitic transition during deformation and the reversion process to austenite upon annealing.

3.2. Microstructure

Figure 4 shows a SEM micrograph and EBSD grain boundary construction map of the microstructure of 301 stainless steel in an as-received condition. LABs are shown in gray, while HABs are shown in black. Three different CSL boundaries are also illustrated (red, yellow, and green for Σ3, Σ9, and Σ27 boundaries, respectively). The as-received microstructure, shown in Figure 4a, is a fully annealed structure with coarse austenitic grains and no formation of a martensitic phase. The EBSD map in Figure 4b indicates a considerable formation of HABs, with a large proportion of those having Σ3 boundaries, demonstrating that annealing twins predominate in the initial microstructure. Table 3 shows the average grain boundary and grain boundary statistics for a sample in the as-received condition. The typical grain size is about 5 mm (including annealing twins). The CSL fraction is 50.2%, representing the sum of the Σ3, Σ9, and Σ27 boundary fractions. Furthermore, the proportion of HABs is exceptionally high (96%), indicating a fully annealed microstructure.
The grain boundary construction maps for samples processed with a strain of 0.30 are shown in Figure 5. The grain refinement in the deformed sample, shown in Figure 5a, is demonstrated by the transformation of the initial coarse austenitic structure into a refined fragmented α′-martensitic structure with the presence of long α′-martensitic laths. With increasing tensile strain, the structure of α′-martensite generally transforms from long laths to a fragmented structure of submicron size. According to Misra et al. [19], the fragmentation of α′-martensite laths implies a transformation from lath-type martensite to dislocation-cell-type martensite in the microstructure. The deformed structure also reveals the HABs of the martensitic-phase, whereas the austenitic phase’s remnant has LABs. Reversion to austenite occurred during annealing, resulting in a moderate development of the austenitic phase, as illustrated in Figure 5b. A mixture of fine and coarse grains constitutes the austenitic grains. Fine grains mainly have HABs with CSL features of Σ3 boundaries, whereas coarse grains mostly have LABs presenting subgrain boundaries. On the other hand, the residual martensitic phase still contains many LABs.
Figure 6 presents the grain boundary construction maps for samples processed using a strain of 0.20. The deformed structure in cycle 1 underwent a moderate martensitic transformation, as shown in Figure 6a, and was primarily composed of long laths of LABs. Some of the laths were fragmented, resulting in the development of fine martensitic grains of HABs. Additionally, there were still remnants of the original austenitic grains. Annealing in cycle 1 caused a partial reversion to austenite. In the austenitic grains, the annealing twins with Σ3 boundaries formed moderately. Many very fine grains with Σ3 boundaries were also present at the intersection of the shear bands, which were initially generated in the deformed structure. Significant improvement in martensitic transformation occurred upon deformation in cycle 2, resulting in a strong fragmentation of the martensitic laths (Figure 6c). Subsequent annealing provided near complete reversion to austenite, as shown in Figure 6d. Compared to the 0.30 strain scheme, the grain refinement in the austenitic phase is greater, resulting in a more pronounced mixture of fine and coarse grains as a sign of bimodal grain distribution. Most of the fine grains are free of sub-boundaries and have Σ3 boundaries. In contrast, many subgrain boundaries with LABs are present in the coarse austenitic grains. Additionally, the residual martensitic phase was mainly characterized by LABs.
Furthermore, Figure 7 shows the grain boundary construction maps for samples processed with a strain of 0.15. Compared to a strain of 0.20, the deformed structure in cycle 1 exhibited less martensitic transformation, as previously demonstrated in Figure 3a and Table 2. The map shown in Figure 7a indicates that long laths of LABs account for most of the martensitic phase. Yet, the formation of tiny martensitic grains of HABs also occurs due to lath fragmentation. The remnants of the original austenitic grains are largely present, and they contain many fragmented laths. Subsequent annealing in cycle 1 resulted in a modest reversion to austenite. A moderate formation of annealing twins was also observed in the remnants of the original austenitic grains. There are also a few fine grains with Σ3 boundaries. In cycle 2, a considerable fragmentation of martensitic laths occurred in the deformed structure, resulting in an increase in the fraction of HABs for the martensitic phase and the initial austenitic grains becoming martensitic. Upon annealing, the complete reversion formed an austenitic structure comprising fine and coarse grains (Figure 7d). Compared to the 0.2 strain scheme, the fine grains are generally free of sub-boundaries and possess Σ3 boundary characters, while coarse grains are present in a higher quantity and bigger size. Cycle 3 produced a deformed structure compared to that obtained in cycle 2 but with a slightly higher fragmentation of the martensitic phase. An almost complete reversion to austenite occurred upon annealing. Nonetheless, annealing in cycle 3 increased the population of fine austenitic grains while decreasing their coarse counterparts (Figure 7f).
In addition, the SEM micrographs presented in Figure 8 further illustrate the microstructure evolution using the 0.15 strain scheme. In the case of samples in deformed conditions, the resultant microstructure in cycle 1 is characterized by the formation of shear bands in the retained austenite and moderate transformation into long α′-martensite laths (Figure 8a). With an increasing number of cycles, remarkable strain-induced martensitic transformation occurred, resulting not only in an increase in the volume fraction of α′-martensite but also in breakage of the lath-type martensite and formation of dislocation-cell-type martensite (Figure 8c,e). For the annealed sample in cycle 1, a partial reversion occurred, producing a mainly austenitic microstructure of shear-banded pancaked grains, while the residual microstructure of the martensitic phase was partially preserved (Figure 8b). Due to the massive martensitic transformation in the deformed structure of cycles 2 and 3, subsequent annealing produced an almost fully reverted austenitic structure of two major types: the austenitic microstructure of shear-banded pancaked coarse grains and equiaxed recrystallized grains of fine size (Figure 8d,f).
Moreover, the grain boundary characteristics and average grain size for TMP samples are summarized in Table 4. For samples processed with a strain of 0.30, the martensitic transformation was sufficient to demolish the initial austenitic microstructure and refine the grain size to 1.30 and 1.74 mm for the austenitic and α′-martensitic phases, respectively. The proportion of HABs in the austenitic phase was deficient (1.38%), and most CSL boundaries were destroyed. In addition, the fragmentation of α′-martensite laths resulted in a moderate formation of HABs (33%). HABs in the austenitic phase were partially restored to a moderate degree (~55%) by reversion annealing, while HABs in the martensitic phase were reduced to about 17.5%. The average grain size of the austenitic and α′-martensitic phases increased slightly after annealing to about 2.0 and 1.9 mm, respectively. In addition, using the 0.20 strain scheme increased HABs formation for the austenitic phase (~65%) during cycle 2. Using a strain of 0.20 enhanced grain refinement, resulting in an average grain size of roughly 1.8 and 1.6 mm for the austenitic and α′-martensitic phases, respectively. In comparison, applying a strain of 0.15 led to a higher formation of HABs in the austenitic phase (63%) for samples processed in cycle 3. The use of three cycles of grain refinement via the 0.15 strain yielded an average grain size of 1.9 and 1.7 mm for the austenitic and α′-martensitic phases, respectively.
Figure 9 plots the grain size distribution for the TMP samples considering grain boundaries possessing a misorientation of 5° and above. For samples processed by a strain of 0.15 (Figure 9a), an increasing number of cycles resulted in progressive grain refinement, as demonstrated by the increase in frequency for grains with grain size in the range of 1–3 µm at the expense of coarse grains (≥3 µm). An essential factor for the increase in grain refinement is grain fragmentation of the coarse retained austenite via recrystallization. Another possible rationale is the coarsening of the reverted grains during annealing. During the initial stages of the reversion process, most of the reverted grains are expected to have a fine size in the submicron range [8]. With increasing annealing time, recovery and recrystallization occur for reverted grains initially formed via shear reversion (and grain growth for grains nucleated via diffusional reversion). Yet, the size of the reverted grains remains fine, typically smaller than the size of the martensitic block before reversion [8]. Moreover, the grain size distribution indicates a slight decrease in the frequency of submicron grains with the increasing number of cycles. This suggests further coarsening of reverted austenite grains and an additional increase in the frequency of grains in the range of 1–3 µm. Moreover, a comparison of the grain refinement among the three schemes is presented in Figure 9b. The 0.15 strain scheme was shown to result in the highest frequency for grain size in the 1–2 µm range compared to samples processed by a strain of 0.20 and 0.30. The increase in cycles also provided more substantial grain refinement, as less coarse grains (≥3 µm) were shown to form for samples processed by a strain of 0.15 with three cycles.
