Research on Microstructures and Properties of High-Strength Anticorrosion Twinning-Induced Plasticity Steels

Abstract

The microstructures, mechanical properties, and anticorrosion properties of Fe–25Mn–Cr(12, 18 wt.%)–CN TWIP steels compared with nominal Fe–22Mn–0.6C TWIP steel (TWIP-ref) in the solid-solution state were investigated in the present study by using X-ray diffraction (XRD), electron backscattered diffraction (EBSD), transmission electron microscopy (TEM), tensile tests, and potentiodynamic polarization measurement, etc. The results show that the Fe–Mn–Cr–CN TWIP steels have excellent mechanical properties and anticorrosion properties compared to the TWIP-ref steel. The strain-hardening rate of the 12CrN steel is lower than that of the TWIP-ref steel because of the relatively late generation of high-density deformation twins, which resulted in less tensile strength increments during the tensile test. Moreover, M₂₃C₆ precipitates enriched with Mn and Fe formed in the Fe–Mn–Cr–CN steel as the Cr content increased, which had an important influence on the mechanical properties and anticorrosion properties. In the Fe–Mn–Cr–CN TWIP steels, because of the influence of the M₂₃C₆ precipitation in the 18CrN sample, the yield strength (σ_YS) is increased slightly, while the ultimate tensile strength (σ_UTS) and elongation to failure are decreased. In addition, 18CrN has poor corrosion resistance because of the formation of M₂₃C₆ precipitation (i.e., it has relatively negative corrosion potential (E_corr) and a smaller corrosion current density (I_corr)).

1. Introduction

Twinning-induced plasticity (TWIP) steel is regarded as one of the structural materials with the most potential for automobiles because it meets the demands for safety and light weight, which are attributed to its excellent combination of high strength and high ductility [1,2,3,4,5]. Fe–Mn–C-series [5,6,7,8] TWIP steel is one type of TWIP steel that has received widespread attention because of its hardening and strengthening mechanisms. Previous studies have confirmed that TWIP steels have outstanding mechanical properties that are attributable to their grain refinement by deformation-induced twinning, and to dynamic strain aging [9,10,11]. The Fe–Mn–C system, with high Mn (15–25 wt.%) and C (0.4–1.0 wt.%) elements, possesses low stacking-fault energy (SFE) (20–40 mJ·m⁻²) and a full austenitic structure [12,13]. However, all TWIP steels, including the Fe–Mn–C system, have excellent mechanical properties but are they easily corroded. Therefore, the development of corrosion-resistant TWIP steel with high CrN content can greatly expand the applications of conventional Fe–Mn–C TWIP steels.

In an attempt to produce a TWIP steel with not only with excellent mechanical properties, but also with enhanced anticorrosion properties, a series of studies were conducted [7,12,13,14,15,16,17,18,19], such as with N and/or Cr addition. Manganese is used as an austenite stabilizer and to provide the plasticity effect. Chromium is used to improve the corrosion resistance of the material. Cr has more positive electrode potential than Mn (i.e., the standard reduction potential of Cr ($E_{\mathrm{Cr}3+/\mathrm{Cr}}^0=-0.74$ V vs. standard hydrogen electrode (SHE)) and it is more positive than that of Mn ($E_{\mathrm{Mn}2+/\mathrm{Mn}}^0=-1.18$ V vs. SHE) [20,21]. Carbon and nitrogen are used as interstitial elements to obtain a stable austenitic structure. On the one hand, nitrogen improves the resistance to pitting corrosion [22], and, on the other hand, the incorporation of chromium increases the solubility of the nitrogen in the melt during casting [16].

Yuan et al. [14] found that the amount of the intragranular precipitated phases of Cr₂₃C₆ increased with the increase in the Cr content (1.13~3.95 wt.%) in Fe–Mn–C–Al–Cr–N TWIP steels, and that the yields and ultimate tensile strengths were enhanced by increasing the Cr₂₃C₆ precipitation. Roncery et al. [16] report a new series of Fe–Mn–Cr–C–N steels that were developed on the basis of thermodynamic calculations, and the results are fully austenitic microstructures without δ-ferrite or pores. The diffusion calculations that were made for the prediction of the diffusion annealing process indicate the tendency of C and N to cosegregate with Cr and Mn.

