The influence of the heat generation during deformation on the mechanical properties and microstructure of the selected TWIP steels

The TWIP (Twinning Induced Plasticity) steels are one of the most promising materials in reducing the weight of vehicles. Despite a lot of research on TWIP steel, there are some issues that are not explained enough. Due to the future use of TWIP steel and the manufacturing of the final part by metal forming, three issues still need to be clarified. The first one, which is the most important, is the increase of the temperature due to the conversion of the deformation work into heat. TWIP steel has a high limit strain, strength and lower conductivity than conventional steel, therefore the heat generation of TWIP steel is greater than for other materials. The second and third issues are combined. They concern the influence of V microadditions on the stress–strain curves, the strain hardening coefficient n and the strain rate sensitivity coefficient m under cold deformation conditions. These properties determine the cold formability of TWIP steels. In the research, two TWIP steels were used with and without V microadditions (MnAl and MnAl-V steel). The special methodology using strain and temperature measurement systems as well as light and scanning electron microscopy (SEM) were applied. Research shows a significant increase of the temperature in the material due to high plastic deformations as well as a high level of yield stress. In the neck area, for the highest strain rate of 0,1 s -1, at the moment of rupture, the temperature reaches more than 200 °C. The difference between the average temperature in the rupture area and the maximum temperature is equal to 100° C. Its high increase can lead e.g. to changes in the deformation mechanism from twinning to dislocation gliding, which is also connected with a worsened workability, and thus also energy consumption of the bodywork elements. MnAl-V steel has better or similar ductility for the deep drawing in comparison to MnAl steel at low strain rates for almost isothermal conditions (constant temperature during deformation). However the MnAl steel has better ductility for the larger strain rates over 0.1 s−1 then there is large heat concentration in a very narrow area for MnAl-V steel. The obtained results are very important from an application point of view. The strain rate sensitivity coefficient m of the steel MnAl has very low, and even negative, values, which can make the production of complicated drawpieces difficult. Higher values of the strain rate sensitivity coefficient are exhibited by steel MnAl-V, i.e. at the level of 0,05, which is almost constant in the whole range of the obtained deformations.