Increasing TMP cycles enhanced the formation of α′-martensite during tensile deformation, implying that austenite stability against strain-induced martensitic transformation was weakened. The progressive increase in α′-martensite formation was also correlated with the enhancement of microstructure refinement with increased cycles during subsequent annealing. Therefore, it is suggested that the weakening of austenite stability is highly connected to the grain refinement of the austenite phase. Some studies reported that the refinement of the austenite grain size causes a drop in austenite stability, which leads to an increase in the transformation rate to α′-martensite with deformation [38,39,40]. For example, Sinclair et al. [40] showed that reducing austenite grain size below 0.9 µm reversibly increased the α′-martensite transformation and strain hardening. On the other hand, Lee et al. [32] indicated that the austenite stability against the strain-induced martensitic transformation in austenitic stainless steels was improved with increasing TMP cycles owing to grain refinement. Other investigations have also found that when the grain size decreases, the austenite stability increases, implying that grain refinement lowers the transformation temperature required for martensitic transformation [41,42,43,44]. Since no quantitative relationship between the mechanisms of α′-martensite formation and grain size has been found, such contradictory behaviors have remained unsolved. However, the current findings imply that not only the size of austenitic grains impacts the α′-martensite transformation during tensile deformation but also the characteristics of the austenitic structure resulting from subsequent annealing by either α′→γ reversion or retained from the deformed austenitic structure. The development of a mixture of the austenitic structure of reverted ultra-fine grains and retained coarse grains is believed to play a significant role in lowering austenite stability rather than only developing a fully reverted fine austenitic structure. This was previously reported by Kisko et al. [45], who studied strain-induced martensitic strain transformation in 204 Cu austenitic stainless steel and indicated that the formation of α′-martensite tends to occur more frequently in a mixture of the austenitic structure of reverted ultra-fine grains and retained coarse grains rather than in a sample with a fully reverted structure of austenitic grains. It was suggested that dislocations in the retained austenite assist further martensite formation upon tensile straining, while refining the austenite size in the range of 0.5–1.5 µm induces α′-martensite transformation at grain boundaries and mechanical twins rather than at shear band intersections. Such a mixed-type structure was also found to enhance the strain hardening rate upon subsequent deformation. Moreover, Spencer [46] reported that pre-straining at room temperature promoted α′-martensite transformation during subsequent tensile deformation of 316 L austenitic stainless steel at −196 °C, suggesting that the rate of α′-martensite transformation is sensitive to the microstructure features, such as dislocations. The current findings also indicate that producing fine austenitic grains via reversion annealing becomes more effective when using repetitive TMP schemes. Compared to the one-cycle TMP scheme, repetitive TMP schemes with moderate strains of 0.15 and 0.20 not only maintained the development of bimodal grain distribution but also resulted in stronger enhancement in austenite reversion and grain refinement upon annealing.
Moreover, annealing the TMP samples was shown to result in the formation of a shear-banded structure within the coarse pancaked austenitic grains, as observed in particular for the 0.15 scheme in Figure 8d,f, which implies that the reversion from martensite to austenite is a shear-type reversion. Somani et al. [47] examined the reversion mechanisms in various metastable austenitic stainless steels and indicated that the shear-type reversion for 301 stainless steel occurred at short times (1–100 s) in the range of 650–900 °C, where the shear-type reversion was found to be characterized by a pancake-banded structure. At longer annealing times, fine austenite grains were formed by static recrystallization. The findings in the work of Somani et al. [47] agree, to a large extent, with those obtained in the current study by subjecting the 301 stainless steel samples to repetitive TMP, which support the occurrence of reversion via a martensitic shear. According to Tomimura et al. [8], the newly reverted austenite block formed via a martensitic shear mechanism, containing a high density of dislocations. The microstructure of reverted austenite is similar to that of the ά-martensite lath. During the early stage of annealing, the boundaries of newly reverted austenitic laths rapidly disappear as they are bounded by small-angle boundaries to form a relatively coarse austenite block bounded by large-angle boundaries. Dislocations within the austenite block form a cellular structure that changes to subgrains during recovery. With increasing annealing time, a coalescing of the subgrains occurs due to recovery and recrystallization to form ultra-fine grains within the austenite block. Misra et al. [18] also reported that phase reversion could occur upon annealing by either nucleation and growth of fine austenite grains at martensite lath boundaries or by forming reverted intra-lath of austenite layers as thin plates. However, one cannot eliminate the possibility of a diffusion-type reversion taking place, since the heating rate during annealing is 22 °C/min, which is considered not high enough to prevent reversion by diffusion. This was reported by Kisko et al. [24], who demonstrated that it is possible for both shear and diffusional reversion mechanisms to occur during the reversion annealing of different austenitic stainless steels.
Concerning the development of HAB/CSL boundaries, the difference between the two repetitive schemes, i.e., using strains of 0.15 and 0.20, is minimal. Nevertheless, the TMP scheme utilizing the strain of 0.15 contributed to a slight improvement in austenite reversion after three cycles (99.2%) compared to that obtained in the sample processed by a strain of 0.20 using two cycles (98%). Additionally, the formation of a few recrystallized austenitic grains with CSL boundaries suggests the occurrence of partial recrystallization, which might originate at the shear bands of the reverted austenite and/or in the retained (original) austenitic grains. Table 4 shows that samples deformed by a strain of 0.30 developed fewer CSL boundaries. Using a relatively large strain (0.30) can generate higher strain energy during tensile deformation and, upon annealing, favor the development of grain boundaries by conventional recrystallization over strain-induced boundary migration (SIBM). The latter mechanism of grain boundary development is known to occur more frequently when annealing fcc metals of low to medium stacking fault energy that have been deformed to small strains, resulting in further development of Σ3 boundaries and their alternatives [48].
Furthermore, the microstructure examination demonstrates that reversion annealing of samples used in repetitive TMP produced a mixture of fine and coarse austenitic grains in austenitic stainless steels [14,16,18,19]. According to Johannsen et al. [14], annealing deformed stainless steels resulted in fast reversion of the dislocation-cell-type martensite, whereas the lath-type martensite was retained due to its slower reversion. The fast reversion of the dislocation-cell-type martensite was attributed to the heavy deformation, which led to an increase in the number of nucleation sites for austenite via the increase in dislocation density and intrusion of slip bands. Due to the difference in reversion rate, it is possible to have a wide variation in grain size, particularly at higher annealing temperatures, where recrystallization of retained austenite is greatly possible. This suggests the necessity of forming dislocation-cell-type martensite during tensile deformation of samples in the current TMP schemes, not only for grain refinement but also for the production of a bimodal grain distribution.