So far, there have been many studies on the microstructures and mechanical properties of TWIP steels. However, there are nearly no investigations that report the effect of high Cr content on the microstructures and properties of the Fe–Mn–C-system TWIP steels. It is expected that the present study will provide some new information for a better understanding of the relationship between the chromium content and the mechanical and anticorrosion properties of Fe–25Mn–Cr(12, 18)–CN TWIP steels. On the basis of this, we discuss the tensile properties and corrosion resistance in association with the microstructures and the M₂₃C₆ precipitates of TWIP steels that contain various contents of Cr.

2. Experimental Procedures

It has been reported [15,16,17] that, in the Cr-alloyed TWIP steels, the contents of C and N, as well as the C/N ratio, are important factors to the increase in the solubility of the C in the austenite matrix. Therefore, suitable additions of C and N will not only guarantee the stability of the austenitic phase, but they will also keep the precipitation of the carbides that contain Cr from the austenitic matrix. On the basis of these findings, both the contents of C and N are set at 0.3 wt.% in the present TWIP steels. In addition, from the principle of corrosion resistance, the addition of Cr is usually 12 wt.% or more, and, thus, two additions of Cr were considered (i.e., 12 and 18 wt.%). Finally, the nominal composition (wt.%) of the current TWIP steels is Fe–25Mn–xCr–0.3C–0.3N (x = 12 and 18).

The TWIP steel was melted and cast in a 25 kg vacuum induction furnace under an argon atmosphere. After solidification, the ingots were heated to 1100 °C to ensure a homogenous microstructure, and they were then hot forged to round bars that were 40 mm in diameter. Before use, the round bars were annealed at 1100 °C for 4 h in order to obtain fully recrystallized grains and the necessary annealing twins. Finally, the annealed round bars were cut by an electric spark machine to form dog-bone-shaped tensile samples with a gauge size of 10 × 2.5 × 1.5 mm³. The measured composition is presented in

Table 1. The contents of main elements of TWIP steels (wt.%).

Notation	Mn	Cr	C	N	P	S	Si	Ni	Fe
TWIP-ref	21.89	0.052	0.52	-	0.011	0.009	0.609	0.057	Bal.
12CrN	26.16	12.38	0.32	0.31	0.009	0.001	0.001	0.081	Bal.
18CrN	26.28	17.29	0.28	0.32	0.011	0.001	0.001	0.020	Bal.

.

The tensile tests were carried out with a material test system (Instron 3369, Norwood, MA, USA) with a strain rate of 5 × 10⁻³ s⁻¹ at room temperature. The phases were analyzed by an X-ray diffraction apparatus (XRD, X′Pert Pro MPD, PANalytical, Zwolle, Holland), using CuKα radiation with the scan angles from 20° to 100°, at a step of 0.03349°, at room temperature. The microstructures were observed and analyzed by electron backscattered diffraction (EBSD) equipment (HKL/Channel 5 system, Oxford Instruments, Bognor Regis, UK) and transmission electron microscopy (TEM, Tecnai G2 F20, FEI, Bryan, TX, USA). After they were mechanically ground and polished, the EBSD samples were electrolytically polished in a solution of 20% perchloric acid, 10% acetic acid and balanced ethanol by using 1~2 A of current and 20 V of voltage. The TEM samples were prepared by punching a disk sample first, with a diameter of 3 mm, from the tensile sample, and then mechanically thinning it down to 70 μm, and finally polishing it in a solution of 5% HClO₄ and 95% C₂H₅OH at −25 °C by a double-jet electro-polishing apparatus. The mean grain size was measured by using the line intercept method on the EBSD inverse pole figure (IPF) maps, where all of the high-angle grain boundaries (HAGBs) and twin boundaries (TBs) were counted.

The corrosion behavior of the samples in 3.5 wt.% of NaCl solution was measured using potentiodynamic polarization tests (CHI 760E, Chenhua, China). A platinum electrode, a saturated calomel electrode (SCE), and the sample with an exposed area of 0.75 cm² were separately used as the counter, the reference, and the working electrode, respectively. Before the test began, the working electrode was immersed in a 3.5 wt.% NaCl solution for one hour in order to obtain a stable surface state. The test was recorded at a scanning rate of 1.667 mV·s⁻¹ at room temperature.

