Journal of Materials Science & Technology  2020 , 38 (0): 170-182 https://doi.org/10.1016/j.jmst.2019.07.049

Research Article

Solidification microstructure of Cr4Mo4V steel forged in the semi-solid state

Weifeng Liuabc, Yanfei Caoc, Yifeng Guoc, Mingyue Sunac*, Bin Xuac, Dianzhong Lic

aKey Laboratory of Nuclear Materials and Safety Assessment, Institute of Metal Research, Chinese Academy of Sciences, Shenyang 110016, China
bCollege of Materials Science and Engineering, University of Science and Technology of China, Hefei 230026, China
cShenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese Academy of Sciences, Shenyang 110016, China

Corresponding authors:   ∗Corresponding author at: Key Laboratory of Nuclear Materials and Safety Assessment, Institute of Metal Research, Chinese Academy of Sciences, Shenyang 110016, China.E-mail address: mysun@imr.ac.cn (M. Sun).

Received: 2019-05-15

Revised:  2019-07-12

Accepted:  2019-07-28

Online:  2020-02-01

Copyright:  2020 Editorial board of Journal of Materials Science & Technology Copyright reserved, Editorial board of Journal of Materials Science & Technology

More

Abstract

Semi-solid forging of iron-based alloys during solidification has unique characteristics distinct from those of the classical hot forging. With the aim of acquiring precise knowledge concerning the microstructural evolution of bearing steel Cr4Mo4V in this process, a series of semi-solid forging experiments were carried out in which samples were wrapped in a designed pure iron sheath. The effects of forging temperature and forging reduction on the grain morphology and liquid flow behavior were investigated, respectively. By forging solidifying metal (FSM), bulky primary dendrites were broken and spheroidal grains with an average shape factor of 0.87 were obtained at 1360 °C. With the decreasing forging temperature to 1340 °C, the microstructural homogeneity can be improved. On the other hand, it shows that a higher forging reduction (50%) is essential for the spheroidization of grains and elimination of liquid segregation. Those microstructural characteristics are related to different motion mechanisms of solid and liquid phases at different forging temperatures. Additionally, the effect of semi-solid forging on the eutectic carbides was also investigated, and the results demonstrate that the higher diffusion capacity and less liquid segregation jointly lower the large eutectic carbides and consequently cause its uniform distribution during FSM.

Keywords: Solidification ; Semi-solid forging ; Microstructure ; Eutectic carbides

0

PDF (8905KB) Metadata Metrics Related articles

Cite this article Export EndNote Ris Bibtex

Weifeng Liu, Yanfei Cao, Yifeng Guo, Mingyue Sun, Bin Xu, Dianzhong Li. Solidification microstructure of Cr4Mo4V steel forged in the semi-solid state[J]. Journal of Materials Science & Technology, 2020, 38(0): 170-182 https://doi.org/10.1016/j.jmst.2019.07.049

1. Introduction

Iron-based heavy forgings with better compactness and homogeneity in composition and microstructure, are widely used in metallurgy, petroleum, and aviation industries. The conventional manufacturing forging process is an energy- and time-consuming chain consisting of steelmaking, casting ingots, demold (after complete solidification), reheating, and forging. On the other hand, the dendritic grains during solidification cause the inevitable metallurgical defects such as coarse grain, porosity and segregation in the forgings and final products. In order to eliminate these defects, a large number of methods have been proposed in practice, such as multiple warm forging for ultrafine grained microstructures [1], high cooling rate to decrease microsegregation [2], and hot-top design and optimization [3]. However, it should be stressed that these remedial measures cannot change the natural dendrite feature of ingot and hence, solidification defects still exist to some extent. Therefore, some innovative routes for manufacturing high-quality forgings with the lower energy input should be adopted to break dendrites during solidification rather than during forging.

