Dilatometric and Microstructural Study of Martensite Tempering in 4% Mn Steel

materials

Article Dilatometric and Microstructural Study of Martensite Tempering in 4% Mn Steel

Adam Grajcar 1,* , Mateusz Morawiec 1 , Jose Antonio Jimenez 2 and Carlos Garcia-Mateo 2

1 Department of Engineering Materials and Biomaterials, Silesian University of Technology, 44-100 Gliwice, Poland; [email protected] 2 Department of Physical Metallurgy, National Center for Metallurgical Research, 28040 Madrid, Spain; [email protected] (J.A.J.); [email protected] (C.G.-M.) * Correspondence: [email protected]; Tel.: +48-3223-72933

Received: 8 September 2020; Accepted: 2 October 2020; Published: 7 October 2020

Abstract: This paper presents the results of martensite tempering resistance in 4% Mn steel. The material was quenched and tempered at 350 ◦C for 15, 30, and 60 min. The analysis of the quenching and tempering was carried out using dilatometric and microstructural approaches. The phase composition was assessed using X-ray diffraction. The Ms temperature and tempering progress were simulated using JMatPro software. The dilatometric analysis revealed a small decrease in the relative change in length (RCL) during tempering. This decrease was connected to the precipitation kinetics of cementite within the martensite laths. The microstructure investigation using a scanning electron microscope showed a very small amount of carbides, even for the longest tempering time. This showed the high tempering resistance of the martensite in medium-Mn steels. The hardness results showed an insignificant decrease in the hardness depending on the tempering time, which confirmed the high tempering resistance of martensite.

Keywords: quenching; tempering; medium-Mn steel; martensitic transformation; dilatometry; cementite precipitation

1. Introduction Medium manganese steel is a grade of steel that uses Mn as an important alloying element for the stabilization of retained austenite at room temperature. This is because manganese is an austenite stabilizer [1]. This microstructural feature is used in different types of heat treatments, which usually result in ferrite, bainite, or martensite formation (as a matrix), with some fraction of retained austenite. The typical heat treatment for this grade of steel is the intercritical annealing of cold-rolled steel [2,3]. Here, the formation of a mixture of α + γ phases from low-carbon martensite takes place. The formation of the ferrite leads to the diffusion of carbon and manganese (for sufficiently long times) into the austenite. This element increases its thermal stability, which favors higher amounts of the γ phase at room temperature. Another possible heat treatment is austempering for the bainite formation after austenitization [4,5]. The formation of bainite stimulates a higher carbon content of the remaining austenite. Recently, some research has focused on the use of a two- or three-step heat treatment using intercritical annealing or isothermal holding [6–8]. Very complex microstructures can be obtained in this way, leading to better mechanical properties. All of these treatments use the rejection phenomenon of carbon and manganese from the alpha phases. Recently, a very popular heat treatment is quenching and partitioning (QP), where first, some fraction of martensite is formed during partial quenching. Then, the material is heated above the martensite start temperature (Ms), resulting in martensite tempering and carbon partitioning [9]. Depending on the partitioning temperature (from 200 to 500 ◦C), different phases, such as bainite or

Materials 2020, 13, 4442; doi:10.3390/ma13194442 www.mdpi.com/journal/materials Materials 2020, 13, 4442 2 of 10 pearlite, can be formed [10]. The higher the temperature of the isothermal holding after the partial quenching, the larger the grain size of the matrix. However, when the temperature is close to Ms, the diffusion needs more time for carbon partitioning into the austenite. The tempering behavior is well known for conventional low-alloyed or high-alloyed steels [11,12]. However, there are no systematic studies on the tempering resistance in advanced medium-Mn steels. The QP process uses martensite tempering as a source of carbon and manganese (at higher partitioning temperatures) for austenite stabilization. So far, there is not enough knowledge regarding how the tempering time affects the martensite resistance in steels containing above 3% Mn. Therefore, this work was focused on the analysis of martensite tempering near the Ms temperature for different tempering durations.

2. Material and Experiments

2.1. Material The chemistry of the experimental laboratory-produced medium-Mn steel is shown in Table1. The steel belongs to the newly developed third-generation of AHSS (advanced high strength steel) [1,3]. The chemical composition was analyzed using glow discharge optical emission spectrometry (GDOES). This steel was designed for the bainite transformation heat treatment to obtain a microstructure composed of a bainitic matrix with some fraction of retained austenite. The manganese content of 3.6% was added as the γ-phase stabilizer for the retained austenite stabilization [1]. A high Al content was introduced to reduce the negative eﬀect of Si on the sheet wettability by liquid zinc during the hot-dip galvanizing [13]. These elements were used to prevent carbide precipitation during the bainite transformation. The small Nb content was added to limit the prior austenite grain growth during austenitization [14].