Introduction
The effort to reduce the weight of vehicles results in the use of various groups of materials such as composites, polymers or light alloy materials [1], however, the energyabsorbing car body components are still made of steel [2]. The approach to the design of modern steels with a very high strength and excellent plastic properties has recently become more and more important [3]. The TWIP (Twinning Induced Plasticity) steels are one of the most promising materials in this class.
The phenomenon of the mechanical (deformation) twins formation in the microstructure of a manganese steel was ultimately revealed by Raghavan et al. [4]. The TEM (transmission electron microscopy) technique was used to observe this work-hardening mechanism. Then it was confirmed that the deformation twins constitute obstacles during the dislocation glide [5].
In the early 1990s, information about the high manganese steel containing 0.5% C, 25% Mn, 1.5% Al and 0.1% N appeared, pointing to the possibility of its application for thin-walled elements in the automotive industry [6]. Grassel and Fromayer proposed for the first time a concept of TWIP steel [7]. During the analysis of steels with 15 and 25% Mn, the authors established that the ideal combination of high mechanical properties and high plasticity is determined by the effect of the formation of mechanical twins in the austenitic matrix [8]. The subsequent research was carried out by the same authors with steels that have a density lower than 7 g/cm 3 and very good mechanical properties: the ultimate strength of up to 1100 MPa and the limit strain of up to 60% [9]. It was demonstrated that in order to obtain an advantageous combination of mechanical and plastic properties of steels in which the twinning mechanism is dominant during cold forming, the Mn content should be within the range of 20 to 30% wt., the Al content -from 3 to 5% wt. and the C content -not more than 0.6% wt. [10]. The TWIP effect is responsible for a range of advantageous properties, such as those which can be used in transport vehicles or military technologies for ballistic shields. The compression behaviour of the TWIP steels was manifested by an excellent combination of strength and ductility as well as with continuous strain hardening to the high strain [11,12].
There are three main mechanisms that particularly affect the manganese steels deformation: i.e. the martensitic phase transformation causing the Transformation-Induced Plasticity (TRIP) effect, the mechanical twinning of the austenite -the Twinning Induced Plasticity (TWIP) effect, and formation of shear microbands in the austenite-Microbands Inducted Plasticity (MBIP) effect [13]. These effects influence uniquely the combination of mechanical and plastic properties. Their occurrence depends on the value of stacking fault energy (SFE), which is determined by the chemical composition (Fig. 1).
The value of stacking fault energy depends not only on the chemical composition but also on the deformation temperature. Correlation of these two factors determines directly the dominant deformation mechanism, especially for the MnAl steel.
The formation of twins during deformation increases the mechanical properties without a loss of ductility, or even with its increase [16]. An analysis of the literature referring to steels with the TWIP effect has revealed that one of the most important subjects in this area is the evaluation of their mechanical properties as well as ductility. These features decide on the possibilities of TWIP steels application for responsible vehicle structural elements. In this context, TWIP steels are characterized by the high work hardening, which is mostly the result of mechanical twinning, but also other factors, such as dynamic strain ageing (DSA) and hardening through the formation of LC barriers, which have only been observed in steels containing Nb, V and Ti microadditions [13].
The research concerns mainly the effect of a vanadium on microstructure changes and hot formability of TWIP steels. There is no information in the literature about the influence of vanadium on the workability at ambient temperature. Presented research is very important because in the future TWIP steel will be used as a construction material for the car bodies which are formed mainly in the cold sheet metal forming operations. Vanadium microadditions can accelerate or inhibit softening processes such as dynamic recrystallisation or recovery. For example, the significant effect of a vanadium addition on the grain size was revealed in [17]. In the six investigated steels the fine vanadium carbide (VC) precipitates allow to get the grain size refinement from ∼1250 µm to ∼700 µm, and with boron addition even to ∼500 µm. It is caused by the segregation of VC on the boundaries that results grain refinement during solidification. Fine VC precipitations are mostly the responsible for decrease of the ductility. In the investigated steels, Fig. 1 Correlation schematic presentation of privileged deformation mechanisms depending on SFE changes in the manganese steels [11,14,15] formability of non-recrystallised austenite was reduced gradually with the increase of temperature to 1000 °C. On the other hand, an increase of hot ductility in the TWIP steel with vanadium, in comparison to TWIP steel with a similar composition but without vanadium, in the temperature range of 800-900 °C, was observed. The largest limit strain results from the onset of dynamic recrystallization [18]. Reyes-Calderon et al. [8] studied the influence of microalloying elements such as V, Ti and Nb on hot formability of a highmanganese TWIP steel. They proved that the microalloying elements addition slightly increases the peak strain and the peak stress, particularly at low strain rates and depending on temperature.
The main factor responsible for the strain-hardening capability of manganese steels is the strain rate. These materials belong to a group of alloys with significant sensitivity to this parameter. The strain rate is one of the basic parameters influencing any metal forming process, which may vary in a broad range, while the other parameters are constant or negligibly small. In this context, studies of material may be carried out in a broad range of the strain rate [19].
Even in the low-manganese steel, the twinning begins to play an important role in deformation under the dynamic deformation conditions (high strain rates) [11]. The dynamic tensile behaviour of TWIP steel study experimentally in a broad range of strain rates from 0.001 to 400 s −1 , confirms the high strength approximately 1500 MPa over a strain of 0.1-0.5 and the significant strain hardening. Additionally, the work hardening rate has an increasing tendency with strain rate. The strain rate effect for the TWIP steel could be described by Johnson-Cook hardening model [12].
The experimental results indicate that the work-hardening exponent, ultimate tensile strength and the uniform elongation decrease with the increasing strain rate. This phenomenon exhibits negative strain rate sensitivity (NSRS), and the strain rate sensitivity value (m) is observed to be higher in the aluminium added FeMnC TWIP steel [19,20].
Another important factor responsible for the strain-hardening capability of the manganese steels is the deformation temperature. An increase in the deformation temperature significantly influences the microstructure evolution, resulting in a remarkable alteration in the work-hardening behaviour. Consequently, the mechanical twinning is gradually deteriorated with the deformation temperature [21].
It is important to consider not only strain rate, but also the increase of temperature as a result of a change of the deformation work into heat. Unfortunately, the literature in this area is very limited. Some tests were carried out for TRIP and DP steels. For example Gao et al. (2015) claimed [22], that the temperature rises during adiabatic heating of TRIP and DP steels with increasing strain rate. For the experiments carried out with strain rates of 0.1 up to 2000s −1 , the temperature increase of TRIP steels was in the range of 100 ÷ 300 ºC, while that of DP steel was in the range of 100 ÷ 220 ºC. More detailed research was carried out by the authors, who showed that at the strain rate of 1000 s −1 temperature increases almost by 100ºC and at a moment of necking, it can increase locally to over 300ºC [23].
The heat generated during the plastic deformation of TWIP steels has higher values than in the case of conventional steels, due to their higher yield stress and deformation. This heat has a very important effect on the microstructure of the deformed material and thus also on its properties. Its high increase can lead e.g. to changes in the deformation mechanism from twinning to dislocation glide, which is also connected with a worsened deformability and thus also energy consumption of the bodywork elements. The presented literature review suggests that the studies on the subject concern mainly the influence of Nb, V and Ti microadditions on hot formability of TWIP steels due to the fact that the grain size in TWIP steels can be controlled by adding different elements and consistently inhibiting softening processes such as dynamic recrystallization or recovery. There is no information about the effect of vanadium on the stress-strain curves, the strain hardening coefficient n as well as the strain rate sensitivity coefficient m at cold deformation conditions. These properties determine cold formability of TWIP steels. It is very important because in the future TWIP steel will be used as a construction material for the car bodies that are formed mainly in the cold sheet metal forming operations. Vanadium microadditives significantly affect the mechanical properties of TWIP steel and its susceptibility to cold forming through the deep drawing stamping process determined by the strain hardening coefficient n as well as the strain rate sensitivity coefficient m.
The main aim of the study is to determine the temperature increase during the deformation of TWIP steel at various strain rates (from 10 -3 s -1 to 10 -1 s -1 ) as a result of a change of the plastic deformation work in heat and then, to analyze the effect of this temperature rise on the microstructure evolution that take place in the investigated steel. An additional aim was to determine the basic factors affecting the limit deformation in the press forming processes-the hardening coefficient and the strain rate sensitivity coefficient-as well as to analyze the effect of a vanadium addition on the properties and microstructure of TWIP steel in the case when it is dissolved in a solid solution.