Semi-solid forming (SSF) is a promising near net shaping technology innovated by Flemings [4], which allows the production of complex-shape parts with one single step and smaller forging force. A new solidification mechanism, nondendritic solidification in this process, was first put forward. Flemings [4] revealed that dendritic fragments, rosettes and spheroids occur sequentially after the dendrites are broken by continuous shearing, and hence a unique non-dendritic microstructure, i.e., spherical solid grains, surrounded by a liquid matrix, can be obtained. The coexistence of solid and liquid phases makes the alloys appear as liquid-like flow behavior and fill the complex-shape dies under smaller resistance. The special non-dendritic microstructure in SSF also shows a great potential in eliminating those common casting defects caused essentially by the dendrites [5]. Although the SSF of low melting Al and Mg alloys has already been industrialized, the SSF of ferrous alloys is still a matter of research. Atkinson and Rassili [6] pointed out that the main obstacle hindering the direct application of steel SSF is the high working temperatures, which is a rigorous challenge for the tool lifetime.

Forging solidifying metal (FSM) is a novel method based on SSF technology for manufacturing high-quality forgings [7], in which the benefits of SSF are partially exploited. In this short semi-solid forging process, the ingot is demolded before complete solidification and forged with a mushy core. The temperature gradient between the core of the ingot (semi-solid state) and outer skin (solid state) results in a significant decrease in the forging force. The bulky dendrites are broken during solidification and, thus, a spherical microstructure may be obtained, which is similar to SSF. More importantly from a practical standpoint, as the essential dies in SSF are replaced by the solidified skin of ingot, direct contact of tools with the liquid melt can be avoided. It has been validated that the tool lifetime and part quality can be improved by such a thermal gradient [8]. It seems that the application of FSM faces fewer challenges, such as the thermal shock applied on the tool, and the oxidization of the semi-solid core.

In the present work, with the aim of verifying the feasibility of FSM process, more details on the microstructural behavior of ferrous alloys during the semi-solid forging process were investigated by physical simulation in the laboratory scale. Steels that show a large semi-solid interval and low-temperature sensitivity are desired for SSF [9]. Therefore, Cr4Mo4V, the high-temperature-bearing steel, was examined owing to its wide solidifying range. The effect of the semi-solid forging temperature and forging reduction on the microstructure was fully investigated. Furthermore, Cr4Mo4V is a hypereutectoid steel with high content of alloying elements, and plenty of eutectic carbides form at interdendritic regions during solidification. To eliminate eutectic carbides, high cooling strength [10] and relative motion between mold and consumable electrodes in ESR process [11] were investigated. But the study related to the eutectic carbides formation mechanism in the semi-solid interval has been rarely reported. Thus, the effect of semi-solid forging on the amount and distribution of eutectic carbides was also investigated in this work.

2. Experimental procedure

2.1. Material

Commercial Cr4Mo4V steel with the chemical composition given in Table 1 was used as the feedstock. The original microstructure is tempered martensite, and short-bar-shape primary carbides and uniformly distributed granular secondary carbides can be observed in the α -Fe matrix, as depicted in Fig. 1. According to the results of Differential Scanning Calorimetry (DSC), the solidus and liquidus temperatures of Cr4Mo4V are 1308 °C and 1428 °C, respectively. The wide solidifying range (more than 100 °C) makes the steel well-suited for FSM.

Table 1   Chemical composition of the commercial Cr4Mo4V steel (wt%).

CCrMoVMnSiFe
0.824.204.241.010.220.21Bal.

New window

Fig. 1.   Optical micrograph (a) and SEM image (b) of the commercial Cr4Mo4V steel.