Table 1. Chemical composition (wt%) of the analyzed steel.

C Mn Si Al Mo Nb 0.18 3.6 0.23 1.7 0.2 0.04

2.2. Experimental Details The main purpose of this work was to analyze the martensite tempering resistance during a heat treatment near the Ms temperature. The work consisted of two parts, including quenching and tempering of the investigated steel. The dilatometric study of the Ms temperature and the tempering kinetics were carried out using a high-resolution BAHR Dilatometer DIL805A/D (TA Instruments, Wetzlar, Germany) with induction heating of the samples in a vacuum. The critical temperatures were determined based on the ASTM A1033-04 standard [15]. The samples were 4 mm in diameter and 10 mm in length and were machined for the dilatometric analysis. The cooling was conducted using helium. The steel was austenitized at 1100 ◦C for 300 s and cooled down at a rate of 60 ◦C/s to room temperature to obtain a fully martensitic microstructure. Then, the dilatometric samples were tempered at 350 ◦C for 15, 30, and 60 min, as presented in Figure1. After this, the microstructure analysis using a light microscope and scanning electron microscope, as well as X-ray diﬀraction, were performed. For metallographic purposes, the samples were cut in half, ground, and polished using 3 and 1 µm diamond pastes, after which, etching in 2% Nital (the solution of 2% nitric acid with the alcohol) was conducted. The hardness measurements were carried out using the Vickers method with a load of 100 N for 15 s. In this way, the softening behavior of the martensite during tempering was assessed. JMatPro (Version 11.2, Sente Software, Guildford, UK) simulations during quenching and tempering steps were carried out for further analysis of the microstructure changes [16]. The database version 11.2 and the general steel module were used for the simulations. Materials 2020, 13, 4442 3 of 10 Materials 2020, 13, x FOR PEER REVIEW 3 of 11

Figure 1. A schematic of the heat treatment.

X-ray diffraction diffraction (XRD) measurements were carr carriedied out using a Bruker AXS D8 didiffractometerffractometer (Bruker AXS,AXS, GmbH,GmbH, Karlsruhe, Karlsruhe, Germany) Germany) equipped equipped with with a Co a Co X-ray X-ray tube tube working working at 40 at kV 40 andkV and 30 mA, 30 asmA, well as aswell a LynxEye as a LynxEye Linear PositionLinear Position Sensitive Sens Detectoritive (BrukerDetector AXS, (Bruker GmbH, AXS, Karlsruhe, GmbH, Germany),Karlsruhe, whichGermany), allowed which for dataallowed acquisition for data up acquisition to 450 times up faster to 450 than times a conventional faster than pointa conventional detector system. point Wedetector used Cosystem. radiation We used to avoid Co radiation the strong to fluorescent avoid the strong radiation fluorescent that iron radiation emits when that copper iron emits radiation when is used,copper which radiation causes is aused, significant which decreasecauses a insignificant the intensity decrease of the in di theffraction intensity peaks of andthe diffraction the low-intensity peaks linesand the to low-intensity be submerged lines into to the be contributionsubmerged into of fluorescencethe contribution to the of background.fluorescence to In the the background. equipment configurationIn the equipment used, aconfiguration parallel beam used, of monochromatic a parallel beam X-ray of radiation monochromatic was produced X-ray by radiation introducing was a parabolicallyproduced by bentintroducing graded a multilayer parabolically Göbel bent mirror grad ined the multilayer path of theGöbel primary mirror beam. in the Conventional path of the θprimary–2θ scans beam. were Conventional carried out over θ–2θ a 2scansθ range were of carried 45–135◦ outwith over a step a 2θ size range of 0.01of 45–135°◦ on samples with a preparedstep size followingof 0.01° on the samples metallographic prepared procedures following described the metallographic before. The experimentalprocedures described setup used before. allowed The for obtainingexperimental high-resolution setup used allowed diffraction for patterns obtaining that high-resolution showed over 15,000 diffraction counts patterns for the mostthat showed intense lines.over Therefore,15,000 counts the for limit the of most detection intense (LoD) lines. of aTherefore, phase could the be limit set atof 0.3detection wt% when (LoD) its of strongest a phase peaks could did be notset at overlap 0.3 wt% with when diffraction its strongest peaks ofpeaks other did phases. not overlap Furthermore, with diffraction the use of thepeaks Rietveld of other method phases. to evaluateFurthermore, these the patterns use of allowed the Rietveld for performing method quantitativeto evaluate these analyses patterns with relativeallowed errors for performing lower than 20%quantitative in stable analyses fits with with good relative precision errors only lower for contentsthan 20% higher in stable than fits 1.0 with wt%. good In precision this work, only the 4.2for versioncontents of higher the Rietveld than 1.0 analysis wt%. In program this work, TOPAS the 4.2 (Bruker version AXS, of GmbH,the Rietveld Karlsruhe, analysis Germany,) program was TOPAS used for(Bruker the XRD AXS, data GmbH, refinement. Karlsruhe, Germany,) was used for the XRD data refinement.

