Effect of Initial Intergranular Ferrite Size on Induction Hardening Microstructure of Microalloyed Steel 38MnVS6

Abstract

In this study, we investigated the effect of grain size of an initial microstructure (pearlite + ferrite) on a resulting microstructure of induction-hardened microalloyed steel 38MnVS6, which is one topical medium carbon vanadium microalloyed non-quenched and tempered steel used in manufacturing crankshafts for high-power engines. The results show that a coarse initial microstructure could contribute to the incomplete transformation of pearlite + ferrite into austenite in reaustenitization transformation by rapid heating, and the undissolved ferrite remains and locates between the neighboring prior austenite grains after the induction-hardening process. As the coarseness level of the initial microstructure increases from 102 μm to 156 μm, the morphology of undissolved ferrite varies as granule, film, semi-network, and network, in sequence. The undissolved ferrite structures have a thickness of 250–500 nm and appear dark under an optical metallographic view field. To achieve better engineering applications, it is not recommended to eliminate the undissolved ferrite by increasing much heating time for samples with coarser initial microstructures. It is better to achieve a fine original microstructure before the induction-hardening process. For example, microalloying addition of vanadium and titanium plays a role of metallurgical grain refinement via intragranular ferrite nucleation on more sites, and the heating temperature and time of the forging process should be strictly controlled to ensure the existence of fine prior austenite grains before subsequent isothermal phase transformation to pearlite + ferrite.

1. Introduction

Microalloying technology, developed in the 1960s, provided a theoretical basis for the non-quenched and tempered steel. The emergence of microalloyed non quenched and tempered steel was directly accelerated by the oil crisis. In 1972, THYSSEN in Germany developed novel non-quenched and tempered steels, and ferrite-pearlite steel 49MnVS3 was classically provided to automotive crankshaft manufacturers to replace quenched and tempered CK45 steel. Its advantages include a remarkably improved yield strength, higher production efficiency, higher fatigue performance, better machinability, and helping cut the cost of forged steel components [1]. This group of steels has been used to produce crankshafts, connecting rods, steering arms, and different axles in the auto industry [2]. For example, Mercedes Benz in Germany manufactured crankshafts using non-quenched and tempered steels instead of 40CrMn quenched and tempered steel, and Volvo in Sweden employed over 30,000 t of the steels annually in the early 1990s, with the goal of producing all forgings using non quenched and tempered steels, except for carburized parts. Subsequently, microalloyed non-quenched and tempered steels achieved rapid development globally and became the preferred material for automotive crankshafts all over the world. For example, it dominates 90% of crankshafts in the current market of Japan’s automotive industry and more than 70% of forgings in the German automotive industry. Microalloyed steels can be strength-enhanced by dealing with aspects of precipitation which controls grain size and dispersion strengthening in ferrite–pearlite steels [3]. It mainly depends on the solid solution strengthening effect of alloying elements and the precipitation strengthening of resulting carbides, nitrides, and carbonitrides, since the solid solution atoms cause lattice distortion, and the precipitation of carbonitrides hinders dislocation movement [4]. Microalloying with V element is an efficient method of dispersion strengthening [5] for medium carbon ferrite–pearlite steels, in addition to grain refinement by ferrite nucleation on VN particles [6], thus providing improved mechanical properties [7]. The degree of precipitation strengthening of ferrite at a given vanadium content depends on the available quantities of carbon and nitrogen [8]. Reheating temperature and direct-cooling rate after forging and compositions of microalloyed steels highly influence the microstructure and therefore play an important role on final microstructure and resulting mechanical properties [9].