2.2. Experimental setup and process program

The heating and semi-solid forging experiments were implemented on the Gleeble 3500 thermal simulator. To simulate the process of FSM accurately (the forging is performed during solidification rather than heating to the semi-solid state), the solid steel first needs to be re-melted. It was designed such that the cylindrical sample 10 mm in length and 6 mm in diameter was put into a hollow pure iron sheath. A small pure iron cylinder was welded upon the hollow sheath by argon arc welding to avoid the leaking of molten metal. More dimensional details of the assembled sample are illustrated in Fig. 2(a). After the solid Cr4Mo4V sample was first re-melted, it was then cooled from the above liquidus temperature to the semi-solid interval. Owing to the higher melting point of pure iron (1538 °C), the outer sheath was still in solid state with sufficient strength when the assembled sample was heated to the liquidus temperature of Cr4Mo4V (1428 °C), which thus can firmly support and protect the aimed sample well in the whole experiment. As depicted in Fig. 2(b), the sample was clamped horizontally via tungsten carbide (WC) anvils in the thermal simulator, in which lubricated tantalum and graphite plate were placed between the sample and anvils to ensure the uniform distribution of temperature. A Type-S thermocouple was welded on the midpoint of the sample to measure and monitor its temperature. Argon was used as an inert atmosphere to avoid the oxidation of samples at high temperature. Cooling copper pipe was arranged above the sample, and cold water was sprayed from the pipe to freeze the microstructure at predetermined temperatures. The heating routes and semi-solid forging parameters were controlled by the QuikSim software implanted in Gleeble 3500 system.

Fig. 2.   Schematic diagram of assembled sample (a) and experimental setup (b).

The microstructure of the Cr4Mo4V before and after semi-solid forging were investigated separately. In the solidification test, the sample was heated to 1430 °C at a rate of 5 °C/s and held for 30 s until the inside Cr4Mo4V steel was completely re-melted. Then, the sample was cooled down to the predetermined temperatures (1380, 1360, 1340, and 1320 °C) at a cooling rate of 1 °C/s and held isothermally for 10 s before quenching. The liquid fraction (fl) evolution of Cr4Mo4V as a function of temperature was determined by image analysis. Three pictures at each temperature were selected for calculation, and the average values of liquid fraction (fl ) at predetermined temperatures and the corresponding standard deviations (σ) are listed in Table 2. In the semi-solid forging test, the same heating and cooling strategies mentioned above were adopted, and the forging was carried out after isothermal holding, followed by quenching to freeze the microstructure. The experiments were conducted under different temperatures (corresponding to liquid fraction (fl) and forging reductions (K)).

Table 2   The corresponding liquid fraction (fl) and standard deviations (σ) at predetermined temperatures (Tpre).

(°C)fl (%)σ
138026.13.55
136017.182.05
135016.23.4
134013.473.06
13207.121.25

New window

2.3. Microstructural analysis

The quenched samples were cut longitudinally by wire electrical discharge machining and subjected to standard metallographic procedures. Fig. 3(a) and (b) represents the longitudinal section before and after semi-solid forging, respectively. After being polished and etched in 10% nitric acid/alcohol solution for 30 s, the microstructure was observed by Zeiss MC63 optical microscope (OM) and FEI INSPECT F50 scanning-electron microscope (SEM). The eutectic carbides were etched by potassium ferricyanide, while grain boundaries were not etched. To quantitatively investigate the microstructural evolution, the liquid fraction (fl), shape factor (F), and grain size (D) of solid particles are defined as:

$f_l=\frac{A_l}{A_a}$ (1)

$F=\frac{∑4πA_i/P_i^2}{N}$ (2)

$D=\frac{∑2(A_i/π)^{1/2}}{N}$ (3)

where Al and Aa are the areas of liquid phase and analyzed zone, respectively. Ai and Pi are the area and perimeter of each grain on the sectioned surface, respectively, and N is the number of solid particles. For F = 1, the shape of a particle is a perfect circle, while for the decreasing value (F < 1) the shape of particle is irregular. Energy-dispersive X-ray spectroscopy (EDS) was used to detect the distribution of alloying elements. High-resolution X-ray tomography experiments were conducted to investigate the three-dimensional (3D) distribution of eutectic carbides directly and quantitatively, in which samples with dimensions of Fig. 3(e) were machined from the unforged and forged Cr4Mo4V samples, as shown in Fig. 3(c) and (d).

Fig. 3.   Longitudinal sections of unforged (a) and forged samples (b). The schematic of the sample locations for X-ray tomography observation from unforged (c) and forged (d) samples and its dimension is shown in (e).