TWIP steel
For the tests, two grades of steel were used-TWIP MnAl and MnAl-V -with similar chemical compositions, presented in Table 1 (chemical composition of the examined steels). Alloy MnAl-V was a modification of alloy MnAl, through an addition of vanadium in the amount of 0.1%. The alloys were melted in a vacuum induction furnace VSG100S, which ensures high metallurgical purity and makes it possible to obtain the required chemical composition of steel. The basic iron bearing charge for the melts was Armco iron grade 04 J as well as alloy additions in the form of pure metals (Mn, Al) and C and ferroalloy (FeB). As a rare earth metal carrier, a mischmetal with the composition: 50% Ce, 20% La and 20% Nd was used. Due to high losses of manganese as a result of its evaporation during degassing in vacuum, an additional portion of manganese with a low sulfur content was loaded into the crucible, so that the condition of its high content could be met. The melting was carried out initially in vacuum and at the moment of the appearance of a "liquid metal pool" -in the atmosphere of argon. The casting was also performed in the argon atmosphere, into a copper crystallizer cooled with water, and the size of the ingot was 100 × 100 × 1100 / 835 mm.
The rolling was carried out in module B of a line for semi-industrial simulation of metal and alloy product manufacturing processes at the Institute for Ferrous Metallurgy.
The ingots were preliminarily rolled in 5 passes with relative deformation in the range of about 20-25% to flat bars with the intermediate thickness of 26 mm. At the further stage, after the flat bars were divided into 3 sections, each material was rolled in 4 passes with the relative deformation in the range of about 12-25% to flat bars with the thickness of 11,7-11,8 mm. The subsequent rolling to the final thickness was carried out in two stages. At the first stage, after the rearrangement of the bands into two sections, the material was rolled into intermediate bands with the thickness of about 4,1-4,5 mm with the use of 5 passes with relative deformation of about 17-27%. Before the final rolling, the flat bars were subjected to chemical etching in order to remove the oxide layers from their surfaces, formed during the heating for the plastic treatment. The last stage of rolling the flat bars into the final thickness for each steel grade consisted of 5 passes, in which the total relative deformation equalled about 34%. The flat bars, right after rolling, underwent free cooling in air.
The chemical analyses of the content of: C, Mn, Al, B, P and S in the steel were made by means of the optical emission spectroscopy method with spark excitation (OES spark) with the use of the tool Magellan Q8 Bruker, whereas those of: Ce, Nd and La -by the OES-ICP method (inductively coupled plasma optical emission spectrometry) with the use of an Agilent 5100 ICP-OES spectrometer. The nitrogen content was determined by the high temperature extraction method by means of the TCHEN 600 Leco tool.
The final thermal treatment ensuring an austenitic structure consisted in supersaturation from 1100 °C/2 h and cooling in water. The microstructure of the obtained steels MnAl and MnAl-V after supersaturation is presented in Fig. 2.
A measurement of the mean grain diameter was made with the use of the MET-ilo program [24], which, for steel MnAl, equalled 312 µm, and for MnAl-V-263 µm. In this case, the material was supersaturated from 1100 °C, which, according to the calculations of the ThermoCalc program, proves that the only small amount vanadium carbide precipitation can occur (Fig. 3). The obtained small difference in the grain size is a result of the recrystallization processes at the consecutive rolling stages.
The stacking fault energy of the examined materials was determined on the basis of analytical calculations. For the determination of SFE, Eq. (1) was applied. The stacking fault energy is a thermodynamic function if the stacking fault is considered as a boundary between and [14].
where: A -the molar density of the most dense atom packing along plane (111), dependent on the austenite lattice constant and the Avogadro number, with the value of 2,94 × 10 -5 J × mol 2 [14,25] → -the free energy on the phase boundary equalling 0.015 J × mol −2 [14,25] The SFE value of the investigated steels was determined by means of a modification of the ΔG → value expressed by relation (2) [25] (2) where: ΔG → -the change of the molar free energy of the elements between phases and : ΔG → [26,27] Ω → FeMn -the parameter of the component's effect between the phases: [26,28] The stacking fault energy value determined analytically for steel MnAl equals 39 mJ/m 2 and for steel MnAl-V, it is 42 mJ/m 2 . As the basic reference material with a tendency for twinning, the Hadfield steel is usually assumed.
The obtained SFE values for MnAl and MnAl-V confirm that both examined high manganese steels characterize in a stacking fault energy in the scope of 20 mJ/m 2 ÷ 60 mJ/m 2 therefore they belong to materials demonstrating a tendency to form mechanical twins induced by plastic deformation [29].

