Intervention mechanism of carbide precipitation during tempering and its impact on residual stress and mechanical properties of wear-resistant steels

Abstract

The production of low-stress, high-performance wear-resistant steels poses a formidable challenge in the steel industry. This article delves into the carbide precipitation behaviour of wear-resistant steel NM300TP during tempering and its subsequent impact on residual stress and mechanical properties. Our findings reveal a crucial aspect: prolonging the holding time can interfere with the precipitation behaviour of carbides, thereby altering the relaxation mechanism of residual stress and its influence on material properties. Remarkably, we discovered that substituting the formation of θ-cementites with the precipitation of ε-transition carbides can yield a steel with significantly lower residual stress while maintaining exceptional mechanical properties. Specifically, when θ-cementites precipitate, the residual stress in the wear-resistant steel diminishes by 54.23%, accompanied by a hardness measurement of 299.58 HV_0.5. However, when a substantial quantity of ε-transition carbides is dispersed and precipitated, the residual stress is further diminished by 69.51%, and the hardness reaches 317.2 HV_0.5. Furthermore, this precipitation of ε-transition carbides exhibits superior wear resistance compared to θ-cementites. These groundbreaking discoveries offer critical theoretical insights for the development of low-stress, high-performance wear-resistant steels, paving the way for innovative advancements in the steel industry.

Keywords

wear-resistant steel residual stress mechanical property tempering carbide precipitation

Introduction

Due to its rapid three-stage cooling process, low-alloy wear-resistant steel introduces high-amplitude residual stress within its structure, which can result in material deformation and failure due to the presence of residual stress.^1,2 It has been demonstrated in practice that tempering serves as a beneficial heat treatment method for reducing the residual stress and enhancing the plate shape during the production process.³ Past experiences have indicated that augmenting the tempering temperature and prolonging the holding time can relax the residual, although concurrently causing a reduction in the material's mechanical properties. However, low-alloy wear-resistant is widely used in the production and manufacturing of construction machinery, mining machinery, and commercial vehicles due to its exceptional comprehensive mechanical properties.⁴ As a result, it is an important research topic to relieve residual stress of wear-resistant steel by optimising the tempering heat treatment process while taking into account the excellent mechanical properties.

Ritter et al.⁵ believed that the change in the mechanical behaviour of materials during the tempering process is the main reason for residual stress relaxation, and they attempted to reveal the inherent mechanism of residual stress evolution by investigating the changes in the elastic modulus, yield strength, and creep behaviour of materials during the tempering process. This research also provides a theoretical basis for reducing the residual stress distribution by increasing the tempering temperature and extending the tempering time. Morra et al.⁶ found that irreversible plastic deformation similar to phase transformation plasticity occurs when the nanoscale carbides in the material are coarsened. This study provides a new entry point for exploring the change of residual stress during tempering. Yamazaki et al.^7,8 found that during the tempering process, when the applied load is lower than the yield strength of the material, the material produces irreversible plastic behaviour in the direction of the load. This plasticity was named as ‘tempering plasticity’. It is used as the main influence factor to predict the residual stress during the tempering process. Subsequently, Kaiser et al.^9–11 found that plastic behaviour similar to phase transformation plasticity occurs in cementite precipitation stage during the tempering process and predicted the evolution of residual stress during tempering by the Greenwood–Johnson model. Ding et al.^12,13 confirmed the existence of irreversible plastic behaviour during carbide precipitation. They further pointed out that this plastic behaviour caused by carbide precipitation is the main reason for the relaxation of residual stress during tempering. It is worth noting that the precipitation, coarsening and dissolution of carbides can lead to different plastic behaviours, resulting in differences in the relaxation degree of residual stress. However, previous studies have mostly focused on the relaxation of residual stress during tempering, ignoring the changes in material mechanical properties during this process. Therefore, we need to comprehensively investigate the comprehensive impact of the tempering process on material properties. It is widely acknowledged that during tempering, there are multiple overlapping and distinct phases of organisational change, including carbon segregation and aggregation, transition carbide precipitation, decomposition of residual austenite, cementite precipitation and precipitation of the alloy carbides.^14–16 As the parameters are changed during the tempering process, the nature of the carbides undergoes a transformation, resulting in a profound effect on the mechanical properties. Specifically, as the tempering temperature gradually increases, the coarsening of carbides inevitably leads to deterioration of the mechanical properties of wear-resistant steels.¹⁷ Liang et al.¹⁸ found that as the tempering temperature increases, the activity of the carbon atoms increases, the carbide size becomes larger, and the solid solution strengthening and precipitation strengthening are weakened, resulting in a decrease in strength. On the other hand, Yi et al.¹⁹ found that with the extension of the tempering holding time, the carbides undergo coarsening and aggregation, the carbide size increases and the volume fraction decreases, leading to a decrease in impact toughness and wear resistance. However, when it comes to low-alloy wear-resistant steel, its wear resistance and hardness are crucial performance indicators. Therefore, how to control the precipitation of carbides while optimising the tempering process, thereby improving residual stress while maintaining high mechanical properties, remains an urgent problem facing researchers. Currently, there is no reasonable method that can perfectly solve this issue.