As an excellent representative of Mn-V series microalloyed steels, steel 38MnVS6 has become a research hotspot and has achieved remarkable results in recent years [10,11,12,13,14,15,16]. Manganese sulfides usually form because of more S addition, which can improve the machinability of this steel. MnS is detrimental to anti-pitting behavior [17,18] and it could be improved by tellurium treatment to increase pitting resistance [19]. On the other hand, MnS behaves as internal stress concentrations and lowers the fatigue strength and life due to an effect characterized by a function of shape factor γ (length-to-width ratio) of the MnS [12]. The application of induction hardening treatment plays a vital role for enhancing fatigue life because of compressive residual stresses incorporated in the component to a significant extent [13]. The microstructure and resultant mechanical properties of the hardened case produced in steels have a dependence on heating and cooling rate, maximum temperature of the treatment, composition, and initial microstructure [20]. Effects of the initial microstructure on the induction hardening behavior of microalloyed steels have been reported. However, the samples of microalloyed steels in the previous research are relatively designed with remarkable differences in initial microstructures involving bainite–martensite [21], or ferrite–martensite/austenite [21], isothermal upper and lower bainite [22]. There is no literature reporting the size effect of the same initial microstructures on induction hardening of microalloyed steels. In this research, samples are prepared with same initial microstructure of pearlite + ferrite, and the difference between them is the network dimension of intergranular ferrite. The effect of intergranular ferrite size on the resulting microstructure of induction-hardened microalloyed steels is investigated. This investigation could be provided with important reference for engineering practice of induction hardening in high-power gasoline engine crankshafts.

2. Materials and Methods

2.1. Raw Material

The experimental material in this study is a medium carbon microalloyed steel 38MnVS6 produced via continuous casting. The chemical composition of steel 38MnVS6 meets the requirements of standard DIN EN 10267-1998 [23], as shown in

Table 1. Chemical composition of experimental steel 38MnVS6 (wt.%).

Element	C	Si	Mn	P	S	Cr	Mo	V	N
Standard	0.34–0.41	0.15–0.80	1.20–1.60	≤0.025	0.02–0.06	≤0.30	≤0.08	0.08–0.20	0.01–0.02
Measured	0.38	0.58	1.39	0.016	0.056	0.14	0.02	0.11	0.015

. The crankshafts prepared for the study were treated to a hardness range of 250–300 HBRW by controlled cooling after hot forging. The obtained structure was pearlite + ferrite, without bainite. Three batches of crankshafts were produced with different initial intergranular ferrite sizes, named Batch A, Batch B, and Batch C. The differences between these batches were produced by the process parameters in hot forging, including controlled heating temperature, holding time, forging ratio, and finishing temperature. The journals were machined with a finished diameter of 55 mm in the study, ready for the induction hardening process.

2.2. Induction Hardening

In total, three pcs of crankshafts from each batch were selected randomly for induction hardening. Half journals of each crankshaft were subsequently quenched by quenching coolant. The other half were kept for raw microstructure inspection after heat treatment. Intermediate frequency induction hardening was carried out by a BAZ-2 typed machine. The average power was 58 kW, and the frequency was 11 kHz. Induction heating time was set at 10 s, with 88% intermediate frequency voltage. The temperature of the quenching coolant was 22 °C in an ambient environment and the coolant cooling time was 4 s.

2.3. Microstructural Characterization

Metallurgical specimens were cut from the crankshaft journals along the radial direction. Hot mounting was used with powder resin material. The samples were ground and polished, following the details in

Table 2. Description of grinding and polishing procedure.

Step	Procedure	Description of Cloth Type/Size	Liquid	Time/min
1	Rough grinding	SiC paper 220 grit	Water	2
2	Fine grinding	SiC paper 800 grit	Water	2
3	Cleaning	Ultrasonic bath	Ethanol	2
4	Rough polishing	DAC cloth	3 μm diamond suspension	4
5	Fine polishing	NAP cloth	1 μm diamond suspension	4
6	Cleaning	Ultrasonic bath	Ethanol	2

, and then etched by 4% nital solution for several seconds until the surfaces appeared gray in color. The prepared samples were analyzed using an optical microscope. A scanning electron microscope (SEM) was employed to observe the morphological features of quenched microstructures with high resolution under 6000 magnifications. Electron backscatter diffraction (EBSD) was used to further identify the dark phase.