3. Results and Discussion

3.1. Influence of Cr on the Stacking-Fault Energy (SFE) of Alloy

The stacking-fault energy (SFE) can be calculated by the following equation [23,24]:

$${\gamma }_{SFE}=2\rho \Delta G^{\gamma \to \epsilon }+2\sigma$$

(1)

where $\rho$ is the molar surface density along the {111} planes; $\Delta G^{\gamma \to \epsilon }$ is the molar Gibbs energy related to the γ→ε phase transition; and σ is the interfacial energy between the γ and ε phases. The molar surface density ($\rho$) is determined by the lattice parameter (a) and Avogadro’s constant (A):

$$\rho =\frac{4}{\sqrt{3}}\frac{1}{a^{2}A}$$

(2)

The $\Delta {\mathrm{G}}^{\gamma \to \epsilon }$ for the FCC alloy steels is determined by the following equation:

$$\Delta {\mathrm{G}}^{\gamma \to \epsilon }={\displaystyle \sum }_i{x}_i\Delta G_i^{\gamma \to \epsilon }+{\displaystyle \sum }_{ij}{x}_i{x}_j{\mathsf{\Omega }}_{ij}^{\gamma \to \epsilon }$$

(3)

where $x_i$ is the molar fraction of all the elements in the steel; $\Delta G_i^{\gamma \to \epsilon }$ is the changes in the Gibbs free energy; and ${\mathsf{\Omega }}_{ij}^{\gamma \to \epsilon }$ is the interaction parameter between the i and j elements. If twinning deformation is to arise, the SFE of the steel should be in the range of 18 to 45 mJ/m² [24,25].

In terms of the above equations, it is calculated that the SFE of 12CrN alloy is approximately 31mJ·m⁻² [15]. In addition, Dumay et al. [23] and Mosecker et al. [26] studied the effect of the Cr contents on the SFE of TWIP steel by computational thermodynamics. on the basis of the thermodynamic stacking-fault energy (SFE) calculations, the SFEs of 12CrN and 18CrN are between 30 and 35 mJ·m⁻². Therefore, the twinning deformation should be the predominant plastic mechanism in the present TWIP steels (i.e., when the content is equal or below 18 wt.%, the Cr addition will not change the plastic deformation modes of TWIP steels).

3.2. Influence of Cr on the Microstructures of Alloy

The XRD spectra of the samples in the solid-solution state before and after the tensile tests are shown in Figure 1 and Figure 2, respectively. Only the austenite phase existed in the microstructures, which verifies that the austenite is stable in the present alloys. From Figure 2, it is seen that the γ(111), γ(200), and γ(220) peaks are very strong, while the γ(311) and γ(222) peaks are quite weak, which suggests that there are preferential orientations of the γ phase during the tensile deformation.

Figure 3 presents the inverse pole figure (IPF) maps of the samples that were obtained by the EBSD analysis. It can be seen that there are many annealing twins that appear in the samples, which are similar to those in the common TWIP steels without the Cr addition. Therefore, it is suggested that an addition of Cr up to 18 wt.% does not affect the stability of the austenite and the twinning behavior of TWIP steels.

3.3. Influence of Cr on the Mechanical Properties of Alloys

Figure 4 shows the engineering stress–strain curves of the TWIP steels, with or without the Cr addition, and the typical properties are listed in

Table 2. Tensile properties of Fe–Mn–Cr–CN steel in comparison to TWIP-ref steel.

Samples	σYS/MPa	σUTS/MPa	δ/%
TWIP-ref	308	890	63.4
12CrN	496	904	67.2
18CrN	512	876	50.3

. It is obvious that, when the Cr content was 12 wt.%, the alloy showed the best overall mechanical properties among the three TWIP steels. The average yield and tensile strength are 496 MPa and 904 MPa, respectively, and the average elongation is 67.2%. As a comparison, the common TWIP steel shows a yield strength of 308 MPa, a tensile strength of 890 MPa, and an elongation of 63.4%. All of the properties are lower than those of the 12CrN TWIP steel. On the other hand, by increasing the Cr content to 18 wt.%, the elongation of the TWIP steel decreased, although the strength still maintained relatively high values.

Figure 5 shows the TEM images of the 12CrN and 18CrN samples in the solution state, and they all have relatively low-density dislocations with significant twins. Figure 6 shows the precipitation, the inset diffraction pattern, and the EDS analysis of the 18Cr TWIP steel in a solution state. From the results of the EDS analysis and the diffraction pattern, the precipitation was identified as M₂₃C₆ enriched with Mn and Fe.

