Revealing the Role of Pre-Strain on the Microstructure and Mechanical Properties of a High-Mn Austenitic Steel

Abstract

The effects of different pre-strain levels on the dislocation density, twinning behavior, resultant tensile properties, and cryogenic impact toughness of a high-manganese austenitic steel for low-temperature service were investigated. The results indicate that the dislocation density and volume fraction of twins are sharply increased when the pre-strain exceeds 15%, leading to an increase in yield strength and a decrease in impact toughness. At a 5% pre-strain level, few mechanical twins are observed while the dislocation density increases, resulting in enhanced yield strength whilst maintaining the toughness. The dislocation and grain refinement strengthening effects dominate the yield strength at various pre-strain levels. The initial mechanical twins and increased dislocations induced by pre-straining adversely affect the impact toughness. These findings validate the potential of controlling the mechanical twins and dislocations via pre-strain treatment as an effective approach to tailoring the mechanical properties of high-manganese austenitic steel.

1. Introduction

In recent years, under the world “dual carbon” policy, the demand for liquefied natural gas (LNG) storage and transportation containers has grown rapidly [1,2]. Storing and transporting LNG require container materials to maintain excellent toughness at ultra-low temperatures [3,4]. Traditional low-temperature materials are mainly nickel-based steel and stainless steel, both of which have a high nickel content and are costly. As an alternative material, high-manganese austenitic steel has received much attention because of its excellent comprehensive properties and low price. Moreover, high-manganese austenitic steel may prospectively be used as a liquid hydrogen container material due to its superior resistance to hydrogen embrittlement [5]. However, the face-centered cubic crystal structure of high-manganese austenitic steels results in a relatively low yield strength (typically 200–400 MPa [6]), which cannot meet the practical application and development trend of high-strength metallic structural materials.

Improving the yield strength whilst maintaining high ductility in high-manganese steels has always been a hot topic. Previous studies have shown that fine grains can help to improve the yield and tensile strength of twinning-induced plasticity (TWIP) steel [7,8]. Moreover, it was found that the contribution of twinning to yield strength is limited, with dislocation strengthening predominantly governing the flow stress in TWIP steel [9,10]. For high-manganese austenitic steel, mechanical twinning is a primary plastic deformation mechanism, except for a dislocation slip, which is significantly influenced by stacking fault energy (SFE) [11,12]. The TWIP effect is also affected by other factors. Chen et al. [13] found that grain refinement can reduce mechanical twinning activity during tensile deformation, thereby decreasing impact toughness. A practical and effective approach to increasing strength is through pre-straining. Tensile pre-strain has proven to be a successful technique for strengthening metal materials and is widely utilized in both scientific research and industrial applications [14,15,16].

However, the microstructure evolution and mechanical behavior of high-manganese austenitic steel are inadequately studied. Additionally, high-manganese austenitic steel inevitably undergoes processing deformation during container manufacturing. As a result, it is actually in service with pre-strain. Therefore, it is crucial to understand the effect of pre-strain on the microstructure and mechanical properties of high-manganese austenitic steel. In this work, the impacts of different pre-strain levels on the microstructures and mechanical properties of high-manganese austenitic steel were investigated. X-ray diffraction (XRD) and electron backscatter diffraction (EBSD) were used to characterize the dislocation density, grain boundary density, and twin volume fraction. The multiple strengthening mechanisms for yield strength and the changes in cryogenic impact toughness were analyzed.

2. Materials and Methods

2.1. Materials

The composition of the experimental steel was Fe–0.41C–0.17Si–23.50Mn–3.60Cr–0.55Ni–0.33Cu–0.11V (wt.%). The cast ingots with a thickness of 100 mm were reheated to 1200 °C for 2 h for homogenization and then were hot-rolled to a thickness of 20 mm through seven passes. The starting rolling temperature was 1100 °C, and the finishing rolling temperature was 900 °C. At the final rolling temperature of 900 °C, the alloy atoms were almost completely soluted in the matrix, and thus precipitation strengthening can be neglected [17,18]. The SFE of the steel was evaluated using the modified Olsen–Cohen thermodynamic model [19]. The physical parameters required for this calculation can be obtained from the literature [20,21,22,23,24]. The experimental steel has SFEs of 32.3 mJ/m² at room temperature (20 °C) and 22.0 mJ/m² at liquid nitrogen temperature (−196 °C). To prepare the pre-strained samples, large tensile samples with gauge sections of 80 mm × 20 mm × 12 mm were cut from the hot-rolled plate along the longitudinal direction; then, five true strains (0%, 5%, 15%, 25% and 35%) were applied to the samples through quasi-static tension. The pre-strained samples with true strains of 0%, 5%, 15%, 25%, and 35% were labelled as S0, S5, S15, S25, and S35, respectively.