3.3. Mechanical Properties

Figure 10 shows the true stress vs. true strain and the strain hardening rate (dσ/dε) vs. true stress for TMP samples and a sample deformed at room temperature by a tensile strain of 0.30. For the room temperature tensile test, the plastic deformation is mainly governed by the slip, resulting in an increase in dislocation density and thus strengthening by conventional strain hardening, as shown in Figure 10a,b. A small amount of the hard phase of α′-martensite was formed during the deformation, and it had a limited effect on enhancing the strain hardening rate at true stress values greater than 800 MPa (Figure 10b). On the other hand, tensile deformation at −50 °C promoted a large formation of the α′-hard martensite, which resulted in a change in the shape of the stress–strain curve from parabolic, typically exhibited at room temperature tensile deformation, to sigmoidal “S-shaped” for the TMP samples shown in Figure 10a. The sigmoidal behavior was strongly related to the development of α′-martensite, as it followed a sigmoidal function of tensile strain for austenitic stainless steels [49]. In addition, the strain hardening plots shown in Figure 10b indicate three distinct stages of strain hardening. However, it should be noted that the initiation of each hardening stage depends on the amount of strain and the number of cycles applied to the TMP samples.
Table 5 summarizes the yield strength (YS) of the TMP samples. Compared to the sample tested at room temperature, which has YS = 355 MPa, the yield strength for TMP samples in cycle 1 is approximately 455 MPa. In cycle 2, the sample of the 0.20 strain scheme has an increase in YS to 500 MPa, while an insignificant change occurs in the YS for the sample processed by a strain of 0.15 (445 MPa). This might be attributed to the greater refinement of the reverted austenitic grains and higher dislocation density in the retained austenitic structure for the sample deformed by a strain of 0.20 in cycle 1. Processing by three cycles of the 0.15 strain scheme resulted in the highest increase in YS (515 MPa). Such an increase was highly related to the former cycle (cycle 2), which produced a near complete reverted structure of a mixture of fine equiaxed austenitic grains and shear-banded pancaked coarse grains.
Generally, the stress–strain curves of TMP samples show that plastic deformation is initially dominated by a slip of the austenitic phase during stage I hardening, as demonstrated by the low strain hardening (Figure 10b). Nevertheless, the formation of α′-martensite might occur in this stage to accommodate localized strain and thus contribute to the low strain hardening [50]. With increasing tensile deformation, a remarkable increase in the strain hardening occurs, as marked by stage II, which is strongly related to the extensive α′-martensite transformation occurring from austenite and/or ε-martensite [51]. The α′-martensitic formation is thermomechanically much more stable than the ε-phase, and thus, its formation becomes more dominant at higher levels of plastic deformation [47,52,53]. One can also observe that the rate of strain hardening in stage II becomes higher as the number of TMP cycles increases, implying a higher amount of α′-martensite formed for samples processed by three cycles of the 0.15 strain as compared to samples processed by one or two cycles using a strain of 0.30 and 0.20, respectively. However, the remarkable increase in the strain hardening rate can ease the occurrence of intermittent plastic instability in stage II. Consequently, the end of stage II is marked by a decline in the strain hardening rate, most likely attributed to the limitation in the formation of α′-martensite and/or the strain hardening of existing α′-martensite [54]. Therefore, it is possible for intermittent plastic instability to occur at high stress values (>800 MPa), as demonstrated in Figure 10b.
Interestingly, further deformation resulted in a sharp increase in strain hardening (stage III), which occurred at a stress of 960 MPa for a sample processed by a strain of 0.30. It can also be observed that with the increasing number of TMP cycles, the sharp increase in hardening in stage III is initiated at lower stresses: 900 MPa for the sample processed by two cycles using the 0.20 strain and 825 MPa for the sample treated by three cycles using the 0.15 strain. The strain hardening rate in stage III gradually declines with a further increase in the tensile deformation, being steepest for the sample processed by three cycles of the 0.15 strain, indicating stronger plastic instability. It should also be noted that further transformation of α′-martensite at a high stress level is possible, but it most likely involves different transformation mechanisms. Misra et al. [16] indicated that additional transformation of α′-martensite in 301 stainless steel occurred with increasing tensile strain due to the break-up of the α′-martensite lath structure into finer laths and ultimately into dislocation-cell-type martensite. Hedström et al. [55] also reported similar transformation over large strains due to the rapid growth of α′-martensite embryos by autocatalytic transformation. Consequently, it is assumed that breakage of the α′-martensite laths into very fine laths resulted in the initial increase in stage III hardening, while the latter formation and deformation of dislocation-cell-type martensite led to a decrease in strain hardening in stage III, presumably related to the saturation in dislocation density in the cellular martensitic structure.
In addition, the tensile behavior of TMP samples shows that stage II hardening is initiated at a smaller strain by increasing the number of cycles. This was also accompanied by an increase in α′-martensite formation with increasing TMP cycles, as demonstrated by the increase in the volume fraction of α′-martensite in Table 2. As a result, a greater increase in the hardening rate of stage II occurred for samples deformed by a strain of 0.15 and 0.20. As discussed in Section 3.2, the microstructure development via reversion annealing is believed to significantly weaken austenite stability and thus enhance α′-martensite transformation during subsequent deformation. A recent study on TMP of 304 stainless steel by Al-Fadhalah and Aleem [37] reported sigmoidal tensile stress–strain behavior and strain hardening characterized by three stages, similar to that observed in the current study. An enhancement in strain-induced α′-martensitic transformation and austenitic grain refinement upon subsequent annealing was indicated to occur with the increase in the number of cycles. The application of repetitive TMP on 304 stainless steel was also found to result in a mixture of fine and coarse austenitic grains produced by shear-type reversion and recrystallization.
Moreover, Figure 11 compares the microhardness development of TMP samples. In cycle 1, the microhardness reaches a maximum (545 Hv) for samples deformed with a strain of 0.30, whereas the microhardness of samples deformed with a strain of 0.20 and 0.15 is 498 Hv and 464 Hv, respectively. Upon annealing in cycle 1, samples deformed by a strain of 0.15 and 0.20 have hardness reductions of 266 and 270 Hv, respectively. However, annealing a sample subjected to tensile deformation by a strain of 0.30 resulted in a significant decline to roughly 236 Hv. The drop in hardness is most likely related to the higher amount of stored energy available for reversion and recrystallization in a sample processed by a strain of 0.30. Furthermore, a drop in hardness from 498 Hv in cycle 1 to 466 Hv in cycle 2 occurred for deformed samples using a strain of 0.20. Such an unexpected drop was also accompanied by a drop in the strain hardening rate in stage III (Figure 10b). This implies that the change in the formation mechanisms of α′-martensite in stage III, from a lath-type into a dislocation-cell-type structure, hindered the accommodation of dislocations and consequently resulted in a lower hardness increase in cycle 2 compared to cycle 1. As reported by Ma [56], ultra-fine grains in metals generally cannot accommodate a high density of dislocations during deformation, which significantly reduces the strain hardening ability. In addition, the microhardness values for samples deformed with a strain of 0.20 and 0.15 are comparable in cycle 2 (466 Hv and 472 Hv), suggesting a similar martensitic transformation. Table 2 indicates the formation of a martensitic phase with a volume fraction of 95.4% and 99.7% for samples deformed by strains of 0.15 and 0.20, respectively, which is consistent with the observed hardness development in cycle 2. Subsequent annealing in cycle 2 resulted in a comparable decline in hardness values (261 Hv for the 0.15 strain and 256 Hv for the 0.20 strain). On the other hand, deformation by a strain of 0.15 in cycle 3 reduced hardness to 452 Hv, regardless of the extensive formation of martensite (98.4%). Following the same pattern of sample deformation by a strain of 0.2, a drop in strain hardening in stage III is recorded for the deformed sample using a strain of 0.15 in cycle 3 (Figure 10b), supporting the possibility of a change in the α′-martensite structure from a lath type into a dislocation cell type. Subsequent annealing in cycle 3 resulted in a drop in hardness to 263 Hv, comparable to that recorded in cycle 2.
The martensitic transformation and reversion processes during TMP cycles are generally consistent with the recorded microhardness development. It should be emphasized that stage II hardening, characterized by the massive formation of the lath-type martensitic structure, is primarily responsible for the increase in hardness. On the other hand, stage III hardening was shown to cause a reduction in the strain hardening rate due to a change in the structure of α′-martensite from lath type into dislocation cell type. Consequently, it eased the dislocation motion, resulting in a drop in the hardness of deformed samples with the increase in the number of cycles.

