Effects of Vacuum-Carburizing Conditions on Surface-Hardened Layer Properties of Transformation-Induced Plasticity-Aided Martensitic

The effects of carbon potential in vacuum-carburization on the surface-hardened layer properties of the 0.2%C-1.5%Si-1.5%Mn-1.0%Cr-0.05%Nb transformation-induced plasticity-aided martensitic steel were investigated for the fabrication of precision gears. The volume fraction of retained austenite and hardness in the surface hardened layer of the steel increased with increasing carbon potential. Subsequent fine-particle peening enhanced the hardness and the compressive residual stress via severe plastic deformation and strain-induced martensite transformation, especially under a high carbon potential. The severe plastic deformation mainly contributed to increased hardness and compressive residual stress and the contribution of the strain-induced martensitic transformation was relatively small.


Introduction
Recently, the demand for precision transmission gears with a high torque capacity and high wear resistance has increased, with the gears also being downsized to reduce energy consumption [1].Next-generation structural steels such as ultrahigh-strength transformation-induced plasticity (TRIP)-aided steels, i.e., "TRIP-aided martensitic (TM) steel" [2][3][4], "quenching and partitioning steel" [5,6] and "nanostructured bainitic steel" [7,8] are expected to be increasingly used for fabricating precision gears because of their high toughness and superior fatigue strength [3][4][5][6][7][8][9].Sugimoto et al. [10,11] reported that fine-particle peening (FPP) further increases the rotational bending and torsional fatigue strengths of heat-treated TM steel by imparting a significant hardness and compressive residual stress to the precision-gear surface layer.Numerous researchers have reported that high-energy shot peening also increases the fatigue strength of these steels [12,13].Higher fatigue strengths of TM steel may be achieved by gas-or vacuum-carburizing with shot peening and/or FPP [14][15][16][17][18].To date, there have only been a few reports on the microstructure and the fatigue strength of carburized TM steel subjected to FPP [19,20].However, no investigation has been conducted into the effects of vacuum-carburization conditions on the surface-hardened layer properties of TM steel.
In this study, the surface-hardened layer properties of 0.2%C-1.5%Si-1.5%Mn-1.0%Cr-0.05%NbTM steel subjected to vacuum-carburization under various carbon potentials and subsequent FPP were investigated.In addition, the roles of strain-induced martensitic transformation of the retained austenite and severe plastic deformation on the surface-hardened layer properties were analyzed.

Materials and Methods
Steel slabs were vacuum-melted, hot-forged, and hot-rolled into bars with a 13 mm diameter.The chemical composition of the steel slabs is given in Table 1.Round bars 5 mm in diameter and 90 mm in length were machined from the 13 mm bars and ground using Emery paper #2000.Carbon potentials were varied in a range between 0.20 mass % and 0.80 mass %.The heat-treated TM steel [11] which was subjected to quenching to 50 • C, isothermal transformation at 50 • C for 1000 s and partitioning at 350 • C for 1000 s had a carbon potential of 0.20 mass %.The specimens corresponding to a carbon potential range between 0.70 mass % and 0.80 mass % were subjected to vacuum-carburization in enriched gas with various concentrations of hydrocarbons and then quenching in oil at 80 • C followed by tempering at 180 • C for 90 min (Figure 1) in a vacuum-carburizing batch furnace (VCB; IHI Machinery and Furnace Co., Ltd., Tokyo, Japan).For comparison, commercial JIS-SNCM420 steel bars were prepared using the same vacuum-carburization conditions as the TM steel.For SNCM420 steel corresponding to a carbon potential of 0.20 mass %, the heat-treatment of quenching to 50 • C and tempering at 200 • C for 3600 s was conducted.