3. Results and Discussion 3.1. Quenching Behavior 3.1. Quenching Behavior The ﬁrst step of the experiment was the quenching of the medium manganese steel. The JMatPro The first step of the experiment was the quenching of the medium manganese steel. The JMatPro simulation was carried out to determine the martensite start temperature (Ms) of the steel. simulation was carried out to determine the martensite start temperature (Ms) of the steel. Then, the Then, the dilatometric analysis of the quenching process was conducted. Two samples were quenched dilatometric analysis of the quenching process was conducted. Two samples were quenched to to ensure the reproducibility of the experiment. According to the computational and dilatometric results ensure the reproducibility of the experiment. According to the computational and dilatometric results (Figure2), the calculated M s temperature of the steel was 352 ◦C. The dilatometric curves showed the (Figure 2), the calculated Ms temperature of the steel was 352 °C. The dilatometric curves showed the same bulk Ms temperature for the analyzed steel. However, there were some diﬀerences in the curves. same bulk Ms temperature for the analyzed steel. However, there were some differences in the curves. First, it can be seen that the martensite transformation started at a slightly higher temperature of 390 C First, it can be seen that the martensite transformation started at a slightly higher temperature of 390◦ (MsII—represents a temperature at which small amount of lower stability austenite transforms from γ °C (MsII—represents a temperature at which small amount of lower stability austenite transforms to α’ just before the bulk transformation). Then, at 352 C, the bulk martensitic transformation occurred. from γ to α’ just before the bulk transformation). Then,◦ at 352 °C, the bulk martensitic transformation At the same time, a small increase in the relative change in length (RCL) in both samples occurred occurred. At the same time, a small increase in the relative change in length (RCL) in both samples after the bulk transformation (Ms0—represents a temperature just after the bulk transformation, occurred after the bulk transformation (Ms0—represents a temperature just after the bulk at which another small amount of highly stable austenite is transformed into martensite). This means transformation, at which another small amount of highly stable austenite is transformed into martensite). This means that at around 335 °C, another martensite transformation occurred. This

Materials 2020, 13, 4442 4 of 10

Materials 2020, 13, x FOR PEER REVIEW 4 of 11 that at around 335 ◦C, another martensite transformation occurred. This result means that the austeniteresult means undergoing that the the austenite martensitic undergoing transformation the martensitic had adi transformationfferent chemical had composition a different chemical in various composition in various microregions. We confirmed this effect in the microstructural investigations microregions. We confirmed this effect in the microstructural investigations (banding appeared). (banding appeared). This behavior resulted in a three-step transformation during quenching. First, a This behavior resulted in a three-step transformation during quenching. First, a small amount of small amount of austenite with the lowest thermal stability underwent a transformation; then, the austenite with the lowest thermal stability underwent a transformation; then, the bulk transformation bulk transformation of the austenite took place, and finally, the austenite with the highest thermal of the austenite took place, and finally, the austenite with the highest thermal stability was transformed. stability was transformed. This effect could be the result of local variations in the chemical Thiscomposition effect could due be theto the result segregation of local variationsof Mn during in the solidification. chemical composition The hot-working due to theenhanced segregation the of Mnmanganese during solidification.redistribution in The segreg hot-workingation bands. enhanced This difference the manganese in the local redistribution manganese incontent segregation also bands.influenced This the diff erencelocal Ms in temperature. the local manganese According to content Hidalgo also et influencedal. [17], Mn-rich the localregions M sdecreasetemperature. the Accordinglocal martensite to Hidalgo start temperature et al. [17], Mn-rich compared regions to Mn-poor decrease regions. the Therefore, local martensite Mn can startinfluence temperature the Ms comparedtemperature to Mn-poor of the steel regions. similarly Therefore, to that Mn presen canted influence for carbon the Mins Figuretemperature 2. Moreover, of the steelmanganese similarly to thatreduces presented the C foractivity carbon and in its Figure diffusion2. Moreover, rate [18] manganese, which affects reduces the themartensite C activity formation and its diduringffusion ratequenching. [18], which affects the martensite formation during quenching.

FigureFigure 2. 2.Analysis Analysis of of the the martensitic martensitic transformationtransformation according to to the the JMatPr JMatProo simulation simulation and and the the dilatometricdilatometric study. study. RCL: RCL: relative relative change change in in length. length.