Tensile tests
2,5 mm thick metal sheets were used to cut out samples for the tensile tests with the dimensions given in Fig. 4. The dimensions were determined on the basis of the standard ISO PN-EN ISO 6892-1. The samples were cut out by means of a CNC wire cut EDM machine, parallel to the rolling direction.
The tensile tests were carried out on a dual column universal testing machine Instron 3369, equipped with the 50 kN capacity load cell. For the tests, an extensometer was used, with the initial measurement length of 50 mm. Three strain rates were applied: 0.001, 0.01 and 0.1 s -1 . The tests were performed at ambient temperature (about 21ºC 士5 ºC). For each applied rate, three samples were used.

Temperature measurement
For the analysis of the effect of the strain rate on the temperature increase in the area of sample rupture, during the test, the temperature was measured by means of a thermovision camera FLIR T840 with the frame rate of 30 fps. Before the tension, the specimens were covered on one side with black enamel in aerosol assigned for metal surfaces. This procedure aimed at increasing the emissivity and thus also the measurement precision [30]. For a surface prepared this way, the assumed emissivity was n = 0.95. Fig. 5 shows an exemplary thermogram at the moment of sample rupture, with a measurement rectangle 8 x 8 mm marked in red (denoted as B1).
Due to the fact that the structural tests are performed on the basis of sample sections from the vicinity of the rupture area, it was deemed the most reliable to assume the determination is not only the maximum but also the average temperature in the area B1.

