Influence of High-pressure Quenching on the Microstructure, Martensite Transformation, and Mechanical Properties of 0.2 Mass% C Steel

Abstract

Low-carbon steel (0.2 mass%) samples were austenitized and quenched at a cooling rate of 10°C/s under GPa level high pressure. The morphology, lattice constant, and order degree of C atom distribution of high-pressure quenching martensite were characterised and analyzed by TEM, EBSD, XRD, and Mössbauer. Besides, the transformation characteristics and strengthening mechanisms were discussed. The results show that the microstructure of 0.2 mass% C steel is fine hierarchical lath martensite with almost no residual austenite, and its laths mostly follow {112} <111> twin relationship, indicating the self-accommodation effect among martensite variants. Compared to atmospheric pressure, the order degree of carbon atom distribution increases in high-pressure quenched martensite, meanwhile the tetragonality (c/a) of martensite lattice increases from 1.009 at atmospheric pressure to 1.012 at 4 GPa. The significant promotion of hardness in 0.2 mass% C steel subjected to high-pressure treatment can be ascribed to a large number of dislocations in the structure, grain refinement strengthening caused by twin boundary, and solution strengthening caused by large distortion due to the increase of the order degree for C atom distribution and the decrease of lattice constant. These findings provide new insights into the carbon steel martensite transformation mechanism, and a new martensite transformation technique can be developed.

1. Introduction

Martensite transformation is a main heat treatment method for steel material reinforcement. Much work has so far focused on the morphology, crystalline structure of martensite, as well as martensite transformation characteristics.^1,2,3) The hardness of martensite depends on the carbon content, while the strength is affected by dislocations, twins, grain size and atomic solution, etc.^4,5) Although high-carbon acicular martensite has high hardness and strength, its substructure is dominated by twins with low toughness. The substructure of low-carbon lath martensite is dislocation, which has good toughness, but low strength and hardness.^6,7) Besides, martensite is usually obtained by the rapid cooling of austenite, but it is difficult to obtain martensite structure in the center of large workpieces due to the low cooling rate. Therefore, finding a low cooling rate transformation method to obtain high strength and toughness martensite has become a long-term goal pursued by engineering material researchers.^7,8,9,10)

Pressure, temperature, and composition are three fundamental physical parameters that can be used to control the properties of materials. The role of pressure is irreplaceable by other means. It can not only change the various internal interactions of material but also change the structure and properties.^11,12,13) During the phase transformation of metal material, pressure can change the phase stability and the energy barrier of metal material transformation.¹⁴⁾ Previous investigations demonstrated that pressure can influence some aspects of martensitic transformations, such as transformation temperatures, crystallography, and the amount and morphology of the product martensite. Kang et al.¹⁵⁾ obtained a fully martensitic structure of IF steel (Fe-0.004C-0.9Mn-0.01Ti-0.03Nb) quenched by use of the high-pressure high-temperature (HPHT) technology, while only diffusion transformation can occur at normal pressure. Zhang et al.¹⁶⁾ found that martensitic transformation was induced in pure iron under a pressure of 3 GPa at a slow cooling of 10°C/s, and the martensite was characterized by twin-related laths with an average thickness of 3.8 nm, resulted in an unprecedented strengthening (800 HV). Lu et al.¹⁷⁾ believed a metastable fcc phase couldn’t exist in the transition from the bcc to hcp phase in Fe under ultra-high pressure, since the pressure tends to impede the transformation from the fcc-hcp phase. It was studied the effect of hydrostatic pressure on the microstructure evolution and mechanical properties of Zr and its alloys. And an excellent tensile strength of 1387 MPa was obtained with fine needle-like martensite containing a large number of dislocations and stacking faults when TZ-30 alloys subjected to 2 GPa pressure with a cooling rate of 15°C/s.^18,19)

Although the effect of high pressure on the martensitic transformation have been observed, the mechanism has not yet been realized, which restricts the further research and application of high-pressure martensitic transformation. Hence, 0.2 mass% C steel was employed as a model system and quenched under GPa level pressure. Thereafter, the impact of pressure on martensite morphology, substructure, lattice constant, and order degree of C atom distribution was discussed by the method of scanning electron microscope (SEM), transmission electron microscope (TEM), Mössbauer, etc, and the mechanism of high-pressure martensitic transformation and hardness enhancement have been investigated. These researches will not only clarify the physical mechanism of quenched martensite under high pressure but also provide experimental evidence for the manufacture of large block martensite.