AnotherAnother factor factor influencing influencing the the M Mss temperaturetemperature is related to to the the partial partial dissolution dissolution of of niobium niobium carbides.carbides. According According to to Garcia Garcia de de AndresAndres etet al.al. [[1199],], the occurrence of of the the steps steps during during a amartensite martensite transformationtransformation could could be be the the result result of of the the partial partial dissolutiondissolution of carbides carbides during during austenitization. austenitization. In Inthe the analyzedanalyzed case, case, this this could could be be the the resultresult ofof thethe partialpartial dissolution of of niobium niobium carbides carbides during during the the austenitizationaustenitization step step at at 1100 1100◦ C.°C. The The austenite austenite wherewhere the carbides were were dissolving dissolving should should have have higher higher carboncarbon concentrations concentrations compared compared to to the the areasareas wherewhere the carbides did did not not dissolve. dissolve. It Itseems seems that that this this effecteffect was was smaller smaller compared compared to to the the one one related related to to thethe MnMn segregation.segregation. According to the calculation results shown in Figure 3, the Ms0 temperature of 335 °C indicates According to the calculation results shown in Figure3, the M s0 temperature of 335 ◦C indicates that thethat austenite the austenite should have should a carbon have contenta carbon of ~0.2conten wt%.t of At ~0.2 the samewt%. time,At the the same small time, transformation the small of austenitetransformation at the beginning of austenite means at the that beginning it had a lowermeans carbon that it contenthad a lower compared carbon to content the bulk compared composition. to the bulk composition. However, after comparing the C and Mn effects on the austenite However, after comparing the C and Mn effects on the austenite decomposition, it seems that decomposition, it seems that the manganese segregation can be a critical issue in terms of the Ms the manganese segregation can be a critical issue in terms of the Ms temperature in medium-Mn temperature in medium-Mn steels [20,21]. steels [20,21].

3.2. Tempering Behavior The next step of the study was to analyze the tempering process of the steel after quenching. The tempering was carried out at 350 ◦C for 15, 30, and 60 min to investigate whether it is possible

Materials 2020, 13, 4442 5 of 10 for martensite to undergo tempering at this relatively low temperature. According to the dilatometry results presented in Figure4, the change of the RCL was negligible for all selected times. This change was calculated via the subtraction of the lowest value of the RCL from the higher value for a selected time. The drop in RCL was correlated with the precipitation of carbides during tempering. The obtained experimental results were in good agreement with the simulation data performed using JMatPro. The simulation was performed with the martensitic microstructure being heated up at a rate of 3 ◦C/s to 350 ◦C and tempered for 60 min. The types of carbides taken into account in the simulations for the Fe alloys model were M6C, M7C3, and M23C6. The results of the simulation presented a neglectable shrinkage during tempering over 60 min. One can see that with a longer tempering time, a higher drop in the RCL took place. However, the small level of this decrease indicated a lack (or a very limited extent) of any precipitation of carbides. The carbides could be formed locally but the amount of them should be very small. Therefore, they could not influence the properties of the martensite in any way. The reason for this can be related to the Mn effect on the carbon diffusion. As mentioned above, Mn decreases the carbon diffusion rate in steel [18]. This means that the selected time (60 min) was too short for carbon to have enough time to form carbides. This was in good agreement with the dilatometricMaterials 2020, 13 results., x FOR PEER REVIEW 5 of 11

Materials 2020, 13, x FOR PEER REVIEW 6 of 11 Figure 3. Martensite start temperature calculated using JMatPro.

3.2. Tempering Behavior The next step of the study was to analyze the tempering process of the steel after quenching. The tempering was carried out at 350 °C for 15, 30, and 60 min to investigate whether it is possible for martensite to undergo tempering at this relatively low temperature. According to the dilatometry results presented in Figure 4, the change of the RCL was negligible for all selected times. This change was calculated via the subtraction of the lowest value of the RCL from the higher value for a selected time. The drop in RCL was correlated with the precipitation of carbides during tempering. The obtained experimental results were in good agreement with the simulation data performed using JMatPro. The simulation was performed with the martensitic microstructure being heated up at a rate of 3 °C/s to 350 °C and tempered for 60 min. The types of carbides taken into account in the simulations for the Fe alloys model were M6C, M7C3, and M23C6. The results of the simulation presented a neglectable shrinkage during tempering over 60 min. One can see that with a longer tempering time, a higher drop in the RCL took place. However, the small level of this decrease indicated a lack (or a very limited extent) of any precipitation of carbides. The carbides could be formedFigure locally 4. but Tempering the amount progress of them according should toto the thebe dilatometricvedilatomery small.tric resultsresultsTherefore, andand the thethey JMatProJMatPro could simulation. simulation.not influence the properties of the martensite in any way. The reason for this can be related to the Mn effect on the carbonTo diffusion.estimate the As possible mentioned amount above, of carbidesMn decreases at a temperature the carbon of diff 350usion °C, calculationsrate in steel were [18]. made This meansand the that obtained the selected results time are (60 presented min) was in too Figure short 5. for The carbon maximum to have amount enough oftime carbides to form (cementite carbides. Thisaccording was in to good the simulations) agreement with that couldthe dilatometric be precipit atedresults. during 60 min of tempering was around 2.5%. This value was constant from the tempering temperature of 275 °C and did not increase up to 350 °C. This means that the contraction of the dilatometry curve could be the result of the formation of ~2.5% carbides during tempering.

