Influences of Strain on the Microstructure and Mechanical Properties of High-Carbon Steel

Abstract

The effects of strain on the microstructure and mechanical properties of 0.81C-0.22Si-0.31Mn (wt%) high-carbon steel were investigated by thermal simulation, scanning electron microscopy, high-resolution transmission electron microscopy (HRTEM), and an electron backscatter diffractometer (EBSD). It was found that when the steel was deformed at 670 °C (a temperature between A₁ and A_r1), a deformation-induced pearlite transformation and cementite spheroidization occurred. The volume fraction of pearlite and the spheroidization ratio of cementite increased with a strain increase from 20% to 75%. The microstructure mainly consisted of pearlite when the deformation strain exceeded 40%. The aspect ratio was at its maximum (5.3) at 40% strain and decreased to 1.4 at 75% strain. In addition, the strength of the steel decreased and the elongation increased rapidly with the increase in strain from 20% to 60% due to the spheroidization of cementite. However, as the strain further increased to 75%, the strength increased slightly due to the refinement of the ferrite matrix. The comprehensive performance of the investigated steel can be improved by applying a strain between A₁ and A_r1.

1. Introduction

High-carbon steels have excellent comprehensive performance and are widely used in aerospace, automobile engineering, etc. [1,2]. The microstructure of hot-rolled high-carbon steel is generally dominated by flaky pearlite, which has high strength and low elongation [3]; thus, it is difficult to use hot-rolled high-carbon steels directly in engineering applications. In order to improve the comprehensive performance of high-carbon steels, they are generally treated by spheroidizing annealing after hot rolling [4,5]. Flaky pearlite particles are transformed into micro-ferrite and nano-granular cementite after spheroidizing annealing [6]. However, spheroidizing annealing generally takes a long time, which greatly increases the production time and cost of high-carbon steels [7]. Therefore, how to realize the online spheroidization of high-carbon steels during hot rolling has been one of the main research issues in recent years.

It is reported that deformation-induced pearlite transformation (DIPT) and cementite spheroidization (CP) can cause the transformation and spheroidization of flaky pearlite in a short time during hot rolling [8,9]. Previous studies on the DIPT and spheroidization of high-carbon steels have mainly focused on the deformation rate and temperature [10,11,12,13]. It is known that the deformation rate has a significant effect on pearlite transformation and cementite spheroidization. Normally, with an increase in deformation rate, pearlite transformation and cementite spheroidization accelerate [14,15]. In addition, the microstructure of high-carbon steels depends on the deformation temperature. When the deformation temperature is high, austenite is transformed into a ferrite matrix and lamellar pearlite, whereas at a low deformation temperature, austenite is transformed into a ferrite matrix, pearlite and grain-boundary cementite [16]. A composite structure of the ferrite matrix and finely disseminated cementite can be obtained at a deformation temperature near the pearlite transformation temperature [17]. Moreover, it is reported that with the increase in strain, DIPT and CP accelerate in medium-carbon steels [18,19,20]. However, the effects of strain on the DIPT and CP of high-carbon steels are rarely explored. Therefore, in the present study, the DIPT and cementite spheroidization of 0.81C-0.22Si-0.31Mn (wt%) high-carbon steel were investigated. The effects of strain on the microstructure and mechanical properties were discussed to provide a theoretical and experimental basis for the production of low-cost high-carbon steels.

2. Materials and Methods

The investigated steel was melted in a vacuum induction furnace and cast into a 50 kg ingot with a diameter of 150 mm. The ingot was then heated to 1200 °C for two hours in a furnace, and subsequently hot-rolled into a plate of 60 mm thickness and 100 mm width on a laboratory hot-rolling mill. The steel plate was cut into billets of 250 × 100 × 60 mm (length × width × thickness). The billets were heated to 1200 °C for 4 h and then cooled to room temperature so as to make the elements dissolve evenly in the steel. Finally, the billets were machined into thermal simulation samples of 8 mm in diameter and 12 mm in length. The chemical composition of the investigated steel is presented in

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

C	Si	Mn	P	S	N	Fe
0.81	0.22	0.31	0.011	0.003	0.004	Bal.

.