2. Material and Methods

An industrial low-carbon alloyed steel with a composition of Fe-0.2C-0.45Mn-0.23Si mass% was investigated. Cylindrical specimens with a gauge section of Φ6×13 mm were used for atmospheric (10⁻⁴ GPa) and high-pressure (2, 3, 4 GPa) quenching. High-pressure quenching was conducted on a hexahedral top ultra high-pressure device (CS-1 V), which could produce high pressure by equally compressing the six sides of a cubic sample assembly consisting of a specimen, BN crucible, graphite tube, etc. (Fig. 1). The temperature was measured with K-type thermocouples, and the pressure was calculated from calibration curves of the multi-anvil high-pressure apparatus.²⁰⁾ The specimens were heated together with the assembly under the pressure of 2, 3 or 4 GPa to 900°C, held for 20 min, and then cooled at 10°C/s to room temperature. For comparison, specimens were austenitized at 900°C for 20 minutes under atmospheric pressure, followed by cooling to room temperature in 10% NaCl solution, at a cooling rate of over 50°C/s.

Fig. 1.

Schematic illustration of the hexahedron pressure facility. (Online version in color.)

The microstructures were characterized by optical microscopy (OM; Axio Scope A1 Pol). Patterns for electron backscattered diffraction (EBSD) were acquired by scanning electron microscopy (SEM; SUPRA-55) equipped EDAX TSL (Mahwah. NJ) OIM EBSD system, and the data were analyzed with Channel 5 software. Specimens for EBSD were ground with 800#–5000# sandpaper. Subsequently, electropolishing was carried out for 10 s of 20 V in a solution of 10% perchloric acid and 90% ethanol vol.%. Nanoscale martensite and lattice constants were analyzed by TEM (Tecnai G2 F20). The TEM samples were prepared by mechanically thinning down to 0.05 mm. 3 mm diameter discs were punched out from the thin foils and thinned down to 0.025 mm. And it was subsequently subjected to ion beam thining (Gatan 691 PIPS) with a voltage of 4 KeV and grinding angle of 8°–2°. Mössbauer spectra were measured in an equal acceleration mode at room temperature, and measurements were performed in transmission geometry with a 57Co source in an Rh matrix. Calibration was performed with several samples of α-Fe foils, and hyperfine parameters were fitted by Mosswinn software.

Vickers hardness was measured on an HV-1000A tester with a load of 5N. Specimens with a gauge section of Φ1.8×6 mm were used for tensile testing at a strain rate of 5×10⁻⁴/s at room temperature.

3. Results and Discussion

3.1. Microstructure of High-pressure Quenched Martensite in 0.2 mass% C Steel

Figure 2 shows the microstructures of 0.2 mass% C steel austenitized at 900°C for 20 min and cooled to room temperature at 10°C/s under diffrent pressure. Under atmospheric pressure, martensite structure of hierarchical structure is obtained for 0.2 mass% C steel (Fig. 2(a)). A large number of dislocations are entangled in martensite matrix, and laths with twin orientation relationships of {112} <111> are observed. However, this type of lath is rare and difficult to observe at atmospheric pressure in 0.2 mass% C steel. The typical features of lath-block-packet martensite is also observed in 3 GPa-steel and 4 GPa-steel, but the size of the packets is much smaller. The growth of martensite displays symmetrically in 4 GPa steel at arrow A (Fig. 2(g)), indicating the self-accommodation²¹⁾ of martensite transformation. According to optical micrographs, the prior austenite grain sizes are estimated in 0.2 mass% C steel with and without pressurization. The grain size of the prior austenite grain is approximately 140.66 μm under atmospheric pressure, which is refined under GPa level high pressure (51.35 μm of 3 GPa).

Fig. 2.

Microstructures characterization of 0.2 mass% C steel under 10⁻⁴ GPa (a–c), 3 GPa (d–f), and 4 GPa (g–i). (a, d, g): Optical micrographs; (b, e, h): bright-field (BF) images; (d, f, i): dark-field (DF) SAED patterns. (Online version in color.)