Materials and Methods
Steel slabs were vacuum-melted, hot-forged, and hot-rolled into bars with a 13 mm diameter.The chemical composition of the steel slabs is given in Table 1.Round bars 5 mm in diameter and 90 mm in length were machined from the 13 mm bars and ground using Emery paper #2000.Carbon potentials were varied in a range between 0.20 mass % and 0.80 mass %.The heat-treated TM steel [11] which was subjected to quenching to 50 °C, isothermal transformation at 50 °C for 1000 s and partitioning at 350 °C for 1000 s had a carbon potential of 0.20 mass %.The specimens corresponding to a carbon potential range between 0.70 mass % and 0.80 mass % were subjected to vacuumcarburization in enriched gas with various concentrations of hydrocarbons and then quenching in oil at 80 °C followed by tempering at 180 °C for 90 min (Figure 1) in a vacuum-carburizing batch furnace (VCB; IHI Machinery and Furnace Co., Ltd., Tokyo, Japan).For comparison, commercial JIS-SNCM420 steel bars were prepared using the same vacuum-carburization conditions as the TM steel.For SNCM420 steel corresponding to a carbon potential of 0.20 mass %, the heat-treatment of quenching to 50 °C and tempering at 200 °C for 3600 s was conducted.FPP was carried out on the surface of the cylindrical specimens under the conditions listed in Table 2.According to a previous study [20], these conditions achieve the highest fatigue strength for vacuum-carburized TM and SNCM420 steels.The microstructure of the steels was observed by field emission-scanning electron microscopy (FE-SEM; JSM-6500F, JEOL Ltd., Tokyo, Japan), which was performed using an instrument equipped with an electron backscatter diffraction system (EBSD; OIM system, TexSEM Laboratories, Inc., Draper City, UT, USA).The specimens for the FE-SEM-EBSD analysis were first ground with alumina powder and colloidal silica, then subjected to ion thinning.Vickers hardness was measured by a micro Vickers hardness tester (HVM, Shimadzu Co., Kyoto, Japan).The testing load was 4.9 N. Surface roughness was measured by laser scanning microscopy (LSM; VK-8510, Keyence Co., Milton Keynes, UK) and was evaluated by the maximum roughness height (Rz, JIS B0601:2001).
The volume fraction of the retained austenite phase was quantified from the integrated intensity of the (200)α, (211)α, (200)γ, (220)γ, and (311)γ peaks in the X-ray diffraction pattern (RINT2000, Rigaku Co., Tokyo, Japan) [21].The cosα method [22] was applied to X-ray studies of residual stress in a longitudinal direction using an X-ray residual stress analyzer (μ-X360, Pulstec Ind. Co. Ltd., Hamamatsu, Japan).The measurement conditions and material constants used are described in detail in [22].FPP was carried out on the surface of the cylindrical specimens under the conditions listed in Table 2.According to a previous study [20], these conditions achieve the highest fatigue strength for vacuum-carburized TM and SNCM420 steels.The microstructure of the steels was observed by field emission-scanning electron microscopy (FE-SEM; JSM-6500F, JEOL Ltd., Tokyo, Japan), which was performed using an instrument equipped with an electron backscatter diffraction system (EBSD; OIM system, TexSEM Laboratories, Inc., Draper City, UT, USA).The specimens for the FE-SEM-EBSD analysis were first ground with alumina powder and colloidal silica, then subjected to ion thinning.Vickers hardness was measured by a micro Vickers hardness tester (HVM, Shimadzu Co., Kyoto, Japan).The testing load was 4.9 N. Surface roughness was measured by laser scanning microscopy (LSM; VK-8510, Keyence Co., Milton Keynes, UK) and was evaluated by the maximum roughness height (R z , JIS B0601:2001).

Surface-Hardened Layer Properties of Vacuum-Carburized Steels
Results from the EBSD analyses of TM steel subjected to vacuum-carburization under various carbon potentials are shown in Figure 2. The microstructure of the TM steel was characterized as having a mixture of α'-martensite structure matrix with wide lath size, martensite-austenite (MA)-like phases and retained austenite islands.Two types of retained austenite were observed; island type of retained austenite isolated in the matrix and filmy type of retained austenite located along fine and narrow martensite lath boundaries in the MA-like phase [2,3].The volume fractions of the retained austenite and the MA-like phase increased with increasing carbon potential in the TM steel.The TM steel was found to possess a larger amount of retained austenite and a smaller amount of MA-like phase than the SNCM420 steel [19].