Therefore, in this article, we employed a strategy of extending the holding time at low temperature to intervene in the precipitation behaviour of carbides during the tempering process of wear-resistant steel NM300TP. The characterisation of carbide precipitation is carried out through dilatometry curve analysis, dislocation density measurement and examination of the microstructure. The effects of different carbide precipitations on residual stress and mechanical properties during tempering were investigated. Additionally, we aim to explore the intervening plastic behaviour of carbide precipitation during tempering and their effect on residual stress and mechanical properties improvement.

Experimental

Materials

The material for this study is a low-alloy wear-resistant steel NM300TP with a thickness of 4 mm. The chemical composition is shown in Table 1.

Table 1.

Chemical composition of the specimen (mass. Wt-%).

C	Si	Mn	V	Ti	Nb	Cu	Cr	Fe
0.1368	1.2048	1.7126	0.0031	0.0214	0.0009	0.028	0.0439	Bal.

Residual stress measurement

Eight specimens (60 mm × 70 mm × 4 mm) were taken from the middle of the steel plate with a wire cutter. One of the specimens was used for the original specimen. Respectively, four specimens were placed in a box-type heating furnace with stable temperature (100/200/300/350 °C) and held for 30 min. Next, the furnace was stabilised at 200 °C, and the remaining three specimens were placed in turn and held for 30/45/60 min, respectively. The finished specimens after heat treatment were air cooled to room temperature. The crack compliance method (CCM)^20,21 was used to measure the residual stress in the original test before tempering and the specimens after tempering. The principle of the CCM involves introducing a gradually deepening crack with a depth of a_j into the tested object with a thickness of t to release the residual stress in the material. During the process of residual stress release, strain values at certain specific points on the tested object are collected. By comparing the strain values with the compliance function calculated using the finite element software, the residual stress σ_xy can be inversely calculated, as shown in Figure 1.

To ensure accurate stress measurements and minimise the influence of introduced cracks on the stress at the measurement point, a wire electric discharge machine (EDM) was employed to introduce cracks precisely in the centre of the specimen surface. The EDM wire cutting process, specifically using the DK7732SF model with a $φ$ 0.18 mm molybdenum wire, allowed for incremental slitting with precise control. After cutting, the width of the slit measured 0.20 mm. The released strain was accurately read from a commercial bridge-type indicator (DH3818Y) attached to a foil phenolic-acetal resistance strain gage (BX120-2AA), featuring a sensitive grid size of 2.0 × 1.0 mm, ensuring reliable and repeatable measurements.

Transmission electron microscopy characterisation

A size specification of 10 mm × 10 mm × 0.1 mm was cut from the aforementioned heat-treated specimen and ground to a thickness of 60 μm with different grits. In addition, the thin wafer of Φ3 mm was punched out using a special puncher and processed using a twin-jet electropolisher. The electrolyte used during the twin-jet electropolisher was a 5-vol.-% perchloric acid-alcohol solution. The morphology, distribution and composition of carbides of the prepared specimen were observed by JOEL JSM 2100F 200 kV field emission transmission electron microscopy (TEM) and energy dispersive spectrometer (EDS).

Dislocation density measurement

A 10 mm × 10 mm × 4 mm specimen was cut from the aforementioned heat-treated specimen, and its dislocation density was measured using a Bruker D8 intelligent X-ray diffractometer. The specific test parameters were as follows: the target was Cu, the wavelength was λ = 0.1540538 nm, the tube voltage and tube current were 40 kV and 150 mA, the continuous scanning step was 0.01°, the scanning speed was 0.5° min⁻¹, and data were collected in the 2θ diffraction angle range of 30 to 120°, in which the diffraction peaks of the phase crystal plane were observed. The dislocation density was calculated by the Williamson–Hall (WH) method.