At 3 GPa-steel, fine martensitic laths with high-density tangled dislocations are observed in Fig. 2(e). The width of laths is approximately 150–250 nm taken from 10 bright-field (BF) images. The selected area electron diffraction (SAED) patterns are indexed for the selected area in Fig. 2(e) as shown in Fig. 2(f). The adjacent martensite laths most follow symmetric twin orientation relationships of {112} <111>. The twin plane index is determined by trace analysis and related analysis.²²⁾ Comparably, the lath width of martensite quenched at 4 GPa is about 100–200 nm as shown in Fig. 2(h). The twinning relationship of adjacent martensite laths is {112} <111> (Fig. 2(i)).

EBSD is used to observe the microstructure characteristics of atmospheric-pressure and 3 GPa high-pressure quenched martensite, as shown in Fig. 3. In order to truly reflect the internal structure of the sample, the central part was selected for characterization. It can be seen from Fig. 3(a) that the larger martensite packets divided the grains into several regions. Different martensite packets or a collection of blocks can be identified from the color patterns. Several blocks in the same direction form a martensitic packet, and several martensitic laths with similar orientation and parallel to each other form a block.

Fig. 3.

EBSD characterizations of the lath martensite under 10⁻⁴ GPa (a–c), and 3 GPa (d–f). (a, d): the inverse pole figure (IPF) images; (b, e): Band contrast (BC) maps depicting boundary distribution (blue line: 10° > θ > 5°, black line: 45° > θ > 10°, red line: 55°>θ > 45°, green line: θ > 55°); (c, f): Frequency of boundaries with different misorientations. (Online version in color.)

In martensite structure, a unique packet containing blocks of parallel laths (sometimes one lath) can be distinguished from other blocks by high-angle boundaries. Therefore, the size of the martensite blocks can be calculated by EBSD. The results show that the size of the blocks is about 2–8 μm with the mean value of 4.1 μm without pressurization, while 0.45–8.00 μm with the mean value of 1.41 μm in 3 GPa-steel. The size of martensite blocks is related to the grain size of parent austenite. The smaller the prior austenite grain size, the smaller the block size of martensite is.²³⁾ The growth of austenite grain is a spontaneous process when heated at atmospheric pressure. However, the growth of austenite grain is strongly inhibited under high pressure, and the higher the pressure, the greater the inhibition effect.^24,25) After high-pressure quenching, the growth of austenite grain is strongly inhibited, so martensite blocks are refined. Therefore, fine lath martensite can be obtained after high-pressure quenching.

The frequency of boundaries with different misorientations of corresponding regions in the IPF diagram is shown in Figs. 3(c) and 3(f), to describe the proportion of different grain boundaries of atmospheric-pressure and high-pressure quenched martensite. The small-angle grain boundary with a misorientation angle less than 10° can be identified as a dislocation structure, while the grain boundary with a large angle of more than 55° is regarded as a twin structure. Figure 3(c) shows that small-angle grain boundary is more prevalent than twin boundary in atmospheric-pressure quenched martensite. However, the proportion of twin boundaries is 60% and that of small-angle grain boundaries is less than 20% in high-pressure quenched martensite, so the proportion of twin boundaries is much higher than that of small-angle boundaries. Therefore, twin boundary is dominant in 3 GPa-steel, while small-angle grain boundary representing dislocations accounted for a relatively small proportion. It is found that twin misorientations of 60°/<111> are dominant between laths in high-pressure quenched martensite, which confirms the orientation relationships of {112} <111> from the TEM observation of Fig. 2.

Figure 4 shows the XRD spectra of 0.2 mass% C steel quenched under different pressures. Peaks appearing in all samples could be indexed as α-Fe (110), (200), (211), which shows the consistency of the matrix structure. However, as shown in Fig. 4(b), the (110)_α diffraction peak shifts to the right and the symmetry changed after quenching at 3 GPa and 4 GPa.

Fig. 4.

XRD profiles of 0.2 mass% C steel under different pressures. (a): wide scan XRD profile; (b): XRD profile for α-Fe (110). (Online version in color.)

In order to determine the tetragonality (c/a > 1), (110)_α diffraction peak asymmetry factor (A_f), is assessed as follows:²⁶⁾   

$$A_{\text{f}}=m/n$$(1)

Where m, n represented the width of the peak at 10% peak height, measured to the left, and right respectively of the peak maximum position (as shown in the inset in Fig. 4(a)). When quenched at atmospheric pressure, the profile becomes symmetric because of the broadening of the peak width. As for 3 and 4 GPa-steel, the peak shows a fronting nature (A_f > 1) indicating that the tetragonal martensite starts to appear.