4. Conclusions

Grain refinement was achieved using repetitive thermomechanical processing (TMP) in 301 austenitic stainless steel. The TMP consisted of tensile deformation at −50 °C for enhancing α′-strain-induced martensitic transformation, followed by reversion annealing at 850 °C for 5 min. The martensitic transformation and subsequent austenitic reversion were evaluated using three TMP schemes: one cycle with a strain of 0.30, two cycles with a strain of 0.20, and three cycles with a strain of 0.15. The following concluding remarks can be put forward:
(i)
After three TMP cycles, the 0.15 strain scheme yielded the highest increase in the martensitic transformation to about 99 vol.%., resulting in an entirely austenitic structure upon annealing, with an average grain size of 1.9 µm and a HABs proportion of 64 vol.%.
(ii)
Annealing produced a bimodal grain distribution of the austenitic phase for the three schemes, mainly consisting of fine reverted grains (from the α′-martensite structure) and coarse recrystallized grains (from the retained austenite structure). The bimodal grain distribution was less present in samples of the 0.30 strain scheme and most pronounced for samples processed by three cycles of the 0.15 strain scheme.
(iii)
During subzero deformation, the tensile behavior is generally characterized by initial strain hardening by the slip (stage I), followed by a remarkable increase in the strain hardening rate due to the α′-strain-induced martensitic transformation (stage II). The α′-martensite banded structure is broken by further deformation to high values of strain, resulting in dislocation-cell-type martensite. The change in the mechanism of α′-martensite formation/deformation is marked by stage III hardening, characterized by a decrease in the strain hardening rate. Stage III hardening is also found to significantly lower the hardness enhancement of deformed samples subjected to multiple cycles.
(iv)
The optimal way to achieve reversion annealing is to induce martensitic phase transformation during stage II hardening via a small to medium strain of multi-cycle TMP rather than large deformation in stage III hardening using single-cycle TMP. With the increase in the number of cycles, it is possible to have a massive formation of martensite during stage II hardening and subsequently intense grain refinement upon reversion annealing.

Author Contributions

Conceptualization, K.J.A.-F.; methodology, K.J.A.-F. and Y.A.-A.; analysis and experimental investigation, K.J.A.-F., Y.A.-A. and M.A.R.; resources, K.J.A.-F. and M.A.R.; writing—original draft preparation, K.J.A.-F. and Y.A.-A.; writing—review and editing, K.J.A.-F.; supervision and project administration, K.J.A.-F. All authors have read and agreed to the published version of the manuscript.

Funding

The authors would like to thank the College of Graduate Studies at Kuwait University for supporting this research.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, K.J.A.-F., upon reasonable request.