3. Results

3.1. Initial Microstructure

Figure 1, Figure 2 and Figure 3 reveal the initial microstructure prepared in this study. The initial microstructure mainly consists of intergranular ferrite and intragranular pearlite colonies. The order of intergranular ferrite from coarse to fine is Batch A, Batch B, and Batch C, respectively. The smaller the intergranular ferrite size, the more intragranular ferrite granules were present. To quantitatively characterize intergranular ferrite size, a method to collect and calculate mean lineal intercept length l on the measuring lines has been introduced, following standard ISO 643:2024 Steels—Micrographic determination of the apparent grain size [24]. Quantitative statistical results are represented in

Table 3. Mean lineal intercept length of initial intergranular ferrite.

Sample	Measured la/μm	Measured lb/μm	Measured lc/μm	Mean l/μm
Batch A	153	155	160	156
Batch B	104	97	106	102
Batch C	61	67	59	62

. The mean lineal intercept length of initial intergranular ferrite of Batch A is 156 μm, Batch B 102 μm, and Batch C 62 μm. Box diagrams in Figure 4 expound the statistical distribution of mean lineal intercept length of the initial intergranular ferrite.

3.2. Induction Hardened Microstructure

During the heating procedure of the induction-hardening process, the most ideal situation is that that reaustenitization from the initial microstructure of ferrite and pearlite is accomplished in three stages, including nucleation, growth, and homogenization of austenite within a very short period. The crankshaft samples are reheated at fast heating rates, and the austenite formation mechanism and kinetics are controlled by the diffusion of mainly carbon atoms. Martensite would form from austenite at a rapid cooling rate during the following quenching process. Any untransformed ferrite, or non-hardened phase such as sorbite, is not desirable to be attained. As stated above, the original microstructure has an influence on the induction hardening behavior of microalloyed steels. In this study, different induction quenching microstructures were obtained based on the original microstructures of different sizes of initial intergranular ferrite. Figure 5 shows the resulting microstructure of Batch A with initial intergranular ferrite size of 156 μm. It can be seen that dark phases in the form of semi-network and network interestingly appear as prior austenite grain boundaries in the conventional micrographic determination of the apparent grain size. As the size of the initial intergranular ferrite reduces, dark semi-network and network would disappear in the resulting microstructure. However, the presence of dark phase in the form of granules and film is easily to be affirmed as shown in Figure 6, for Batch B with an initial intergranular ferrite size of 102 μm. When initial intergranular ferrite size decreases to 62 μm, no dark phase in any form appears in the resulting microstructure of mainly martensite (M’), with a few retained austenites (RA) in Figure 7. This is the normal induction hardened microstructure, providing higher hardness and wear property to crankshaft journals.

3.3. Identification of Dark Phase

To identify the metallurgical properties of the dark phase in Batch A and Batch B, secondary electron (SE) images obtained in the SEM experiment reveal the details of morphological characteristics of dark phases in the three-dimensional direction. It is more easily etched by 4% nital solution, and appears in a gully shape, as arrowed in Figure 8a,b, which is representative of Batch A and Batch B, respectively. As a comparison, Figure 8c shows a normal induction-hardened microstructure in Batch C. The prior austenite grain boundaries are clear and distinguishable, completely different from the dark phase structure. The speculation that the dark phase might be austenite grain boundaries can be eliminated. Based on metallographic experience, the dark phase should be recognized as ferrite. Note that the dark phases have a thickness range of 250–500 nm. It is too thin for an optical microscope to recognize clearly due to its resolution limitation at this scale. This is the fundamental reason why it looks dark in optical metallographic observation, while similar structures with larger sizes are well known as white ferrite by metallurgists [25,26,27,28,29]. The EBSD results provide more scientific experimental evidence that the dark phase is a BCC crystalline structure (see Figure 9). Image quality (IQ) maps indicate the ferrite (high IQ areas) and the martensite (low IQ regions). Additionally, the percentage of retained austenite phase is only 1.7%. Combining the experimental results of both SEM and EBSD, the dark phase could be accredited as ferrite.

4. Discussion

4.1. Effect of Initial Intergranular Ferrite Size

The resulting microstructures of induction hardening in different batches of crankshafts with different initial intergranular ferrite size are summarized in

Table 4. Summary of induction hardened microstructures.