Figure 5. Cementite precipitation level at different tempering temperatures.

3.3. Microstructure After quenching, the microstructure of the medium-Mn steel was fully martensitic (Figure 6). No retained austenite or any other phases were visible in the microstructure. This was the result of the high hardenability effect of manganese [22]. Note that the austenitization was full despite the high concentration of Al in the steel, which increased the Ac3 temperature.

Materials 2020, 13, x FOR PEER REVIEW 6 of 11

Materials 2020, 13, 4442 6 of 10

Figure 4. Tempering progress according to the dilatometric results and the JMatPro simulation. To estimate the possible amount of carbides at a temperature of 350 ◦C, calculations were made and the obtainedTo estimate results the are possible presented amount inof carbides Figure 5at. a Thetemperature maximum of 350 amount°C, calculations of carbides were made (cementite accordingand to the the simulations)obtained results that are couldpresented be in precipitated Figure 5. The during maximum 60 amount min of of tempering carbides (cementite was around 2.5%. according to the simulations) that could be precipitated during 60 min of tempering was around 2.5%. This value was constant from the tempering temperature of 275 ◦C and did not increase up to 350 ◦C. This value was constant from the tempering temperature of 275 °C and did not increase up to 350 °C. This meansThis that means the that contraction the contraction of the of the dilatometry dilatometry curve could could be the be result the result of the offormation the formation of ~2.5% of ~2.5% carbides duringcarbides tempering. during tempering.

Figure 5.FigureCementite 5. Cementite precipitation precipitation levellevel at at different diﬀerent tempering tempering temperatures. temperatures.

3.3. Microstructure3.3. Microstructure After quenching,After quenching, the microstructure the microstructure of of the the medium-Mnmedium-Mn steel steel was wasfully martensitic fully martensitic (Figure 6). (Figure6). No retainedNo austenite retained austenite or any or other any other phases phases were were visible visible in in thethe microstructure. This This was wasthe result the resultof of the the high hardenability effect of manganese [22]. Note that the austenitization was full despite the high hardenabilityhigh concentration eﬀect of Al manganese in the steel, which [22]. increased Note that the theAc3 temperature. austenitization was full despite the high concentration of Al in the steel, which increased the Ac3 temperature. Materials 2020, 13, x FOR PEER REVIEW 7 of 11

Figure 6. FigureMartensitic 6. Martensitic microstructures microstructures ofof steel after after quenching: quenching: (a) light ( amicroscopy) light microscopy (LM) and (b) (LM) and scanning electron microscopy (SEM). (b) scanning electron microscopy (SEM). After tempering (Figure 7), the microstructure did not change much. The microstructure after 15 min (Figure 7a) did not show any distinct changes in the morphology of the martensite and no cementite precipitates were visible. Looking at the martensite after 60 min of tempering, some small changes in its morphology were visible. Moreover, some precipitates inside the martensite laths could be seen. According to Bhadeshia [23], the carbides with a plate-like shape that are presented in Figure 7b (indicated by a white arrow) come from the tempering process of martensite. This means that some tempering effect took place. However, its effect was not too strong. Thus, the martensite in the analyzed medium-Mn steel had a high tempering resistance at 350 °C. Even 1 h tempering was not enough to significantly start the precipitation process.

Materials 2020, 13, 4442 7 of 10

After tempering (Figure7), the microstructure did not change much. The microstructure after 15 min (Figure7a) did not show any distinct changes in the morphology of the martensite and no cementite precipitates were visible. Looking at the martensite after 60 min of tempering, some small changes in its morphology were visible. Moreover, some precipitates inside the martensite laths could be seen. According to Bhadeshia [23], the carbides with a plate-like shape that are presented in Figure7b (indicated by a white arrow) come from the tempering process of martensite. This means that some tempering effect took place. However, its effect was not too strong. Thus, the martensite in the analyzed medium-Mn steel had a high tempering resistance at 350 ◦C. Even 1 h tempering was not enoughMaterials 2020 to significantly, 13, x FOR PEER start REVIEW theprecipitation process. 8 of 11

Figure 7. The SEM microstructures of the steel after tempering for (a) 15 and (b) 60 min.min.