2.2. Mechanical Tests

Tensile specimens were cut from the uniform deformation zone of the pre-strained samples, and the gauge length dimensions were 16 mm × 5 mm × 1.5 mm. Room temperature tensile tests were conducted on a universal testing machine (Zwick/Roell Z100, Ulm, Germany) at a strain rate of 1 × 10⁻³ s⁻¹. Charpy impact tests were carried out using V-notched specimens with dimensions of 10 mm × 5 mm × 55 mm on a pendulum impact testing machine (PTM2200−D1, SUNS, Shenzhen, China) at 20 °C and −196 °C. Three parallel specimens were processed, and their mean values were taken after the tests.

2.3. Microstructure Characterizations

Microstructure characterization was performed using an FE-SEM (TESCAN, Brno, Czech Republic) and an EBSD detector (Oxford Instruments, Oxford, UK). The samples for SEM characterization were prepared by mechanical polishing and then etching in FeCl₃ hydrochloric acid solution (1 g of FeCl₃, 3 mL of hydrochloric acid, and 12 mL of water). The samples for EBSD analysis were prepared by electrolytic polishing, with a solution of 10% perchloric acid in ethanol. The volume fraction of the twins was quantified and analyzed using the EBSD-based point counting method with ImageJ software (Version 1.51j8) [25,26,27].

The phases and dislocation density of the pre-strained steels were analyzed and evaluated based on the XRD spectra obtained by using a Rigaku SmartLab diffractometer (XtaLAB Synergy-S, Tokyo, Japan) with CuKα radiation. The scanning step and speed were 0.01° and 2°/min, respectively. In this study, a modified Williamson–Hall (MWH) model was used to calculate the dislocation density [28]:

$$\mathsf{\Delta }K={\displaystyle \frac{0.9}{D}}+{\left({\displaystyle \frac{\pi A^2{b}^2}{2}}\right)}^{{\displaystyle \frac{1}{2}}}{\rho }^{{\displaystyle \frac{1}{2}}}\left(K{\overline{C}}^{{\displaystyle \frac{1}{2}}}\right)+O\left(K^2\overline{C}\right)$$

(1)

where ΔK = 2cosθ(Δθ)/λ and K = 2sinθ/λ; Δθ and θ represent the integral width of the diffraction peak and diffraction angle, respectively; M is 2; b denotes the Burgers vector, with b = 0.254 nm [10]; D and ρ are the average grain size and dislocation density, respectively; and C represents the average dislocation contrast factor and can be calculated using the following equation:

$$C=C_{h00}\left(1-q{\displaystyle \frac{h^2{k}^2+k^2{l}^2+h^2{l}^2}{h^2+k^2+l^2}}\right)$$

(2)

where C_h00 represents the average contrast factor corresponding to the C_h00 reflection plane; q is a constant related to the anisotropic elastic factor; h, k, and l are the Miller indices of the diffraction planes; and q depends on the elastic constants of the material’s lattice, C₁₁, C₁₂, and C₄₄. At room temperature, the elastic constants for TWIP steel can be taken as C₁₁ = 169 GPa, C₁₂ = 82 GPa, and C₄₄ = 96 GPa [29]. Thus, C_h00 and q are 0.248 and 2.13, respectively. Finally, the dislocation density formula can be simplified to the following:

$$\rho =2{\displaystyle \frac{m^2}{\pi M^2{b}^2}}$$

(3)

where m represents the slope of the ΔK vs. KC^1/2 plot, which is typically calculated using four or more peaks.