Sample	Measured a	Measured b	Measured c
Batch A	M’ + RA + SNF/NF	M’ + RA + SNF/NF	M’ + RA + SNF/NF
Batch B	M’ + RA + GF/FF	M’ + RA + GF/FF	M’ + RA + GF/FF
Batch C	M’ + RA	M’ + RA	M’ + RA

. It could be concluded that with the increasing size of initial intergranular ferrite, the presence of undissolved ferrite is a solid possibility. As coarseness level increases, the phenomenon of incomplete austenite transformation from initial microstructure gradually appears, and the form of undissolved ferrite varies as granule (GF), film (FF), semi-network (SNF), and network (NF), in sequence. With respect to the overall transformation in a previous study [30], a finer microstructure results in a faster process of dissolution from the initial microstructure, as compared with coarser microstructures.

There are two different transformations in the austenitization of microstructures composed of ferrite and pearlite, involving pearlite dissolution and ferrite-to-austenite transformation, which take place by nucleation and growth processes [31]. Induction hardening is a heat treatment of non-isothermal reactions. The austenite volume fraction ($V_{\gamma }^P$), obtained from pearlite dissolution during continuous heating of a ferrite plus pearlite initial microstructure, could be expressed as follows [31]:

$$V_{\gamma }^P=V_{P_0}\left\{1-exp\left(-{\int }_{A_{c_1}}^T{\displaystyle \frac{4\pi }{3{\left(Ṫ\right)}^4}}ṄG^3\mathsf{\Delta }{T}^{3}dT\right)\right\}$$

(1)

where $V_{P_0}$ is the volume fraction of pearlite present in the initial microstructure, T is heating temperature, ΔT is overheating (ΔT = T − Ac₁), Ṅ and G are nucleation and growth rates of austenite, Ṫ is a constant rate for the heating condition, and Ac₁ is the eutectoid temperature using Andrews’ formula [32].

Pearlite-to-austenite transformation could be completed within one second [33]. Once austenite transformation from the pearlite has been totally accomplished, the α/γ interface keeps advancing into the ferrite grains until all the initial microstructure has been reaustenitized [34]. It is newly shown that the critical temperatures for austenite formation migrate to higher levels when at a rapid heating rate [35]. The increase has no influence on the starting temperature of the austenite formation (Ac₁), but it changes its final critical temperature (Ac₃), which could be increased from 920 °C to 1035 °C [36]. This indicates that it requires a higher temperature for ferrite-to-austenite transformation. The kinetics of the ferrite-to-austenite transformation at higher temperatures (T ≥ 870 °C) become different, and the volume fraction of austenite formed from ferrite after complete pearlite-to-austenite transformation has primarily a time (t) dependence [30]. The volume fraction of austenite formed from ferrite during continuous heating at a given temperature could be expressed as follows [31]:

$$V_{\gamma }^{\alpha }=V_{{\alpha }_0}-{\displaystyle \frac{\left(V_{{\alpha }_0}{-V}_D^{\alpha }\right){\left(Ṫ\right)}^2}{{\left(Ṫ\right)}^2+1.2\times {10}^{-3}\left(V_{{\alpha }_0}{-V}_D^{\alpha }\right)\left[{\left(T-T_C\right)}^2-{\left(T_D-T_C\right)}^2\right]}}$$

(2)

where T_C is the starting temperature of ferrite-to-austenite transformation, T_D is the temperature at which the kinetics of ferrite-to-austenite transformation change under non-isothermal conditions. T_C and T_D temperatures could be determined experimentally by means of dilatometric analysis. $V_D^{\alpha }$ is the austenite volume fraction formed from ferrite at T_D temperature, and $V_{{\alpha }_0}$ is the volume fractions of ferrite present in the initial microstructure.

It Is the time dependence in the heating period of the induction-hardening process that causes the difference in the resulting microstructures in crankshafts with different initial intergranular ferrite size. The increase in temperature is a function of heating time, and the diffusion of carbon atoms in austenite transformation is also a function of heating time. For the coarser initial intergranular ferrite, it needs more time for carbon to diffuse into ferrite with a longer distance during the final stage of ferrite-to-austenite transformation, which is diffusion controlled. From the experimental result, it could be inferred that the designed heating parameters could only achieve the completed reaustenitizating transformation for samples of Batch C. However, it is slightly inadequate for Batch A and Batch B, hence a very small amount of ferrite remains undissolved right between neighboring prior austenite grains which are in growth before the quenching process and exists in final quenching microstructure.

