Dynamic recrystallization behavior and processing maps of 5CrNiMoV steel during hot deformation

Hot deformation tests of 5CrNiMoV steel were performed at deformation temperatures of 700 to 870 °C and strain rates of 0.001 to 0.1 s−1 using the DIL 805D thermomechanical simulator. The critical strain and volume fraction models of Dynamic Recrystallization (DRX) were constructed based on the work hardening theory. The results showed that the critical strain of DRX decreases with increasing deformation temperature and decreasing strain rate, which implies that DRX occurs easily at higher temperatures and lower strain rates. The average DRX grain size model was established to predict grain size changes during hot deformation. Based on the hot processing maps that were constructed using the Dynamic Material Model (DMM) and microstructure observation, the optimum hot working parameters for 5CrNiMoV steel are a deformation temperature of 800 °C–870 °C, a strain rate of 0.001–0.05 s−1.


Introduction
Hot work die steels are mainly used in the production of dies that have to exceed the recrystallization temperature in service, such as hammer forging dies, hot extrusion dies, machine forging dies, and die casting dies [1]. 5CrNiMoV steel is a typical hot work die steel, primarily used for the manufacture of a variety of large forging dies. The contents of Cr, Ni, and Mo are higher than the 5CrNiMo steel, and a small amount of V element is added to it. V is a powerful carbide formation element that is hard to dissolve in solid iron solution and can be used to increase the thermal hardness of the steel, as well as to refine grains and improve wear resistance. Specific deformation parameters determine the final microstructure and performance of the hot work die. Therefore, hot deformation studies are necessary for improving thermomechanical properties, such as thermal stability and oxidation resistance.
By regulating DRX, the high temperature deformation can eliminate a number of defects, effectively refine the prior austenite grains and martensitic structure, adjust the grain distribution and improve the mechanical properties of materials [2,3]. The stored energy of deformation increases continuously as the deformation of materials increases. DRX occurs when energy accumulates within a certain quantity range. DRX grains nucleate on the initial grain boundaries, swallow the surrounding grains and grow up depending on the migration of the grain boundary until the original grains are swallowed by the DRX grain. Eventually, the deformed structure disappears, resulting in uniform and refined DRX grains. In this process, elevated temperatures can facilitate the movement of grain boundaries [4]. While the increased strain rate will lead to the rapid generation of dislocations, and insufficient time for nucleation and growth of DRX grains [5]. The DRX volume fraction and grain size increase as the temperature increases and the strain rate decreases [6,7]. Poliak et al [8] proposed an effective method to determine critical DRX conditions by analyzing the relationship between strain hardening rate ( d d q s e = / ) and flow stress (s), which is known as critical strain ( c e ) [9]. The hot processing map is a new method to study the hot deformation behavior of materials. It consists of a power dissipation map and an instability diagram, which can be used to optimize the hot deformation parameters by combining the region of maximum power dissipation with the microstructure analysis. That's why the hot processing map is widely used in aluminum alloys [10,11], titanium alloys [12,13], superalloys [14,15], and other metallic materials. In the hot processing map, different regions with their respective hot deformation mechanisms can be divided into dynamic recovery (DRV), DRX, superplastic deformation, flow localization, adiabatic shear band, mechanical twin, etc [16][17][18].
Over the past few decades, many researchers have studied the recrystallization behavior of metallic materials and optimized the parameters of the hot deformation process using hot processing maps. Nie et al [19] studied the DRX behavior of 1MnCrMoNi steel, established the DRX model and hot processing maps, and determined the optimum hot deformation process parameters for 1MnCrMoNi steel. Song et al [20] investigated the hot compression tests of high-Ti low-C microalloyed steel at different temperatures and strain rates, studied the hot deformation behavior and the processing maps. Wang et al [21] studied the effects of homogenization and strain rate on the microstructure evolution, mechanical behavior and DRX behavior related to Al6Mn phase and strain rate of Al-6Mg-0.8Mn alloys at 450°C with strain rates ranging from 0.001 s −1 to 1 s −1 . Zhou et al [22] performed hot compression tests on 25CrMo4 steel, obtained the critical strain model, DRX model and processing maps. The effect of strain on the dissipation factor was analyzed and the optimum hot deformation parameters for 25CrMo4 steel were obtained by combining with microstructure analysis.
Most studies have generally concentrated on the relationship between deformation conditions and microstructure by analyzing flow stress curves. The recrystallization model was designed to predict the evolution of DRX from hot deformation tests with a wide range of temperatures and strain rates. Although it has great significance for practical production, there are relatively few relevant DRX behaviors and hot processing maps available for 5CrNiMoV steel.
In this article, hot deformation tests of 5CrNiMoV steel were performed at various temperatures and strain rates using the DIL 805D thermomechanical simulator. The kinetic model of DRX was derived from the flow stress curves, and the hot processing maps were obtained by superimposing an instability map and a power dissipation map. The hot forming parameters optimized for the 5CrNiMoV steel were obtained by combining the processing maps with the observation of the microstructure.