3. Results

3.1. Mechanical Properties

The engineering stress–strain curves of the pre-strained steels are shown in Figure 1. As the pre-strain increases to 5%, 15%, 25%, and 35%, the yield strength of the S5, S15, S25, and S35 steels increases to 621, 846, 1065, and 1216 MPa, respectively. Their tensile strengths increase to 970, 1048, 1175, and 1259 MPa, respectively, while their total elongations decrease to 61.6%, 50.6%, 32.5%, and 19.5%, respectively. The impact energy changes with increasing pre-strain. When the pre-strain increases to 5%, the impact energy of the S5 steel is 85 J at room temperature (20 °C) and 58 J at liquid nitrogen temperature (−196 °C). The impact toughness of the S5 steel is comparable to that of the S0 steel. As the pre-strain further increases to 15%, 25%, and 35%, the impact energy begins to decrease.

3.2. Morphological Structures

The morphological structures of the pre-strained steels are shown in Figure 2, with the white arrows indicating twins or slip bands. The S0 steel contains annealing twins, with austenite grains primarily in an equiaxed form. The S5 steel exhibits a few twins, showing no significant difference in the number of twins compared to the S0 steel. The volume fraction of twins increases with the level of pre-strain, particularly when the pre-strain exceeds 15%, at which the number of twins increases rapidly.

3.3. Crystallographic Features

The grain boundary maps of the pre-strained steels are shown in Figure 3. The green and blue lines represent grain boundaries with misorientation angles (θ) of 5°–15° and >15° (none Σ3), respectively, and the red lines stand for Σ3 grain boundaries. Figure 3f shows the corresponding boundary densities for the pre-strained steels. As the pre-strain increases, the fraction of low-angle grain boundaries (LAGBs, 5° < θ < 15°) gradually increases, indicating a rise in local dislocation density. The high-angle grain boundaries (HAGBs, θ > 15°) also increase with increasing pre-strain. Additionally, many closed-grain boundaries marked by red lines within the austenite grains can be observed, suggesting that the regions with closed-grain boundaries are mechanical twins. There is no significant change in the LAGB, HAGB, and Σ3 boundary densities for S0 and S5 steels. With increasing pre-strain, the densities of the LAGBs, HAGBs, and Σ3 boundaries for S15, S25, and S35 steels increase. This also indicates an increase in dislocation density and the formation of mechanical twins.

The geometrically necessary dislocation (GND) distribution of the pre-strained steels is shown in Figure 4. The black lines represent grain boundaries with θ > 15°, and the GND density is indicated by a rainbow color scale. It is evident that the GND distribution in the S0 steel is uneven. Higher GND densities are observed near small grains and grain boundaries, while lower densities are found far from the grain boundaries. This may be due to strain concentration at small grains and grain boundaries caused by deformation. The results indicate that with increasing pre-strain, the range of GND distribution significantly broadens, demonstrating a substantial increase in GND density. This increase is particularly pronounced when the pre-strain exceeds 15%.

3.4. XRD Characterization

Figure 5a shows the XRD spectra of the pre-strained steels. XRD spectra show that the S0 steel only exhibits austenite peaks, and no other peaks are detected with increasing pre-strain. Figure 5b presents the peak broadening analysis using the MWH method for pre-strained steels. Five diffraction peaks, i.e., (111)_γ, (200)_γ, (220)_γ, (311)_γ, and (222)_γ, were selected to calculate the dislocation density. As the pre-strain increases, the dislocation density also increases, with an increase observed as the pre-strain exceeds 15%.

4. Discussion

4.1. The Effect of Pre-Strain Levels on Microstructure

In general, the deformation of steels involves a combination of dislocation glide and secondary mechanisms such as phase transformation and/or mechanical twinning, which are governed by the SFE. The XRD spectra in Figure 5 reveal that the austenite is highly stable and does not undergo phase transformation during deformation. Consequently, dislocation slip and mechanical twinning are identified as the predominant deformation mechanisms. The dislocation densities for the S0, S5, S15, S25, and S35 steels are 3.40 × 10¹³, 2.01 × 10¹⁴, 5.52 × 10¹⁴, 9.45 × 10¹⁴, and 1.15 × 10¹⁵ m⁻², respectively. It can be observed in

Table 1. Statistics on microstructure features of the pre-strained steels.