4.2. Actions Preventing Dark Ferrite

For the given non-quenched and tempered microalloyed steel with coarser grains, more heat is required to achieve an austenitizing transformation from ferrite and pearlite. To avoid undissolved dark ferrite in the induction-hardening process, heating time has been designed with additional two seconds for rest crankshafts from Batch A and Batch B, and other process parameters remain unchanged. Figure 10 shows the induction hardening microstructure after process improvement. The tempered martensite morphology can be clearly seen with a very small amount of residual austenite, and the previous undissolved dark ferrite has disappeared. This experimental result could serve as evidence to support the previous assertion about insufficient austenitization that caused the existence of undissolved dark ferrite.

However, these original structures coarser than grain size 5 are usually considered to be detrimental to fatigue performance. Hence, the above induction-hardening process optimization is not the best solution. Microstructural optimization shall be carried out before the induction hardening process. According to the results of this experiment, controlling grain size of the intergranular ferrite can suppress the undissolved dark ferrite in induction hardening process. Grain size of intergranular ferrite can be controlled dominantly through the following two methods: metallurgical microalloying via alloy composition design, and, subsequently, deformation degree and deformation temperature controlling in the forging process. The influence of microalloying via the addition of elements V [37,38], Nb [39,40], Ti [41,42], and Te [43] could prevent austenite grain coarsening development; therefore, intergranular ferrite could be refined by the precipitation of carbonitrides [44]. The forging deformation and deformation temperature play an important role in the microstructure of non-quenched and tempered steel. The finer grains can be obtained by increasing the deformation and reducing the deformation temperature [45]. Recent advances indicate that microstructural modification via a novel thermomechanical controlled processing [46] could lead to ferrite grain refinement by dynamic recrystallization and deformation-induced ferrite transformation. An appropriate isothermal holding period could be conducive to the nucleation and growth of fine carbides with FCC crystal structure which lead to ferrite refinement [47]; nevertheless, to decrease or to increase the isothermal holding time too much, precipitates coarseness and decreased grain refinement would take place [48].

5. Conclusions

In this study, the effect of initial intergranular ferrite size on resulting microstructure of induction hardened microalloyed steel 38MnVS6, which is a topical material for gasoline engine crankshaft, was revealed. The results show that the designed heating parameters could only achieve the completed reaustenitization transformation for samples with a fine initial microstructure. As the coarseness level of the initial microstructure increases from 102 μm to 156 μm, the phenomenon of incomplete austenite transformation from pearlite + ferrite initial microstructure gradually takes place by very rapid heating, and a very small amount of ferrite remains undissolved in the final quenching microstructure in the form of granules, films, semi-networks, and networks, in sequence, and exists right between neighboring prior austenite grains which are in growth before the quenching process. The undissolved ferrite structures have a thickness of 250–500 nm and appear dark under optical metallographic view field due to its resolution limitation at such a small scale. In addition, it has been experimentally confirmed that the undissolved ferrite can be eliminated by increasing the heating time for samples with coarse initial microstructures. However, the best optimization measure should be to achieve a fine original microstructure before the induction-hardening process, such as microalloying addition of vanadium and titanium to metallurgically provide more sites for intragranular ferrite nucleation, and strictly controlled heating temperature and time of the forging process before subsequent isothermal phase transformation to pearlite + ferrite. For more accurate characterization of the dark network ferrite discovered under an optical microscope in this study, more advanced and high-resolution characterization techniques and devices should be applied, such as transmission electron microscopy or atomic probe technology, to provide more direct experimental evidence to reveal its phase transition kinetics. On the other hand, the influence of the network ferrite in the induction quenched microalloyed steel on mechanical properties including hardness, toughness, and fatigue performance should be further investigated.