There have been many studies on nitrogen solid-solution strengthening to increase the yield strength [16,27,28,29]. Roncery et al. [16] mention that nitrogen is 1.5 times more effective than carbon as a strengthening element in austenitic steels. Rawers et al. [27] propose an equation for solid-solution strengthening (σ_s) by elemental N: ${\sigma }_s=constent+\alpha \left[N\right]+\beta {\left[N\right]}^{1/2}$, where [N] is the nitrogen concentration (wt.%). Wang et al. [28] studied chromium/nitrogen alloying to improve the yield strength of Fe–Mn–C TWIP steel, and they found that AlN precipitation refined the grain size of the studied steel and yielded greater precipitation strengthening.

The yield strengths of the 12CrN and 18CrN samples are 496 and 512 MPa, respectively, which are much higher than the 308 MPa of the TWIP-ref sample. The carbon element in the 12CrN and 18CrN samples is 0.32 and 0.28 wt.%, respectively, which is far lower than the 0.52 wt.% in the TWIP-ref sample. However, the N contents in the 12CrN and 18CrN samples reached up to 0.31 and 0.32 wt.%, respectively. Considering that the solid-solution strengthening effect of N is 1.5 times that of C [16] (that is, the corresponding C contents are 0.465 and 0.48 wt.%, respectively), so the carbon contents that represent the total interstitial element (C + N) contents of the 12CrN and 18CrN samples are 0.785 and 0.76 wt.%, respectively, which are much higher than the TWIP-ref sample. It can be concluded that the yield strength of Fe–Mn–Cr–CN TWIP steel is significantly higher than that of TWIP-ref steel.

The nominal interstitial element content of 18CrN (carbon content represents total interstitial content) is 0.76 wt.%, which is lower than the 0.785 wt.% of 12CrN. According to the theory of interstitial-element solid-solution strengthening, the yield strength of 18CrN should be lower than that of 12CrN. However, the experimental results show that the yield strength of 18CrN is 512 MPa, which is slightly higher than that of 12CrN (496 MPa). Zheng et al. [30] found that the precipitation of M₂₃C₆ carbides increased the strength modestly and deteriorated the ductility significantly in 0.3C–20Cr–11Mn–1Mo–0.35N steel. Therefore, M₂₃C₆ precipitates formed in the 18CrN sample (Figure 6), which resulted in its higher yield strength. Because of the influence of the M₂₃C₆ precipitation, the elongation to failure of 18CrN (50.3%) is significantly worse than that of 12CrN (67.2%). In addition, the average grain size of the 18CrN sample is 37.7 μm, which is lower than the 38.1 μm of 12CrN. According to the Hall–Petch relation [31,32] (${\sigma }_{YS}={\sigma }_0+K{d}^{-1/2}$, where σ_YS is the yield strength; σ₀ is the resistance of the lattice to the dislocation motion; K is the strengthening coefficient; and d is the average grain size), the smaller the average grain size, the higher the yield strength.

Figure 7 shows the strain-hardening rate against the true strain curves of the TWIP-ref, 12CrN, and 18CrN samples in the solution state. It can be seen that the TWIP-ref sample shows a very different strain-hardening rate vs. true strain (i.e., the strain-hardening rate in the intermediate stage is, at first, kept constant, with an increase in the true strain from 8 to 28%, and then the strain-hardening rate slowly dropped with the increase in the true strain). However, the 12CrN and 18CrN samples showed very similar tendencies of the strain-hardening rate with the increasing true strain. There were three steps in the strain-hardening-rate curve: the strain-hardening rate dropped almost vertically at the beginning and at the end, and it dropped more slowly in the intermediate stage of deformation. The strain-hardening-rate curves of the 12CrN and 18CrN samples are very similar, which indicates that the increase in the Cr content from 12 to 18 wt.% hardly affects the deformation mechanism of Fe–Mn–Cr–CN TWIP steel. Moreover, the strain-hardening rate of Fe–Mn–Cr–CN TWIP steel is significantly lower than that of the TWIP-ref steel. This implies that there will be two distinct deformation behaviors between TWIP-ref and Fe–Mn-Cr–CN TWIP steel.

Figure 8 shows the TEM images of the deformation of the studied steels. High-density mechanical twins are observed in the TWIP-ref steel that was deformed to 0.15, and the thickness of the observed twin lamella ranges from 3 to 30 nm (Figure 8a). However, for the 12CrN steel, only parallel high-density dislocation walls (HDDWs) were observed in the steel that was deformed to 0.15 (Figure 8b); high-density mechanical twins are observed in the steel when the strain is further increased to 0.45, and the thickness of the twin lamella ranges from 3.5 to 35 nm (Figure 8c). During tensile deformation, the twinning behavior of 12CrN steel is not as obvious as that of TWIP-ref steel, and, hence, the former generated fewer twin boundaries (TBs) than the latter during the tensile test, which resulted in the poorer hindering effect on the dislocations than that of the latter, and which resulted in less tensile strength increments during the tensile test, which explains why the strain-hardening rate of the 12CrN steel is significantly lower than that of the TWIP-ref steel (Figure 7). The representative tensile fracture features of the 12CrN and 18CrN samples are presented in Figure 9. The profiles exhibit many refined dimples, which are typical ductile fracture characteristics, and which also indicate that there is no effect on the fracture mechanism in Fe–Mn–Cr–CN TWIP steel with large amounts of Cr content from 12 to 18 wt.%.