The influences of deformation strain on the microstructure and mechanical properties of the investigated steel were studied with a THERMECMASTER-Z thermal simulator. Axisymmetric compression deformation was carried out on the specimens. The thermomechanical processing route is schematically presented in Figure 1. The A₁ temperature (the pearlite transformation temperature at equilibrium) of the investigated steel was calculated as 720 °C by JmatPro software, and the A_r1 temperature (pearlite transformation temperature during cooling) was measured as 610 °C by thermal simulations at a cooling rate of 20 °C/s after austenitizing. The samples were first heated to 1150 °C at 20 °C/s, then kept at this temperature for 300 s, and finally cooled to 670 °C (a temperature between A₁ and A_r1) at 20 °C/s. The samples were then deformed to 0, 20, 40, 60 and 75% strains at a strain rate of 0.01 s^–1 and cooled to room temperature at 50 °C/s. The dilation data for these different paths were recorded.

The dog-bone-shaped specimens, with a gauge dimension of 10 mm × 5 mm × 2 mm (length × width × thickness), were machined and tensiled at room temperature with a strain rate of 2 × 10⁻¹ s⁻¹ using an Instron 8801 universal testing machine (Instron Limited, High Wycombe, Britain). The direction of the tensile specimen was perpendicular to the deformation direction. Duplicate tests were carried out to ensure the accuracy of the experiments. The microstructures of different samples were revealed by a field-emission scanning electron microscope (FE-SEM) at an accelerating voltage of 20 kV. Electron backscatter diffraction (EBSD) was employed for crystallographic characterization and grain size measurement. The cross-section of the tensile specimen was the observation surface of the SEM and EBSD microstructure. The EBSD specimens were electro-polished in an electrolyte consisting of 10 vol% of perchloric acid and 90 vol% of glacial acetic acid at a voltage of 40 mV, and a step size of 0.2 µm was applied to obtain high-quality images. The transmission electron microscope (TEM) was employed to observe the fine microstructure and precipitates. Slices for the TEM were cut from the thermal simulation samples, and were subsequently ground to less than 70 µm and punched into 3 mm discs. Thin foils for TEM observation were prepared by twinjet polishing with an electrolyte solution consisting of 6% perchloric acid and 94% ethyl alcohol below −30 °C. The surface of the tensile specimen was the observation surface of the TEM microstructure. In addition, the TEM observation was carried out using a JEM-2100 field-emission high-resolution transmission electron microscope (HRTEM) operated at a voltage of 200 kV.

3. Results

3.1. Microstructure

The microstructures at different strains are exhibited in Figure 2.

It is observed that when the steel was directly cooled to room temperature at a cooling rate of 50 °C/s without deformation, the microstructure was composed of martensite (Figure 2a). Under deformation with 20% strain, partial DIPT occurred, and the microstructure mainly consisted of pearlite (red arrows) and martensite (yellow arrows) (Figure 2b). Martensite was formed during the cooling process after deformation, and lamellar pearlite appeared during DIPT. When the deformation strain was 40%, a large amount of DIPT appeared during the deformation of the austenite, and lamellar pearlite (red arrows) together with some spherical ones (white arrows) were formed (Figure 2c). When the strain increased to 60%, spherical pearlite (white arrows) with a small amount of short lamellar cementite (red arrows) appeared (Figure 2d). When the deformation strain further increased to 75%, the microstructure was composed of uniform and spherical pearlite grains (Figure 2e).

The TEM microstructures at different strains are presented in Figure 3. It is observed that at 20% strains, the microstructure was dominated by flaky pearlite containing small cementite particles. The vertical distance between two flaky pearlites is the lamellar spacing of pearlite. The Image-Pro software was used to measure the pearlite flake spacing in Figure 3a, and the average lamellar spacing of pearlite was 163 nm. When the strain increased to 40%, pearlite particles deformed and a small proportion of the cementite bent (the white-dotted line in Figure 3b) or fractured (the red-dotted line in Figure 3b). At 60% strain, cementite was distributed as particles or short rods (the red circles in Figure 3c), and when the deformation strain increased to 75%, carbon atoms re-precipitated in the supersaturated ferrite matrix in the form of cementite (Figure 3d). Moreover, two types of cementite particles (fused and spheroidized) were detected in the microstructure. Large spheroidized particles (type 1 in Figure 3d) were mainly distributed on the ferrite matrix grain boundaries, and their average diameter was 129–427 nm, while smaller cementite particles with an average diameter of 87–155 nm (type 2 in Figure 3d) were distributed in the ferrite matrix.