Non-isothermal tempering tests

In order to reveal the characteristics related to the tempering process, the specimen with the size of ø4 mm × 10 mm (Figure 2(b)) was put into the TA DIL805L (TA Instruments Corp., New Castle, PA, USA) dilatometer for test. The test's procedural flowchart is shown in Figure 2(a), the heating rate was 2 °C min⁻¹ to 350 °C and subsequently cooled down to room temperature at a cooling rate of 50 °C min⁻¹ to complete the first tempering, and the second temperature-raising and loading process was precisely the same as the first one. Notably, the expansion curve recorded during the second heating process served as the benchmark reference baseline. Throughout both tempering cycles, the specimen's length variations during the heating and cooling processes were meticulously recorded.

Mechanical properties tests

An 8 mm × 10 mm × 4 mm specimen was cut from the aforementioned heat-treated specimen. After meticulous grinding and polishing, it was put into a 4% nitric acid ethanol solution for etching, and then its hardness was measured using a Netzsch STA449C type Vickers hardness tester, applying a load of 4.9 N for a dwell time of 10 s. To ensure reliability, the average of the results obtained from three separate experiments was calculated and adopted as the final hardness value.

A 10 mm × 20 mm × 4 mm specimen was cut from the aforementioned heat-treated specimen. After grinding and polishing, the sliding friction test was carried out on a BMT-I multifunctional material surface performance comprehensive tester. The friction coefficient of the tested sample was automatically and accurately measured by the computer. The friction coefficient curve of each measurement was recorded and plotted by the sensors integrated in the equipment. Before and after conducting the test, it is crucial to measure the weight of the specimen using an electronic balance with a precision of 0.0001 g. It is essential to clean the specimen with alcohol and thoroughly dry it before each weighing. This will effectively remove any iron filings and stains that may be clinging to the abraded specimen, ensuring the precision of the weight loss resulting from abrasion.

Results

Carbide precipitation behaviour in tempering process

The tempering reaction mainly occurs in the heating process.²² In the non-isothermal double tempering expansion experiment, the central microstructure transformation during the tempering process occurs during the first heating process, and the axial change of the sample during the second heating process is only affected by the temperature change.²³ Therefore, the expansion curve of the second heating process is used as the reference baseline. The expansion curve due to phase transformation (ΔL1 - ΔL2) is obtained by subtracting the first tempering temperature rise curve (ΔL1) from the second tempering temperature rise curve (ΔL2), as shown in Figure 3(a). In order to observe more clearly the changes induced by the precipitation during the tempering process, the first derivative of the difference between the length changes with respect to temperature (d(ΔL1 - ΔL2)/dT) is shown in Figure 3(b). At tempering temperatures around 200 °C and above 300 °C, there is a clear trend in the expansion of the test steels. This indicates the occurrence of phase transformation in these temperature ranges before the change in expansion.

Carbide precipitation during tempering process

Figure 4(a) and (b) displays the initial martensite matrix and TEM microstructure after tempering at 100 °C. Specifically, the clear martensite lath exhibits a substantial distribution of dislocations. The lath shows severe dislocation entanglement, thereby illustrating the high density of dislocations and their consequential impact on dislocation strengthening. It is worth noting that the tempering temperature is relatively low in this context, rendering the thermal activation energy insufficient for any microstructure transformation or carbide precipitation.²⁴

As shown in Figure 4(c), the martensitic lath is still clearly visible. After tempering at 200 °C, the second phase of wave or flake begins to precipitate. The lattice fringe spacing is measured by high-resolution image Fourier transform of TEM. Compared with PDF card data and energy spectrum data, it can be determined that the precipitated phase is the ε-transition carbide. This ε-transition carbide has a close-packed hexagonal structure and is dispersed in the martensitic lath.²⁵ The carbon content in the ε-transition carbide accounts for 30.3%, and its width ranges from 8 to 20 nm. Furthermore, it should be noted that the arrangement direction of the ε-transition carbides in the same martensitic lath is consistent.

As shown in Figure 4(d), it can be observed that the ε-transition carbides underwent coarsening when the tempering temperature was increased to 300 °C. This resulted in a substantial increase in the width of carbides, ranging from 33 to 60 nm. In this case, the coarsening process followed the Oswald ripening mechanism. It is worth noting that as the smaller precipitated phases dissolved, the overall number of precipitated phases within the slats decreased.