Microstructure analysis
Microstructural observations of the examined steel were made on microsections parallel to the sample axis (longitudinal), in area B1 of the sample after rupture (Fig. 6). The tests were performed by means of a light microscope Olympus GX71 and with the use of the scanning electron microscopy technique (SEM), with a Hitachi S-3400N microscope. In order to reveal the microstructure, the material was etched in 6% Nital (94 ml ethyl alcohol, 6 ml HNO 3 acid), once as well as several times. Fractographic examinations of the formed fractures were carried out by way of using the scanning electron microscopy technique (SEM) by means of a microscope equipped with a cold field emission gun, with the accelerating voltage of 15 kV.

Tensile properties
Figures 7 and 8 present the true stress -strain diagrams for both investigated steels and various strain rates. The strain hardening exponent values as well as the average and maximum temperature measured in the area B1 at the moment right before the sample fracture are also given. For the lowest strain rate (0.001 s −1 ) the both steels exhibit a relatively low yield strength of about 250 MPa and a maximum true stress of ca. 950 MPa. The slightly higher increase of yield strength at higher strain rates was obtained for the MnAl-V steel, due to the smaller mean grain size (263 µm) in comparison to the mean grain size in the MnAl steel (312 µm) and the presence of very fine face-center cubic vanadium carbides that were observed in the microstructure (Fig. 11).
The most important issue is that the vanadium addition decreases the limit strain (strain to fracture) at higher strain rates. At the largest applied strain rate (0.1 s −1 ) it is 0.45 for the MnAl steel, while it is only 0.28 for the MnAl-V steel. It is connected with heat generated during deformation. In the case of MnAl-V steel vanadium mainly remains in a solid solution. It lowers the thermal conductivity, which translates to the heat concentrated in a very narrow area in the necking. The increase of temperature for MnAl and MnAl-V steel are different ( Fig. 9). At strain 0.25 for the highest strain rate of 0.1 s −1 for MnAl-V steel, the temperature reaches more than 60 °C, while for MnAl steel, it is only above 45 °C. The large increase of temperature for MnAl-V steel accelerates localization of strain in the necking and causes that the limits strain is much smaller than for MnAl steel. For the smallest strain rates, the limit strain for both materials is similar. The temperature increases for both materials are almost similar and amount to about 40. Then deformation process is very long and it can be assumed that the conditions are almost isothermal.
The course of the strain hardening rate (δσ/δε) in the function of true strain for both examined steels has been shown in Fig. 10.
The strain hardening rate curves for both steels differ slightly. For the MnAl steel four stages can be distinguished, as it is described in the literature for the most of TWIP steels [16,21]. In the stage I, there is an intensive reduction of the hardening rate, which is observed in all materials and is related to a transition from the elastic to plastic region [31]. This stage has a very small strain range; the microstructure should be close to the initial one, with visible slip lines and without any other dominating deformation mechanisms. In this stage, the deformation should be very homogeneous.
The beginning of the second stage corresponds to the presence of stacking faults and generation of the first twins in the deformed microstructure. The third stage corresponds to an increasing intensity of the twins generation in the several systems and their intersections. The fourth stage corresponds to the stabilization of mechanical twins generation in the microstructure.
For the steel with vanadium microaddition, the strain hardening rate curves can be divided into only three distinct stages. The stage II for this steel became so short and it is in fact unnoticeable. However, on the curve obtain for the strain rate of 0.1 s −1 we found the bulk. In the literature [21], we can read that the bulk could be connected with the increasing temperature of sample which effect on the increasing the SFE. Moreover, such an effect visible on the dσ/dε curve could be connected with the interactions the precipitates VC with a dislocations, grain boundaries, and another high angles boundaries like twin boundaries or shear bands. In addition the bulk can be connected with the phenomenon of stronger actions of grain refinement (due to V microaddition) according to the strain rate increasing. A lack of stage II should be bound in steel with V microaddition with faster generation of mechanical twins, which also can be combined with the effect grain refinement and the influence of both solution hardening and the influence of fine carbides precipitates (VC) on this fact. In the tested steel, on the base of Scanning Transmission Electron Microscopy (STEM) technique with Energy-dispersive X-ray spectroscopy analytical technique, VC precipitations can be clearly observed in the microstructure, as shown in Fig. 11.  Fig. 12 presents diagrams that combine the true stress -true strain curves with the strain hardening rate changes and the courses of the average and maximum temperature in the vicinity of the fracture for both examined steels tested at various strain rates. The corresponding series of thermograms recorded during the tensile tests are presented in Fig. 13.