Surface-Hardened Layer Properties of Vacuum-Carburized Steels
Results from the EBSD analyses of TM steel subjected to vacuum-carburization under various carbon potentials are shown in Figure 2. The microstructure of the TM steel was characterized as having a mixture of α'-martensite structure matrix with wide lath size, martensite-austenite (MA)like phases and retained austenite islands.Two types of retained austenite were observed; island type of retained austenite isolated in the matrix and filmy type of retained austenite located along fine and narrow martensite lath boundaries in the MA-like phase [2,3].The volume fractions of the retained austenite and the MA-like phase increased with increasing carbon potential in the TM steel.The TM steel was found to possess a larger amount of retained austenite and a smaller amount of MA-like phase than the SNCM420 steel [19].To eliminate complications, residual stress distribution of TM steel was subjected to vacuumcarburization under a carbon potential of 0.80 mass % in Figure 3e.
Figure 4 shows the carbon potential dependences of the maximum retained austenite fraction, hardness, and compressive residual stress.The maximum retained austenite fraction, hardness, and compressive residual stress were strongly dependent on the carbon potential of vacuumcarburization except for the maximum residual stress.The maximum values of retained austenite fraction and Vickers hardness were slightly higher in the TM steel than in the SNCM420 steel [19].Figure 4 shows the carbon potential dependences of the maximum retained austenite fraction, hardness, and compressive residual stress.The maximum retained austenite fraction, hardness, and compressive residual stress were strongly dependent on the carbon potential of vacuum-carburization except for the maximum residual stress.The maximum values of retained austenite fraction and Vickers hardness were slightly higher in the TM steel than in the SNCM420 steel [19].

Surface-Hardened Layer Properties of Vacuum-Carburized and Fine-Particle Peened Steels
A typical top-view LSM image of the surface and a cross-sectional side view SEM image of the surface-hardened layer in vacuum-carburized TM steel subjected to FPP are shown in Figure 5.The microstructure just below the surface is refined in all steels.A white layer (nanostructure or an amorphous structure) [12,23] was not observed on the surface of either steel subjected to FPP.There was no difference in microstructures of the TM steels subjected to vacuum-carburizing under various carbon potentials.Surface roughness of the TM steel ranged from Rz = 5.0 μm to 5.2 μm, which was the same as that of SNCM420 steel.
Figure 3b,d,f shows the distribution of volume fraction of untransformed retained austenite, Vickers hardness, and X-ray residual stress of the α-bcc phase in the surface-hardened layer of the carburized TM steel subjected to FPP.A large amount of retained austenite transformed to martensite on the surface and at depths less than 20 μm from the surface.Consequently, the volume fraction of strain-induced martensite was highest on the surface or at the subsurface region.Hardness was also highest on the surface under all carbon potentials tested.The compressive residual stress was highest right under the surface, at a depth of 10 μm. , thin lines) steels after vacuum-carburization (VC) or vacuum carburization and subsequent fine-particle peening (VC + FPP).HV max,c is HV max of as-vacuum-carburized steels.

Surface-Hardened Layer Properties of Vacuum-Carburized and Fine-Particle Peened Steels
A typical top-view LSM image of the surface and a cross-sectional side view SEM image of the surface-hardened layer in vacuum-carburized TM steel subjected to FPP are shown in Figure 5.The microstructure just below the surface is refined in all steels.A white layer (nanostructure or an amorphous structure) [12,23] was not observed on the surface of either steel subjected to FPP.There was no difference in microstructures of the TM steels subjected to vacuum-carburizing under various carbon potentials.Surface roughness of the TM steel ranged from R z = 5.0 µm to 5.2 µm, which was the same as that of SNCM420 steel.