Steel	Dislocation Density (1014/m2)	Grain Boundary Density (μm−1)			Volume Fraction of Twins (%)
		LAGBs	HAGBs	Σ3
S0	0.34	0.020	0.308	0.071	0.9
S5	2.01	0.030	0.324	0.082	1.0
S15	5.52	0.067	0.365	0.119	3.1
S25	9.45	0.202	0.584	0.184	9.2
S35	11.50	0.280	0.884	0.341	13.7

that the dislocation density, grain boundary density, and volume fraction of twins increase with increasing pre-strain. When the pre-strain is below 5%, the primary change in the microstructure is an increase in dislocation density, while the changes in the volume fraction of twins and grain boundary density are relatively minor. When the pre-strain exceeds 15%, there is a significant increase in dislocation density, grain boundary density, and volume fraction of twins.

4.2. The Effect of Pre-Strain Levels on Strength

To reveal the influence of pre-strain on the yield strength of experimental steel and its strengthening mechanisms, the contributions of internal friction stress, solid solution, grain boundaries, precipitation, and dislocations to the yield strength were discussed. In the current study, the yield strength of experimental steels depends on the lattice friction, grain boundaries, solution atoms, dislocations, precipitates, and mechanical twins [9]. At the yield point, mechanical twins have not formed yet. Therefore, the yield strength (σ_y) can be expressed as follows [4]:

$${\sigma }_{\mathrm{y}}={\sigma }_0+{\sigma }_{\mathrm{s}}+{\sigma }_{\mathrm{g}}+{\sigma }_{\mathrm{p}}+{\sigma }_{\mathrm{d}}$$

(4)

where σ₀ is the lattice friction, 53.9 MPa; σ_s is the solution strengthening, MPa; σ_g is the grain boundary strengthening, MPa; σ_p is the precipitation strengthening, which is 0 MPa for all samples with a finishing rolling temperature of 900 °C since no precipitates form; and σ_d is the dislocation strengthening, MPa.

Carbide precipitation can reduce the content of dissolved carbon atoms in the austenite matrix. Since no carbide precipitated in the steels with a finishing rolling temperature of 900 °C, σ_s can be expressed as follows [30]:

$${\sigma }_{\mathrm{s}}=290X_{\mathrm{C}}+48.6X_{\mathrm{S}\mathrm{i}}-{1.5X}_{\mathrm{M}\mathrm{n}}+7X_{\mathrm{N}\mathrm{i}}+0.5X_{\mathrm{C}\mathrm{r}}+32X_{\mathrm{M}\mathrm{o}}+17X_{\mathrm{V}}+22X_{\mathrm{A}\mathrm{l}}$$

(5)

where X_i represents the mass fraction of each element in weight percent (wt.%). Based on the above formula, the contribution of σ_s in samples with different final rolling temperatures is 100.0 MPa.

Dislocation strengthening follows the relationship with respect to dislocation density as described below:

$${\sigma }_{\mathrm{d}}=M\alpha Gb\sqrt{\rho }$$

(6)

where M is the Taylor factor, determined through an EBSD analysis, with a value of 3.06; α is the dislocation interaction coefficient, 0.36; G is the shear modulus, with a value of 62,000 MPa for Fe–22Mn–0.6C steel at 20 °C; and b is the Burgers vector, 2.56 × 10⁻⁴ μm [31,32]. By substituting the dislocation density results, the dislocation strengthening in S0, S5, S15, S25, and S35 steels are 104.0, 253.0, 418.9, 548.3, and 604.5 MPa, respectively.

For grain refinement strengthening, the contribution of grain refinement to strength can be described using the Hall–Petch relationship:

$${\sigma }_{\mathrm{g}}={\displaystyle \frac{k_{\mathrm{y}}}{\sqrt{d}}}$$

(7)

The grain boundary size lacks a unified standard, making it difficult to define grain size. Therefore, in this study, the contribution of grain boundary strengthening is represented by the value after subtracting σ₀, σ_s, σ_p, and σ_d from the experimental yield strength.

Figure 6 gives the contributions of each strengthening mechanism to the yield strength of the experimental steels and corresponding weights to the yield strength. One can see that grain boundary and dislocation strengthening play major roles. At 5% pre-strain, the increase in yield strength is primarily due to dislocation strengthening, with minimal change in grain boundary strengthening, which is consistent with the microstructural characterization results. When the pre-strain increases to 15% and higher, the contributions from both dislocation strengthening and grain boundary strengthening increase.