Materials and experiments
The chemical composition (wt%) of 5CrNiMoV steel used in this work is given in table 1. The size of the cylindrical specimens is Φ5×10 mm and machined from a large forging die. The microstructure of the asreceived 5CrNiMoV steel was etched with 4% nitric acid and alcohol. EMPYREAN x-ray diffractometer (XRD) with the target of Cu was used for phase analysis, with a working voltage, current, step-size and scanning angle are 40 KV, 40 mA, 0.025 and 0°-110°, respectively.
The hot compression tests were carried out using the DIL 805D thermomechanical simulator and the test chamber is shown in figure 1(a). An S-type thermocouple (Pt-Pt/Rh 10%) was spot welded in the center of the sample surface and used for closed-loop temperature control. All specimens were heated to an austenitic temperature (870°C) at a heating rate of 2°C s −1 and held for 10 min to obtain a homogeneous microstructure. Subsequently, the specimens were cooled to the predefined deformation temperature with a cooling rate of 40°C s −1 , and the cooling medium was high-pressure helium. The compression tests were performed at strain rates of 0.001, 0.005, 0.01, 0.05, and 0.1 s −1 and temperatures of 700, 750, 800, and 870°C, respectively, with the maximum true strain of 1.1. After deformation, the specimens were cooled to room temperature with a cooling rate of 40°C s −1 . The schematic of hot compression tests is shown in figure 1(b). In order to reveal the grain boundaries, the deformed specimens were sliced along the axial section, abraded with abrasive paper, mechanically polished and etched with a corrosion solution called 'new grain etchant' for two hours. The microstructural observation of deformed specimens was performed using NREEOHY J-X3 optical microscope (OM) and Tescan Vega3 scanning electron microscope (SEM).

Original microstructure
The microstructure of the as-received 5CrNiMoV steel is shown in figure 2(a) and figure 2(b), which consists of ferrite and cementite and the grain size is not uniform, with an average grain size of about 21 μm measured by the intercept method. Figure 2(c) shows the XRD spectra of the as-received steel, only α phase can be found within the matrix.