Also, it can be seen from Fig. 4(b) that the peak width of (110)_α of 0.2 mass% C steel after quenching at atmospheric pressure and 4 GPa was 0.42° and 0.60° respectively. That is, the diffraction peak of the quenched steel at high-pressure is wider than that at normal pressure. This indicates that the microstructure of high-pressure quenched martensite is finer and the defect density is higher.

The lattice constants and tetragonality in 3, 4 GPa-steel are calculated with SAED patterns in Fig. 5. The results are a=0.2846 nm, c=0.2877 nm, c/a=1.011 and a=0.2842 nm, c=0.2876 nm, c/a=1.012 at 3 GPa and 4 GPa, respectively. Comparatively, c/a=1.009, a=0.2858 nm, c=0.2884 nm at atmospheric pressure, according to the relationship proposed by Honda and Nishiyama:²⁷⁾   

$$\begin{array}{l}c=a_0+\alpha \rho \\ a=a_0-\beta \rho \\ c/a=1+\gamma \rho \end{array}$$(2)

Where a₀=2.861Å; α=0.116±0.002; β=0.013±0.002; ρ is the mass percent of carbon.

Fig. 5.

TEM images and SAED patterns of high-pressure quenched martensite. (a): 3 GPa; (b): 4 GPa. (Online version in color.)

Accordingly, the lattice constants a and c decreases with the increase of pressure, while c/a indicated opposite tendency. This suggests that there is high tetragonality in high-pressure quenched martensite.

The tetragonality is closely related to the order degree for C atom distribution. In martensite the carbon atoms occupying an octahedral interstice produce a large local distortion. Figure 6 shows the atomic arrangement of Fe and C atoms in martensite and austenite lattice. The sublattice formed by the possible positions of C atoms in the martensite lattice is shown in Fig. 6(a), where the first and second sublattices are on the a-axis and b-axis respectively, and the third sublattice is on the c-axis. If the probability of distribution of C atoms in the three sublattices is equal, i.e. disordered distribution, then martensite should be cubic lattice; if C atoms preferentially occupy the third sublattice of octahedral site, that is, C atoms are arranged parallel to [001], then martensite should be a tetragonal lattice. The more C atoms occupied the third sublattice, the higher the degree order of C atom distribution is.

Fig. 6.

Atomic arrangements in (a) martensite and (b) austenite.

Mössbauer spectroscopy is utilized to test the distribution of carbon in quenched martensite at atmospheric pressure and high pressure. C atoms occupied the third sublattice were recognized to be the most influenced and thus showed the smallest magnetic hyperfine field.²⁸⁾ The decrease of the hyperfine field of martensite structure indicates that more C atoms preferentially occupy the third sublattice of octahedral site, therefore, the tetragonality of martensite increases with the decrease of hyperfine field. Figure 7 shows the room-temperature Mössbauer spectra of 3, 4 GPa-steel, taking atmospheric quenching steel for comparison. The computed Mössbauer parameters which are obtained from the fitting of the spectra can be seen in Table 1.

Fig. 7.

Mössbauer spectra of 0.2 mass% C steel under different pressures. (Online version in color.)

Table 1. The Mössbauer parameters of 0.2 mass% C steel under different pressures.

Sample (mm/s)	Isomer Shift	Hyperfine field (T)	Line width (mm/s)	Relative content (%)
10−4 Gpa	0.00	32.97	0.325	94.6
	−0.04	–	0.782	5.4
3 GPa	0.00	32.96	0.354	100
4 GPa	0.00	32.94	0.369	100

The line width of martensite peaks in 10⁻⁴, 3 and 4 GPa-steel is 0.325, 0.354, 0.369 mm/s, respectively, which reveals that the degree of lattice distortions gradually enhanced with the increasing pressure. Moreover, the hyperfine field of high-pressure quenched martensite is lower than that of atmospheric pressure and decreased with the increase of pressure. The hyperfine field of martensite in 10⁻⁴, 3, and 4 GPa-steel are 32.97T, 32.96T, and 32.94T. The hyperfine field corresponded to the atomic structure characteristics of the phase. So that the decrease of the hyperfine field of martensite structure indicates that the nuclear magnetic moment sense of Fe nucleus is reduced by the intensity of the external magnetic field, which means the order degree of C atom distribution in martensite increases. Hence, high-pressure quenching enhances the order degree of C atom distribution.