4.3. The Effect of Pre-Strain Levels on Cryogenic Impact Toughness

SFE determines the microstructure, mechanical properties, and deformation mechanism of high-manganese steel, and its value is influenced by the elemental content and deformation temperature. Since the SFE of the studied steel is within the range of 20.0 to 55.0 mJ/m², it is prone to the TWIP effect, contributing to excellent cryogenic impact toughness [33,34]. The fracture morphology of the pre-strained steels at −196 °C is shown in Figure 7. The impact fracture surface exhibits a large number of dimples, with no obvious cleavage fracture morphology, indicating that the fracture mode is typical ductile fracture.

Relevant studies suggest that the presence of dislocation strain fields can interfere with stacking fault regions, resulting in an increase in SFE as the dislocation density increases [35]. As the SFE increases, twinning becomes more difficult, leading to a decrease in cryogenic impact toughness. Additionally, the introduction of mechanical twins and dislocations leads to a decrease in impact toughness. This is primarily because the pre-existing dislocations and mechanical twins in the austenitic matrix effectively hinder dislocation movement and the further initiation of twinning [13]. In the current study, when the pre-strain exceeds 15%, the dislocation density and volume fraction of the twins increase and, thus, the cryogenic impact toughness declines. It is inferred that the initial mechanical twins and increased dislocations triggered by pre-straining will jeopardize the impact toughness.

4.4. General Discussion

It remains a challenge to break through the strength–toughness trade-off dilemma and achieve high strength and toughness in high-manganese steel. In the current work, it has been proven that the strength and toughness of high-manganese austenitic steel can be manipulated by tuning the pre-strain level. A moderate pre-strain (up to 15%) can enhance yield strength without significantly compromising impact toughness, which is primarily through dislocation strengthening. However, a large pre-strain that exceeds 15% will induce high-volume fraction twins and high-density dislocations, which thereby deteriorates the impact toughness. Thus, the mechanical twins and dislocations in the high-manganese austenitic steel are controlled by introducing appropriate pre-strain levels to the plate, through which the strength and toughness are tailored. These findings provide a new strategy in the microstructure design of high-manganese austenitic steel for achieving balanced strength and toughness. Moreover, the high-manganese austenitic steel will inevitably undergo processing deformation during container manufacturing, thus being in service under pre-strain conditions. These findings also guide the application of high-manganese austenitic steel for container manufacturing with large deformation tolerance.

5. Conclusions

To high-manganese austenitic steel, various levels of pre-strain were applied. Through tensile and cryogenic impact tests on steels with different pre-strain levels, the effects of pre-strain on microstructure and mechanical properties were investigated. The main findings can be summarized as follows:

(1)

When the pre-strain increased from 0% to 5%, although the dislocation density increased, there was almost no change in the density of the grain boundaries and volume fraction of the twins. The dislocation density and volume fraction of the twins increased significantly as the pre-strain exceeded 15%.

(2)

With increasing the pre-strain, the yield and tensile strength of the pre-strained steels increased, while the elongation decreased. When the pre-strain exceeded 15%, the yield strength increased significantly. This increase in yield strength is primarily attributed to the contributions of grain boundary strengthening and dislocation strengthening.

(3)

When the pre-strain was 5%, the cryogenic impact toughness was almost the same as that of the steel with 0% pre-strain. However, the cryogenic impact toughness significantly decreased when the pre-strain exceeded 15%. The initial mechanical twins and increased dislocations caused by pre-straining are unfavorable to the impact toughness.

In the current work, the potential of controlling microstructure via pre-strain treatment to tailor the mechanical properties of high-manganese austenitic steel has been proven. The above findings provide new insights into its microstructure design and guide the application of high-manganese austenitic steel for container manufacturing with large deformation tolerance. Considering that high-manganese austenitic steel can be used as a material for storage and transportation containers of liquid hydrogen, the mechanical behavior of high-manganese austenitic steel with different pre-strain levels under hydrogen conditions cannot be ignored. Therefore, further research is urgently needed in this area.