3.4. X-ray Diffraction Diffraction To check the amount and type of carbides in th thee tempered microstructure, X-ray diffraction diffraction was carried out. In Figure 88,, thethe observedobserved (blue(blue points)points) XRDXRD patternspatterns werewere comparedcompared withwith thethe calculatedcalculated (red solid line) line) profiles profiles obtained obtained after after a Rietveld a Rietveld refinement. refinement. The The differential differential curve curve is shown is shown as a gray as a grayline and line andthe background the background removed removed phase phase (α (phase)α phase) is isshown shown in in orange. orange. According According to to the the result presented inin this this figure, figure, no no carbides carbides were were detected detected after 15after (Figure 15 (Figure8a) or 60 8a) min or (Figure 60 min8b) (Figure of tempering. 8b) of Thistempering. means thatThis eithermeans the that carbide either precipitation the carbide didprec notipitation occur did or that not their occur amount or that and their size amount was below and thesize detectionwas below limit the ofdetection the technique. limit of The the lattertechnique. seems The to be latter the case,seems as to the be JMatPro the case, simulations as the JMatPro and SEMsimulations investigations and SEM in Figureinvestigations7 indicate in some Figure precipitation 7 indicate tooksome place. precipitation The XRD took patterns place. in The Figure XRD8 alsopatterns show in that Figure no peaks8 also ofshow retained that no austenite peaks of were reta present,ined austenite but only were the present, following butα onlyphase the peaks following were: (110),α phase (002), peaks (211), were: and (110), (022). (002), Moreover, (211), if and some (022). retained Moreover, austenite if some was presentretained before austenite quenching, was present the γ before quenching, the γ to α transformation should be detected via dilatometry, and as Figure 4 indicates, this was not the case. In brief, no retained austenite was present in the microstructure after either of the tempering conditions that were analyzed.

Materials 2020, 13, 4442 8 of 10 to α transformation should be detected via dilatometry, and as Figure4 indicates, this was not the case. In brief, no retained austenite was present in the microstructure after either of the tempering conditionsMaterials 2020 that, 13, x were FOR PEER analyzed. REVIEW 9 of 11

Figure 8. X-rayX-ray diffraction diffraction patterns patterns of ofthe the samples samples after after tempering tempering for: for:(a) 15 (a )and 15 ( andb) 60 ( bmin) 60 (blue min (blueline—experimental line—experimental pattern, pattern, red line—calculated red line—calculated pattern, pattern, gray line—differential gray line—differential curve). curve). 3.5. Hardness 3.5. Hardness According to the JMatPro simulations of the hardness changes during tempering for different According to the JMatPro simulations of the hardness changes during tempering for different times (Figure9), the hardness should decrease with increasing time. Specifically, the hardness of the times (Figure 9), the hardness should decrease with increasing time. Specifically, the hardness of the steel tempered at 350 ◦C for 15 min should be ~435 HV10, with a progressive decrease to 419 and 400 steel tempered at 350 °C for 15 min should be ~435 HV10, with a progressive decrease to 419 and 400 HV10 as the tempering time was increased to 30 and 60 min, respectively. The experimental results HV10 as the tempering time was increased to 30 and 60 min, respectively. The experimental results show that the hardness of the quenched martensitic microstructure was 468 HV10, which decreased show that the hardness of the quenched martensitic microstructure was 468 HV10, which decreased to 428 HV after 60 min of tempering at 350 ◦C. This means that the value of the decrease was not to 428 HV after 60 min of tempering at 350 °C. This means that the value of the decrease was not significant. Regardless of the exact values, the tendencies of the calculation and experimental results significant. Regardless of the exact values, the tendencies of the calculation and experimental results match but not their magnitudes. While the calculations predicted a more pronounced drop in Vickers match but not their magnitudes. While the calculations predicted a more pronounced drop in Vickers hardness as the tempering time increased, the experimental results showed that this drop was limited, hardness as the tempering time increased, the experimental results showed that this drop was revealing a high tempering resistance of the martensite in the 4% Mn steel studied in this work. limited, revealing a high tempering resistance of the martensite in the 4% Mn steel studied in this work.