Nearly 80% C atoms occupied the third sublattice preferentially in quenched martensite of 0.2 mass% C steel at atmospheric pressure, and 20% C atoms are distributed on the other two sublattices.²⁹⁾ However, the diffusion of C is suppressed when quenched at 3 and 4 GPa. More C remained in octahedral sites which inherited the characteristic in austenite can enlarge the degree of C distribution order.

Furthermore, the matrix martensite phase and the austenite phase can be distinguished, according to Mössbauer spectra, because the former is ferromagnetic while the latter is paramagnetic at room temperature.³⁰⁾ At atmospheric pressure, the green curve in Fig. 7 shows the Mossbauer spectrum characteristics of austenite, that is, it represents the existence of retained austenite. The mixed microstructures of 94.56% martensite and 5.44% austenite are observed under atmospheric pressure from Table 1, identical to the typical microstructure of low carbon steel in previous studies. The amount of retained austenite is too little to be examined by XRD. However, the martensite in the 0.2 mass% C steel after high-pressure quenching accounts for nearly 100%, indicating the complete transformation of martensite. The dominant twin relationship between laths in 3, 4 GPa-steel shown in Figs. 2 and 3 also provides convincing evidences that there is no retained austenite in high-pressure quenched martensite.

Thus, martensite transformation can easily occur with a low cooling rate of 10°C/s under GPa level high pressure (over 50°C/s under atmospheric pressure). The high-pressure martensite has the typical feature of lath martensite, and the size of the blocks is about 0.45–8.00 μm, less than that of atmospheric-pressure quenching (20–35 μm).¹⁰⁾ There is a dominant twin relationship between martensite laths, and there is almost no retained austenite. The tetragonality of martensite after high-pressure quenching increases from 1.009 at atmospheric pressure to 1.012 at 3 GPa and 4 GPa, and the order degree of C atom distribution is also improved.

It is generally accepted that a high cooling rate is critical for martensite transformation in low carbon steel. When 0.2 mass% C steel is quenched under atmospheric pressure, martensite can only be formed with the quenching medium of 10–15% NaOH or NaCl solution and the cooling rate of above 50°C/s because undercooled austenite is easy to decompose during the cooling process. However, martensite transformation can easily occur with a low cooling rate of 10°C/s under the pressure of 3, 4 GPa. The stability of undercooled austenite is improved, and the decomposition of austenite is inhibited when cooled at GPa level high pressure.

The pressure dependence of the diffusion coefficient (D(P)) of vacancies or interstitial atoms, such as carbon and nitrogen, is expressed as³¹⁾   

$$D(P)=D(0)\mathrm{e}\mathrm{x}\mathrm{p}(-P\mathrm{\Delta }V/RT)$$(3)

Where D(0) and D(P) are the diffusion coefficients at pressures of 0 and P, respectively, and ΔV is the activation volume, R the gas constant, and T the absolute temperature. For example, taking R=8.314 J/(K mol), T=700 K, ΔV=5×10⁻⁶ m³/mol, the results show D(P)/D(0)=1.00 (for P=10⁻⁴ GPa) and D(P)/D(0) = 0.032 (for P = 4 GPa).

Clearly, the diffusion coefficient of solute decreases exponentially with an increase in pressure. More ordered structure and less atom diffusion would elevate the chemical stabilization of austenite which impedes all other solid-state transformations such as ferrite and pearlite, so that martensite can be achieved with a quite slow cooling rate. From a thermodynamic point of view, high pressure would shift the transformation-time-temperature curve and lower the critical cooling rate under which martensite forms.

According to the following empirical formula by Capdevila,³²⁾ Martensite-start temperature (Ms) of the Fe-0.2C-0.45Mn-0.23Si (mass%) sample under atmospheric pressure is calculated as 413°C.   

$$\mathrm{M}\mathrm{s}[\mathrm{K}]=764.2-302.6\mathrm{C}-30.6\mathrm{M}\mathrm{n}-14.5\mathrm{S}\mathrm{i}[\mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s}\%]$$(4)

The application of GPa level high pressure will depress Ms in low carbon steels significantly. Fujita²⁵⁾ found that the Ms-temperature was lowered approximately 40°C/GPa in Fe–C Alloys Containing Mn, Si or Cr. Therefore, Ms temperatures under high-pressure of 3 and 4 GPa were 293°C and 253°C, respectively.

