Introduction

SiMo ductile cast iron has higher levels of silicon (4 to 6 pct) and molybdenum (0.5 to 2.0 pct). The microstructure of SiMo ductile iron consists of nodular graphite and an apparent minor range of cell-boundary carbides in a typically ferritic matrix. The carbides have a significant role in increasing the hardness and abrasive wear resistance.1,2 The increase in Si content causes an increase in the yield strength but lowers toughness and elongation. During solidification, Mo partially segregates and forms carbides on grain boundaries. These Mo-rich carbides increase creep, corrosion and wear resistance but reduce ductility. Due to the above combination of characteristics, SiMo iron alloy has been widely used in high-temperature applications such as exhaust manifolds, furnace parts, and turbocharger housing in the past dozen years.1,2,3,4

Several methods were commonly obtained for modifying the properties and structure of ductile iron, such as heat treatment, structural refinement, and alloying by carbide-forming elements, including V, Cr, Al, and Mn.5,6,7,8,9 Riebisch et al.5 examined the impact of Cr, V, and Mn on the properties of ductile iron containing 3.8 wt.% Si and found that V had a significant effect on graphite morphology, promoting the formation of chunky graphite due to its capacity to slow down C diffusion in austenite and prolong solidification time. Conversely, due to their segregation, Cr and Mn chiefly influenced the pearlite area fraction. M. Ibrahim et al.9 examined how the microstructure, high-temperature oxidation, and mechanical properties of SiMo ductile cast iron are influenced by varying levels of Si and Mo, as well as 3 wt.% Al. Aluminum promotes compacted graphite formation in magnesium-treated ductile cast iron. The maximum operating temperature of SiMo alloys can be increased by increasing Si content or adding 3 wt.% Al to low Si-SiMo irons. Increased Si content or adding 3 wt.% Al significantly increases ultimate and yield strengths but decreases elongation and impact energy. It was found that at about 0.7 wt.% Mo, the highest toughness values are obtained for all Si contents.

Cast irons are well-known as typical materials for wear applications, particularly for frictional sliding wear under dry and lubricated conditions.10 A combination of significant variables influencing wear behavior include the following: material, operational, geometrical, and environmental factors. Alloying and heat treatment can enhance abrasion wear resistance to get a harder martensitic or ausferritic matrix.10,11

Extensive efforts have been made to comprehend the abrasion wear behaviors of various types of cast iron in multiple variables and applications. A recent study 12 investigated the effect of Mo contents on the high-temperature wear behavior of SiMo ductile iron at different temperatures up to 750 °C. It was found that the increase in Mo content improved the wear resistance regardless of operational condition. At higher temperatures, the increase in wear resistance was clearly due to the formation of an oxidation layer, which acts as a protective layer.

Some of the applications of SiMo may be exposed to high-temperature wear conditions. Therefore, this current work studies the effect of alloying elements (Cr, V, and Ni) and different cooling rates on the microstructure of SiMo ductile iron alloys. Based on these microstructures, the abrasive wear behavior of SiMo ductile iron was investigated under various testing loads and a temperature of 700 °C.

Experimental Investigation

Material

SiMo ductile iron alloys were melted using a 100 kg induction furnace with a medium frequency (1000 Hz). The primary raw materials used were steel scrap and high-purity pig iron. The furnace was charged with the additives: (carburizer (99% C), FeSi (65% Si), FeMo (70% Mo), FeCr (65% Cr), FeV (80% V), and pure Ni. After the charge completely melted and reached 1520 °C, the molten metal was treated with magnesium as the Fe-Si-Mg (9% Mg) alloy needed for graphite spheroidization, then inoculant (1.3–1.8% Zr) was added as well. The Vortex technology13,14 was used for spheroidization and inoculation purposes (see Figure 1). To create the vortex, the molten iron flows tangentially into a funnel-shaped chamber. Through a calibrated hole in a thermally shock-resistant tube at the center of the produced vortex, the additive is fed into the metal from a hopper. The molten iron was deslagged before being poured into the molds and after the spheroidization and inoculation procedures, as shown in Figure 2. The chemical composition of the SiMo ductile iron samples is as shown in Table 1. The patterns used to cast 200×150 mm plates with thicknesses of (6 and 9 mm with an error range of ±0.25 mm). The utilized molds in this study were green sand and investment molds. The investment casting molds were preheated from room temperature to 300 °C for almost half an hour and then to 900 °C to avoid the thermal shock.

Figure 1
figure 1

A schematic illustration of the vortex unit used.

Full size image
Figure 2
figure 2

Test investment molds and the casting process.