Acknowledgments

The authors would like to acknowledge the support provided by Kuwait University General Facility projects (grant number GE 01/07) for sample preparation and SEM-EBSD measurements.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Shen, J.; Gonçalves, R.; Choi, Y.T.; Lopes, J.G.; Yang, J.; Schell, N.; Kim, H.S.; Oliveira, J.P. Microstructure and mechanical properties of gas metal arc welded CoCrFeMnNi joints using a 308 stainless steel filler metal. Scr. Mater. 2022, 222, 115053. [Google Scholar] [CrossRef]
  2. Oliveira, J.P.; Shamsolhodaei, A.; Shen, J.; Lopes, J.G.; Gonçalves, R.M.; de Brito Ferraz, M.; Piçarra, L.; Zeng, Z.; Schell, N.; Zhou, N.; et al. Improving the ductility in laser welded joints of CoCrFeMnNi high entropy alloy to 316 stainless steel. Mater. Des. 2022, 219, 110717. [Google Scholar] [CrossRef]
  3. Rodrigues, T.A.; Escobar, J.D.; Shen, J.; Duarte, V.R.; Ribamar, G.G.; Avila, J.A.; Maawad, E.; Schell, N.; Santos, T.G.; Oliveira, J.P. Effect of heat treatments on 316 stainless steel parts fabricated by wire and arc additive manufacturing: Microstructure and synchrotron X-ray diffraction analysis. Addit. Manuf. B 2021, 48, 102428. [Google Scholar] [CrossRef]
  4. Maki, T. Stainless steel: Progress in thermomechanical treatment. Curr. Opin. Solid State Mater. Sci. 1997, 2, 290–295. [Google Scholar] [CrossRef]
  5. Raj, A.K.; Kumar, J.S.; Padmanabhan, K.A. Martensitic transformation and plastic flow in metastable 2.5 wt.% nickel austenitic stainless steel sheets. Mater. Lett. 1999, 38, 386–390. [Google Scholar]
  6. Padilha, A.F.; Plaut, R.L.; Rios, P.R. Annealing of cold-worked austenitic stainless steels. ISIJ Int. 2003, 43, 135–143. [Google Scholar] [CrossRef]
  7. Takaki, S.; Tanimoto, S.K.; Tomimura, K.; Tokunaga, Y. Strengthening of Metastable 16-10 Austenitic Stainless Steel by Ultra Grain Refining. Iron Steel Inst. Jpn. 1988, 74, 1058–1064. [Google Scholar] [CrossRef] [Green Version]
  8. Tomimura, K.; Takaki, S.; Tokunaga, Y. Reversion Mechanism from Deformation Induced Martensite to Austenite in Metastable Austenitic Stainless Steels. ISIJ Int. 1991, 31, 1431–1437. [Google Scholar] [CrossRef] [Green Version]
  9. Mirzadeh, H.; Najafizadeh, A. Modeling the reversion of martensite in the cold worked AISI 304 stainless steel by artificial neural networks. Mater. Des. 2009, 30, 570–573. [Google Scholar] [CrossRef]
  10. Di Schino, A.; Salvatori, I.; Kenny, J.M. Effects of martensite formation and austenite reversion on grain refining of AISI 304 stainless steel. J. Mater. Sci. 2002, 37, 4561–4565. [Google Scholar] [CrossRef]
  11. Di Schino, A.; Barteri, M.; Kenny, J.M. Development of ultra fine grain structure by martensitic reversion in stainless steel. J. Mater. Sci. Lett. 2002, 21, 751–753. [Google Scholar] [CrossRef]
  12. Tao, K.; Choo, H.; Li, H.; Clausen, B.; Jin, J.E.; Lee, Y.K. Transformation-induced plasticity in an ultrafine-grained steel: An in situ neutron diffraction study. Appl. Phys. Lett. 2007, 90, 101911. [Google Scholar] [CrossRef] [Green Version]
  13. Jung, Y.S.; Lee, Y.K. Effect of pre-deformation on the tensile properties of a metastable austenitic steel. Scr. Mater. 2008, 59, 47–50. [Google Scholar] [CrossRef]
  14. Johannsen, D.L.; Kyröläinen, A.; Ferreira, P.J. Influence of annealing treatment on the formation of nano/submicron grain size AISI 301 Austenitic stainless steels. Metall. Trans. A 2006, 37, 2325–2338. [Google Scholar] [CrossRef]
  15. Rajasekhara, S.; Ferreira, P.J.; Karjalainen, L.P.; Kyröläinen, A. Hall—Petch Behavior in Ultra-Fine-Grained AISI 301LN Stainless Steel. Metall. Trans. A 2007, 38, 1202–1210. [Google Scholar] [CrossRef]
  16. Misra, R.D.K.; Kumar, B.R.; Somani, M.C.; Karjalainen, L.P. Deformation processes during tensile straining of ultrafine/nanograined structures formed by reversion in metastable austenitic steels. Scr. Mater. 2008, 59, 79–82. [Google Scholar] [CrossRef]
  17. Eskandari, M.; Najafizadeh, A.; Kermanpur, A. Effect of strain-induced martensite on the formation of nanocrystalline 316L stainless steel after cold rolling and annealing. Mat. Sci. Eng. A 2009, 519, 46–50. [Google Scholar] [CrossRef]
  18. Misra, R.D.K.; Nayak, S.; Mali, S.; Shah, J.S.; Somani, M.C.; Karjalainen, L.P. Microstructure and Deformation Behavior of Phase-Reversion-Induced Nanograined/Ultrafine-Grained Austenitic Stainless Steel. Metall. Trans. A 2009, 40, 2498–2509. [Google Scholar] [CrossRef]
  19. Misra, R.D.K.; Nayak, S.; Mali, S.; Shah, J.S.; Somani, M.C.; Karjalainen, L.P. On the Significance of Nature of Strain-Induced Martensite on Phase-Reversion-Induced Nanograined/Ultrafine-Grained Austenitic Stainless Steel. Metall. Trans. A 2010, 41, 3–12. [Google Scholar] [CrossRef]
  20. Kumar, B.R.; Sharma, S. Recrystallization Behavior of a Heavily Deformed Austenitic Stainless Steel During Iterative Type Annealing. Metall. Mater. Trans. A 2014, 45, 6027–6038. [Google Scholar] [CrossRef]
  21. Sharifian, K.; Mirzadeh, H.; Kheiri, S.; Naghizadeh, M. Two-step annealing treatment for grain refinement of cold-worked AISI 316L stainless steel. Int. J. Mater. Res. 2020, 111, 676–680. [Google Scholar]