Martensite transforms from austenite lattice, causing volume expansion. However, because volume expansion is constrained at high pressure, the strain energy would increase, leading to the increase of transformation resistance. In the process of lattice transformation, the two adjacent variants with equal shear component and opposite direction could reduce the total strain energy of the transformation.³³⁾ Therefore, adjacent martensite laths of GPa-steel mostly follow symmetric twin orientation relationships of {112} <111>, and twin plane of {112}b.T//{112}f exists in martensite. The proportion of 60° <111> twin boundaries is 60% in 3 GPa-steel, while the relationship between adjacent laths quenched under atmospheric pressure is mostly non-twin orientation. When quenched at 4 GPa in 0.2 mass% C steel, a distinct hierarchical structure can be presented in martensite, and the symmetric growth of blocks is also observed. Hence, the total energy in the system can be decreased by lowering the strain energy through the self-accommodation effect among martensitic variants. The transformation takes place by twinning as the easiest mode due to the twin symmetry relationship between martensite laths under high pressure.

The decrease of mobility of grain boundaries could attribute to the refinment of parent austenite grains. In previous studies, it was explained that low-angle grain boundaries move by the vacancy diffusion mechanism.^34,35) However, the vacancy concentration in the material decreases with increasing pressure. Hence, the movement of low-angle grain boundaries is strongly restrained under high hydrostatic pressure. In the high-pressure samples there is a high density of nanotwin boundaries viewed as 60<111>. Krawczynska³⁶⁾ confirmed that the migration of twin boundaries is hampered under high pressure. Therefore, the mobility of grain boundaries could be hampered by high pressure, causing the grain refinement of high-pressure samples. Besides, the growth of parent austenite grains is strongly inhibited by high pressure, leading to a significant increase in the grain boundary density. Grain boundary defects are proliferated in austenite, providing more nucleation sites for martensite nucleation. Thus, the transformation rate of martensite nucleation at the initial stage have been improved.³⁷⁾

Generally, as pressure increases, denser phases are favored in the transformation. From this point, it can be easily inferred that the fcc structure (γ-austenite) is more stable than bcc structure (α-ferrite) under pressure. Due to the increasing stabilization of austenite, γ-austenite under high pressure could be retained much more than that under a normal pressure one after cooling to room temperature. However, the retained austenite grains might transform to martensite structure during unloading the pressure from 3 or 4 GPa to the normal pressure. Therefore, there is almost no residual austenite in 0.2 mass% C steel after high-pressure quenching.

3.2. Mechanical Properties

The change in the hardness of 0.2 mass% C steel after the heating and cooling cycle with the pressure is investigated, and the results are given in Fig. 8. The hardness of the steel, austenitized at 900°C (10 min), and quenched in 10% NaCl solution at atmospheric pressure is only 412 HV (42 HRC). Superior performance is exhibited in high-pressure quenching steel with the hardness of 581 HV (54 HRC) at 3 GPa and 654 HV (58 HRC) at 4 GPa, increasing by 58.74% and 70.39% compared with that of atmospheric quenching, respectively. The hardness of 3 GPa-steel is close to that of ordinary 0.45 mass% C steel (HRC 55), while the hardness of 4 GPa quenched martensite is higher than that of 0.45 mass% C steel.

Fig. 8.

The hardness bars of 0.2 mass% C steel under different pressures. (Online version in color.)

The hardness of low carbon steel mainly depends on the content of martensite and the hardness of martensite, and the latter is closely related to the carbon content and distribution in martensite. There is no retained austenite in the microstructure of 0.2 mass% C steel after high-pressure quenching, which partly explains the increase of hardness.

The supersaturated carbon atoms occupy octahedral interstitial sites of α phase, forming a tetragonal lattice distorted dipole stress field centered on the C atom. Under the influence of high-pressure quenching, the radius of flat octahedron decreases caused by the decrease of lattice constant, and the tetragonal lattice distorted dipole stress field enlarge due to the ordered distribution of carbon. Consequently, the hardness of quenched martensite under high pressure is greatly larger than that of under atmospheric pressure.

In summary, there are several possible factors concerned with the hardness promotion. A) Substructures with a high density of dislocations and fine twinning result in phase transformation strengthening. B) The growth of austenitic grain size is constrained under high pressure, which generates fine substructure of martensite leading to grain refinement strength. C) The lattice constant of martensite decreases, but the order degree of C distribution enhances, which bring about a higher solution strengthening effect. All these factors make a comprehensive contribution to the superior performance of the high-pressure quenched 0.2 mass% C steel.