Temperature measurement results
It can be seen in Fig. 12 that the temperature rises almost linearly during the uniform elongation and a significant temperature increase starts at the moment of strain localization (necking). The temperature reached more than 200 °C at the fracture surface of the MnAl steel sample deformed at the highest strain rate, whereas the differences between the average temperature of the fracture area and the maximum temperature reached even 100 °C (Fig. 12c).
In all the thermograms there is temperature increase at the moment of deformation localization; for the lowest strain rate of 0.001 s −1 for steel MnAl, it takes place with the strain of about 0,5 (Fig. 13a), while for steel MnAl-V -with the strain of 0,6 (Fig. 13d). It can be stated that according to the literature, a vanadium addition increases the deformability [32]; however, the results presented in the literature are ambiguous. For the higher strain rates, this relation becomes reversed and for the steel with the vanadium addition, when it is in a solid solution, the deformability becomes reduced. The higher the strain rate, the larger difference in deformability of both steels, e.g. for the highest applied deformation rate, 0.1 s −1 for steel MnAl, the localization occurs with the strain of about 0,42 (Fig. 13c), whereas for steel MnAl-V -with the strain of 0,22 (Fig. 13f). It seems that this fact can be related to the vanadium addition; in the case when it is precipitated in the form of VC particles, it will lead to precipitation hardening and then, according to the literature description, to a deformability increase [17]. In turn, when vanadium remains in a solid solution, it lowers the thermal conductivity. It causes a bigger temperature increase for high strain rates, as the heat is more slowly removed, which leads to temperature localization in a smaller area. This is especially visible for the highest strain rate 0.1 s −1 , when, for steel MnAl-V, the heat is concentrated in a very narrow area around the fracture (Fig. 13f).

Basic parameters affecting the limit of deformation in press forming processes -the hardening coefficient and the strain rate sensitivity coefficient
In the future, TWIP steels could be used as construction materials for car body, which are mainly shaped in deep drawing. The formability of such materials is mainly determined by the hardening coefficient -n as well as the strain rate sensitivity coefficient-m. The higher their values, the easier is to obtain complicated drawpieces without the occurrence of cracking during their production. That is why, it is very important to assume a clear definition of these coefficients. In the case of the hardening coefficient-n, the most literature studies [19,20] do not provide a way of its determination. Based on the included diagrams, it could be concluded that it was calculated based on the Hollomon power equation [33]. In such a case, the curve schematization is very imprecise and the hardening coefficient value n is overestimated, depending on the yield point level.
More accurate values of n are determined with the application of the Ludwik equation [34], that much better represents the actual hardening of the material. The values It should be emphasized that the determined hardening coefficient values at the level of about 0,4 are much higher than for conventional steel, which proves strong reinforcement of the material and delayed deformation (4) p = p0 + C p n localization. Based on the obtained relation, one can conclude that together with a deformation rate increase, the hardening coefficient becomes lower for alloy MnAl-V from 0,39 with 0.001 s -1 , through 0,35 with 0.01 s -1 , to 0,34 with 0.1 s -1 , whereas for alloy MnAl -from 0,42, through 0,3, to 0,28, at the respective strain rates (Fig. 14). This seems to be consistent with the results obtained by other researchers [19,20] and it is explained by the increase of the sample temperature as a result of a change in the plastic deformation work into heat. This phenomenon is especially visible in TWIP steels, which exhibit much higher limit deformations, strength and lover conductivity coefficient than conventional steels.
Another very important parameter affecting the deformation limit in press forming processes is the strain rate sensitivity coefficient expressed by Eq. (5).
where: 1, 2 -flow strass for strain rates ̇1, 2 The higher its value, the lower the material's susceptibility to cracking, e.g. materials with superplastic properties have their coefficient m above 0,3.
For steel MnAl, the strain rate sensitivity coefficient for the deformation of 0.1 equals 0.015, whereas for the deformation of 0,45, it is negative and equals -0.01. For steel MnAl-V, with the deformation of 0.1, it equals 0,05, while with the deformation of 0,25, it is 0,048. For steel MnAl, a very low strain rate sensitivity coefficient, and even its negative value for high deformations, can make difficult the production of complicated drawpieces from the examined steels. MnAl-V steel has higher strain rate sensitivity coefficient, i.e. at the level of 0,05, where it is almost constant in the whole range of the obtained deformations. It should be considered that it is the vanadium addition that can have such a positive effect in the case when it is dissolved in a solid austenite solution.