3. Results

3.1. Unforged semi-solid microstructure

The understanding of the microstructural characteristics of steel before deformation is essential for revealing the effect of semi-solid forging process. As shown in Fig. 4, the typical and dominant dendritic microstructure can be observed obviously in the quenched samples. Parallel columnar crystals (primary dendrites) propagated preferentially into the molten metal with an angle of near 90° to the interface of pure iron sheath and Cr4Mo4V alloy. Meanwhile, axiolitic secondary dendrites nucleated and grew perpendicularly to the primary ones. When the sample was quenched at 1380 °C, spherical solid grains at the center area, inherited from the prior mushy zone, were surrounded by less-developed primary dendrites. The different morphology of the grains at the center and periphery areas highlights the thermal gradient before quenching, namely, the solidification took place from the outside to the inside. As the temperature dropped, the bulky primary dendrites grew continually and formed a skeleton stretching over the entire cross section of samples (Fig. 4(e) and (g)). It is difficult to identify the original solid-liquid interface, owing to the high mobility of phase boundary [12]. However, for high alloying steels, e.g., Cr4Mo4V, there is a distinguished difference in the color or grain morphology between the early and late solid. As shown in Fig. 4(a), (c), (e), and (g), the white blur area among grains at low magnification represents the residual liquid. The liquid phase can transform into fine grains during subsequent quenching (Fig. 4(b)). In addition, some entrapped liquid was observed in the interior of primary dendrites, as shown in Fig. 4(d), and solidified at the final stage of solidification owing to the high solute content [7].

Fig. 4.   Optical micrographs of Cr4Mo4V quenched at 1380 °C (a, b), 1360 °C (c, d), 1340 °C (e, f) and 1320 °C (g, h) during solidification (RL: grains from residual liquid; EL: entrapped liquid; EC: eutectic carbides).

According to the EDS results in Fig. 5, the former liquid regions (white areas) are enriched with C, Cr, Mo and V, while the solid phase is deprived of alloying elements. Consequently, plenty of eutectic carbides (Fig. 4(h)) formed at the intergranular solute-enriched regions. The large eutectic carbides are detrimental for the fatigue properties of steel and cannot be eliminated by a subsequent hot working process [13].

Fig. 5.   SEM-EDS mapping of major elements in the unforged sample.

3.2. Microstructural evolution during FSM

The microstructure of samples forged at different temperatures with a reduction (K) of 30% are presented in Fig. 6. When forged at 1380 °C, bulky primary dendrites were partially broken up as some residual dendrites with small size still existed in Fig. 6(a) and (b). Owing to the higher liquid fraction (fl = 26.1%), solid grains with smooth surfaces were surrounded by the liquid matrix, which were transformed into martensite and austenite during subsequent quenching, respectively [14]. The distribution of the liquid phase was inhomogeneous as some coalescence of solid grains occurred. Needle-like martensite (dark brown area) can be observed within grains, as shown in Fig. 6(b). When forged at 1360 °C, primary dendrites were broken up completely, and globular grains were in contact with each other (Fig. 6(c) and (d)). Yet it is still easy to distinguish the solid and liquid phases (laminar white area) by the color difference. The bright narrow channels among grains indicate that those areas were transformed from the former liquid. As the forging temperature dropped to 1340 °C (fl = 13.47%), spherical grains became more irregular and denser (Fig. 6(e)). The entire section of the sample is almost with the single color, which means its microstructure is homogeneous in the macro scale. There are two different grain boundaries: original grain- boundary (OG) and new grain- boundary (NG). The former is the interface of different grains, while the latter mainly exist in the interior of grains, as shown in Fig. 6(f). When forged at the final stage of solidification 1320 °C (fl = 7.12%), the microstructure can be divided into two different regions, namely dark brown and white regions. More interestingly, the distribution of dark brown area retained the dendrite morphology with distortion to some extent. When the sample was held for 2 min at Tpre after deformation, the new grain-boundary in the interior of grains “disappeared” as the grains grew to more than 100 μm, but the morphology of dark brown area remained unchanged, as presented in Fig. 7.

Fig. 6.   Microstructures of samples forged at 1380 °C (a, b), 1360 °C (c, d), 1340 °C (e, f) and 1320 °C (g, h) with a reduction of 30% (A: austenite; M: martensite; OG: original grain-boundary; NG: new grain-boundary).