Full size image
Table 1 The Chemical Composition of SiMo Alloys
Full size table

Metallographic Observation

The samples were manually ground with Si carbide abrasive paper, polished and ultrasonically cleaned, and then etched in 2% Nital. After that, metallographic examinations were carried out on the prepared samples. EDX and SEM FEI-QuantaTM were used for further investigating the microstructure.

Thermodynamic Calculations

Thermo-Calc Calculations

The Thermo-Calc 2021b database (TCFE 11) was utilized to investigate the phase diagram and for determining the volume fraction of phases and transformation temperatures.

Differential Scanning Calorimeter

Nonetheless, in order to determine the phase transformation temperatures more precisely, the information from the computed phase diagram by Thermo-Calc was compared with that obtained from measurements made using differential scanning calorimetry (DSC). Differential Scanning Calorimetry (DSC) machine NETZSCH STA 409 C/CD, reference Al2O3, mass/mg: 18,000, type Crucible DSC/TG pan Al2O3 were used for the tests. The testing samples were heated to 1300 °C at a rate of 30 °C / min and then cooled to room temperature at a rate of 30 °C / min in helium gas medium.

High-Temperature Abrasive Wear Test

High-temperature ball-on-disk wear experiments were carried out in a dry environment to determine the impact of the alloying elements and thickness on the wear behavior of the SiMo ductile iron alloys. The tests were performed on a wear test machine (type T-21, Poland) (see Figure 3). The heating chamber in which the friction couple was located was adequately insulated to prevent heat loss throughout the tests conducted at 700 °C. The test sample had a 25 mm diameter and a 10 mm thickness, and a ceramic ball with a 10 mm diameter served as the counterpart. All samples were ultrasonically cleaned with absolute ethanol and then dried. The test conditions were three different normal loads (10, 20, 30 N), a fixed rotation speed of 0.25 m/s, and a sliding distance of 200 m for 20 min. The weight loss was measured by a weighing method, and each sample was weighed twice before and after wear testing by a sensitive digital balance. The wear amount was measured by weight loss, averaging at least three runs for accuracy. The friction force curves were also recorded versus time and temperatures during the test. The average coefficient of friction can be calculated using the values of Ff and Fn, which represent the friction force and the normal force, respectively.

Figure 3
figure 3

Test machine setup and schematic illustration of the high-temperature wear test.

Full size image

Results and Discussion

Thermodynamic Calculations

Thermo-Calc Calculations

Theoretical calculations of Thermo-Calc determine the phases and transformation temperatures based on chemical composition, ignoring the various cooling rates caused by the varying thickness and mold material. The solidification process of the carbide phases in equilibrium with ferrite and austenite is described in further details by the Thermo-Calc phase diagrams in Figure 4. During eutectic solidification, eutectic carbides M6C type are formed in unalloyed SiMo alloys; these carbides remain in the SiMo alloys until they reach ambient temperature. Different complex carbides are generated in alloyed SiMo alloys, including Mo-rich carbides M6C, V-carbides VC, and Cr-carbides M7C3, as listed in Table 2. This agrees with Thermo-Calc calculations,19,20 which suggested that the formed carbides in SiMo51 be of types: (M=Mo, Si, and Fe) M6C and (M=Cr) M7C3. The M7C3 carbides precipitated at 620°C for 0.5% Cr and at 770 °C for 1% Cr in the phase diagram.15 The vanadium carbides are VC-type, as mentioned in the phase diagram of the Fe-C-V system.17 From Table 2, the difference in carbide precipitation behavior might be correlated with the stoichiometric ratio of iron in eutectic carbides in alloyed/unalloyed SiMo. Furthermore, the system goes through the eutectoid transformation, which results in the coexistence of ferrite and austenite across the temperature ranges of the transformation process (at which ferrite turns into austenite) Ac1- Ac3 throughout the whole carbon content range. Thermo-Calc calculations show that unalloyed SiMo showed higher eutectoid temperatures, with austenite formation A1 occurring at 963 °C. The eutectoid temperature A1 is roughly 50 °C lower in alloyed SiMo, indicating lower thermal stability than unalloyed SiMo. This may be related to Ni addition, an austenite stabilizer.

Figure 4
figure 4

Phase diagrams predicted by Thermo-Calc: (a) Unalloyed SiMo,16 (b) Alloyed SiMo.