4. Conclusion

(1) Martensite transformation occurs with a low cooling rate of 10°C/s under GPa level high pressure (over 50°C/s under atmospheric pressure). The morphology of 3 GPa-steel is typical lath martensite with the width of laths about 150–250 nm, and adjacent martensite laths follow the {112} <111> symmetrical twinning relationship. The proportion of twin boundaries is as high as 60%, and low angle grain boundaries are 20%. The characteristic of lath-block-packet is presented in 4 GPa-steel with the size of laths 100–200 nm. The growth of martensite displays symmetrically, indicating the self-accommodation among martensitic variants.

(2) High-pressure quenching leads to high tetragonality in martensite. The lattice constants and tetragonality are a=0.2846 nm, c=0.2877 nm, c/a=1.011 and a=0.2842 nm, c=0.2876 nm, c/a=1.012 at 3 GPa and 4 GPa, respectively. The lattice constants a and c are decreased with the increase of pressure, while c/a indicates opposite tendency. Meanwhile, high-pressure quenched martensite has finer grains and higher defect density. The peak width of (110)_α after quenching at 4 GPa is 0.60°, much higher than 0.42° at atmospheric pressure.

(3) High-pressure quenching enhances the order degree of C atom distribution and the degree of lattice distortions of 0.2 mass% C steel. The hyperfine field of martensite in 4 GPa-steel is 32.94 T, lower than that of atmospheric quenching (32.97 T). The peak width of martensite quenched at 4 GPa is 0.369 mm/s, which is higher than that of atmospheric quenching (0.325 mm/s).

(4) Superior performance is exhibited in high-pressure quenched steel with the hardness of 54 HRC at 3 GPa and 58 HRC at 4 GPa, increasing by 58.74% and 70.39%, respectively, compared with that of atmospheric quenching (55 HRC). The main mechanisms of hardness promotion of high-pressure quenched martensite are the increase of lattice distortions and defect density, especially the twin boundary.

Acknowledgments

This work was supported by the National Science Fund for Distinguished Young (Grant No. 51925105) and the National Natural Science Foundation of China (Grant No. 51675092).