As shown in Figure 4(e), the tempering temperature is increased to 350 °C, the martensite matrix recovers. This is accompanied by a series of processes, including inter-atomic diffusion enrichment, merger and reorganisation at the martensite slat boundary. These processes lead to the softening of the martensite slat beams and their gradual fuzziness.²⁶ Additionally, the ε-transition carbide present in the martensite gradually dissolves, while a larger-sized the θ-cementite precipitates in the matrix, with a width ranging from 80 to 100 nm. Furthermore, the carbon content of the tempered martensite decreases gradually, settling at a value of 24.9 At.-%.

As shown in Figure 5, the TEM images reveal the effects of different holding times after tempering at 200 °C on carbide precipitation. It can be observed that the martensite slats are clearly visible, and they do not soften with the increase in holding time. After a holding time of 45 min, a significant number of ε-transition carbides precipitate within the martensite slats, leading to dispersion (Figure 5(b)). However, with a further extension of the holding time up to 60 min, it is noticed that some of the ε-transition carbides undergo coarsening and dissolution (Figure 5(c)).

Residual stress

Figure 6(a) and (b) depicts the distribution of residual stress in the thickness direction for the wear-resistant steel in its original state and after various heat treatments. It can be seen that the residual stress of NM300TP shows a very uneven ‘W’ distribution along the thickness direction. After different heat treatments, this ‘W’ type distribution is improved, and residual stress distribution becomes more uniform.

In order to more clearly reflect the trend of the overall residual stress level after different heat treatments, this paper introduces a physical parameter: elastic strain energy (E_RS).²⁷ And its numerical meaning is the integral of the absolute value of residual stress along the thickness direction. This parameter describes the overall level of residual stress in the rolling direction of a point in the thickness direction, as shown in equation (1).

E_{RS} = \int_{0}^{x} | σ_{RS} | d x

(1)

Where E_RS is the unit elastic strain energy of residual stress with the unit of MPa mm, and

σ_{RS}

is the value of residual stress in MPa. Figure 7 shows the elastic energy of the original specimen and the specimens after different heat treatments. It can be observed that the initial residual level of the specimen is high, as indicated by the maximum value of elastic strain energy, which is 271.62 MPa mm. After tempering at 200 °C for 30 min, the specimen exhibits the first valley point of elastic strain energy, which is 108.3 MPa mm, indicating a decrease of residual stress by 60.13%. After tempering at 200 °C for 45 min, the elastic strain energy of the sample decreases to a minimum value of 82.82 MPa mm, showing the overall lowest level of residual stress with a decrease of 69.51%. After tempering at 200°C for 60 min, the elastic strain energy of the specimen increases to 123.37 MPa mm, indicating a modest deterioration of residual stress. For the specimen tempered at 350 °C for 30 min, a second valley point of elastic strain energy is observed, which is 124.33 MPa mm, showing a decrease in residual stress by 54.23%.

Mechanical property

Wear resistance is an essential index of a material's ability to withstand wear and tear, making it a comprehensive property. In this article, the wear resistance of NM300TP was evaluated by friction coefficient, wear amount and absolute wear resistance. The calculation method of wear amount is shown in equation (2):

X = W_{0} - W

(2)

X is the wear amount (mg);

W₀ is the mass of the sample before the wear test (mg);

W is the mass of the sample after the wear test (mg).

The wear resistance increases as the amount of wear and coefficient of friction decrease. Absolute wear resistance is typically described by the reciprocal of the amount of the weight loss, which is direct proportion to the wear resistance.

Table 2 displays the results of the friction and wear tests of specimens before and after tempering. It is evident from these results that the friction coefficient, wear amount, and absolute wear resistance of the experimental steel vary with an increase in tempering temperature. Specifically, when the tempering temperature is 200 °C with a holding time of 45 min, the experimental steel demonstrates the highest wear resistance. This is evidenced by the smallest friction coefficient (0.672) and the smallest wear amount (3.6). Conversely, when the tempering temperature is increased to 350 °C with a holding time of 30 min, the experimental steel exhibits the poorest wear resistance. This is indicated by the highest friction coefficient (0.937) and the maximum wear amount (5.7).

Table 2.

Experimental results of friction and wear of specimens before and after tempering.