Full size image
Table 2 The Critical Transformation Temperatures and Observed Phases at Room Temperature
Full size table

Figure 5 shows the volume fraction in terms of temperature under an equilibrium state. Unalloyed SiMo’s microstructure is composed of graphite nodules, M6C carbides, and a ferritic matrix. The microstructure of alloyed SiMo contains nodular graphite, complex carbides (Fe, Si, Mo, Cr, and V), and a pearlite-ferrite matrix. In alloyed SiMo, adding Cr increases the amount of pearlite and stable Cr-carbides.

Figure 5
figure 5

Volume fraction of all phases in SiMo castings: (a) unalloyed, (b) alloyed.

Full size image

Differential Scanning Calorimeter

The theoretical Thermo-Calc calculations neglect casting conditions such as the varied cooling rates subsequent from different thicknesses and mold material. Differential Scanning Calorimetry (DSC) test was conducted to get the actual measurements of the phases and transformation temperatures at different casting parameters. Table 3 and Figure 6 reveal temperatures of the phase transformation peaks derived from the DSC curves, and compare these peaks of phase transformation temperatures in cooling and heating to the simulated measurements by Thermo-Calc. The higher cooling rate (6 mm) reduces the values of the eutectoid temperature A1 in comparison to the lower cooling rate of 9 mm, as listed in Table 3. It is notable that the phase changes depicted by the DSC curve corroborate those predicted by the phase diagram produced by Thermo-Calc (see Tables 2 and 3). This confirmation, on the other hand, is more qualitative, with some variances in transformation temperatures to be expected.

Table 3 Transformation Peaks of Unalloyed/alloyed SiMo Ductile Iron During Cooling and Heating
Full size table
Figure 6
figure 6

Cooling DSC curves of unalloyed SiMo samples: (a) 6mm, and (b) 9mm thickness.

Full size image

Microstructure Investigation

For investigating the influence of casting parameters such as thickness and mold material (i.e., cooling rate) in addition to the added alloying elements to the SiMo alloys (Cr, Ni, V) on the microstructure, the contents of pearlite and ferritic matrix, and nodule count are determined by using image analysis software according to EN ISO 945-1 as plotted in Figure 7. The microstructure of unalloyed SiMo ductile iron (see Figure 6) typically consists of nodular graphite with eutectic carbides embedded in different contents of cell boundary precipitates in a fully ferritic matrix. The alloyed SiMo samples exhibit a pearlite-ferrite matrix due to alloying with Cr, Ni, and V, which are strong pearlite promoters and contain an amount of nodular graphite and complex carbides, as shown in Figure 8.

Figure 7
figure 7

Qualitative analysis of (a) Nodule count, (b, c) Ferrite and pearlite contents, at different casting parameters, where U and A are unalloyed and alloyed SiMo DI, respectively.

Full size image
Figure 8
figure 8

Optical micrographs of all examined unalloyed/alloyed SiMo alloys.

Full size image

Effect of Alloying Elements

Unalloyed SiMo

As revealed in Figure 8, the microstructure of unalloyed SiMo ductile iron consists of a ferritic matrix with nodular graphite and the cell-boundary precipitates that are related to Mo-segregation phenomena.12,18,19 Thermo-Calc calculation suggested these eutectic carbides to be of M6C-type, which is in accordance with SEM and EDX analysis. As in Figure 9, the eutectic carbides contain around 43% Mo with variable levels of [Fe, Si, C]. These eutectic carbides are surrounded in the cell boundary by precipitates of the lamellar pearlite with a low Mo content of 3%, as shown in Figure 9.

Figure 9
figure 9

SEM-EDS analysis of the carbides in unalloyed SiMo ductile iron, 10 micron magnification.

Full size image
Alloyed SiMo

In alloyed SiMo, the microstructure contains nodular graphite, pearlite, ferrite, and a wide range of (Fe, Mo, Cr, and V) carbides. At closer magnification by SEM and EDX (see Figure 10), angular and dot-like carbides are noticed. Based on Thermo-Calc software, carbides precipitate at various stages. In previous work,20 these carbides have been classified into the following types: (i) type I precipitates form on the liquid/solid interface, either during or after solidification. Type I precipitates, which are extremely stable precipitates, are most likely angular carbides. (ii) Type II particles precipitate in austenite after the process of solution treatment, as well as during the hot deformation stage when the temperature decreases. (iii) Type III particles (dot-like carbides) formed during and after the austenite to ferrite phase transformation, by nucleation on the γ/α interface, and in ferrite. Vanadium, as a carbide stabilizer, resembles molybdenum, which segregates close to the cell boundary, forming eutectic carbides (see Figure 10). While the fine precipitate consists of a more complex form and has lower Mo-contents, the eutectic carbides are primarily (Fe, Mo, Cr, and V) carbides.