  22. Naghizadeh, M.; Mirzadeh, H. Microstructural Evolutions During Annealing of Plastically Deformed AISI 304 Austenitic Stainless Steel: Martensite Reversion, Grain Refinement, Recrystallization, and Grain Growth. Metall. Mater. Trans. A 2016, 47, 4210–4216. [Google Scholar] [CrossRef]
  23. Kheiri, S.; Mirzadeh, H.; Naghizadeh, M. Tailoring the microstructure and mechanical properties of AISI 316L austenitic stainless steel via cold rolling and reversion annealing. Mat. Sci. Eng. A 2019, 759, 90–96. [Google Scholar] [CrossRef]
  24. Kisko, A.; Hamada, A.S.; Talonen, J.; Porter, D.; Karjalainen, L.P. Effects of reversion and recrystallization on microstructure and mechanical properties of Nb-alloyed low-Ni high-Mn austenitic stainless steels. Mater. Sci. Eng. A 2016, 657, 359–370. [Google Scholar] [CrossRef]
  25. Xu, D.M.; Li, G.Q.; Wan, X.L.; Misra, R.D.K.; Zhang, X.G.; Xu, G.; Wu, K.M. The effect of annealing on the microstructural evolution and mechanical properties in phase reversed 316LN austenitic stainless steel. Mater. Sci. Eng. A 2018, 720, 36–48. [Google Scholar] [CrossRef]
  26. Niu, G.; Wu, H. Microstructural evolution and mechanical behavior of phase reversion-induced bimodal austenitic steels. Mater. Sci. Eng. A 2020, 772, 138669. [Google Scholar] [CrossRef]
  27. Kumar, B.R.; Gujral, A. Plastic Deformation Modes in Mono- and Bimodal-Type Ultrafine-Grained Austenitic Stainless Steel. Metallogr. Microstruct. Anal. 2014, 3, 397–407. [Google Scholar] [CrossRef]
  28. Sabooni, S.; Karimzadeh, F.; Enayati, M.H.; Ngan, A.H.W. The role of martensitic transformation on bimodal grain structure in ultrafine grained AISI 304L stainless steel. Mater. Sci. Eng. A 2015, 636, 221–230. [Google Scholar] [CrossRef] [Green Version]
  29. Jia, Q.; Chen, L.; Xing, Z.; Wang, H.; Jin, M.; Chen, X.; Choi, H.; Han, H.N. Tailoring hetero-grained austenite via a cyclic thermomechanical process for achieving ultrahigh strength-ductility in medium-Mn steel. Scr. Mater. 2022, 217, 114767. [Google Scholar] [CrossRef]
  30. Naghizadeh, M.; Mirzadeh, H. Processing of fine grained AISI 304L austenitic stainless steel by cold rolling and high-temperature short-term annealing. Mater. Res. Express 2018, 5, 056529. [Google Scholar] [CrossRef]
  31. Järvenpää, A.; Ghosh, S.; Khosravifard, A.; Jaskari, M.; Hamada, A. A new processing route to develop nano-grained structure of a TRIP-aided austenitic stainless-steel using double reversion fast-heating annealing. Mater. Sci. Eng. A 2021, 808, 140917. [Google Scholar] [CrossRef]
  32. Lee, C.Y.; Yoo, C.S.; Kermanpur, A.; Lee, Y.K. The effects of multi-cyclic thermo-mechanical treatment on the grain refinement and tensile properties of a metastable austenitic steel. J. Alloy. Compd. 2014, 583, 357–360. [Google Scholar] [CrossRef]
  33. Moallemi, M.; Kermanpur, A.; Najafizadeh, A.; Rezaee, A.; Samaei baghbadorani, H. Formation of nano/ultrafine grain structure in a 201 stainless steel through the repetitive martensite thermomechanical treatment. Mater. Lett. 2012, 89, 22–24. [Google Scholar] [CrossRef]
  34. Sun, G.S.; Du, L.X.; Hu, J.; Xie, H.; Wu, H.Y.; Misra, R.D.K. Ultrahigh strength nano/ultrafine-grained 304 stainless steel through three-stage cold rolling and annealing treatment. Mater. Charact. 2015, 110, 228–235. [Google Scholar] [CrossRef]
  35. He, Y.M.; Wang, Y.H.; Guo, K.; Wang, T.S. Effect of carbide precipitation on strain-hardening behavior and deformation mechanism of metastable austenitic stainless steel after repetitive cold rolling and reversion annealing. Mat. Sci. Eng. A 2017, 708, 248–253. [Google Scholar] [CrossRef]
  36. Sun, G.S.; Du, L.X.; Hu, J.; Misra, R.D.K. Microstructural evolution and recrystallization behavior of cold rolled austenitic stainless steel with dual phase microstructure during isothermal annealing. Mater. Sci. Eng. A 2018, 709, 254–264. [Google Scholar] [CrossRef]
  37. Al-Fadhalah, K.; Aleem, M. Microstructure Refinement and Mechanical Properties of 304 Stainless Steel by Repetitive Thermomechanical Processing. Metall. Mater. Trans. A 2018, 49, 1121–1139. [Google Scholar] [CrossRef]
  38. Shrinivas, V.; Varma, S.K.; Murr, L.E. Deformation-induced martensitic characteristics in 304 and 316 stainless steels during room-temperature rolling. Metall. Mater. Trans. A 1995, 26, 661–671. [Google Scholar] [CrossRef]
  39. De, A.; Speer, J.; Matlock, D.; Murdock, D.; Mataya, M.; Comstock, R. Deformation-induced phase transformation and strain hardening in type 304 austenitic stainless steel. Metall. Trans. A 2006, 37, 1875–1886. [Google Scholar] [CrossRef]
  40. Sinclair, C.W.; Maréchal, D.; Mithieux, J.D. The effect of grain size on the mechanical response of a metastable austenitic stainless steel. In MATEC Web of Conferences; EDP Sciences: Les Ulis, France, 2013; Volume 7, p. 01010. [Google Scholar]
  41. Umemoto, M.; Owen, W.S. Effects of austenitizing temperature and austenite grain size on the formation of athermal martensite in an iron-nickel and an iron-nickel-carbon alloy. Metall. Mater. Trans. B 1974, 5, 2041–2046. [Google Scholar] [CrossRef]
  42. Jun, J.H.; Choi, C.S. Variation of stacking fault energy with austenite grain size and its effect on the MS temperature of γ→ε martensitic transformation in Fe–Mn alloy. Mater. Sci. Eng. A 1998, 257, 353–356. [Google Scholar] [CrossRef]