Heat treatment	Weight loss (mg)	Wear resistance (1/mg)	Average friction coefficient
Original	3.7	0.270	0.765
100 °C – 30 min	3.9	0.256	0.714
200 °C – 30 min	4.2	0.238	0.748
200 °C – 45 min	3.6	0.278	0.672
200 °C – 60 min	3.8	0.263	0.958
300 °C – 30 min	4.9	0.204	0.989
350 °C – 30 min	5.7	0.175	0.937

Figure 8 illustrates the changes in hardness for the original NM300TP after different heat treatments. It can be observed that the initial hardness of the specimen is 322.19HV_0.5, and the hardness of the specimens after heat treatment is reduced. When the tempering temperature is 200 °C with a holding time of 30 min, the hardness is measured at 309.6HV_0.5. However, by extending the holding time to 45 min, the hardness increases to 317.2HV_0.5. Conversely, when the holding time is extended to 60 min, the hardness decreases to 301.3HV_0.5. Alternatively, when the tempering temperature is raised to 350 °C with a holding time of 30 min, the hardness is recorded as 299.58HV_0.5.

Discussion

Residual stress relief mechanism for wear-resistant steel

Transformation plasticity²⁸ is an irreversible strain observed when metallurgical transformation occurs under small external stress lower than the yield stress of the weaker phase. According to Greenwood–Johnson,²⁹ the transformation plasticity of diffusive phase transitions is caused by the difference in densities between the old and new phases. The microscopic stress generated by the phase transition produce small plastic strains in the weak phase. Transformation plasticity model obtained according to the Greenwood–Johnson mechanism is shown in Equation (3)²⁸

ε^{tp} = K (\frac{Δ V}{V}; σ_{1}^{y}) f_{2} (δ) f_{3} (σ)

(3)

Where:

ε^{tp}

is the transformation plastic strain, K is the coefficient of transformation plasticity,

Δ V / V

is the relative volume difference between the parent (

V

) and product phase (

Δ V

σ_{1}^{y}

is the yield strength of the weaker phase,

δ

is the amount of phase transformation,

σ

is the equivalent applied stress.

From Equation (3), the volume difference $Δ V$ between the parent phase and the product phase, and the stress that causes the strain to accumulate in a certain direction are the two necessary conditions for phase transformation plasticity $ε^{tp}$ to occur during the phase transformation process. During the tempering process, ε-transition carbides and θ-cementites precipitate out of martensite, and there is a volume difference between the new phase ε-transition carbides and θ-cementites and the old phase martensite,⁹ which satisfies the volume difference condition. At the same time, the initial residual stress of the material makes it possible to superimpose the phase transition strain in the stress direction. Therefore, according to the Greenwood–Johnson mechanism,²⁹ precipitation of carbides during tempering should also produce phase transformation plasticity. This is in agreement with the conclusions of the study by Morra et al.³⁰ In order to distinguish the phase transformation plasticity in the traditional sense, the phase transformation plasticity produced by carbide precipitation during tempering is referred to as precipitation plasticity $ε^{tp}$ in this paper.

Figure 9 illustrates the intricate relationship between microstructural transformation and residual stress during the tempering process. Notably, at 200 and 350 °C, the expansion curve exhibits significant volume contraction. This contraction is attributed to the precipitation of carbides from the matrix, which leads to the reduction of lattice distortion and subsequently results in volume shrinkage. At these specific temperatures, the residual stress in steel reaches two minima. Residual stress, simply put, is the uneven plastic strain that remains within the material. It's worth noting that the precipitation of carbides does not occur independently but rather under the influence of initial stress. During this process, the strain generated by the carbides accumulates in the direction of the initial stress. This volumetric change is intricately linked to the state of residual stress in steel.

Effect of carbide type on residual stress and mechanical properties

After tempering at 350 °C and holding for 30 min, the specimen exhibited the precipitation of spherical θ-cementites, which led to the shrinkage phenomenon in the expansion curve (Figure 3(b)). During this process, from Figure 10(a), the elastic strain energy of the specimen decreased significantly, from the original 271.62 to 124.33 MPa mm, indicating a 54.23% reduction in the overall level of residual stress compared to before tempering. Simultaneously, the hardness and absolute wear resistance of the specimen also decreased, primarily due to the consumption of carbon in the martensite matrix by the precipitation of θ-cementites, thereby weakening the solid solution strengthening effect of martensite. It is worth noting that the high solid solution strengthening of the carbon atoms in the matrix is the main contributor to the high hardness and wear resistance in wear-resistant steels.¹⁷ Notably, the softening and recovery of martensite laths were observed, and the second-phase precipitation strengthening effect of θ-cementites was unable to fully compensate for the decline in performance caused by the solid solution strengthening and softening recovery of martensite.