Fig. 7.   Microstructure of sample forged at 1320 °C with a reduction of 30% and held isothermally for 2 min after deformation.

The average grain size (D) and shape factor (F) of solid grains under different forging temperatures were calculated and exhibited in Fig. 8. As the forging temperature decreases, the average grain size decreases monotonically. When forged at 1380 °C, the highest average value is 55 μm. On the contrary, the average shape factor at 1380 °C is only 0.57 owing to the coexistence of dendrites and spherical grains. The average grain size and shape factor at 1360 °C are 45 μm and 0.87, respectively. Grains of the size of approximately 25 μm and shape factor near 0.7 were obtained at 1340 °C and 1320 °C.

Fig. 8.   Grain size (a) and shape factor (b) of solid grains forged at different temperatures.

Apart from the forging temperature, the effect of forging reduction (K) on the morphology of grains and flow behavior of liquid was also investigated. The microstructural evolution with different forging reductions at 1350 °C (fl = 16.2%) is depicted in Fig. 9. As shown in Fig. 9(a), bulky primary dendrites were developed in the solidification and subsequent quenching process when the forging was not carried out, and the average grain size and shape factor of grains were 82 μm and 0.33 in Fig. 10, respectively. In the case of 10% forging reduction, the bulky dendrites were broken partially owing to the collision with each other, as indicated by the yellow rectangle in Fig. 9(b), in which coarse primary dendrites were broken from the middle site. The average grain size decreased to 65 μm and the shape factor was 0.52. This is caused by both the flow of interdendritic liquid phase and its aggregation, as indicated by the yellow circles in Fig. 9(b). When the forging reduction increased to 30%, the primary dendrites were broken further, and more spherical grains were obtained, but primary dendrites with small size still existed, as shown in Fig. 9(c). The dendrites were broken completely until the forging reduction reached 50% in Fig. 9(d), and its shape factor of solid grains was 0.89. Sporadic coalescence of grains was observed after the liquid film between them drained out.

Fig. 9.   Microstructures of samples under different forging reductions (K) at 1350 °C (PD: primary dendrite): (a) 0; (b) 10%; (c) 30%; (d) 50%.

Fig. 10.   Grain size (a) and shape factor (b) of solid grains under different forging reductions (K) at 1350 °C.

3.3. The eutectic carbides during FSM

In the original microstructure of the sample subjected to conventional rolling and heat treatment, the preferential alignment of carbides demonstrated the rolling texture, and their distribution in 3D space was extracted and depicted in Fig. 11(a) and (d). In the unforged sample, the distribution of coarse eutectic carbides has notable dendritic features (Fig. 11(b) and (e)). Nevertheless, semi-solid forging broke the dendrite structure and changed the spatial distribution of eutectic carbides, as shown in Fig. 11(c) and (f). More details were obtained through OM images in Fig. 12. Owing to the small size of the sample, no primary carbides greater than 50 μm formed, though they appear commonly in the industrial steel ingot. The main type of carbides was eutectic carbides in the present work. In the unforged sample (Fig. 12(a)), rod and granulated carbides with rough surface were observed along grain boundaries, and the average thickness of rod-carbide was 2 μm. When deformed at 1380 °C, a large number of lamellar carbides with a width of 1 μm and length of 50 μm at most were formed at intergranular areas. When deformed at lower temperatures (1360 and 1340 °C), only small-scale and dispersive eutectic carbides were observed. However, the connected regions where eutectic carbides existed were observed again at 1320 °C, which is similar to the unforged sample. The volume fraction of eutectic carbides (Fec) was calculated through extraction software (3D extraction) and image analysis (eutectic carbide and eutectic structure), respectively, as shown in Fig. 13.

Fig. 11.   The eutectic carbides of samples and the corresponding 3D distribution: (a, d) original microstructure; (b, e) microstructure of unforged state; and (c, f) 30% forging reduction at 1380 °C.

Fig. 12.   OM images of eutectic carbides of samples in the unforged state (a) and forged at 1380 °C (b), 1360 °C (c), 1340 °C (d), and 1320 °C (e) with a reduction of 30%.