Figure 10
figure 10

SEM-EDS analysis of the carbides in alloyed SiMo ductile iron, 10 micron magnification.

Full size image

Effect of Casting Parameters (Thickness and Mold Material)

The mold material and thickness (e.g., cooling rate) influence the nodule count and the relative volume fraction of ferrite, pearlite, and carbides, as represented in Figures 5 and 6. Alloyed/unalloyed SiMo samples, cast in greensand molds with a thinner thickness, have a higher count of nodules because of the faster cooling rates. The cooling rate controls the amount of ferrite and pearlite in the microstructure, as described in Figure 6. According to,21,22 thin-walled DI castings during solidification generally exhibit high cooling rates, which lead to changes in the microstructure, especially (a) the precipitation of iron carbides and (b) an increase in nodule count. As displayed in Figure 11, fine precipitates of lamellar pearlite formed at higher cooling rates (6 mm sections), whereas at slower cooling rates, spheroidal or rod-like structure was revealed. The morphology of the eutectic carbides depends on the cooling rate. At the slower cooling rate, the eutectic carbides have fish-bone structure morphology with higher levels of molybdenum due to the higher degree of segregation (see Figures 9, 11b, d). When the cooling rate increases (i.e., 6 mm), reducing the amount of molybdenum segregation, intercellular precipitates of carbides have a Chinese script pattern formed (see Figures 9, 11a, c). This agrees with Youssef et al.16 who stated that the casting’s cooling rate affects the final microstructure and the morphology and content of the carbide precipitates.

Figure 11
figure 11

SEM micrograph showing the carbide morphology with apparent pearlite in the intercellular area, (a, b) 40 micron magnification, (c, d), 30 micron magnification.

Full size image

High-Temperature Abrasive Wear Behavior

The material’s microstructure is one of the main factors influencing and controlling the wear behavior. Environmental parameters, such as in the current work high temperature and applied load, also affect wear. The calculation of weight loss is one of the standard methods in the evaluation of wear behavior. Figure 12 shows the weight loss of all unalloyed/ alloyed SiMo samples at different casting conditions.

Figure 12
figure 12

The weight loss of all examined SiMo samples, where U and A are unalloyed and alloyed SiMo DI, respectively, under different loads.

Full size image

Influence of the Microstructure

According to Figure 12, unalloyed SiMo samples exhibit higher weight loss than alloyed ones in both thicknesses and molds under different loads. So, alloyed SiMo alloys demonstrate higher wear resistance than unalloyed SiMo alloys. Based on Thermo-Calc and EDX analysis, alloyed SiMo alloys contain a higher carbide content than unalloyed ones due to Cr and V, which formed M7C3 and VC carbides besides the eutectic carbides M6C. As the carbide fraction increases, the wear resistance increases and the weight loss decreases to the lowest values. In both alloyed and unalloyed SiMo samples, the thinner-thickness samples have higher wear resistance than the thicker samples due to increased carbide contents. Consequently, sample A-6mm, containing the highest carbide volume fraction, showed the best wear resistance. The assumed explanation is that the carbides (M6C in Unalloyed SiMo) and (M7C3 and M6C carbides in alloyed SiMo), which have relatively high hardness and chemical stability, make it difficult for abrasive particles to be implanted in the wear surface. This agrees with Gao23 finding that eutectic carbides significantly improve the alloy’s wear resistance. When the matrix is fully ferritic (unalloyed SiMo), the wear resistance is lower than alloyed SiMo alloys, which attain a multiphase (pearlite-ferrite) matrix (see Figure 6). The pearlite has good wear resistance to friction and modest abrasion.10 Moreover, the wear results show that the increased nodule count reduces the wear resistance of unalloyed SiMo samples. However, this reduction effect does not appear in alloyed SiMo samples due to the low nodule count. This agrees with R. Dommarco,24 who revealed that an increase in nodule count deteriorates the abrasion resistance of ferritic ductile iron. Nevertheless, minor variations in nodule count could be ignored without affecting the abrasive wear. According to Cimenoglu,25 graphite nodules function as a solid lubricant that increases wear resistance by lowering the friction coefficient and the contact surface. However, this graphite nodule behavior may not be observed at high temperatures because graphite is no longer effective as a lubricant. This is consistent with the current study, which was conducted at 700 °C. These results conclude that the microstructure of SiMo alloys is a critical parameter in controlling wear loss.