References

• 1)   E. A.  Wilson: ISIJ Int., 34 (1994), 615.
• 2)   S.  Morito,  J.  Nishikawa and  T.  Maki: ISIJ Int., 43 (2003), 1475.
• 3)   N.  Gu,  X.  Song,  Z.  Ma,  R.  Wang and  F.  Yin: ISIJ Int., 37 (1997), 914.
• 4)   W. C.  Leslie: Metall. Trans., 3 (1972), 5.
• 5)   Z. N.  Zhou,  H.  Akamine,  J.  Li,  L.  Yang,  T.  Higashizono,  Q. D.  Hu,  M.  Nishida and  J. G.  Li: J. Alloy. Compd., 817 (2020), 153150.
• 6)   J. B.  Seol,  J. E.  Jung,  Y. W.  Jang and  C. G.  Park: Acta Mater., 61 (2013), No. 2, 558.
• 7)   G.  Krauss: Mater. Sci. Eng. A, 273–275 (1999), 40.
• 8)   B. B.  Wu,  Z. Q.  Wang,  X. L.  Wang,  W. S.  Xu,  C. J.  Shang and  R. D. K.  Misra: Mater. Sci. Eng. A, 759 (2019), 430.
• 9)   J.  Sun,  T.  Jiang,  Y.  Wang,  S.  Guo and  Y.  Liu: Mater. Sci. Eng. A, 726 (2018), 342.
• 10)   Y.  Wang,  J.  Sun,  T.  Jiang,  Y.  Sun,  S.  Guo and  Y.  Liu: Acta Mater., 158 (2018), 247.
• 11)   J. C.  Slater: Science, 148 (1965), 805.
• 12)   J. V.  Badding,  L. J.  Parker and  D. C.  Nesting: J. Solid State Chem., 117 (1995), 229.
• 13)   L.  Zhang,  Y.  Wang,  J.  Lv and  Y.  Ma: Nat. Rev. Mater., 2 (2017), 17005.
• 14)   T.  Kakeshita,  T.  Saburi and  K.  Shimizu: Mater. Sci. Eng. A, 273–275 (1999), 21.
• 15)   J. Y.  Kang,  S. C.  Kim,  J. O.  Oh,  H. N.  Han,  K. H.  Oh and  H. C.  Lee: Scr. Mater., 66 (2012), 45.
• 16)   H.  Zhang,  Y.  Wang,  Y.  Peng,  P.  Zhu,  J.  Liu,  Z.  Feng,  G.  Wu and  X.  Huang: Mater. Res. Lett., 7 (2019), 354.
• 17)   Z.  Lu,  W.  Zhu,  T.  Lu and  W.  Wang: Model. Simul. Mater. Sci. Eng., 22 (2014), 025007.
• 18)   Y.  Zhang,  Z.  Liu,  Z.  Zhao,  M.  Ma,  Y.  Shu,  W.  Hu,  R.  Liu,  Y.  Tian and  D.  Yu: J. Alloy. Compd., 695 (2017), 45.
• 19)   Y.  Zhang,  Y.  Ren,  P.  Wang,  Z.  Zhao,  Z.  Liu,  J.  He,  B.  Xu,  Y.  Tian and  D.  Yu: High Press. Res., 37 (2017), 278.
• 20)   B.  Yuan,  Q.  Tao,  X.  Zhao,  K.  Cao,  T.  Cui,  X.  Wang and  P.  Zhu: Rev. Sci. Instrum., 85 (2014), 013904.
• 21)   N. J.  Gu,  X. Y.  Song,  J. X.  Zhang,  F. X.  Yin and  R. X.  Wang: Metall. Mater. Trans. A, 26 (1995), 1979.
• 22)   W. X.  Liu: Acta Metall. Sin. (Engl. Lett.), 14 (1988), 397 (in Chinese).
• 23)   S.  Morito,  H.  Yoshida,  T.  Maki and  X.  Huang: Mater. Sci. Eng. A, 438–440 (2006), 237.
• 24)   Y.  Li,  Z.  Wang,  S.  Zhang,  N.  Liu,  Y.  Wang,  P.  Zhu,  D.  Yu,  Y.  Peng and  H.  Zhang: Scr. Mater., 187 (2020), 163.
• 25)   M.  Fujita,  I.  Uchiyama and  M.  Suzuki: Tetsu-to-Hagané, 60 (1974), 58 (in Japanese).
• 26)   T.  Tanaka,  N.  Maruyama,  N.  Nakamura and  A. J.  Wilkinson: Acta Mater., 195 (2020), 728.
• 27)   L.  Xiao,  Z.  Fan,  J. X.  Zhang,  M. X.  Zhang,  M. K.  Kang and  Z. Q.  Guo: Phys. Rev. B, 52 (1995), 9970.
• 28)   R.  Kaplow,  M.  Ron and  N.  DeCristofaro: Metall. Trans. A, 14 (1983), 1135.
• 29)   J.  Speer,  D. K.  Matlock,  B. C.  De Cooman and  J. G.  Schroth: Acta Mater., 51 (2003), 2611.
• 30)   J. I.  Besoky,  C. A.  Danon and  C. P.  Ramos: J. Mater. Res. Technol., 8 (2019), 1888.
• 31)   Y.  Minamino,  T.  Yamane,  A.  Shimomura,  M.  Shimada,  M.  Koizumi,  N.  Ogawa,  J.  Takahashi and  H.  Kimura: J. Mater. Sci., 18 (1983), 2679.
• 32)   C.  Capdevila,  F. G.  Caballero and  C.  García de Andrés: ISIJ Int., 42 (2002), 894.
• 33)   P.  Zhang,  Y.  Chen,  W.  Xiao,  D.  Ping and  X.  Zhao: Prog. Nat. Sci.: Mater. Int., 26 (2016), 169.
• 34)   L. E.  Tanner and  S. V.  Radcliffe: Acta Metall., 10 (1962), 1161.
• 35)   V.  Sursaeva,  S.  Protasova,  W.  Lojkowski and  J.  Jun: Textures Microstruct., 32 (1999), 175.
• 36)   A. T.  Krawczynska,  S.  Gierlotka,  P.  Suchecki,  D.  Setman,  B.  Adamczyk-Cieslak,  M.  Lewandowska and  M.  Zehetbauer: J. Mater. Sci., 53 (2018), 11823.
• 37)   C.  Celada-Casero,  J.  Sietsma and  M. J.  Santofimia: Mater. Des., 167 (2019), 107625.