The critical model of dynamic recrystallization
Softening of metallic materials at elevated temperatures is a prerequisite for hot deformation. Most peaks in the flow stress curve are caused by DRX. The occurrence of a peak is usually used to determine whether DRX has occurred. However, an increasing number of experimental findings have shown that DRX may occur even if there is no peak in the flow stress curve [23]. The critical condition for DRX can be obtained from the inflection point on the q-s curve or minimum value of theq s ¶ ¶ / -s curve [5,24,25]. In addition, the critical stress ( c s ),  peak stress ( p s ), steady stress ( ss s ), saturated stress ( sat s ), and maximum dynamic recrystallization stress ( * s ) can be obtained from the work hardening curve [26]. The relationship between the work hardening rate (q) and the flow stress (s) is shown in figure 3. Generally, with the increase of stress, the work hardening rate values decrease first and then increase, and there are two intersections with zero. The first intersection represents the peak stress ( p s ) and the second represents the steady stress ( ss s ), which imply that the dynamic equilibrium occurs internally when the 5CrNiMoV steel reaches the peak or steady state. The lowest point in the curve is the maximum dynamic recrystallization stress ( * s ).
The minimum point of the d d q s -/ -s curve (figure 4) corresponds to the inflection point on the q-s curve.
The critical stress, peak stress, steady stress, saturated stress, maximum dynamic recrystallization stress and the corresponding strain under different deformation conditions were obtained, as shown in table 2. It can be seen that in all deformation conditions, the critical stress and the corresponding strain are lower than the peak stress and the corresponding strain, which indicates that dynamic recrystallization started prior to the peak stress. Moreover, with the increase of deformation temperature and the decrease of strain rate, the DRX critical strain decreases, that means increasing the deformation temperature or decreasing the strain rate is beneficial to the DRX of 5CrNiMoV steel. There is a linear relationship between the critical strain and the peak strain, with the critical strain being about 0.42 times the peak strain.
To express the linear relationships between characteristic parameters and deformation conditions, they were adapted to the Z parameter. The equation is as follows: Where A, B are the material constants, C is the characteristic parameter, and Z is the Zener-Hollomon parameter. The method for resolving the Z -parameter can be found in [27].
The linear regression results of stress and strain using the Z-parameter are shown in figure 5. The DRX parameters increase significantly with increase of the Z parameters and the correlation between the stress parameter and the Z parameter is higher than that between strain parameter and Z parameter. The values of A and B in equation (1) were obtained based on the slope and intercept of the different fitting lines, and the relationships between the characteristic parameters and the Z parameters for the 5CrNiMoV steel were expressed as follows:  Where X drex is the volume fraction of DRX, c e is the critical strain, p e is the peak strain, k is the material constant and m is the Avrami constant. Meanwhile, the volume fraction of DRX can also be expressed as:  Where wh s and s correspond to the assumed stress resulting from DRV alone and experimental flow stress, as shown in figure 6.
The wh s can be expressed as equation (5)  Where 0 s is the yield stress, which can be calculated directly from the flow stress curve, and W is the parameter related to the shape of the DRV curve. It can be expressed as equation (6).  (6), the value of W can be obtained under different deformation conditions. The relationship between ln W and Z ln is illustrated in figure 7 and can be given as equation (7). Taking the logarithm of both sides of equation (3), it can be written as equation (8), it is easy to obtain the relationship between X ln ln 1  The average DRX grain size model can be expressed as equation (10): Where D drx is the average DRX grain size (μm), a 3 and m 3 are the material constants, Q 3 is the activation energy for grain growth (J/mol), R is the gas constant (8.314J/(mol·K)),  e and T are strain rate (s −1 ) and deformation temperature (K), respectively. Taking the logarithm of both sides of equation ( Figure 10 shows the relationship between the predicted and experimental grain size, it can be seen that the square of the correlation coefficient reaches 0.9813, which indicates that the proposed model can be used to predict the evolution of the average DRX grain size of 5CrNiMiV steel during hot deformation.

Establishment of hot processing maps
According to the dissipative intrinsic theory, the total energy P input from the exterior of metallic materials during high temperature deformation consists of two parts. One is the power dissipation of plastic deformation (G), the other is the power dissipation of microstructure transition (J ) [30,31]. Therefore, the total energy can be expressed as equation (13).
When the deformation temperature is established, the relationship between the flow stress and strain rate can be expressed by the equation (14) [32].
Where m is the strain rate sensitivity coefficient of the material, and k is the material constant. The energy in the evolution of microstructure can be expressed as a dimensionless power dissipation efficiency factor , h expressed as equation (15) [33].
The power dissipation map determined by the dissipation factor alone can't accurately predict the material's safety region and the plastic instability region. Therefore, Prasad [34] proposed a rheological instability criterion, expressed as equation (16).
The relationship between log s and  log e was established by a cubic spline interpolation function based on the experimental data. The expression of the function is indicated in equation (17), and the results are shown in figure 11.  Figure 12 shows the effects of strain rate and deformation temperature on h of 5CrNiMoV steel at strains of 0.5 and 0.8, respectively. There are differences in the power dissipation factor of the 5CrNiMoV steel under different temperatures and strain rates. In the range of 0.005-0.1 s −1 , the power dissipation factor increases with the increase of deformation temperature, indicating that the hot workability will also improve. When the true strain increases from 0.5 to 0.8, the maximum values of h increase from 36.04% to 36.85%, which indicates that the energy consumed by microstructure evolution is increasing, and the percentage of DRV as well as DRX within the alloy are increasing [35]. The minimum values of h are 8.29% and 9.49% for strains of 0.5 and 0.8, respectively. However, higher values of h are not always better, and regions of instability may also appear in the higher h region. Therefore, it is necessary to determine the extent of the regions of instability during hot deformation of metallic material. The power dissipation maps and the instability maps were superimposed to obtain the hot processing maps of 5CrNiMoV steel. As can be seen in figure 13, the colored areas represent the region of instability, and the contour numbers represent the power dissipation factor. The results show that the deformation temperature, strain, and strain rate have a significant effect on the hot processing maps. When the strain increases from 0.5 to 0.8, the instability region has a relatively obvious transformation, and the areas of instability region sharply decrease at elevated temperature and large strain rate. Because the increase of strain leads to the occurrence of DRX sufficiently, which inhibits the occurrence of stress concentration and adiabatic shear band processes. It is generally agreed that when h>30%, DRX occurs more easily and the hot deformation of materials is greater [36]. As shown in figure 13, hot processing maps have a safe processing zone with the power dissipation factor h greater than 30% that corresponds to the high temperatures and medium strain rates (blue frame areas). The dissipation factor values in the instability region are all low, mainly concentrated in the low deformation temperatures, which indicates that the thermal workability within this range is low and flow instability is easy to occur [37]. Therefore, the processing parameters in areas of instability should be avoided.