Fig. 13.   The volume fraction of carbides of samples subjected to different forging conditions.

In the eutectic structure, lamellar or granular carbides were embedded in the α-Fe matrix, which is difficult for the calculation software to distinguish. The extracted and calculated “carbides” via 3D tomography actually included carbides and α-Fe matrix near them. Thus, the Fec value of 3D extraction was similar to that of the eutectic structure but higher than that of eutectic carbides. The Fec of samples forged at 1380 °C (6.24%) was lower than that in the as-casting state (7.23%), while the eutectic structure in the former is higher than that in the latter. As the forging temperature decreased, Fec was reduced to 1.29% at 1340 °C. However, the volume fraction of the eutectic structure and carbides increased when semi-solid forging was implemented at 1320 °C.

4. Discussion

4.1. Effect of semi-solid forging temperature on microstructure

Cr4Mo4V steel solidifies with a typical columnar dendritic structure, as depicted in Fig. 4. FSM broke the dendritic structure and spherical grains were obtained similar to SSF. The forging behavior of steels in the semi-solid state is deeply affected by the liquid phase [15], so the morphology of solid grains after quenching significantly varies with forging temperatures. For instance, at 1380 °C, the dendrite skeleton is “soaked” in the liquid phase owing to the high liquid fraction (26.1%). The large deformation first breaks the skeleton, namely primary dendrites. Then the solid particles from fragmented primary dendrites or spherical secondary grains may rotate and move under external force [16]. The liquid phase flows along the original gaps among grains or new paths penetrating the broken dendrites. The forced-convection of liquid phase makes the solid grains smoother simultaneously due to the re-melting of sharp edges. Nevertheless, the shape factor of those grains is scattered as dendrites and spherical grains coexist (as indicated in Fig. 6(b)), which results in a lower average shape factor (0.57). There are two sources of residual dendrites: unbroken primary dendrites and newly formed dendrites. The former is caused by the lower forging reduction (30%), while the latter results from the normal solidification of residual liquid after forging. In the case of 1360 °C (fl = 17.18%), the rotation of spherical grains is restricted, and the sliding mechanism becomes dominant in which the residual liquid functions as a lubricant instead of completely surrounding the grains. The dendrites are adequately broken up and a part of the solid grains begin to come into contact with each other as the liquid phase decreases. The lubricant liquid fills the gap among grains and promotes the spheroidization of grains, so the highest shape factor 0.87 is finally obtained. When deformed at the final stage of solidification (1340 °C and 1320 °C), the rotation or sliding mechanism is limited owing to the decreasing in the lubricant liquid, and the direct plastic deformation of primary dendrites skeleton plays a dominant role. The above mentioned three motion mechanisms under different temperatures are similar to that in SSF which was reported [17], but are quite different from the classical hot forging, owing to the coexistence of solid and liquid. It is noteworthy that irregular grains with rough surfaces and sharp edges, i.e., lower shape factor, can arrange compactly at high magnification, as shown in Fig. 6(d), (f), and (h). This phenomenon is attributed to the high mobility of the phase boundary in spite of the usage of cold water quenching [12].

Similarly, Meng et al. studied the martensite transformation of SKD11 and SKD61 during rapid cooling from semi-solid temperature, and demonstrated the coexistence of austenite and martensite in high alloying steel SKD11 [18]. Such phase transformation also occurs in Cr4Mo4V during quenching, since the alloying element content of the steel is greater than 10%. As shown in Fig. 14(a) and (b), after etched in nital, the areas among grains was not etched and remained bulged. The analysis of EDS demonstrated that those areas are enriched with C, Cr, Mo, and V, which could affect the phase transformation during quenching. According to the empirical equation [19], the martensite starting temperature (Ms: °C) can be described as a function of contents of alloying elements:(4)Ms=635-475WC-17WCr-33WMnWhere WC, WCr, and WMn are the content of C, Cr, and Mn (wt%), respectively. The residual liquid with higher contents of C and Cr transforms into metastable austenite owing to the lower Ms, while the carbon-poor solid fraction transforms into martensite after rapid cooling from the semi-solid state. Thus, the distribution of austenite and martensite in Fig. 6 reflects the segregation of alloying elements or microstructural inhomogeneity to some extent. Semi-solid forging results in liquid convection and thus plays an important role in the diffusion-controlled homogenization of alloying elements. Nevertheless, when the deformation is conducted at 1380 °C, owing to the high liquid fraction, the elemental redistribution continues during the subsequent normal solidification, which causes alloying segregation and an inhomogeneous microstructure. When deformed at 1360 and 1340 °C, the distribution of martensite is homogeneous on a large scale, which confirms the full diffusion of alloying elements and the absence of severe liquid segregation. When the forging is conducted near the solidus temperature, the distribution of martensite presents the characteristic of dendrite. Since the solidification completed, deformation does not affect the distribution of alloying elements, and only the dendrite skeleton is broken via severe plastic deformation (Fig. 6(g) and Fig. 8). In other words, the forging around the solidus temperature makes no difference to the less-segregation of alloying elements.