  43. Maki, T.; Tomota, Y.; Tamura, I. Effect of Grain Size on the Transformation-Induced Plasticity in Metastable Austenitic Fe-Ni-C Alloy. J. Jpn. Inst. Met. 1974, 38, 871–876. [Google Scholar] [CrossRef]
  44. Jimenez-Melero, E.; van Dijk, N.; Zhao, L.; Sietsma, J.; Offerman, S.; Wright, J.; van der Zwaag, S. Martensitic transformation of individual grains in low-alloyed TRIP steels. Scr. Mater. 2007, 56, 421–424. [Google Scholar] [CrossRef]
  45. Kisko, A.; Misra, R.D.K.; Talonen, J.; Karjalainen, L.P. The influence of grain size on the strain-induced martensite formation in tensile straining of an austenitic 15Cr–9Mn–Ni–Cu stainless steel. Mater. Sci. Eng. A 2013, 578, 408–416. [Google Scholar] [CrossRef]
  46. Spencer, K. The Work-Hardening of Austenitic Stainless Steel, Applied to the Fabrication of High-Strength Conductors. Ph.D. Thesis, McMaster University, Hamilton, ON, Canada, 2004. [Google Scholar]
  47. Somani, M.C.; Juntunen, P.; Karjalainen, L.P.; Misra, R.D.K. Enhanced Mechanical Properties through Reversion in Metastable Austenitic Stainless Steels. Metall. Mater. Trans. A 2009, 40, 729–744. [Google Scholar] [CrossRef]
  48. Bellier, S.P.; Doherty, R.D. The structure of deformed aluminium and its recrystallization—investigations with transmission Kossel diffraction. Acta Metall. 1977, 25, 521–538. [Google Scholar] [CrossRef]
  49. Hecker, S.S.; Stout, M.G.; Staudhammer, K.P.; Smith, J.L. Effects of strain state and strain rate on deformation-induced transformation in 304 stainless steel: Part I. Magnetic Measurements and Mechanical Behavior. Metall. Mater. Trans. A 1982, 13, 619–626. [Google Scholar] [CrossRef]
  50. Talonen, J.; Hänninen, H.; Nenonen, P.; Pape, G. Effect of strain rate on the strain-induced γ→α′-martensite transformation and mechanical properties of austenitic stainless steels. Metall. Mater. Trans. A 2005, 36, 421–432. [Google Scholar] [CrossRef]
  51. Petit, B.; Gey, N.; Cherkaoui, M.; Bolle, B.; Humbert, M. Deformation behavior and microstructure/texture evolution of an annealed 304 AISI stainless steel sheet. Experimental and micromechanical modeling. Int. J. Plast. 2007, 23, 323–341. [Google Scholar] [CrossRef]
  52. Rouseau, D.; Blanc, G.; Tricot, R.; Gueussier, A. Structure Stability under Deformation and at Low Temperatures for Austenitic Stainless Steels with Cr-Ni. Mem. Sci. Rev. Metall. 1970, 67, 315–334. [Google Scholar]
  53. Tavares, S.M.; Gunderov, D.; Stolyarov, V.; Neto, J.M. Phase transformation induced by severe plastic deformation in the AISI 304L stainless steel. Mater. Sci. Eng. A 2003, 358, 32–36. [Google Scholar] [CrossRef]
  54. Maréchal, D. Linkage Between Mechanical Properties and Phase Transformations in a 301LN Austenitic Stainless Steel. Ph.D. Thesis, The University of British Columbia, Vancouver, BC, Canada, 2011. [Google Scholar]
  55. Hedström, P.; Lienert, U.; Almerb, J.; Odén, M. Stepwise transformation behavior of the strain-induced martensitic transformation in a metastable stainless steel. Scr. Mater. 2007, 56, 213–216. [Google Scholar] [CrossRef]
  56. Ma, E. Instabilities and ductility of nanocrystalline and ultrafine-grained metals. Scr. Mater. 2003, 49, 663–668. [Google Scholar] [CrossRef]
Metals 12 01690 g001 550
Figure 1. Phase maps for TMP samples processed using one cycle and strain of 0.30: (a) Deformed structure and (b) annealed structure. Austenite is blue, α′-martensite is red, and ε-martensite is yellow.
Figure 1. Phase maps for TMP samples processed using one cycle and strain of 0.30: (a) Deformed structure and (b) annealed structure. Austenite is blue, α′-martensite is red, and ε-martensite is yellow.
Metals 12 01690 g001
Metals 12 01690 g002 550
Figure 2. Phase maps for TMP samples processed using two cycles and strain of 0.20: (a,b) Cycle 1 and (c,d) Cycle 2. Austenite is blue, α′-martensite is red, and ε-martensite is yellow.
Figure 2. Phase maps for TMP samples processed using two cycles and strain of 0.20: (a,b) Cycle 1 and (c,d) Cycle 2. Austenite is blue, α′-martensite is red, and ε-martensite is yellow.
Metals 12 01690 g002
Metals 12 01690 g003 550
Figure 3. Phase maps for TMP samples processed using three cycles and strain of 0.15: (a,b) Cycle 1, (c,d) Cycle 2, and (e,f) Cycle 3. Austenite is blue, α′-martensite is red, and ε-martensite is yellow.
Figure 3. Phase maps for TMP samples processed using three cycles and strain of 0.15: (a,b) Cycle 1, (c,d) Cycle 2, and (e,f) Cycle 3. Austenite is blue, α′-martensite is red, and ε-martensite is yellow.
Metals 12 01690 g003
Metals 12 01690 g004 550
Figure 4. Microstructure of as-received 301 stainless steel: (a) SEM micrograph and (b) EBSD grain boundary construction map. HABs and LABs are shown in black and gray. Σ3, Σ9, and Σ27 boundaries are denoted in red, yellow, and green, respectively.
Figure 4. Microstructure of as-received 301 stainless steel: (a) SEM micrograph and (b) EBSD grain boundary construction map. HABs and LABs are shown in black and gray. Σ3, Σ9, and Σ27 boundaries are denoted in red, yellow, and green, respectively.
Metals 12 01690 g004
Metals 12 01690 g005 550
Figure 5. Grain boundaries construction maps for TMP samples processed using one cycle and strain of 0.30: (a) Deformed structure and (b) annealed structure.