Microstructure evolution during hot deformation
To verify the accuracy of the processing maps based on Prasad, the microstructure of the 5CrNiMoV steel was observed and analyzed, as shown in figure 14. The power dissipation value is approximately 0.16 when the  deformation parameters were 750°C and 0.1 s −1 , and its microstructure is shown in figure 14(b). The microstructure of the as-received steel was elongated in the direction of deformation and DRX grains begin to appear at the grain boundaries during the deformation process. However, the number of DRX grains is relatively low, the energy consumption is limited and therefore the power dissipation value is low. Figure 14(c) shows the microstructure at a temperature of 800°C and a strain rate of 0.005 s −1 . The as-received grains were almost entirely replaced by DRX grains, leaving only some elongated deformed grains. The recrystallization nucleation process was substantially complete, and the recrystallized grains require energy consumption during the growth process, so the power dissipation value is approximately 0.3. Figure 14(d) shows the microstructure at 870°C and a strain rate of 0.001 s −1 , corresponding to a power dissipation factor of 0.24. The as-received grains have all been replaced by DRX grains and the grain growth is nearly complete, so the energy consumed in the evolution of the microstructure is relatively low and the majority of the energy is converted to heat, resulting in a low power dissipation value.   Figure 15 shows the microstructure of the 5CrNiMoV steel under different deformation conditions. A typical 'necklace structure' can be seen in figure 15(a), this is because DRX does not occur sufficiently at lower deformation temperatures, and there are a large number of dislocation clusters that accumulate at the grain boundaries during hot deformation, increasing the deformation energy and forming a large number of DRX grains at the grain boundaries. It is a typical microstructure that can be seen in DRX of many FCC metals and alloys [7,38]. The inhomogeneity of this microstructure can lead to mechanical instability. Figure 15(b) shows the microstructure that deformed at a temperature of 700°C with a strain rate of 0.05 s −1 . There are numerous localized deformation bands formed by elongated coarse grains. This is because that the short deformation time and a large amount of latent heat during deformation cannot be dissipated, resulting in the local temperature rise and local deformation bands during plastic deformation [39].

Conclusions
In this study, the dynamic recrystallization behavior and processing maps of the 5CrNiMoV steel were investigated using hot compression tests. The main results are summarized as follows: (1)Based on the q-s curves and d d q s -/ -s curves, the critical stress and strain, peak stress and strain, saturation stress, steady stress, and maximum dynamic recrystallization stress under different deformation conditions were obtained. The models for the critical strain of DRX were developed, as follows: The average DRX grain size under different deformation conditions was obtained, which increases with decreasing the strain rate and increasing the deformation temperature. The average DRX grain size model of DRX can be used to predict the DRX grain size of 5CrNiMoV steel during hot deformation, as follows: (4)The hot processing maps of the 5CrNiMoV steel obtained in this article consist of two instability regions. Excluding regions of instability, the appropriate deformation area is between 800 and 870°C, 0.001 and 0.05 s −1 .