In contrast, specimens tempered at 200 °C and held for 30 min exhibited the precipitation of fine, flaky ε-transition carbides, also accompanied by the shrinkage of the expansion curve (Figure 3(b)). At this point, from Figure 10(a), the elastic strain energy of the specimen decreased to 108.30 MPa mm, representing a 60.13% reduction in residual stress compared to before tempering. From Figure 10(b), when the holding time was extended to 45 min, the elastic strain energy further decreased to 82.82 MPa mm, indicating a 69.51% reduction in residual stress compared to the original state. This suggests that, compared to θ-cementite, the precipitation of ε-transition carbides is more effective in regulating residual stress, improving the effect by 15.28%.

In terms of mechanical properties, from Figure 10(b), specimens tempered at 200 °C and held for 45 min exhibited higher hardness and wear resistance. The hardness reached 317.2HV_0.5, and the absolute wear resistance was 0.278, showing significant improvements compared to the precipitation of θ-cementite. This can be attributed to the fact that although the precipitation of ε-transition carbides consumes the carbon concentration in the matrix, their second-phase precipitation strengthening effect effectively compensates for the loss of solid solution strengthening, maintaining a balanced carbon concentration within the matrix. Additionally, the martensite laths did not undergo softening and recovery during low-temperature tempering, which also contributed to maintaining the hardness and wear resistance of the specimen.

Therefore, after the tempering treatment at 200 °C, in order to optimise the mechanical properties of steel and reduce its residual stress, we can adopt an appropriate extension of the holding time to promote the precipitation of carbides. Extending the holding time has two positive effects: on the one hand, it can increase the effective carbon concentration in the matrix, thereby inducing the diffuse precipitation of a large number of ε-transition carbides. The precipitation plasticity of these carbides helps to relax residual stress. On the other hand, during the tempering process, as the solid solution content of carbon atoms in martensite gradually decreases, the hardness and wear resistance of steel will indeed decrease. However, it is noteworthy that under low-temperature tempering conditions, the small-sized ε-transition carbides precipitated diffusely have a much smaller impact on the supersaturated solid solution carbon in the matrix compared to the continuously growing θ-cementites formed at high temperatures. At the same time, the strengthening effect of the precipitation of fine carbides can effectively compensate for the loss of solid solution strengthening, thereby significantly improving the mechanical properties of the steel.^31,32 In addition, it is worth mentioning that martensite does not undergo softening and recovery during low-temperature tempering, which further ensures the stability and durability of steel.

Mechanism of precipitation of induced ε-transition carbides

Figure 11(a) and (b) clearly demonstrates the changes in dislocation density in the wear-resistant steel during the tempering process. The dislocation density was obtained by the WH^33,34 method. This method is based on the diffraction peak broadening model caused by grain size and microscopic distortion to calculate the dislocation density in mental materials. The specific equation is as follows:

β_{s} = 2 ε \tan θ

(4)

Where:

β_{s}

is the broadening due to dislocations,

θ

is the position of diffraction peak.

β_{d} = \frac{α λ}{D \cos θ}

(5)

Where:

β_{d}

is the broadening caused by grain refinement, D is the grain width, and

λ

is he wavelength of X-ray.

The W–H method assumes that the broadening caused by physical effects, grain refinement and micro-distortion conforms to the Cauchy function, which includes:

β = β_{s} + β_{d}

(6)

Substituting Eq. (1) and Eq. (2) into the above equation, there is:

β \frac{\cos θ}{λ} = \frac{α}{D} + 2 ε \frac{\sin θ}{λ}

(7)

Where, when

β

represents the integral width,

α

is 1; when

β

represents the half width and height,

α

is 0.9. In this paper, the half width and height are calculated,

α

is 0.9, and the half width and height used to calculate the dislocation density are directly obtained by Jade software. In this XRD diffraction experiment, the Cu target was used, and its wavelength was 1.45056 Å, which was converted into 0.1 nm. The dislocation density calculation equation is as follows:

$ ρ = 14.4 \frac{ε^{2}}{b^{2}}

(8)

Where:

ρ

is dislocation density, unit m⁻², b represents the Burgers vector of dislocation, 0.248 nm.

By plotting the $β \cos θ / λ$ and $2 \sin θ / λ$ of different diffraction peaks, the slope obtained by linear fitting is the average effective macrostrain $ε$ . In order to improve the calculation accuracy, only the diffraction surface in the same direction is selected for calculation.