Materials 2020, 13, 4442 9 of 10 Materials 2020, 13, x FOR PEER REVIEW 10 of 11

Figure 9. ComparisonComparison of of the hardness results from th thee JMatPro simulations and the experiments. 4. Conclusions 4. Conclusions The dilatometric and microstructural analysis of the tempering process in Al-rich, medium-Mn The dilatometric and microstructural analysis of the tempering process in Al-rich, medium-Mn steel was performed. The martensite in the analyzed 4% Mn steel had a high resistance to the tempering steel was performed. The martensite in the analyzed 4% Mn steel had a high resistance to the over 15, 30, and 60 min at 350 C. An increase in the time from 15 to 60 min led to a higher contraction tempering over 15, 30, and 60 ◦min at 350 °C. An increase in the time from 15 to 60 min led to a higher during tempering but this eﬀect was neglectable. This means that a very small amount of cementite contraction during tempering but this effect was neglectable. This means that a very small amount of precipitated during tempering. According to the microstructure investigation, some interlath carbides cementite precipitated during tempering. According to the microstructure investigation, some were visible in the martensite laths after 60 min. A small thickening eﬀect of the martensite laths with interlath carbides were visible in the martensite laths after 60 min. A small thickening effect of the increasing tempering time was observed. The relatively small drop in hardness level by 40 HV after martensite laths with increasing tempering time was observed. The relatively small drop in hardness tempering for 60 min compared to the quenched state indicated the strong resistance of the 4% Mn level by 40 HV after tempering for 60 min compared to the quenched state indicated the strong steel to the tempering at 350 C. resistance of the 4% Mn steel◦ to the tempering at 350 °C. The results show that the martensite in the investigated medium-Mn steel was relatively thermally The results show that the martensite in the investigated medium-Mn steel was relatively stable at tempering treatments up to 350 C. It preserved good mechanical properties in this temperature thermally stable at tempering treatments◦ up to 350 °C. It preserved good mechanical properties in region. This knowledge is important for designing novel heat treatment schedules, such as quenching this temperature region. This knowledge is important for designing novel heat treatment schedules, and partitioning or austempering, especially for the heat treatments performed near the M temperature such as quenching and partitioning or austempering, especially for the heat treatmentss performed of medium-Mn steels. near the Ms temperature of medium-Mn steels.

Author Contributions: Conceptualization,Conceptualization, A.G. and and C.G.-M.; C.G.-M.; methodology, methodology, M.M., J.A.J. and and C.G.-M.; C.G.-M.; validation, validation, M.M. and J.A.J.; formal analys analysis,is, A.G. and C.G.-M.; investigation, M.M. and J.A.J.; data curation, curation, M.M. M.M. and and J.A.J.; J.A.J.; writing—original draft preparation, A.G. and M.M.; writing—review and editing, J.A.J. and C.G.-M.; visualization, writing—original draft preparation, A.G. and M.M.; writing—review and editing, J.A.J. and C.G.-M.; M.M. and J.A.J.; supervision, A.G. and C.G.-M.; project administration, A.G.; funding acquisition, A.G. All authors havevisualization, read and M.M. agreed and to J.A. theJ.; published supervision, version A.G. of and the C.G.-M.; manuscript. project administration, A.G.; funding acquisition, A.G. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by a rector grant in the area of scientific research and development works, SilesianFunding: University This research of Technology, was funded grant by a number rector grant 10/010 in/RGJ19 the area/0271. of scientific research and development works, Acknowledgments:Silesian University ofJose Technolo Antoniogy, Jimenez grant number and Carlos 10/010/RGJ19/0271. Garcia-Mateo would like to express their gratitude for the support of the XRD and phase transformation laboratories at CENIM. Acknowledgments: Jose Antonio Jimenez and Carlos Garcia-Mateo would like to express their gratitude for the Conflictssupport of of the Interest: XRD andThe phase authors transf declareormation no conflict laboratories of interest. at CENIM.

Conflicts of Interest: The authors declare no conflict of interest. References

1.ReferencesDe Moor, E.; Matlock, D.K.; Speer, J.G.; Merwin, M.J. Austenite stabilization through manganese enrichment. Scr. Mater. 2011, 64, 185–188. [CrossRef] 1. De Moor, E.; Matlock, D.K.; Speer, J.G.; Merwin, M.J. Austenite stabilization through manganese 2. Schneider, R.; Steineder, K.; Krizan, D.; Sommitsch, C. Eﬀect of the heat treatment on the microstructure and enrichment. Scr. Mater. 2011, 64, 185–188, doi:10.1016/j.scriptamat.2010.09.040. mechanical properties of medium-Mn-steels. Mater. Sci. Technol. 2018, 35, 2045–2053. [CrossRef] 2. Schneider, R.; Steineder, K.; Krizan, D.; Sommitsch, C. Effect of the heat treatment on the microstructure 3. Hu, B.; Luo, H.-W. Microstructures and Mechanical Properties of 7Mn Steel Manufactured by Diﬀerent and mechanical properties of medium-Mn-steels. Mater. Sci. Technol. 2018, 35, 2045–2053, Rolling Processes. Metals 2017, 7, 464. [CrossRef] doi:10.1080/02670836.2018.1548957.