Surface-Hardened Layer Properties of Vacuum-Carburized and Fine-Particle Peened Steels
A typical top-view LSM image of the surface and a cross-sectional side view SEM image of the surface-hardened layer in vacuum-carburized TM steel subjected to FPP are shown in Figure 5.The microstructure just below the surface is refined in all steels.A white layer (nanostructure or an amorphous structure) [12,23] was not observed on the surface of either steel subjected to FPP.There was no difference in microstructures of the TM steels subjected to vacuum-carburizing under various carbon potentials.Surface roughness of the TM steel ranged from Rz = 5.0 μm to 5.2 μm, which was the same as that of SNCM420 steel.
Figure 3b,d,f shows the distribution of volume fraction of untransformed retained austenite, Vickers hardness, and X-ray residual stress of the α-bcc phase in the surface-hardened layer of the carburized TM steel subjected to FPP.A large amount of retained austenite transformed to martensite on the surface and at depths less than 20 μm from the surface.Consequently, the volume fraction of strain-induced martensite was highest on the surface or at the subsurface region.Hardness was also highest on the surface under all carbon potentials tested.The compressive residual stress was highest right under the surface, at a depth of 10 μm. Figure 3b,d,f shows the distribution of volume fraction of untransformed retained austenite, Vickers hardness, and X-ray residual stress of the α-bcc phase in the surface-hardened layer of the carburized TM steel subjected to FPP.A large amount of retained austenite transformed to martensite on the surface and at depths less than 20 µm from the surface.Consequently, the volume fraction of strain-induced martensite was highest on the surface or at the subsurface region.Hardness was also highest on the surface under all carbon potentials tested.The compressive residual stress was highest right under the surface, at a depth of 10 µm.
The maximum values obtained for the surface-hardened layer properties are listed in Table 3 and their corresponding carbon potential dependences are shown in Figure 4.With increasing carbon potential, the maximum volume fraction of the retained austenite, the maximum hardness, and the maximum compressive residual stress all increased, although the maximum harness under carbon potentials of 0.70 mass % to 0.80 mass % was nearly the same.These maximum values are slightly higher than those of the SNCM420 steel [19].The maximum volume fraction of strain-induced martensite of the steels subjected to FPP was obtained on the surface or at the subsurface region under all carbon potentials tested.

