1 Introduction

Aluminum alloys are one of the most widely used materials after steel in automotive manufacturing, aerospace, and other fields due to their low density, high specific strength, and good thermal and electrical conductivity [1,2,3]. It is well known that even with all of these advantages, binary aluminum alloys are not suitable for all applications. Therefore, various methods are applied to increase the strength and ductility properties of aluminum alloys used in industrial applications. The most important of these methods is element addition to the primary aluminum phase [4,5,6,7]. Aluminum's ability to form compounds with other chemical elements has led to the development of various alloys (including Al–Cu, Al–Si–Mg, Al–Mg–Zn, Al–Mg, and Al–Si) [8,9,10]. Aluminum–silicon alloys are currently the most widely used of these alloy types. Because of their exceptional qualities, including high specific strength, low thermal expansion coefficient, high fluidity, and high resistance to corrosion and wear, Al–Si alloys are extremely important in the automotive industry [5,6,7,8,9,10]. These alloys contain elements like copper and magnesium, which dissolve in the α-Al phase and form precipitates [10,11,12,13]. The θ (CuAl2) phase consisting of Cu and Al and the β-Mg2Si phase consisting of Mg and Si are the main phases observed from these alloys. Addition of more than 1% of Mg results in a volume fraction of the β-Mg2Si phase [10,11,12,13]. The hard and brittle β-Mg2Si phase in the microstructure reduces the tensile strength of the alloy and causes crack tendency. For this reason, the magnesium content of aluminum–silicon alloys is kept below 1% [10,11,12,13]. It has been determined that the microstructure of the Al–10Si–Mg alloy in as-cast state consists of α-Al phase, needle silicon particles, and skeletal Mg2Si phase [5,6,7,8, 11,12,13,14,15,16]. After applying an air-cooling heat treatment to the Al–10Si–Mg alloy, it was observed that the Mg2Si skeletal phase was largely eliminated. Subsequently, the quenching and aging heat treatments caused the needle silicon particles to become spherical. In industrial applications, many alloying elements are added to aluminum–silicon alloys [14,15,16,17]. These additive elements can be listed as copper, magnesium, and zinc, respectively. 1–4 (wt%) copper additive forms the θ (CuAl2) phase in the α-Al phase and increases the strength and hardness of the alloy [18,19,20,21,22,23]. In addition, zinc improves the strength properties of the alloy as it dissolves in the α-Al phase. The addition of magnesium to this alloy causes the formation of the Mg2Si phase in the matrix, which increases the alloy's wear resistance [18,19,20,21,22,23]. Studies have shown that adding up to 1% Mg to the Al–Si alloy results in the creation of the β-Mg2Si phase, alters the morphology of silicon particles, and converts the plate-like AlFeSi phase into a more ordered π-Al8Mg3FeSi6 phase. It has also been reported that these changes increase the alloy's strength [20,21,22]. In another study, it was determined that the addition of more than 1.5% Mg increased the ratio of β-Mg2Si and π phases in the microstructure [15, 16, 24, 25]. These phases reduce the load-bearing ability of the alloy and form internal cracks, negatively affecting the tensile properties [15, 16, 24, 25].

There are numerous studies in the literature on the production of Al–10Si–Mg alloys. However, in these studies, the effect of process conditions on the microstructure and mechanical properties of the alloys was investigated [24,25,26,27]. Based on the literature studies, it was concluded that applying different heat treatments to the Al–10Si–Mg alloy did not fully reveal its friction and wear properties. Therefore, the aim of this study is to investigate the effect of heat treatment on the microstructure, strength, wear behavior, and oxidation resistance of Al–10Si–Mg alloy and to determine optimum data for tribological applications.

2 Experimental

The initial Al10SiMg alloy used in this study was fabricated by using gravity die casting method. The chemical composition of the alloy was detected by the spectral analysis method (Table 1).

Table 1 Chemical composition of Al–10Si–Mg alloy
Full size table

For microstructural investigations, alloy samples were subjected to sanding, polishing, and etching processes. Etching was performed by electro-etching with fluoroboric acid using the Struers Lectropol device. Al10SiMg alloy was melted at 700 °C using an electrical resistance furnace. The melted alloy was mixed for 5 min and then poured into a mild steel mold at 690 °C. The dimensions of the casting mold are 180 mm long and conical, with a lower area of 52 mm and an upper area of 72 mm. The technical drawing of the casting mold is shown in Fig. 1. The spectral analysis method was used to ascertain the alloy's chemical composition as soon as solidification was finished. To determine the alloy's properties, it was subjected to 2 hours of solution heat treatment at 560 °C, followed by 5 hours of quenching and aging at 180 °C [28, 29]. The microstructures of the alloy were investigated using a Zeiss Axio Observer A1m optical microscope. Density was determined by dividing the mass of the alloy by the sample volume. Microhardness values of the alloy were measured with the STRUERS DURAMIN A/S DK-2750 device, and Brinell hardness values were determined with the Innovatest Nemesis 9000 Universal Hardness Tester. Average values were obtained by taking at least five measurements from each sample in the microhardness and hardness tests. The yield and tensile strengths of the alloy were measured by making at least five measurements at a deformation rate of 5 × 10–4 s−1 using the MTS Criterion Model 45 universal testing device in accordance with ASTM:E8 standards.

Fig. 1
figure 1

Shape and dimensions of the casting mold

Full size image

The friction and wear properties of the alloy were investigated using a UTS Design Tribology wear device, based on ball-on-disk type and meeting ASTM: G99 standards. The schematic image of the friction and wear test device is given in Fig. 2. 100Cr6 steel balls were used to abrade the samples. In friction and wear tests, the test speed and load were 0.16 m/s and 5 N, respectively, and the sliding distance was between 250 and 1500 m. Each sample that was tested for friction and wear was cleaned in an ultrasonic cleaning device using a mixture of alcohol + trichlorethylene. The mass loss of the worn samples was measured with a balance with a sensitivity of ± 0.01 mg. The volume loss was detected by dividing the measured mass losses by the density of the alloy. SEM images and EDS analysis of the sample surfaces were investigated using the Zeiss Evo LS10 device. Thermogravimetric analysis (TGA) is used to determine the oxidation behavior of samples. PerkinElmer TGA4000 was used for TGA analysis. The experimental procedure of the TGA was specified as 30–650 °C temperature, 20 °C/min heating rate and air.

Fig. 2
figure 2

Schematic of the wear test device

Full size image

3 Results and Discussion

3.1 Microstructure

Figure 3 shows the microstructure of the as-cast Al–10Si–Mg alloy comprised of α-Al and Mg2Si phases and Si fragments. The Si particles in the alloy's microstructure were found to be needle-like and block-shaped appearance (Fig. 3a), whereas the Mg2Si phase appeared to be skeletal, as shown in Fig. 3b. It was shown that the Si particles in the microstructure of the alloy exhibited a needle-like and block-shaped appearance, while the Mg2Si phase exhibited a skeletal appearance. It is known that needle-shaped Si particles in the alloy are formed as a result of eutectic transformation, and block-shaped Si particles are formed directly spontaneously during solidification [30,31,32]. When the solidification behavior of Al–10Si–Mg alloys in the casting state is examined, the first α(Al) nucleus is formed when the molten alloy begins to cool. As the temperature decreases, the α(Al) phase begins to grow. As the chemical composition of the Al–10Si–Mg alloy approaches, the eutectic point, silicon, and magnesium react to form β-Mg2Si, and solidification is completed [26, 27]. The formation of the β-Mg2Si phase reduces silicon concentration in molten solution. This prevents the growth of silicon particles and causes the formation of finer silicon particles [23,24,25,26,27]. There is no important change in the microstructure of the air-cooled Al–10Si–Mg alloy samples compared to the as-cast condition (Fig. 3b). However, it was determined that there was a relative increase in the α(Al) phase ratio in the internal structure of the air-cooled sample. It is thought that this situation may be related to the continuous withdrawal of the Mg2Si phase from the liquid metal during solidification.

Fig. 3
figure 3

Microstructure of Al–10Si–Mg alloy: a as-cast, b air-cooled, c quenched, and d aged treatment state

Full size image

The microstructure of the Al–10Si–Mg alloy in the quenched and aged state shows that the silicon particles become relatively spherical and the β-Mg2Si phase disappears (Fig. 3c, d). However, it was observed that there were more primary silicon particles in the quenched alloy and these particles tended to cluster. The reason why the primary silicon particles in the quenched state of the Al–10Si–Mg alloy is higher than in the cast state of the alloy may be due to the growth of needle-shaped eutectic silicon particles during the dissolution and quenching process. The disappearance of the Mg2Si phase in the structure can be explained by the high diffusion rate of magnesium in the α(Al) phase [26, 27]. During aging, Mg and Si atoms are withdrawn from solution, forming unstable fine Mg2Si precipitates [23,24,25,26,27]. On the other hand, it has been stated in other studies that silicon atoms display a faster diffusion rate than Al atoms [14,15,16,17,18,19, 26, 27]. This causes the needle-shaped silicon particles to break down and turn into a relatively spherical form and be distributed homogeneously throughout the structure.

3.2 Mechanical Properties

The yield and tensile strength, hardness, and microhardness curves of the as-cast, air-cooled, quenched, and aged Al–10Si–Mg alloy samples are given in Fig. 4, respectively. Moreover, Table 2 shows numerical values of mechanical properties of the Al–10Si–Mg alloy samples. It is seen that the hardness and microhardness values of the alloy exhibit similar changes with yield and tensile. Additionally, as shown in Fig. 4, samples that were allowed to cool in air demonstrated a slight decrease for all values, whereas samples that were quenched and aged demonstrated a continuous increase. These changes can be explained by examining the microstructures of the Al–10Si–Mg alloy in different heat treatments [14,15,16,17,18,19]. In contrast to the cast state, the amount of primary Si in the Al–10Si–Mg alloy decreases when it is air-cooled. Considering the load-carrying effect of primary silicon particles in the microstructure, it is expected that there will be some decrease in hardness and strength as this effect decreases. In the quenched and aged states of the alloy, the relatively spherical homogeneous distribution of silicon particles within the microstructure increases the hardness and strength of the alloy samples. In addition, the disappearance of the Mg2Si phase, which has a skeletal phase appearance in the microstructure, reduces the crack tendency and increases the strength values [23,24,25,26,27]. In addition, it was observed that after quenching, the primary silicon content decreased and there was a relative increase in the amount of α(Al) in the microstructure (Fig. 3c). The presence of primary silicon in the alloy caused a partial increase in the yield strength (Fig. 4). The aging process of the sample resulted in a partial improvement in microhardness due to the growth of the α(Al) phase in its internal structure, but it also significantly increased the hardness values (Fig. 4b).

Fig. 4
figure 4

The change of the tensile, yield strength, hardness, and microhardness of Al–10Si–Mg alloy samples

Full size image
Table 2 Yield, tensile strengths, hardness, and microhardness values of Al–10Si–Mg alloy samples
Full size table

Fracture surfaces of alloys, which subjected to various heat treatments, are shown in Fig. 5. It was observed that the fracture surfaces of the alloy comprise of a small amount of cleavage planes (CP) and mostly tear ridges (TR). It should be noted that the CP consists of hard phases with silicon and magnesium, while the tear ridges have soft α(Al) phases. Similar fracture surface characteristics were determined in all alloys subjected to different heat treatments. However, the transformation of silicon particles in the microstructure of quenched and aged samples into a relatively spherical shape causes more cleavage planes on the fracture surfaces (Fig. 5c, d).

Fig. 5
figure 5

SEM photographs of fracture surfaces of the alloys with different heat treatments: a as-cast, b air-cooled, c quenched, and d aged treatment state

Full size image

3.3 Wear Behavior

Figure 6 shows the variation in the Al–10Si–Mg alloy's friction coefficient values depending on the sliding distance in both the as-cast and heat-treated states. The friction coefficient of the alloy initially increased sharply for each sample and reached a relatively equilibrium state. This situation can be explained by the tribological behavior of the materials [33,34,35]. At the beginning of friction and wear experiments, the contact surface between the ball and the sample is point-like gave rise to high pressures. The adhesion bonds formed between the contact surfaces create resistance to sliding, causing the friction coefficient to increase suddenly [33,34,35]. As friction continues on the sample surfaces, the contact area increases and reaches equilibrium after a certain period of time. However, particles broken off from the material surface during friction enter between the ball and the sample surface. This causes a variable coefficient of friction.

Fig. 6
figure 6

The change of friction coefficient of as-cast, air-cooled, quenched, and aged state Al–10Si–Mg alloy samples

Full size image

Figure 7 shows the volume loss values for the Al–10Si–Mg alloy in its cast and heat-treated states as a function of sliding distance. From this Fig. 7, it can be noted that the volume loss increases linearly with increasing sliding distance for each case. It was determined that the alloy that was subjected to aging heat treatment, quenched, cast, and cooled in air exhibited the highest wear resistance, respectively. This observation can be explained depending on the microstructure of the alloy, its hardness, and Archard's equation [31,32,33, 36,37,38,39]. As mentioned above, when heat treatment is applied to the Al–10Si–Mg casting alloy, silicon particles turn into a relatively spherical shape and spread homogeneously into the microstructure. This increases the hardness in the microstructure of the alloy and improves the wear resistance due to the load-carrying ability of hard and brittle silicon particles. Moreover, according to Archard's equation, volume loss increases as sliding distance increases and decreases as hardness and strength increase [36,37,38,39]. Since the hardness and strength increase when heat treatments are applied to the alloy, volume loss is expected to increase as the sliding distance increases [36,37,38,39].