At a tempering temperature of 200 °C, the extension of holding time significantly impacts the precipitation and coarsening of ε-transition carbides. Specifically, when the holding time is extended from 30 to 45 min, a large amount of ε-transition carbides starts to precipitate diffusely. This phenomenon is closely related to the diffusion behaviour of carbon atoms. Within the 45-min holding time, carbon atoms have sufficient time to diffuse and migrate, resulting in a significant increase in the concentration of movable carbon atoms in the structure. Simultaneously, the dislocation density in the matrix also increases (as shown in Figure 11(b)), providing more nucleation sites for effective carbon and enabling more locations in the matrix to reach the saturation carbon concentration for ε-transition carbide precipitation. And a large amount of ε-transition carbide precipitation decreases the subsequent nucleation rate of θ-cementites.³⁵

However, it is noteworthy that an increase in precipitation does not always correlate positively with the holding time. In fact, when the holding time is further extended to 60 min, ε-transition carbides undergo coarsening and gradual dissolution, leading to a significant decrease in their number (as shown in Figure 5(c)). This change indicates that extended holding time does not always facilitate the precipitation of ε-transition carbides. As ε-transition carbides dissolve, their pinning effect on dislocations weakens, resulting in a decrease in dislocation density (as shown in Figure 11(b)).

In summary, during the low-temperature tempering process at 200 °C, there exists an optimal holding time window of 45 min, during which ε-transition carbides can precipitate diffusely in large quantities. Although extending the holding time can promote the diffusion of carbon atoms and increase the dislocation density, excessively long holding times lead to coarsening and dissolution of ε-transition carbides, thereby reducing their precipitation amount and pinning effect on dislocations.

Conclusion

Based on the dislocation density and expansion curves, it can be determined that there are two critical transformation stages, the precipitation stages, the precipitation of ε-transition carbides and θ-cementites, during the tempering process of NM300TP steel. The precipitation plasticity produced in these two stages can effectively decrease the residual stress.

Extending the holding time by low-temperature interferes with the precipitation behaviour of carbides and promotes the concentration of effective carbon in the matrix. This process induces the massive and diffuse precipitation of fine flaky ε-transition carbides at low temperatures, while suppressing the precipitation of coarse spherical θ-cementites at high temperatures.

The substitution of ε-transition carbides for the precipitation of θ-cementites allows for high performance at low residual stress in wear-resistant steels. Specifically, when θ-cementites precipitate, the residual stress of the wear-resistant steel is reduced by 54.23%, and the hardness reaches 299.58HV_0.5. However, when ε-transition carbides precipitate through diffuse precipitation, they reduce the residual stress by 69.51%, and the hardness is 317.2HV_0.5. Moreover, this precipitation of ε-transition carbides also leads to increased wear resistance compared to precipitation of θ-cementites.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Wenhong Ding

References

Zhang

, et al. Detecting milling deformation in 7075 aluminum alloy aeron nautical monolithic components using the quasi-symmetric machining method. Metals 2016; 6: 80.

Heinz

Haszler

Keidel

, et al. Recent development in aluminum alloys for aerospace applications. Mater Sci Eng A 2000; 280: 102–107.

Lerchbacher

Zinner

Leitner

. Direct or indirect: influence of type of retained austenite decomposition during tempering on the toughness of a hot-work tool steel. Mater Sci Eng A 2013; 564: 163–168.

Wang

Zhang

Yuan

, et al. Microstructure and properties evolution of HB450 ultra-high strength low-alloy abrasion resistant steel during tempering. Iron Steel 2011; 46: 60–63.

Ritter

. A modified thickness criterion for fracture toughness testing. Eng Fract Mech 1977; 9: 529–540.

Morra

Radelaar

Yandouzi

, et al. Precipitate coarsening-induced plasticity: low temperature creep behavior of tempered SAE 52100. Int J Plast 2009; 25: 2331–2348.

Yamazaki

Okamura

Suzuki

. High accurate prediction of residual stress considering plasticity during tempering process. J Soc Mater Sci Jpn 2015; 64: 266–273.

Yamazaki

Okamura

Shibutani

. Evaluation of influencing factors on plasticity during tempering process. J Soc Mater Sci Jpn 2020; 69: 555.

Kaiser

de Graaff

Jung

, et al. A dilatometric study on the influence of compressive stresses on the tempering pf martensitic AISI 4140 steel-evidence of transformation induced plasticity during cementite precipitation. Mater Sci Eng A 2017; 705: 114–121.

10.

Kaiser

de Graaff

Dietrich

, et al. Investigation of the precipitation kinetics and microstructure evolution of martensitic AISI 4140 steel during tempering with high heating rates. Rev Metallur-Cahiers D Inform Tech 2018; 115: 404.