Maximum Hardness
In general, vacuum-carburization enhances the hardness and volume fraction of retained austenite in the surface-hardened layer, although the retained austenite fraction on the surface is decreased somewhat.Subsequent FPP increases both the hardness and the compressive residual stress in the surface-hardened layer of steels via severe plastic deformation and strain-induced martensitic transformation [3,9,12,[14][15][16][17][18]23].
Figure 6a shows the relationship between the maximum hardness and the strain-induced martensite fraction in the TM and SNCM420 steels.An increment in hardness after vacuum-carburization and FPP (∆HV max = HV max − HV max,c ) is the sum of the increments in hardness due to the severe plastic deformation and the strain-induced martensitic transformation (∆HV max = ∆HV p + ∆HV t ), respectively.HV max and HV max,c are Vickers hardness after vacuum-carburization and FPP and after vacuum-carburization, respectively (Table 3).The ∆HV max increases with increasing strain-induced martensite fraction.∆HV t can be estimated using Equation (1), where ∆f α m,max is the maximum volume fraction of strain-induced martensite after FPP (Table 3), HVα m * is Vickers hardness of the strain-induced martensite with carbon concentration in the retained austenite of 0.70−0.80mass % (Table 4).The carbon concentration of the retained austenite (Cγ = 0.80 mass %) determined in Reference [2] was used as the value for the heat-treated steel, corresponding to carbon potential of 0.20 mass %.Vickers hardness of martensite determined in Reference [24] was substituted as HVα m * in Equation ( 1).As shown in Figure 6a and Table 4, the calculated ∆HV t values of the TM and SNCM420 steels are between 10 and 21 in a carbon potential range between 0.70 mass % and 0.80 mass %.These values are much smaller than ∆HV max and the ratios of ∆HV t to ∆HV max in carburized TM steel subjected to FPP are between 0.05 and 0.11 (Figure 7 and Table 4), although the ratios increase with increasing carbon potential in vacuum-carburization, except for a carbon potential of 0.2 mass %.Therefore, it can be assumed that ∆HV p is much higher than ∆HV t in TM and SNCM420 steels and is the major factor that affects the change in hardness after FPP.In Figures 3d and 4b, the effect of carbon potential on the maximum hardness was negligibly small in a carbon potential range between 0.70 mass % and 0.80 mass %.This may be caused by the low ratio of ∆HV t to ∆HV max which high hardness of matrix structure weakens the influence of the strain-induced martensite transformation.In Table 4, the ratios of ∆HV t to ∆HV max in a case of a carbon potential of 0.2 mass % were relatively high (0.23 to 0.25).This is associated with low hardness of matrix structure, and it differs from the cases of high carbon potentials.
Metals 2017, 7, 301 7 of 11 7 and Table 4), although the ratios increase with increasing carbon potential in vacuum-carburization, except for a carbon potential of 0.2 mass %.Therefore, it can be assumed that ΔHVp is much higher than ΔHVt in TM and SNCM420 steels and is the major factor that affects the change in hardness after FPP.In Figures 3d and 4b, the effect of carbon potential on the maximum hardness was negligibly small in a carbon potential range between 0.70 mass % and 0.80 mass %.This may be caused by the low ratio of ΔHVt to ΔHVmax which high hardness of matrix structure weakens the influence of the strain-induced martensite transformation.In Table 4, the ratios of ΔHVt to ΔHVmax in a case of a carbon potential of 0.2 mass % were relatively high (0.23 to 0.25).This is associated with low hardness of matrix structure, and it differs from the cases of high carbon potentials.and E = 206 GPa into Equation (3), ∆σ Xα,t can be calculated as shown in Figure 6b and Table 4.The calculated ∆σ Xα,t values for TM and SNCM420 steels linearly increase with the carbon potential, in the same manner as the maximum compressive residual stress (σ Xα,max ).The ∆σ Xα,t values, however, are much smaller than σ Xα,max and the ratios of ∆σ Xα,t to ∆σ Xα,max of vacuum-carburized TM steel are between 0.10 and 0.13 (Figure 7 and Table 4), although the ratios increase with increasing carbon potential of vacuum carburization.Therefore, high compressive residual stress is likely to be caused by the compressive residual stress, primarily due to large plastic strain (∆σ Xα,p = ∆σ Xα,max − ∆σ Xα,t ).
In addition, it is considered that the ∆σ Xα,t mainly brought on the carbon potential dependence of the maximum compressive residual stress, in connection with that of the strain-induced martensite fraction.

Estimation of Fatigue Limit
Matsui et al. [14,15] reported that the sum of estimated yield stress (σ Y,est ) and the absolute value of maximum compressive residual stress {abs(σ Xα,max )} is related to the rotational bending fatigue limit (σ w ) of gas-carburized steel subjected to shot peening, as follows: where the value of σ Y,est can be estimated from HV max by: σ Y,est = (HV max /3) × 9.80665 × (YS/TS), where YS/TS is the yield ratio defined by the ratio of yield stress to tensile strength (assumed by Matsui et al. [14,15] to be 0.95).
Figure 8 shows the relationship between σ Y,est + abs(σ Xα,max ) and carbon potential of vacuum-carburization for the TM and SNCM420 steels.The value of σ Y,est + abs(σ Xα,max ), given in Table 4, increases with increasing carbon potential in both steels, although the values for TM steel are higher than for SNCM420 steel.This indicates that the fatigue limits of both steels are increased by vacuum-carburization under high carbon potential.In general, the fatigue limit of ultrahigh strength steel is also controlled by the initiation depth of fish-eye cracks.The relationship between σ w and σ Y,est + abs(σ Xα,max ) is described in detail elsewhere [28].
Figure 8 shows the relationship between σY,est + abs(σXα,max) and carbon potential of vacuumcarburization for the TM and SNCM420 steels.The value of σY,est + abs(σXα,max), given in Table 4, increases with increasing carbon potential in both steels, although the values for TM steel are higher than for SNCM420 steel.This indicates that the fatigue limits of both steels are increased by vacuumcarburization under high carbon potential.In general, the fatigue limit of ultrahigh strength steel is also controlled by the initiation depth of fish-eye cracks.The relationship between σw and σY,est + abs(σXα,max) is described in detail elsewhere [28].