The transformation fraction of pearlite and the aspect ratio of cementite at different strains are presented in Figure 4 and Figure 5, respectively. Image-Pro software was used to calculate the phase proportion of the different samples. It is clear from Figure 4 that the fraction of pearlite transformation increased significantly with the increase in strain from 0 to 40%. The microstructure mainly consisted of pearlite when the deformation strain exceeded 40% with the proportion of pearlite more than 97%. Figure 5 reveals that the aspect ratio of cementite decreased rapidly from 40% to 60% strain and then slowly decreased from 60% to 75% strain. The aspect ratio was the maximum (5.3) at 40% strain and decreased to 1.4 at 75% strain. The smaller the aspect ratio, the higher the spheroidization ratio of cementite. The particle size distribution of cementite at 75% strain is displayed in Figure 6. The cementite particles smaller than 400 nm were predominant and accounted for 85.71% of size distribution.

Figure 7 presents the calibration diagram of the bright-field image, dark-field image, and diffraction pattern of the precipitated particles at 75% strain. The precipitated particles were calibrated by connecting adjacent diffraction spots through the central spot of the diffraction pattern to form a parallelogram with its diagonal [21]. The two sides and diagonal of the diffraction pattern were denoted as R₁, R₂, and R₃, respectively (vectors correspond to the diagonal), and the corresponding crystal plane spacings were denoted as d₁, d₂, and d₃, respectively (crystal plane spacing corresponding to the diagonal). The angles between the diagonal and two sides were defined as <R₁, R₃> and <R₂, R₃>, respectively. The crystal plane indices were detected as (31$\overline{1}$), (102), and (411); hence, the precipitated phase could be confirmed as Fe₃C.

3.2. Grain Size and Orientation

An EBSD was performed at the cross-section of the tensile specimens to quantify their mean crystallographic unit size. The approximate surface area of the microstructure was 30,000 µm². Figure 8 displays the variation in crystallographic orientation and grain size distributions at different deformation strains. It is observed that the structure of the (101) crystallographic orientation increased significantly with the increase in deformation strain. The fractions of (101) crystallographic orientation at 20, 40, 60 and 75% strains were 11.8, 22.3, 29.3 and 40.9%, respectively. It is seen that the crystallographic orientation of the ferrite matrix was not random at high strain. When analyzing the grain size map, the critical orientation angle was set to be 10°, and the pixel pairs exceeding the critical orientation angle were identified as the grain boundaries. Hence, the areas of each grain can be measured, and then transferred to a circle of equal area. The grain size of each grain is the diameter of the equal area circle. The grain size of the larger strain samples (such as 60 and 75% strain) should be replaced by the width of the grain. It is obvious that the average grain size of the ferrite matrix decreased gradually with the increase in deformation strain (Figure 8a’–d’). The average grain sizes of the ferrite matrix were 10.43 and 3.55 µm at 20% and 75% strains, respectively, indicating that the grains were refined at a larger strain.

Figure 9 displays the ferrite matrix grain size distributions at different strains. When the deformation strain increased from 20% to 75%, the grain size distribution curve moved to the upper left corner, indicating that the ferrite matrix was refined gradually with the increase in deformation strain.

Figure 10 presents the grain boundary distribution maps at different strains, where large-angle grain boundaries (≥15°), small-angle grain boundaries (5–15°), and sub-angle grain boundaries (2–5°) are indicated by blue, green and red lines, respectively. The lengths of the large-angle grain boundaries at 20, 40, 60 and 75% strains were 1.20, 1.61, 2.14 and 2.55 cm, respectively, further proving that the ferrite matrix was refined gradually with the increase in strain.

3.3. Mechanical Properties and Fracture Surface

Figure 11 and

Table 2. Mechanical properties at different strains.

Strain (%)	20	40	60	75
Yield stress (MPa)	1297 ± 13	734 ± 8	683 ± 15	728 ± 6
Tensile stress (MPa)	1443 ± 11	898 ± 13	802 ± 7	883 ± 9
Elongation (%)	3.1 ± 0.3	4.8 ± 0.2	15.3 ± 0.5	14.5 ± 0.7

present the tensile test results of the investigated steel at different strains. The yield and tensile strengths first decreased and then increased with the increase in strain, whereas the total elongation first increased and then decreased. At 20% strain, the yield (1297 ± 13 MPa) and tensile strengths (1443 ± 11 MPa) were found to be the largest. The maximum total elongation (15.3 ± 0.5%) was obtained at 60% strain. This indicates that an improved drawability can be achieved. This is also evidence of an enhanced formability after deformation.