3.4. Influence of Cr on the Corrosion Resistance of Alloy

Figure 10 shows the potentiodynamic polarization curves of the samples in 3.5 wt.% of NaCl solution, and the relevant corrosion potential (E_corr) and corrosion current density (I_corr) data are listed in

Table 3. Characteristic data obtained from polarization curves.

Samples	Ecorr/mV	Icorr/10−6 A·cm−2
TWIP-ref	−845	2.9027
12CrN	−483	1.4707
18CrN	−494	2.3920

. In these experiments, each polarization test was repeated at least three times until stable values were reached, and the data that are shown in Figure 10 and Table 3 are the representative results.

It can be seen from Figure 10 and Table 3 that the corrosion-resistant properties of the samples containing Cr are much better than those of the common TWIP steel sample. The corrosion potential (E_corr) of the common TWIP steel is −845 mV, while the E_corr values of the samples with 12Cr and 18Cr are −483 and −494 mV, respectively. In terms of the standard reduction potential, Cr has a standard reduction potential of −0.74 V at a standard hydrogen electrode, while Mn has a standard reduction potential of only −1.18 V [20,21]. On the other hand, the corrosion current density (I_corr) of the TWIP steel containing 12Cr is 1.4707 × 10⁻⁶ A·cm⁻², while that of the common TWIP steel is 2.9027 × 10⁻⁶ A·cm⁻². The former has a much lower corrosion current density than the latter. Therefore, the Cr addition is significantly beneficial for improving the corrosion-resistant properties of TWIP steels.

It can be noticed that, when the Cr content is increased to 18 wt.%, the corrosion resistance of the TWIP steel was, instead, decreased. The E_corr shifted towards the negative and the I_corr increased to a much larger positive value. In order to understand this change, the microstructures of the 18CrN samples were further analyzed. It was found that there were many small (Fe, Cr)₂₃C₆ carbides in the microstructures of the 18CrN samples, as is shown in Figure 6. It would be reasonable to assume that the precipitation of the carbides that contained Cr not only reduced the Cr content of the austenite matrix, but also increased the phase interfaces. Both the changes are unfavorable for the corrosion resistance of TWIP steels.

4. Conclusions

The microstructures, mechanical properties, and anticorrosion properties of Fe–Mn–Cr–CN TWIP steels, compared to a TWIP-ref steel solution treated at 1100 °C for 4 h, were investigated in the present study. The following conclusions were obtained:

(1)

The Fe–Mn–Cr–CN TWIP steels still had the stable single austenite phase before and after the tensile test, even with Cr contents as high as 12 and 18 wt.%;

(2)

Compared with the TWIP-ref steel, the Fe–Mn–Cr–CN TWIP steels have better mechanical properties and corrosion resistance. The strain-hardening rate of 12CrN steel is lower than that of the TWIP-ref steel because of the relatively late generation of high-density deformation twins, which resulted in less tensile strength increments during the tensile test. The yield strengths of the 12CrN and 18CrN samples are 496 and 512 MPa, respectively, which are much higher than that of the TWIP-ref sample (308 MPa). Moreover, the Fe–Mn–Cr–CN steels have more excellent anticorrosion properties. The E_corr values of the 12CrN and 18CrN samples are −483 and −494 mV, respectively, which are more positive than that of the TWIP-ref sample (−845 mV), and the I_corr of the 12CrN sample (1.4707 × 10⁻⁶ A·cm⁻²) is significantly smaller than that of the TWIP-ref sample (2.9027 × 10⁻⁶ A·cm⁻²);

(3)

The higher Cr content of the 18CrN sample leads to the precipitation of M₂₃C₆, which is harmful to the ductility and corrosion resistance of Fe–Mn–Cr–CN TWIP steels. Therefore, compared to the 12CrN sample, the 18CrN sample has poor ductility and anticorrosion properties.