11.

Kaiser

Damon

Mühl

, et al. Experimental investigation and finite-element modeling of the short-time induction quench-and-temper process of AISI 4140. J Mater Process Technol 2020; 279: 116485.

12.

Ding

. Study on the evolution of residual stress during continuous cooling annealing process of high strength steel. Beijing: University of Science and Technology Beijing; 2020.

13.

Ding

, et al. Effect of carbide precipitation on “precipitation plasticity” and residual stress during tempering. Steel Res Int 2022; 93: 1–9.

14.

Speich

Leslie

. Tempering of steel. Metall Trans 1972; 3: 1043–1054.

15.

Olson

Cohen

. Early stages of aging and tempering of ferrous martensites. Metall Trans 1983; 14: 1057–1066.

16.

Taylor

Olson

Cohen

, et al. Carbide precipitation during stage i tempering of Fe–Ni–C martensites. Metall Trans 1989; 20: 2749–2765.

17.

Cao

Wang

, et al. Effects of tempering temperature and Mo/Ni on microstructures and properties of lath martensitic wear-resistant steels. J Iron Steel Res Int 2013; 20: 70–75.

18.

Liang

Yan

, et al. Effect of heat treatment on microstructure and mechanical properties of low-alloy wear-resistant steel NM 450. Mater Res Expr 2021; 8: 1–6.

19.

Han

Xiao

, et al. Influence of tempering time on the microstructure and mechanical properties of AISI M42 high-speed steel. Metallurg Mater Trans 2018; 49: 5976–5986.

20.

Prime

Hill

. Residual stress measurement by successive extension of a slot: the crack compliance method. Appl Mech Rev 1999; 52: 75–96.

21.

Prime

Hill

. Residual stress, stress relief, and inhomogeneity in aluminum plate. Scr Mater 2002; 46: 77–82.

22.

Ritter

Mopherson

. An isothermal stress relaxation in a carbon-manganese steel. Iron Steel Inst 1970; 10: 935–941.

23.

Priming

Leitner

. Separation of overlapping retained austenite decomposition and cementite precipitation reactions during tempering of martensitic steel by means of thermal analysis. Thermochem Acta 2011; 526: 111–117.

24.

Naghizadeh

Mirzadeh

. Elucidating the effect of alloying elements on the behavior of austenitic stainless steels at elevated temperatures. Metall Mater Trans A 2016; 47: 5698–5703.

25.

Eguchi

Kato

. On the precipitation of ε-carbides during martensite tempering of C-steel. J Electron Microsc 1961; 10: 54–58.

26.

Liu

Tang

, et al. Microstructural evolution and yield strength prediction of quench-tempered 42CrMo steel. Trans Mater Heat Treatm 2020; 41: 144–150.

27.

Ding

Liu

Xie

, et al. Effect of carbide precipitation on the evolution of residual stress during tempering. Metals-Open Access Metallurgy J 2019; 9: 709–716.

28.

Taleb

Cavallo

Waeckel

. Experimental analysis of transformation plasticity. Int J Plast 2001; 17: 1–20.

29.

Greenwood

Johnson

. The deformation of metals under small stresses during phase transformations. Proc R Soc Lond 1965; 283: 403–422.

30.

Morra

Radelaar

Yandouzi

, et al. Precipitate coarsening-induced plasticity: low temperature creep behavior of tempered SAE 52100. Int J Plast 2009; 25: 2331–2348.

31.

Asadabad

Kheirandish

Novinrooz

. Microstructural and mechanical behavior of 4.5Cr–2W–0.25V–0.1C steel. Mater Sci Eng A 2010; 527: 1612–1616.

32.

Yao

Zhang

, et al. Understanding the mechanical properties of vanadium carbides: nano-indentation measurement and first-principles calculations. J Alloys Compd 2013; 548: 60–64.

33.

Williamson

Hall

. X-ray line broadening from filed aluminum and wolfram. Acta Metallurg 1953; 1: 22–31.

34.

Williamson

Smallman

. Dislocation densities in some annealed and cold-worked metals from measurements on the X-ray Debye Scherrer spectrum. J Theor Exp Appl Phys 1956; 1: 34–46.

35.

Cao

Wang

Z.D.

Deng

X.T.

, et al. Carbide precipitation during tempering of low alloy wear-resistant steel. Trans Mater Heat Treatm 2018, 39: 8–12.