After testing, the representative fracture surfaces are examined and illustrated in Figure 12, which exhibits the fracture morphologies of the tensile specimens at the macroscopic and microscopic levels with different strains. It is noted that the fractographs at 20% and 40% strains (Figure 12a’,b’) predominantly comprise cleavage sheaths (shown by the yellow arrows) along with a few dimples (shown by the red arrow). There are some dimples, but the presence of a higher amount of cleavage sheaths clearly indicates that the specimen predominantly exhibited a brittle fracture, while the fractographs at 60% and 75% strain reveal the formation of plentiful dimples (shown by the red arrows in Figure 12c’,d’). It is clear that the tensile fractures were dominated by ductile fractures at 60% and 75% strain. When the deformation strain increased from 20% to 75%, dimples in the fractured specimens became smaller and more uniform with the increase in strain.

4. Discussion

The microstructure of supercooled austenite mainly experienced pearlite dynamic transformation and pearlite dynamic spheroidization during the deformation process (Figure 2, Figure 3 and Figure 4). The expansion curves during the cooling process after deformation at 20% to 75% strains are presented in Figure 13. The straight line during cooling shows that no pearlite transformation occurred during the cooling process, indicating all the pearlite was formed during the deformation process. The deformation temperature was 670 °C, which was in the range of A₁ and A_r1. Pearlite transformation could not occur without deformation in this temperature range [22]. During the deformation process, stress was applied; thus, internal energy was generated in the deformed steel. A small proportion of this energy was transferred to the environment in the form of heat energy, and the remaining energy existed in dislocations and defects in the form of deformation energy [23]. Deformation generated high internal energy in the steel, making the material unstable. In order to restore the equilibrium state, the steel tended to spontaneously shift to a lower energy state; therefore, the deformation energy increased the driving force for phase transformation and caused pearlite transformation [24]. In addition, deformation increased the dislocation density, the number of nucleation sites and the nucleation rate, and consequently accelerated the pearlite transformation. Therefore, when the deformation strain increased to 20%, DIPT occurred during the deformation process.

The internal stress in the microstructure was nonuniform when the stress was small in the initial stage of deformation; hence, the stress concentration occurred near the original austenite grain boundaries, and subsequently [25] the stored deformation energy at these grain boundaries increased [26]. Therefore, when the deformation strain was small, the phase transformation of pearlite preferentially occurred at the grain boundaries. As strains increased and deformation gradually extended to the inside of the austenite, the driving force for phase transformation was increased [27]. Subsequently, the density of internal defects in the austenite increased; thus, the pearlite volume fraction increased with strain (Figure 2 and Figure 4).

With the increase in strain, pearlite particles became spheroidized (Figure 2 and Figure 5). Two types of granular cementite (type 1 and type 2) were formed during spheroidization (Figure 3d). As the strain increased, due to the difference in the mechanical properties of the ferrite matrix (soft phase) and the cementite (hard phase) in pearlite, the deformation of these two phases did not occur in a coordinated state; thus, the cementite lamellar fractured and bent. The curvature radius in the sag areas formed after the fracture of cementite lamellar was significantly smaller than that in the flat areas, causing higher carbon concentration near the smaller curvature radius and forming a carbon concentration gradient at the phase interface [28]. Carbon atoms diffused from the high-carbon zone to the low-carbon zone at the phase interface, promoting the dissolution and spheroidization of cementite lamellar and forming coarse cementite particles on the ferrite matrix boundary (type 1) (Figure 3d).

When the formation energy of the carbon atoms was lower than the binding energy of carbon atoms with dislocations, the carbon atoms dissolved in the dislocations; thus, the dissolution of cementite occurred, and the carbon atoms that had combined with the dislocations dissolved in the ferrite matrix to form a Cottrell gas mass. As the deformation strain increased, dislocations continued to move, and the ferrite matrix recovered and recrystallized [29]. As the dislocation density decreased, the carbon atoms that had dissolved in the dislocations precipitated in the form of fine cementite and were distributed in the ferrite matrix (type 2) (Figure 3d).