Fig. 14.   SEM image of sample forged at 1380 °C with a reduction of 30% (a) and (b). (c) and (d) are the locations of line-scan analysis. (e), (f), (g), (h), and (i) are the contents of Fe, C, Cr, Mo, and V in line-scan 1, respectively. (j), (k), (l), (m), and (n) are the content of Fe, C, Cr, Mo, and V in line-scan 2, respectively.

4.2. Effect of semi-solid forging reduction on microstructure

In the conventional SSF process, the cooling slope process [20], electromagnetic stirring [21], strain-induced melt activation (SIMA) [22], and recrystallization and partial melting (RAP) [23] were widely adopted to fabricate billets with a fine-grained spherical microstructure. The role of those methods is replaced by the forging force in the current FSM. As illustrated in Fig. 15, such dynamic spheroidization process can be summarized as three stages: dendrite fragmentation, flow and aggregation of liquid, and spheroidization of grains. As depicted in Fig. 15(a), coarse dendrites nucleate and grow from the outside owing to the thermal gradient between the core and edge areas. At the initial stage of forging (Fig. 15(b) with a reduction of 10%), a portion of coarse dendrites is broken. According to the dendrite fragmentation criterion established [24], the partial re-melting of dendrite arms by solute-rich liquid leads to the fragmentation of dendrites in electromagnetic stirring (EMS). However, the forced-convection of the liquid phase caused by forging is much weaker than that in EMS. On the contrary, the flow of liquid is restricted by the partly-broken dendrites and then solidifies spontaneously, which even causes liquid segregation to some degree. Hence, the fragmentation of dendrites in FSM is largely dependent on the mechanical force, and the force is transmitted by the direct contact of solid dendrites. As the forging reduction increases to 30% (Fig. 15(c)), completely-broken dendrites promote the flow of liquid, and the interstitial space among grains is also “fed” by the liquid to avoid the formation of porosity or even fracture. When the forging reduction increases to 50% (Fig. 15(d)), the dendrites are broken further and transformed into polygonal grains. The resistance for liquid flow is further reduced, thus, the distribution of solid grains and liquid become more homogeneous. Those polygonal grains surrounded by liquid films grow and come into contact with each other during the subsequent quenching. This process is similar to the spheroidization of dendrites in SSF where shearing of stirring replaces the role of forging reduction. In the thixoforging process, the outflow of liquid from the center at higher forging reduction (more than 50%) is inevitable [25]. Extensive research has been conducted to control liquid segregation, such as back-pressure thixoextruding [26] and multi-stage thixoforging [27]. In the present study, that case is alleviated to a large extent from two aspects since no sever liquid segregation are observed in Fig. 9. On the one hand, the morphology of bulky dendrites increases the resistance of liquid flow at the initial stage of forging, which prevents the aggregation of liquid, as illustrated by the black arrows in Fig. 15(b). On the other hand, the outflow of liquid is limited by the pure iron sheath, where a three-dimensional compressive stress state is created. Because there is no contractive or dilatational volume variation in this stress state, the separation of solid and liquid is suppressed.

Fig. 15.   Schematic diagram of dynamic forging process.