Summary
The effects of the carbon potential of vacuum-carburization on the surface-hardened layer properties of a 0.2%C-1.5%Si-1.5%Mn-1.0%Cr-0.05%NbTM steel were investigated for the fabrication of precision gears.The main results can be summarized as follows: (1) Vacuum-carburization resulted in a large amount of retained austenite and high Vickers hardness in the surface hardened layer of the steel with a low residual stress from 0 MPa to 60 MPa.The retained austenite fraction and the hardness increased with increasing carbon potential during vacuum-carburization, although the residual stress was independent of the carbon potential.(2) Subsequent FPP enhanced the compressive residual stress and the volume fraction of straininduced martensite, especially in the surface hardened layer, with a further increase in hardness.Relationship between the sum of the estimated yield stress and the absolute value of maximum compressive residual stress {σ Y,est + abs(σ Xα,max )} and carbon potential in vacuum-carburization (C P ) in TM ( ) and SNCM420 ( ) steels subjected to fine-particle peening.

Summary
The effects of the carbon potential of vacuum-carburization on the surface-hardened layer properties of a 0.2%C-1.5%Si-1.5%Mn-1.0%Cr-0.05%NbTM steel were investigated for the fabrication of precision gears.The main results can be summarized as follows:

Figure 1 .
Figure 1.Vacuum-carburization and tempering diagram, in which "O.Q." represents quenching in oil at 80 • C, C P is carbon potential.

Figure
Figure 3a,c,e shows the distribution of volume fraction of untransformed retained austenite, Vickers hardness, and X-ray residual stress in the surface-hardened layer of the vacuum-carburized TM steel.The retained austenite fraction and the carburization depth increased with increasing

Figure 2 .
Figure 2. (a-c) Phase maps and (d-f) orientation maps of the α-bcc phase of a cross-sectional side view of the surface-hardened layer of transformation-induced TRIP-aided martensitic (TM) steel subjected to vacuum-carburization under various carbon potentials of (a,d) C P = 0.70 mass %, (b,e) C P = 0.75 mass % and (c,f) C P = 0.80 mass %.In (a-c), yellowish green, gray and red regions represent the α'-martensite (α m ) matrix, MA-like (MA) phase and retained austenite (γ R ) islands, respectively.

Figure
Figure 3a,c,e shows the distribution of volume fraction of untransformed retained austenite, Vickers hardness, and X-ray residual stress in the surface-hardened layer of the vacuum-carburized

Figure 5 .
Figure 5.Typical (a) laser scanning microscopy (LSM) image of the top-view of the surface and (b) SEM image of the cross-sectional side view of the surface-hardened layer of the TM steel subjected to vacuum-carburization under a carbon potential of 0.8 mass % and subsequent fine-particle peening.

Figure 8 .
Figure 8. Relationship between the sum of the estimated yield stress and the absolute value of maximum compressive residual stress {σY,est + abs(σXα,max)} and carbon potential in vacuumcarburization (CP) in TM (•) and SNCM420 (▲) steels subjected to fine-particle peening.

Figure 8 .
Figure 8.Relationship between the sum of the estimated yield stress and the absolute value of maximum compressive residual stress {σ Y,est + abs(σ Xα,max )} and carbon potential in vacuum-carburization (C P ) in TM ( ) and SNCM420 ( ) steels subjected to fine-particle peening.

Table 1 .
Chemical composition [mass %] of the steels used in this study.TM: TRIP-aided martensitic.

Table 1 .
Chemical composition [mass %] of the steels used in this study.TM: TRIP-aided martensitic.
Figure 1.Vacuum-carburization and tempering diagram, in which "O.Q." represents quenching in oil at 80 °C, CP is carbon potential.

Table 3 .
Measured surface-hardened layer properties of TM and SNCM420 steels subjected to vacuum-carburization under various carbon potentials and subsequent fine-particle peening.

Table 4 .
Estimated surface-hardened layer properties of TM and SNCM420 steels subjected to vacuum-carburization under various carbon potentials and subsequent fine-particle peening.