The spheroidization of cementite was a thermal activation process [30]. The internal energy and the dislocation density both increased with strain. The generation of a large number of dislocations promoted the diffusion of carbon atoms and accelerated the dissolution and reprecipitation of carbon atoms; hence, when the deformation strain increased from 40% to 60%, the spheroidization rate of cementite rapidly increased. With the further increase in strain, the increase in internal energy and the dislocation density tended to be saturated; thereby, the spheroidization rate of cementite increased slowly as the strain increased from 60% to 75% (Figure 4).

The ferrite matrix grain size is fine (see Figure 9) due to the uniform dispersion of the fine cementite particles in the microstructure. These cementite particles pin the grain boundaries (i.e., Zener drag effect [31]) and drastically reduce the mobility of the grain boundaries, thus inhibiting the recrystallization process [32]. Hence, it is obvious that the ferrite matrix grain size decreases with the increase in the spheroidization ratio of cementite. The results of the EBSD (Figure 8, Figure 9 and Figure 10) also indicated that the ferrite matrix was refined gradually as the deformation strain increased from 20% to 75%. The refinement of the ferrite matrix was beneficial for improving the comprehensive performance of the investigated steel.

It is noticeable from Figure 11 that the tensile and yield strengths of the steel decreased rapidly, and the total elongation increased as the deformation strain increased from 20% to 40%. When the strain was 20%, partial DIPT occurred in the steel, and the microstructure mainly consisted of lamellar pearlite and martensite. When the strain reached 40%, more DIPT occurred, and the microstructure consisted of only lamellar pearlite (Figure 2). In comparison to martensite, lamellar pearlite had lower strength and better plasticity [33]; therefore, the strength decreased rapidly, and the total elongation increased. Furthermore, with the increase in strain from 40% to 60%, cementite particles became spheroidized. The larger the deformation strain, the higher the proportion of cementite spheroidization. On the one hand, in comparison to lamellar pearlite, granular cementite has lower strength and better plasticity [17]; hence, the strength continued to decrease, and the total elongation increased. On the other hand, the refinement of the ferrite matrix was beneficial for increasing the strength and elongation of the investigated steel. Therefore, when the strain increased from 40% to 60%, the strength decreased slightly, while the elongation increased significantly. When the strain reached 60%, the volume fraction of pearlite and the spheroidization rate of cementite tended to be saturated (Figure 5). As the deformation strain further increased to 75%, it had a higher spheroidization rate and a finer ferrite matrix. On the one hand, in comparison to martensite and lamellar pearlite, the higher spheroidization rate had lower strength and better plasticity. On the other hand, the finer ferrite matrix was beneficial for increasing the strengths and elongations of the investigated steel. Therefore, the comprehensive properties of the steels were affected by the strain amount. When the deformation reached 75%, the better comprehensive properties were obtained.

In addition to strength and ductility, toughness is another important index of steel performance. The toughness is normally estimated by Charpy impact tests, but the impact tests could not be performed due to the thinness of the samples. The area encompassed by the stress–strain curve was used to present the toughness [34,35]. The toughness value at 75% strain was higher than others, as shown in Figure 11. This is because the spheroidized structure suppresses the stress concentration and as a result delays the onset of fracture, thus increasing both ductility and toughness. It is expected that toughness increases with the increase in strain due to enhanced spheroidization. It is also the reason why the fracture of the tensile specimen changes from a brittle fracture to a ductile fracture with the increase in deformation strain.

5. Conclusions

The effects of strain on the microstructure and mechanical properties of high-carbon steel were investigated. The main conclusions can be drawn as follows.

(1) When the deformation temperature was between A₁ and A_r1, deformation-induced pearlite transformation took place as the deformation strain increased to 20%. The volume fraction of pearlite increased with strain. When the deformation strain exceeded 40%, the microstructure mainly consisted of pearlite.

(2) When the strain was less than 40%, the microstructure of the investigated steel was mainly composed of lamellar pearlite. With the increase in strain, the spheroidization of cementite occurred, the aspect ratio of cementite decreased rapidly from 40% to 60% strain, and then slowly decreased from 60% to 75% strain. The ferrite matrix was refined gradually and the structure of the (101) crystallographic orientation increased significantly as the deformation strain increased from 20% to 75%.

(3) The yield and tensile strengths of the steel decreased, and the total elongation increased as the deformation strain increased from 20% to 60%. As the strain further increased to 75%, the strength increased slightly. The strain can effectively improve the comprehensive properties of high-carbon steels.