4.3. The elimination mechanism of eutectic carbides

The formation of eutectic carbides strongly depends on the segregation of alloying elements, such as C, Cr, Mo, and V. The segregation of alloying elements has a strong relationship with the flow of interdendritic solute-enriched liquid [10]. The segregation of alloying elements at grain boundary areas has been proved by the line-scan analysis in Fig. 14. As illustrated in Fig. 16, eutectic carbides mainly form along grain boundaries where solute-enriched liquid aggregates. In the unforged sample, eutectic carbides nucleate and precipitate at interdendritic regions without agitation, so plenty of angular carbides are observed in Fig. 12(a). As the coarse primary dendrites are broken by semi-solid forging, the nucleation and growth conditions of eutectic carbides are altered. The elimination of eutectic carbides can be analyzed from two aspects. First, the diffusion length for alloying elements is shortened since the dendrites are transformed into globular grains. Furthermore, the plastic deformation at low temperatures creates excess vacancies and dislocations [28], which enhance the diffusion of elements. The full diffusion of alloying elements causes the content to deviate from the nucleation of eutectic carbides. Thus, the nucleation of eutectic carbides is suppressed to some extent in its infancy. Next, the fragmentation of primary dendrites causes the contact area between solid grains and liquid phase to increase greatly, so the average size of areas among grains where liquid aggregates decreases. As the distribution of liquid became more homogeneous, the carbide growth tendency decreases because there is no severe liquid segregation to supply enough alloying elements [29]. Consequently, only dispersive carbides with small size are formed, as illustrated in Fig. 16(b). That explains why the amount of eutectic carbides decreased continuously from 1360 °C to 1320 °C. However, as described in Section 3.2, globular grains with smooth surface decrease the resistance of liquid flow, and the residual liquid tends to flow and aggregate under the external force until the occurrence of liquid phase segregation, especially at high forging temperatures. For instance, when forged at 1380 °C (fl = 26.1%), the solute-enriched liquid aggregated at the grain boundary and solidified normally, thus, the area where the eutectic reaction occurred increases, resulting in a higher volume fraction eutectic structure. But meanwhile the short-distance diffusion of alloying elements promotes the refinement of carbides and results in a lower Fec than the unforged sample. To summarize, contrary to the breaking mechanism of primary carbides in classical hot forging, FSM aims to eliminate eutectic carbides by inhibiting the nucleation and growth process, and thus highlights a new method to produce superior bearing steel with homogenized microstructure, composition and properties.

Fig. 16.   Eutectic carbides with large size (a) or small size (b) formed at intergranular regions.

5. Conclusions

The present work demonstrates that semi-solid forging during solidification has an important effect on the microstructural evolution of Cr4Mo4V. The main conclusions are summarized as follows.

(1) Semi-solid forging broke bulky dendrites and, thus, globular grains with the shape factor of 0.87 were fabricated when the sample was deformed at 1360 °C (fl = 17.18%), while dendrites with small size still existed in the case of 1380 °C (fl = 26.1%). The semi-solid forging at 1340 °C is beneficial to the microstructural homogeneity and composition.

(2) The spheroidization of grains during FSM was revealed experimentally. High forging reduction (at least 50%) is essential in the process. The bulky dendrites inhibit the flow of liquid phase and consequently reduce the liquid segregation.

(3) The volume fraction of eutectic carbides was reduced from 7.23% (as-unforged state) to 1.29% when forged in the end of solidification (1340 °C). Different from conventional remedial methods, eutectic carbides are eliminated by promoting the diffusion of alloying elements and alleviating liquid segregation during semi-solid forging.

Acknowledgements

This work was supported financially by the National Key Research and Development Program (No. 2018YFA0702900), the National Natural Science Foundation of China (Nos. U1508215, 51774265 and 51701225), the National Science and Technology Major Project of China (No. 2019ZX06004010), the Key Program of the Chinese Academy of Sciences (No. ZDRW-CN-2017-1) and the Program of CAS Interdisciplinary Innovation Team. The author deeply appreciate Gleeble engineer Jiajun He for her technical support in superhot experiments.


/