Microstructure analysis
In the microstructure of the examined MnAl steel, after static tensile tests, effects of deformation are revealed in the form of austenite grains elongated towards the tensile force direction, inside which the formation of mechanical twins and slip bands occur (Fig. 15). A distinct influence of the strain rate on the microstructure can be observed. For the strain rate of 0.001 s −1 , both the glide lines and the twins in deformed grains usually nucleate in one glide and twinning system ( Fig. 15a and d). The strain rate increase results in intensification of the microstructural effects in the form of initiation of subsequent glide and twinning systems (Fig. 15 b,e and c,f). For the strain rate of 0.01 s −1 , it is characteristic to observe the formation of single twin bundles in one twinning system (Fig. 15 b,e). The strain rate of 0.1 s −1 leads to the activation of several glide and twinning systems as well as mutual intersection of the mechanical twins. The microstructural observations of the MnAlV steel samples revealed similar effects of strain rate increase as in the case of MnAl steel (Fig. 16). Near the fracture of a sample deformed at strain rate of 0.001 s −1 , the formation of a slip line as well as primary mechanical twins in one system were observed (Fig. 16a, d). For the strain rate of 0.01 s −1 , effects of the initiation of a subsequent twinning system occurs in single grains (Fig. 16 b, e). The strain rate of 0.1 s −1 leads to the activation of many glide and twinning systems as well as intersection of the mechanical twins and the twin bundles.
The effects of microstructure evolution at higher strain rates should be related to the temperature increase generated as a result of a change of the deformation work into the heat in a deformation zone. This is suggested by the study [21], which analyzes the changes in the deformation mechanism of manganese steels caused by a temperature increase to Transmission microscopy images for two strain rates revealed that for the lowest strain rate, where the average temperature equalled about 40 ºC, the deformation twins are dominant for both materials (Fig. 17a and  c). There are many mechanical twins in two twinning systems visible and first microbands as an effect of matrix-twins connections are formed [35]. For the highest strain, where the average temperature reached about 100 ºC and the maximum exceeded 150 ºC, one can see a lower participation of twinning as well as noticeable effects pointing to an evolution of the dislocation structure in form of dislocation cells creation as well as non-crystalographic micro bands forming. (Fig. 17b and  d). For steel MnAl-V, rebuilding a dislocation structure and microbands creation in the non-twinning matrix are observed [36]. This fact could be an effect of the increase of the stacking fault energy together with the temperature increase caused by the plastic deformation as well as a higher stacking fault energy for the vanadium containing steel. This can also have an effect on other courses of the curve describing the change in the strain hardening rate of the material in the function of strain for the vanadium containing steel in respect of the steel without vanadium (Fig. 10).

Fracture features observation
After the performed tensile tests, fractographic examinations were conducted on the formed fractures (Figs. 18 and 19). Ductile fractures are observed for the both steels, regardless of the applied strain rate. On the fracture surfaces, characteristic dimples and craters can be observed. The elongated dimples on the fracture surface of the sample deformed at the strain rate of 0.1 s −1 , can also be identified (Fig. 18c). The common feature, regardless of the deformation state, is the presence of a well-developed cracking surface. Figure 20 shows selected fracture images with the examples of measurements of the craters diameter for both steels. Taking into account the diameter of the created craters, it is similar in both steels, despite the smaller grain size of steel with V. The measurements of the average craters diameter on several microphotographs indicate that the diameters value are even slightly higher in the MnAl-V steel.