Figure 5. Grain boundaries construction maps for TMP samples processed using one cycle and strain of 0.30: (a) Deformed structure and (b) annealed structure.
Metals 12 01690 g005
Metals 12 01690 g006 550
Figure 6. Grain boundaries construction maps for TMP samples processed using two cycles and strain of 0.20: (a,b) Cycle 1 and (c,d) Cycle 2.
Figure 6. Grain boundaries construction maps for TMP samples processed using two cycles and strain of 0.20: (a,b) Cycle 1 and (c,d) Cycle 2.
Metals 12 01690 g006
Metals 12 01690 g007 550
Figure 7. Grain boundaries construction maps for TMP samples processed using three cycles and a strain of 0.15: (a,b) Cycle 1, (c,d) Cycle 2, and (e,f) Cycle 3.
Figure 7. Grain boundaries construction maps for TMP samples processed using three cycles and a strain of 0.15: (a,b) Cycle 1, (c,d) Cycle 2, and (e,f) Cycle 3.
Metals 12 01690 g007
Metals 12 01690 g008 550
Figure 8. SEM micrographs of TMP samples processed using three cycles and strain of 0.15: (a,b) Cycle 1, (c,d) Cycle 2, and (e,f) Cycle 3.
Figure 8. SEM micrographs of TMP samples processed using three cycles and strain of 0.15: (a,b) Cycle 1, (c,d) Cycle 2, and (e,f) Cycle 3.
Metals 12 01690 g008
Metals 12 01690 g009 550
Figure 9. Grain size distribution for TMP samples: (a) scheme 3 (ε = 0.15) and (b) Last cycle of schemes 1, 2, and 3 (ε = 0.15, 0.20, 0.30).
Figure 9. Grain size distribution for TMP samples: (a) scheme 3 (ε = 0.15) and (b) Last cycle of schemes 1, 2, and 3 (ε = 0.15, 0.20, 0.30).
Metals 12 01690 g009
Metals 12 01690 g010 550
Figure 10. Tensile behavior of TMP samples: (a) True stress vs. true strain and (b) strain hardening rate (dσ/dε) vs. true stress. Three strain hardening stages are illustrated for the sample processed by a strain of 0.30.
Figure 10. Tensile behavior of TMP samples: (a) True stress vs. true strain and (b) strain hardening rate (dσ/dε) vs. true stress. Three strain hardening stages are illustrated for the sample processed by a strain of 0.30.
Metals 12 01690 g010
Metals 12 01690 g011 550
Figure 11. Microhardness measurements for samples processed using TMP: (a) deformed samples and (b) annealed samples.
Figure 11. Microhardness measurements for samples processed using TMP: (a) deformed samples and (b) annealed samples.
Metals 12 01690 g011
Table
Table 1. Details of thermomechanical processing of 301 stainless steel samples.
Table 1. Details of thermomechanical processing of 301 stainless steel samples.
Scheme No.Cycle Details
Scheme# 1 (Three cycles)Cycle 1 (ε = 0.15 at −50 °C + furnace heating at 850 °C/5 min)
Cycle 2 (ε = 0.15 at −50 °C + furnace heating at 850 °C/5 min)
Cycle 3 (ε = 0.15 at −50 °C + furnace heating at 850 °C/5 min)
Scheme# 2 (Two cycles)Cycle 1 (ε = 0.20 at −50 °C + furnace heating at 850 °C/5 min)
Cycle 2 (ε = 0.20 at −50 °C + furnace heating at 850 °C/5 min)
Scheme# 3 (One cycle)Cycle 1 (ε = 0.30 at −50 °C + furnace heating at 850 °C/5 min)
Table
Table 2. Volume fraction of phases (%) for TMP samples processed.
Table 2. Volume fraction of phases (%) for TMP samples processed.
StrainSample *γ-Austenite (%)α′-Martensite (%)ε-Martensite (%)
ε = 0.30Cycle-1 D2.697.20.227
Cycle-1 A73.726.30.015
ε = 0.20Cycle-1 D20.378.41.33
Cycle-1 A79.620.40.075
Cycle-2 D0.2899.70.015
Cycle-2 A98.01.950.007
ε = 0.15Cycle-1 D46.652.11.28
Cycle-1 A71.228.80.054
Cycle-2 D4.595.40.052
Cycle-2 A97.92.030.111
Cycle-3 D1.698.40.040
Cycle-3 A99.20.710.134
* D: Deformed, A: Annealed.
Table
Table 3. Average grain size and grain boundary statistics for as-received material.
Table 3. Average grain size and grain boundary statistics for as-received material.
SampleΣ3 (%)Σ9 (%)Σ27 (%)CSL (%)HAB (%)Grain Size (μm)
As received48.41.40.450.296.24.96
Table
Table 4. Average grain size and grain boundary statistics for TMP samples.
Table 4. Average grain size and grain boundary statistics for TMP samples.
StrainSampleγ-CSL (%)γ-HAB (%)α′-HAB (%)γ-GS (μm)α′-GS (μm)
ε = 0.30Cycle-1 D0.561.3832.891.301.74
Cycle-1 A22.1255.2917.512.021.87
ε = 0.20Cycle-1 D3.166.7039.281.651.76
Cycle-1 A12.9225.166.662.151.91
Cycle-2 D0.270.2733.811.141.48
Cycle-2 A27.1265.317.151.761.63
ε = 0.15Cycle-1 D5.4210.9439.042.021.68
Cycle-1 A8.4819.988.642.682.37
Cycle-2 D1.452.0534.481.471.68
Cycle-2 A17.0540.456.742.061.81
Cycle-3 D0.500.6838.781.441.52
Cycle-3 A23.9063.392.191.901.74
Table
Table 5. Yield strength of 301 stainless steel samples.
Table 5. Yield strength of 301 stainless steel samples.
Tensile Test ConditionYield Strength (MPa)
Cycle 1Cycle 2Cycle 3
ε = 0.30, T = 25 °C355??
ε = 0.15, T = −50 °C450445515
ε = 0.20, T = −50 °C450500?
ε = 0.30, T = −50 °C450??
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Al-Fadhalah, K.J.; Al-Attal, Y.; Rafeeq, M.A. Microstructure Refinement of 301 Stainless Steel via Thermomechanical Processing. Metals 2022, 12, 1690. https://doi.org/10.3390/met12101690

AMA Style

Al-Fadhalah KJ, Al-Attal Y, Rafeeq MA. Microstructure Refinement of 301 Stainless Steel via Thermomechanical Processing. Metals. 2022; 12(10):1690. https://doi.org/10.3390/met12101690

Chicago/Turabian Style

Al-Fadhalah, Khaled J., Yousif Al-Attal, and Muhammad A. Rafeeq. 2022. "Microstructure Refinement of 301 Stainless Steel via Thermomechanical Processing" Metals 12, no. 10: 1690. https://doi.org/10.3390/met12101690

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop