Introduction

SiMo ductile cast irons with a silicon content of 4–6% and a molybdenum content of 0.5–2% are used in exhaust manifolds in the automotive industry, gas turbine components, molds for casting titanium, brass, and zinc alloys, holders, and elements of furnaces for heat treatment.1,2,3 These applications require oxidation resistance, structural stability, high-temperature mechanical properties (e.g., high strength, ductility), and resistance to fatigue during thermomechanical cycles to ensure better performance under operating conditions.4,5,6,7,8

Casting studies and solidification models have revealed that depending on the critical thickness of the casting, an appropriate carbon equivalent (C.E.) can control the porosity and shrinkage tendency of high Si and SiMo ductile iron.9,10,11 The processing parameters of chemical composition, inoculation treatment, and cooling rate significantly influenced the final microstructure, such as nodule count and carbide volume fraction, e.g., cell boundary carbide precipitation in SiMo alloys.11,12,13

The microstructure of SiMo ductile iron consists of ferrite, carbides at the cell boundaries, and nodular graphite. Limiting pearlite is necessary to avoid alloy elongation and growth resulting from the decomposition of pearlite at high operating temperatures.14 The addition of Si stabilizes the ferritic matrix and increases A1 (ferrite/austenite) transformation temperature. The molybdenum partially segregates, forming Mo-carbides at grain boundaries during eutectic solidification.3,14,15,16 The high Si and Mo content increases the eutectoid transformation temperature, preventing phase transformation at high application temperatures.

Ni-based austenitic ductile iron grades such as Ni-Resist D5S can achieve the requirements at higher exhaust temperatures of up to 950 °C. But Nickel, which is heavily used, has recently fluctuated in market price. Economically, it is essential to develop cheaper heat-resistant materials, i.e., SiMo alloys at high temperatures, instead of more expensive ones. Therefore, besides the standard SiMo, grades alloyed with about 0.5 to 1.0% of Al, Cr, or Ni are produced (SiMo1000, SiMoCr, and SiMoNi, respectively), with the characteristic of having a stronger scaling resistance.17,18 The SiMo1000 is developed up to an exhaust gas temperature of 900 °C by George Fischer Eisenguss GmbH,19 which has a significant amount of compacted graphite and Al-rich carbides.

During solidification, the Mo-rich carbides formed in the ferritic matrix enhance the creep behavior and increase the high-temperature strength.2,3 Chavan and Khandelwal20 studied the influence of Mo on the mechanical properties of SiMo ductile iron. As the Mo percentage increased from 0.4 to 0.8 wt%, the hardness increased from 202 BHN to 230.7 BHN. With a higher Mo%, the ultimate tensile strength (UTS) and yield strength (YS) increased while the elongation decreased. The UTS and YS for the alloy with 0.8 wt% Mo were roughly 600 MPa and 510 MPa, respectively, with a percentage elongation of 10.7 percent. Méndez et al.21 reported that increasing Si also improves the mechanical properties of SiMo ductile cast iron. The Si content of up to 5.2 wt% promotes a significant hardening effect of the ferritic matrix, resulting in UTS and YS values of approximately 700 MPa and 625 MPa, respectively, while still retaining a small amount of elongation at rupture (1–2%). In addition to the enhancement of high-temperature oxidation resistance due to forming these protective Fe–Si layers, which is effective due to the high silicon content.15,21,22

With such a combination of advantageous properties, SiMo alloys are an excellent choice as wear-resistant materials, besides being basically heat-resistant. However, published works that studied the high-temperature wear performance of SiMo ductile iron are scarce. This work examines the microstructure of SiMo ductile iron alloyed with various molybdenum contents. It explains how factors such as microstructure and structural stability influence the wear behavior of SiMo ductile iron at various high temperatures up to 750 °C.

Experimental Work

Material

The melting process of four alloys with different Mo contents was carried out in a medium-frequency furnace with a 100-kg capacity. High-purity pig iron and steel scrap were the main raw materials determining the final chemical composition. Synthetic graphite carburizer (99% carbon) and Fe–Si (65% Si) were the charged additives in the furnace. Spheroidization and inoculation were carried out when the molten metal temperature reached 1520 °C after the full melting and deslagging procedures. Both spheroidization and inoculation treatments were performed using the sandwich process, where the molten metal was treated with a 1.4% Mg–Fe–Si spheroidizing alloy with 9–10% Mg and inoculated with foundry-grade ferrosilicon containing 1.5–2% Zr. This method involved introducing the Fe–Si–Mg alloy into a pocket constructed into the ladle bottom and covered with high-quality low carbon steel turnings. This created a physical barrier between the Fe–Si–Mg alloy and the molten iron, which slowed the reaction time, improved alloy use efficiency (magnesium recovery), and decreased flame and flaring. The greensand molds were used to cast rods with a diameter of 30 mm. The chemical composition of the SiMo ductile iron samples is listed in Table 1.

Table 1 The Chemical Composition of SiMo Iron Samples
Full size table

Metallographic Investigation

The samples were polished and ground before being etched in 2% Nital. Samples were taken from the mid-height of the cast rod and then, subjected to metallographic examination to investigate and characterize the microstructure. EDX and SEM FEI-QuauntaTM were used to investigate the precipitates at the cell boundaries.

Thermodynamic Calculations

In the present work, the Thermo-Calc 2021b database, TCFE 11, was used to investigate the phase diagram and figure out the volume fraction of phases and the transformation temperatures.

Dilatometry Studies

The dilatometry experiments aimed to verify the results predicted by Thermo-Calc thermodynamic calculations. A quenching dilatometer, the LINSEIS L78/RITA, was used for the dilatometer tests (Linseis Messgeraete GmbH, Selb, Germany), as illustrated in Figure 1. In this test, cylindrical samples with a 3 mm diameter and 10 mm length were used, and during the computer-controlled test cycle, the length change was measured versus temperature and time. All thermal cycles were carried out with a high-frequency induction heating generator in a vacuum of 5 * 10−3 Pa. For cooling, high-purity helium gas was used.

Figure 1
figure 1

Schematic photograph of the quenching dilatometer test.

Full size image

Hardness Measurements

As-cast samples previously prepared for the microstructure investigation were used to evaluate the impact of the various Mo contents on the hardness of the SiMo alloys. Vickers hardness measurements were done in accordance with ASTM E384 using a global tester (NEMESIS 9100, INNOVATEST). A typical load of 3 Kf was used, and an average of 5 readings was taken into account.

Hot Abrasion Wear Test

SiMo components may be subjected to high-temperature wear conditions in applications such as furnaces and gas turbine elements. High-temperature ball-on-disk wear tests were carried out in a dry environment to evaluate the impact of the microstructure on the wear behavior of SiMo alloys. The T-21 machine (model T-21, Poland), as in Figure 2, can determine test samples' displacement and wear resistance based on temperature, sliding velocity, load, and wear track radius. In the present work, the test was done at different temperatures of 25 °C, 250 °C, 500 °C, and 750 °C. The applied load was 15 N, and the sliding speed was 0.22 m s−1 with a total sliding distance of 200 m. The track radius was 7 mm with a rotation speed of 300 rpm. The pin was a ceramic ball with a 10 mm diameter to avoid any softening or deformation that may occur when using steel balls. The test samples were machined with a 25 mm diameter and 6 mm thickness to be the disk. For each condition, the average weight loss (3 times) was measured. The samples were examined by an optical microscope before and after the wear test to assess any change in microstructure, especially at 750 °C. The worn surface was also photographed to observe the wear track.

Figure 2
figure 2

Schematic illustration of the ball-on-disk high-temperature wear test.

Full size image

Results and Discussion

Thermodynamic Calculations

Thermo-Calc phase diagrams of the examined SiMo ductile irons are shown in Figure 3, demonstrating just how their structure is formed during cooling and solidification at ambient temperature. In SiMo alloys, carbides of the M6C type were formed during eutectic solidification, and these carbides remain in the alloy until it reaches room temperature. Moreover, the system undergoes eutectoid transformation, leading to the coexistence of ferrite and austenite within the temperature ranges of the transformation points (at which ferrite transforms into austenite) Ac1–Ac3 across the entire carbon content range. According to Thermo-Calc calculations, S1-0% Mo exhibited the highest eutectoid temperatures as austenite formation started at (Ac1) 906 °C and ended at (Ac3) 988.7 °C. The eutectoid temperatures of SiMo samples decrease approximately 7–12 °C compared with Mo-free one until reaching the lowest values at S4-1.5% Mo, as shown in Figure 4. Based on the calculations, it can be deduced that Mo decreases the eutectoid temperatures at the equilibrium system.

Figure 3
figure 3

Thermo-Calc phase diagram of examined samples: (a) 0%Mo, (b) 0.5%Mo, (c) 1%Mo, and (d) 1.5%Mo.

Full size image
Figure 4
figure 4

Volume fraction of formed phases as a function of Mo%, (a) 0%Mo, (b) 0.5%Mo, (c) 1%Mo, and (d) 1.5%Mo.

Full size image

Principally, Thermo-Calc analysis assesses equilibrium conditions and a volume fraction of carbide-forming elements. The volume fraction at equilibrium conditions in terms of temperature is illustrated in Figure 4. The microstructure mainly consists of graphite, ferrite, and carbide phases with different amounts according to chemical composition. As shown in Figure 4, the volume fraction of ferrite and graphite decreases with the increase in Mo content in contrast to the carbide phases, which increase. Indeed, the actual microstructure of SiMo ductile iron is composed of graphite nodules, a ferritic metallic matrix, and a trace of pearlite and Fe–Mo carbides in the intercellular regions, as agreed with previous studies.14,23

Dilatometry Studies

The influence of Mo contents on the critical transformation temperatures was evaluated by means of the dilatometric experiments (as described in Section "Dilatometry Studies"). The four SiMo ductile iron alloys under investigation were heated slowly to 1000 °C with a heating rate of 0.1 K/s. The samples were maintained at the target temperature for 10 min before being slowly cooled at 0.1 K/s. When the sample was heated to the austenite zone, the curve showed a deviation in linearity at Ac1, indicating the beginning of the eutectoid transformation (austenite formation). The curve returns linear at Ac3, indicative of the end of the transformation. During cooling, the austenite starts at Ar3 to decompose into ferrite and pearlite, and this decomposition leads to a nonlinearity. The austenite decomposition occurs slowly because of the ferrite formation. After that, the dilatometric curve restores to linearity again at Ar1 as the end of ferrite formation. The start and end of eutectoid transformation temperatures, calculated by the deviation from linearity using the tangent method, describe the austenite's eutectoid transformation. These temperature ranges (during heating Ac1–Ac3 and Ar1–Ar3 during cooling) achieve the inter-critical region at which the ferrite, austenite, and graphite are present together. Figure 5 demonstrates an enlarged part of the dilatation curve of the tested SiMo samples. The eutectoid transformation temperatures decrease in trend with the increase in Mo contents; this agrees with the Thermo-Calc calculations. The temperatures of the eutectoid transformation during cooling (Ar1–Ar3) are lower than those during heating (Ac1–Ac3), as shown in Figure 5. The sluggish kinetics of austenite decomposition during cooling may be related to the noticeably lower Ar1 and Ar3 temperatures than the Ac1 and Ac3 temperatures.

Figure 5
figure 5

Dilatation temperature curves of all the samples were heated to 1000 °C at 0.1 K/s, held at that temperature for 10 min, then cooled with 0.1(K/s).

Full size image

Table 2 lists the results illustrating the critical temperatures of eutectoid transformation by Thermo-Calc and dilatometer methods. In light of the results in Table 2, the prediction of the influence of Mo addition in decreasing Ac1 and Ac3 by Thermo-Calc is verified experimentally by the dilatometric measurements. Moreover, the dilatometric method of continuous heating provides a more accurate calculation of the range of these transformation temperatures than the Thermo-Calc method.

Table 2 Results of the Eutectoid Transformation Temperatures Determined by Thermo-Calc Simulation and Dilatometry Measurements
Full size table

Metallographic Analysis

Mo content ranges from 0 to 1.5%, according to Table 1. The microstructure of the Mo-free sample exhibits nodular graphite embedded in a fully ferritic matrix, as shown in Figure 6a. The microstructures of the SiMo alloy samples are similar, with carbides (Mo-rich carbides) surrounded by varying amounts of carbides at the grain boundaries and graphite nodules dispersed in the ferrite matrix (Figure 6b–d). The nodule count decreases inversely with the increase in Mo content according to qualitative measurements, as shown in Figure 7. The cell-boundary precipitates increased with increasing Mo content, indicating segregation-related processes. Consequently, adding Mo increases carbide contents while decreasing the ferrite percentage in the matrix, which is compatible with the Thermo-Calc results.

Figure 6
figure 6

Microstructure of examined samples: (a) 0% Mo, (b) 0.5% Mo (c) 1% Mo, (d) 1.5% Mo.

Full size image
Figure 7
figure 7

Nodule count versus Mo contents.

Full size image

Nature of Eutectic Carbides

Thermo-Calc calculations and numerous earlier studies proposed that the eutectic carbides were M6C-type.15,24,25,26 In this current work, the EDS analysis also confirms that the precipitated carbides are M6C (M: Mo, Si, and Fe) type. Figure 8 shows SEM micrographs of such carbides, which have morphologies similar to fishbone or Chinese script and are relatively coarse in size. In both morphologies, the Mo content in the eutectic carbides ranges widely from 26.25 to 57.8%, with varying levels of Si, indicating that these carbides contain a broad range of Fe, Mo, and Si.

Figure 8
figure 8

SEM-EDS analysis of the M6C carbides rich Mo and Fe in SiMo ductile iron: (a) 0.5%Mo, (b) 1%Mo, and (c) 1.5%Mo.

Full size image

Nature of Fine Precipitates

Typically, eutectic carbides are embedded in the fine precipitates at the cell boundaries. The mechanism of these fine precipitates' formation is still up for discussion. The study of Choe et al.25 proposed that these precipitates appear to be complex carbides of Fe and Mo. Due to the different solubilities of Mo, Fe, and C atoms, they were segregated from the austenite into residual liquid in the graphite-austenite eutectic reaction. In the final stage of the reaction, the complex carbides are formed at the austenite cell boundary. Youssef et al.'s study26 suggested that such fine precipitates may be due to (i) austenite decomposition during cooling as one of the eutectoid transformation phases and (ii) the decomposition of pearlite after its reaction. The present study confirmed that the lamellar precipitates are pearlite and the fine spherical phase is a complex carbide of Fe and Mo, as confirmed by SEM and the EDX analysis for these phases (Figure 9).

Figure 9
figure 9

SEM-EDS analysis of the fine precipitates in SiMo ductile iron: (a) spheroidal carbides, (b) lamellar pearlite.

Full size image

As mentioned above, the lamellar and fine spherical phases were found next to the carbides. During cooling, pearlite appears to form in the lamellar phase; and carbide precipitates in the spherical phase. Unlike most pearlite, which has a typical lamellar structure as shown in (Figure 10a), the pearlite formed in the S2-0.5%Mo sample has an irregular lamellar structure (similar to amorphous glass) at the grain boundaries, as revealed by SEM (Figure 10b).

Figure 10
figure 10

SEM photomicrographs of precipitated pearlite in SiMo ductile iron: (a) typical lamellar structure, (b) irregular lamellar structure.

Full size image

Hardness Measurements

The results of the hardness tests performed on the SiMo alloys with different microstructures are shown in Figure 11, which illustrates how Mo additions affected hardness. Vickers hardness increased as molybdenum content increased. The lowest hardness value is 275 Hv in S1-0 % Mo; then, the hardness value increased in trend to 283, 286, and 295 Hv with the Mo contents (0.5 %, 1 %, and 1.5 %) in SiMo samples, respectively. The primary reason for the increase in hardness value with increasing Mo content is Mo-carbides' micro-segregation around the cell boundaries.14,16,19

Figure 11
figure 11

Vickers hardness values versus Mo contents.

Full size image

High-Temperature Abrasive Wear Behavior

Numerous variables influence the wear process, but understanding their effect on friction and wear aids in controlling and evaluating wear behavior. Materials (i.e., chemical composition, properties, and microstructure) and operational, geometric, and environmental factors are considered crucial variables affecting wear.27 In the current research, abrasive wear behavior was studied in view of microstructure and operational conditions.

Figure 12 displays the average values obtained from the wear test, explaining the effect of Mo on the average wear weight loss of SiMo samples at RT, 250, 500, and 750 °C. The average wear weight loss of the Mo-free sample is the highest. After adding Mo, the average wear weight loss of the samples decreased and continued to decrease with increasing Mo content.

Figure 12
figure 12

The weight loss versus Mo contents at different temperatures.

Full size image

Wear Behavior at Room Temperature The RT wear resistance followed the trend of hardness, where increased Mo content decreases weight loss gradually, in a range of 40–70% compared with free-Mo alloy. Following is an explanation for that result: Mo effects in the formation of carbides, which can increase hardness and wear resistance; a reduction in the amount of nodular graphite can lessen the formation of voids brought on by graphite shedding. Such results are confirmed by Gahr,28 who summarized the relative wear resistance of various microstructures of the cast iron matrix as a function of hardness. Carbides exhibit a significantly high wear resistance. Consequently, the sample with 1.5%Mo has the highest wear resistance. According to Figure 13, which shows the track of some specimens at various test temperatures, the wear surface morphology of the S1-0% Mo has deeper furrows and more voids because it contains more nodular graphite than the rest of the samples. Additionally, this sample has lower hardness due to the lower amount of carbides than samples that do contain Mo. Graphite can act as a lubricant, but the produced wear particles during the wear process will erode the lubrication film.29

Figure 13
figure 13

Wear track of some samples at various test temperatures.

Full size image

Wear Behavior at High Temperatures The wear tests were performed at different temperatures under air-atmosphere conditions. The air atmosphere causes oxidation of the samples at high temperatures during the test. This type of wear is typically known as “oxidative wear”.30 The wear resistance of SiMo samples decreases with increasing temperature remarkably at 250 °C. The average wear weight loss at 250 °C and 500 °C increased by 40–65% more than at RT, as shown in Figure 12. The high-temperature wear resistance has the same trend as that at RT, where Mo increases the wear resistance. This result agreed with,31 as the abrasive wear resistance of as-cast ductile iron decreased significantly from 100 to 600 °C. However, the graphite nodule behavior may not be seen at high temperatures because graphite is no longer effective as a lubricant at temperatures higher than 100 °C.32 As shown in Figure 12, the samples with Mo concentration from 0.5 to 1.5 % exhibited a decrease in the depth and width of furrows and the number of voids, owing to an increase in carbides and a decrease in nodular graphite.

Unexpectedly, it was observed that the weight loss of all SiMo samples lessened at 500 °C and 750 °C compared to 250 °C. This increase in wear resistance, which was more significant at 750 °C, may be oxidative wear. The formation of an oxidation layer during the wear test is confirmed by the SEM and EDX analysis for the worn surface at those high temperatures, as illustrated in Figure 14. The wear track revealed oxides on the sample surfaces that may have been produced due to the high temperature causing oxidation; then, the oxide debris partially fell throughout the test. Given that almost all of the samples were covered in oxide, weight loss values were close, and both abrasive and oxidative wear mechanisms played a role in the wear. Therefore, forming an oxide layer can be the reason for the reduction in weight loss. Ekström33 determined the composition and phase fraction in the transition layer in SiMo ductile irons by Thermo-Calc calculations. The high silicon content contributes to forming a SiO2 barrier layer at the oxide/metal interface. This SiO2 layer, denser than the layer of ferrous oxides, served as the main obstruction to oxygen diffusion.34 The oxide layer acts as a protective film on the worn surface, obstructing metal-metal (i.e., test sample–ball) contact.

Figure 14
figure 14

SEM and EDX analysis for the worn surfaces: (a) S4-1.5% at 750°C, (b) S3-1% Mo at 500 °C.

Full size image

Investigation of Structural Stability After Wear Test The microstructure of the samples was evaluated after the wear test to detect any changes that might affect the wear behavior. All alloys generally showed structural stability of phases according to quantitative measurements of phases. However, the ferrite slightly increased due to pearlite transformation into ferrite (not exceeding 2%) at 750 °C, as illustrated in Figure 15. This was in agreement with a previous study35; since the molybdenum carbides are embedded in lamellar pearlite in the ferrite matrix, SiMo has longer reliability and a more stable microstructure. Furthermore, these molybdenum carbide precipitates have a low decomposition rate because they are stable at high temperatures, even at 800 °C.35

Figure 15
figure 15

Microstructures of examined samples: (a) 0% Mo, (b) 0.5% Mo, (c)1% Mo, and (d) 1.5% Mo after the wear test at 750 °C.

Full size image

Friction Behavior Figure 16 shows the friction coefficient versus sliding at constant load and sliding speed but different temperatures for the four SiMo alloys. The average values of the friction coefficient differ according to the different surrounding temperatures during the test. For all samples, the temperatures of 250 and 500 °C show a higher coefficient of about 0.3 than at RT. The friction behavior of the samples at RT is relatively stable. At 750 °C, samples give the lowest friction coefficient level of about 0.19~0.35. Furthermore, the friction behavior reveals instability at high temperatures up to 750 °C. In the current work, the average friction coefficient takes the same behavior as the average weight loss. The more weight loss, the higher the friction coefficient.

Figure 16
figure 16

Friction curves of all the tested samples at various test temperatures.

Full size image

Worn Surface

Figures 17 and 18 represent a closer magnification of optical micrographs for the worn surfaces of SiMo ductile iron samples. The typical characteristics of micro-cutting and plastic deformation mechanisms are apparent (see Figures 17 and 18). The material was directly worn away by the micro-cutting effect of the abrasive grains, and plastic deformation was caused by the plowing action of the abrasive particles in lower wear resistance samples. The damage process may be related to the plastic deformation and instability of the nodular graphite and the surrounding ductile iron matrix (ferrite), leading to crack initiation at the matrix-graphite nodule interfaces and the matrix itself. After the graphite particles were worn away, the dark-colored pits formed. The spalling of graphite nodules from surfaces may be due to the voids in the matrix-graphite interface; graphite nodules have a weak bonding strength, which agrees with Yıldızlı et al. study.36 Additionally, a ferrite layer surrounding the graphite nodule may be harmful. The wear rates increase even as the graphite-matrix interface area increases since such areas are preferred sites where subsurface cracks are most likely to begin and propagate.

Figure 17
figure 17

Photomicrographs of the worn surface of some tested samples at different test temperatures.

Full size image
Figure 18
figure 18

SEM for the worn surface of S4-1.5% at 750 °C.

Full size image

As discussed in the wear results, the possibility of oxidation during friction increases at high temperatures, and a thin oxide layer is formed. This thin layer acts as a lubricant between the test specimen and the counterpart. On the other hand, the oxide layer thickened and started to spall, and partially or fully uncovered the test sample (see Figures 17 and 18), forming oxide-particle-containing wear debris. The wear debris accumulates as a layer after being wedged between the sliding surfaces. The generated oxide layer may become thicker due to the continued removal, preventing the erosion of new layers.

Conclusions

  1. 1.

    Thermo-Calc software was used to calculate the phase diagrams of SiMo ductile irons with 5% Si and various Mo contents ranging from (0%:1.5%). The solidification and solid-state phase transformation sequence was proposed using the data extracted from the diagrams.

  2. 2.

    The dilatometric measurements experimentally verified the influence of Mo addition in reducing the eutectoid transformation temperatures (Ar1–Ar3) that were predicted by Thermo-Calc calculations.

  3. 3.

    The SiMo ductile cast iron microstructure consists of a ferrite matrix with some carbides embedded in the fine precipitates in the intercellular regions. These carbides are suggested to be eutectic carbides of M6C type according to Thermo-Calc calculations and EDX analysis.

  4. 4.

    Increased amounts of the micro-segregation of Mo-carbides at the intercellular regions resulted in a slight increase in hardness from 275, 283, 286, and 295 Hv with increased Mo- contents from 0.0.5 to 1.1.5%, respectively.

  5. 5.

    Wear resistance increases with increased Mo contents due to the formation of higher amounts of intercellular carbides which increase the hardness.

  6. 6.

    Wear resistance decreases with increased temperature up to 500 °C compared to RT. However, the wear resistance unexpectedly increased at higher temperatures up to 750 and this may be related to the metal-metal (ball and sample) contact due to the formed oxidation layer which was verified by EDX.

  7. 7.

    The worn surfaces showed many factors like plastic deformation and nodular graphite instability, and the surrounding SiMo ductile iron matrix (ferrite) contributed to crack initiation in the matrix-graphite nodule contacts and the matrix itself.