Summary
During the deformation of the TWIP steel, a significant increase of the temperature in the material due to high plastic deformations as well as a high level of yield stress are observed. In the necking area, for the highest strain rate 0,1 s −1 , at the moment of rupture, the temperature reaches over 200 °C, and the difference between the average temperature in the rupture area and the maximum temperature equals as much as 100 °C. In all the thermograms, a rapid temperature increase at the moment of the deformation localization is observed. The heat generated during the plastic deformation of TWIP steels has higher values than in the case of conventional steels, due to their higher yield stress, deformation and lower conductivity coefficient.
The temperature distribution is significantly affected by the vanadium addition, when, being in a solid solution, it lowers the thermal conductivity and leads to a bigger temperature increase for high strain rates. The heat is then removed more slowly, which leads to temperature localization in a smaller area. This is especially visible for the highest deformation rate 0,1 s −1 , where, for steel MnAl-V, the heat is concentrated in a very narrow area around the fracture. The temperature increases have a very significant effect on the properties and microstructure of the deformed materials.
In the future, TWIP steels are to be used as construction materials for car body, which are mainly shaped in deep drawing. The formability of such materials is mainly determined by the hardening coefficient-n as well as strain rate sensitivity coefficient-m. The obtained values of hardening coefficients at the level of about 0,4 are much higher than for conventional steel, which proves strong material reinforcement as well as delayed deformation localization, and this positively affects the deep drawing formability of these materials. Together with an increase of the strain rate, the work hardening exponent decreased for alloy MnAl from 0,39 with 0,001 s −1 , through 0,35 with 0,01 s −1 , to 0,34 with 0,1 s −1 , whereas for alloy MnAl-V -from 0,42, through 0,3, to 0,28, at the respective strain rates, which should be explained by the temperature increase of the samples as a result of a change of the plastic deformation work into heat.
The strain rate sensitivity coefficient of steel MnAl has very low, and even negative, values, which can make the production of complicated drawpieces from the researched steels difficult. Higher values of the strain rate sensitivity coefficient are exhibited by steel MnAl-V, i.e. at the level of 0,05, which is almost constant in the whole range of the obtained deformations. It should be considered that it is the vanadium addition that can have such a positive effect in the case when it is dissolved in a solid austenite solution.
In the analysed MnAl steel with the 0.1% V addition the strain-induced twinning effect is still the main plastic deformation mechanism in a whole investigated range of deformation parameters. The activation of several slip and twinning systems as well as mutual intersection of the deformation twins was observed for all strain rates. V microaddition affects the initial grain size during solution annealing by inhibiting the austenite grain growth.
From the point of application of the both steels, MnAl-V has better ductility for the deep drawing than MnAl at very low strain rate, then there is so large heat concentrated in a very narrow area of strain localization. However the MnAl has better ductility for the larger strain rate over 0.1 s −1 .
For the examined steels, effects of microstructure reconstruction were observed for different strain rates, which should be related to the temperature increase of the material as a result of the change of the deformation work into heat in the deformation zone. For the highest deformation rate, where the average temperature reached about 100 ºC and the maximum temperature exceeded 150 ºC, a significant participation of twinning as well as visible effects pointing to reconstruction of the dislocation structure are observed. For steel MnAl-V, there are even subgrains, in which no deformation twins are present. This can be explained by the increase in the stacking fault energy together with the temperature increase caused by the plastic deformation as well as a slightly higher stacking fault energy for the steel containing vanadium.
Funding Work carried out within the framework of the project UMO-2019/35/B/ST8/02184 " Effect of the heat generated during deformation at high strain rates on the structure and properties of high manganese steels with twinning as the dominant deformation mechanism" financed by The NCN.

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.
Code availability Not applicable.

Declarations
Ethics approval Not applicable.

Conflicts of interest
The authors have no relevant financial or nonfinancial interests to disclose.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.