Influence of the operational factors on wear behavior

The topography measurements of the samples reveal the impact of different loads and the high temperature of 700 °C of the surroundings. Wear tests were performed under variable normal loads (10, 20, 30 N) and at a fixed rotation speed of 0.25 m/s. Such applied pressure prominently influences the friction between the pin (ceramic ball) and the rotating disk (test samples). Figure 13a shows the friction coefficient as a function of the applied loads. The friction coefficient (COF) increases with the increase in applied loads. At 30 N, the friction coefficient exhibits its highest values, ranging from 0.99 to 1.27. On the other hand, under a 10N load condition, the COF values decrease by almost half that in 30N. The COF may increase or decrease during the wear test depending on whether debris with hard particles accumulated in the wear track or whether the counterface ceramic ball polished the wear track. According to the results, the applied load plays two roles in impact wear. As the impact load increased, there was a significant increase in friction between the material (sample) and the abrasive, which increased the wear rate. However, when the impact load increased, the friction behavior became unstable, which resulted in an irregular wear rate and a wavy wear track surface (see Figure 13b). Under low load, friction behavior is steady throughout the test cycles, leading to a primarily stable wear rate and relatively smooth wear track surface, as seen in Figure 13b.

Figure 13
figure 13

The friction coefficient vs. (a) the applied loads, (b) time, where U and A are unalloyed and alloyed SiMo DI, respectively.

Full size image

Determining the wear track width on the worn surface is another way to study the wear behavior. Figure 14 shows the wear track of some selected (alloyed/unalloyed) samples under the applied loads. The more applied load, the wider the wear track is. The unalloyed SiMo samples (see Figure 14a–c), which demonstrated more weight loss, have wider wear tracks in comparison to alloyed SiMo samples (see Figure 14d–f). Thus, the values of wear track width are consistent with the trends observed for average weight loss.

Figure 14
figure 14

Worn surfaces of some SiMo samples in macroscale under different loads.

Full size image

The worn surface has been covered with a brownish layer that appears to be an oxidized surface. This suggests that oxidational wear could be the likely wear process due to the iron oxide particles on the worn surfaces under the high-temperature environment of 700 °C. In the current research, the wear surface of the investigated alloys did not contain any abrasive particles. SEM and EDX are used to investigate the observed layer on the worn surface, as illustrated in Figure 15. The EDS analysis reveals the presence of oxygen at the surface of wear debris, suggesting that most of the wear debris is exposed to consistent oxidation. The pearlitic microstructure is more likely to develop an oxide layer than the ferritic or austenitic microstructures.26 The formation of the protective (oxide) layer will also likely be impacted by the pearlite content in these alloys. The applied load remarkably influences the behavior of oxidational wear. The formed oxide layer is predicted to act as a protective layer at low loads, preventing excessive weight loss. Under higher loads, the oxide layer does not have enough time to form and grow because of its strong tendency to break down. Iron oxide particles have sufficient hardness to function as abrasives and accelerate the wear rate.27,28 On the macromorphologies (Figure 14), the wear scars can be seen as roughly parallel.

Figure 15
figure 15

SEM and EDX analysis for the worn surfaces detecting oxide layer.

Full size image

Some worn surfaces were examined by SEM for further investigation, as shown in Figure 16. The worn surface of an unalloyed SiMo sample under 30 N applied load displays deep abrasive grooves, delamination wear, and residual oxide layer.

Figure 16
figure 16

SEM micromorphologies of worn surfaces of unalloyed SiMo at 30N.

Full size image

Conclusions

  1. 1.

    Thermo-Calc software was used for calculating the phase diagrams of SiMo ductile irons from which the sequence of phase transformation and volume fractions of phases were predicted.

  2. 2.

    The unalloyed SiMo ductile iron microstructure is composed of a ferrite matrix and Mo-rich carbides embedded within the fine precipitates in the intercellular areas.

  3. 3.

    The alloyed SiMo ductile iron microstructure consists of a multiphase ferrite-pearlite matrix and a combination of carbides of eutectic M6C, VC, and M7C3 carbides.

  4. 4.

    The alloyed samples showed a noticeable reduction in the weight loss more than that of alloyed ones at variable loads (10, 20, 30 N), which means that the wear resistance of alloyed samples is higher than unalloyed ones in both greensand and investment molds.

  5. 5.

    According to the wear results, the microstructure is critical to wear loss. Comparatively to a ferritic matrix with lamellar pearlite and eutectic M6C carbides, the multiphase ferritic/pearlitic matrix with eutectic M6C and M7C3 carbides appears more wear-resistant.

  6. 6.

    As wear tests were at 700  °C, the oxidational wear affects the wear behavior due to the iron oxides on the worn surface acting as a protective layer and improving wear resistance.