Materials 2020, 13, 4442 10 of 10

4. Caballero, F.G.; Bhadeshia, H.K.D.H.; Mawella, K.J.A.; Jones, D.G.; Brown, P. Very strong low temperature bainite. Mater. Sci. Technol. 2002, 18, 279–284. [CrossRef] 5. Yoozbashi, M.; Yazdani, S.; Wang, T. Design of a new nanostructured, high-Si bainitic steel with lower cost production. Mater. Des. 2011, 32, 3248–3253. [CrossRef] 6. Shen, Y.; Qiu, L.; Sun, X.; Zuo, L.; Liaw, P.; Raabe, D. Effects of retained austenite volume fraction, morphology, and carbon content on strength and ductility of nanostructured TRIP-assisted steels. Mater. Sci. Eng. A 2015, 636, 551–564. [CrossRef] 7. Shen, Y.; Liu, Y.; Sun, X.; Wang, Y.; Zuo, L.; Misra, R. Improved ductility of a transformation-induced-plasticity steel by nanoscale austenite lamellae. Mater. Sci. Eng. A 2013, 583, 1–10. [CrossRef] 8. Long, X.; Zhang, F.; Kang, J.; Lv, B.; Shi, X. Low-temperature bainite in low-carbon steel. Mater. Sci. Eng. A 2014, 594, 344–351. [CrossRef] 9. Jing, C.; Ding, X.; Ye, D.; Zhao, J.; Lin, T.; Xu, S. Effect of Different Isothermal Time on Microstructure and Mechanical Property of the Low-Carbon Steel Treated by Dual-Stable C-Mn Partitioning Process. Scanning 2020, 2020, 5931721. [CrossRef] 10. Ayenampudi, S.; Celada-Casero, S.; Sietsma, J.; Santofimia, M.J. Microstructure evolution during high-temperature partitioning of a medium-Mn quenching and partitioning steel. Materialia 2019, 8, 1–11. [CrossRef] 11. Bala, P.; Pacyna, J. The kinetics of phase transformations during tempering of the new hot working tool steel designed for a large size forging dies. Steel Res. Int. 2008, 2, 407–413. 12. Górka, J.; Opiela, M. Structure and Properties of High-Strength Low-Alloy Steel Melted by the Laser Beam. Mater. Perform. Charact. 2019, 8, 1–10. [CrossRef] 13. Grajcar, A.; Radwanski, K. Microstructural comparison of the thermomechanically treated and cold deformed Nb-microalloyed TRIP steel. Mater. Tehnol. 2014, 48, 679–683. 14. Opiela, M.; Grajcar, A. Elaboration of forging conditions on the basis of the precipitation analysis of MX-type phases in microalloyed steels. Arch. Civ. Mech. Eng. 2012, 12, 427–435. [CrossRef] 15. ASTM. Standard Practice for Quantitative Measurement and Reporting of Hypoeutectoid Carbon and Low-Alloy Steel Phase Transformations; A1033-04; ASTM International: West Conshohocken, PA, USA, 2004. Available online: https://www.astm.org/ (accessed on 25 March 2020). 16. A Collection of Free Downloadable Papers on the Development and Application of JMatPro; Practical Software for Materials Properties. Sente software Ltd., 2005. Available online: http://www.sentesoftware.co.uk/biblio.html (accessed on 1 September 2020). 17. Hidalgo, J.; Celada, C.; Santofimia, M.J. Fracture mechanisms and microstructure in a medium Mn quenching and partitioning steel exhibiting macrosegregation. Mater. Sci. Eng. A 2019, 754, 766–777. [CrossRef] 18. Kral, L.; Million, B.; Cermˇ ák, J. Diffusion of Carbon and Manganese in Fe-C-Mn. Defect Diffus. Forum 2007, 263, 153–158. [CrossRef] 19. Garcia de Andres, C.; Caballero, F.G.; Capdevila, C.; Alvarez, L.F. Application of dilatometric analysis to the study of solid-solid phase transformation in steels. Mater. Charact. 2002, 48, 101–111. [CrossRef] 20. Liang, J.-H.; Zhao, Z.; Tang, D.; Ye, N.; Yang, S.; Liu, W. Improved microstructural homogeneity and mechanical property of medium manganese steel with Mn segregation banding by alternating lath matrix. Mater. Sci. Eng. A 2018, 711, 175–181. [CrossRef] 21. Kalsar, R.; Sanamar, S.; Schell, N.; Brokmeier, H.-G.; Saha, R.; Ghosh, P.; Bhagat, A.; Suwas, S. Elemental partitioning in medium Mn steel during short-time annealing: An in-situ study using synchrotron X-rays. Materialia 2020, 9, 100594. [CrossRef] 22. Calcagnotto, M.; Ponge, D.; Raabe, D. On the Effect of Manganese on Grain Size Stability and Hardenability in Ultrafine-Grained Ferrite/Martensite Dual-Phase Steels. Met. Mater. Trans. A 2011, 43, 37–46. [CrossRef] 23. Bhadeshia, H.K.D.H.; Christian, J.W. Bainite in steels. Met. Mater. Trans. A 1990, 21, 767–797. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).