Fig. 7
figure 7

The change of volume loss of as-cast, air-cooled, quenched, and aged treatment state Al–10Si–Mg alloy samples by sliding distance

Full size image

The change in volume loss value of Al–10Si–Mg alloy in both cast- and heat-treated states depending on pressure is given in Fig. 8. This alloy samples exhibited a gradually increasing volume loss characteristic in all cases, and it was determined that the lowest volume loss was the sample with aging heat treatment. This observation can be explained based on hardness and strength. Hard and brittle silicon particles dispersed homogeneously and relatively spherically within the microstructure have a load-bearing effect, causing an increase in wear resistance [40,41,42,43]. In addition, the needle-like and block-shaped silicon particles in the microstructure of the as-cast and air-cooled samples may have created a notch effect and caused the particles to break off due to the increasing pressure from the sample surface. This situation negatively affects the wear resistance of cast and air-cooled samples compared to quenched and aged samples.

Fig. 8
figure 8

The variation of volume loss of as-cast, air-cooled, quenched, and aged state Al–10Si–Mg alloy samples by pressure

Full size image

SEM images showing the wear surfaces of the as-cast and heat-treated states of the Al–10Si-SEM images showing the wear surfaces of the as-cast and heat-treated states of the Al–10Si–Mg alloy, the particles morphologies obtained from the wear tests, and the ball surface used in the wear tests in the as-cast state are given in Figs. 9, 10, and 11, respectively. Since the wear surfaces, debris pictures and ball surface pictures in different heat-treated states of the alloy are similar, representative as-cast alloy pictures are given. As a result of friction and wear tests, it is observed that the wear surfaces of the sample consist of scratches and delamination layers (Fig. 9). It can be seen that the ball surface is relatively covered with wear material, and the wear particles comprise of large and small particles. The EDS analysis results of the materials adhering to the sample and ball surfaces and the debris are given as a table within the SEM images. According to this table, it was determined that the chemical composition consists of the elements aluminum, silicon, magnesium, iron, chrome, and oxygen. These examinations can be associated with the microstructures of the Al–10Si–Mg alloy in its as-cast and heat-treated states. The microstructure of Al–10Si–Mg alloy consists of soft α(Al) matrix, hard silicon particles, and Mg2Si phase [26, 27, 30,31,32]. The hard and brittle Si particles are easier to smear onto the sample surface thanks to their superior load-carrying abilities. On the contrary, since the α(Al) phase is soft, it is more smeared to the ball surface. In friction wear tests, initially point contact and high-pressure cause particles to break off from the sample surface. Some of these particles adhere to the sample surface and some to the ball surface, causing an adhesion layer to form on the surface. This adhesion layer becomes brittle as oxidation increases. Thus, delamination layers are formed. In addition, hard and brittle particles within the microstructure of the alloy cause cracks on the wear surfaces. These cracks progress under the influence of oxidation and break off from the sample surface. Some of the broken pieces move away from the sample surface under the effect of centrifugal force, while the other part turns into dust between the ball and the sample surface.

Fig. 9
figure 9

SEM images with EDS analysis of worn surface of the alloy: a as-cast, b air-cooled, c quenched, and d aged treatment

Full size image
Fig. 10
figure 10

SEM images with EDS analysis of debris of the alloy: a as-cast, b air-cooled, c quenched, and d aged treatment

Full size image
Fig. 11
figure 11

SEM images and EDS analysis of ball surface after wear test of 1500 m for just as-cast Al–10SiMg alloy

Full size image

3.4 Oxidation

Figure 12 shows the results of thermogravimetric analysis of Al10SiMg alloys subjected to casting, air cooling, quenching, and aged treatment processes, respectively. As can be seen in Fig. 12, no significant mass increase was observed in the TGA experiments of all test specimens up to 650 °C. As a result of the oxidation experiments, the highest mass increase was 0.12%. This can be attributed to the Al2O3 layer formed on the Al surface with increasing temperature. Oxide layers formed on the metal surfaces can be evaluated as protective and non-protective oxide layers. The most concrete expression related to this observation can be explained with the Pilling–Bedworth ratio. The Pilling–Bedworth ratio is a rule used to determine the oxide films formed by metals as protective or non-protective and was presented by N.Pilling and E.Bedworth. The formula for the Pilling–Bedworth ratio is given in Eq. 1 [44, 45].

$${\text{Pilling}} {-} {\text{Bedworth Ratio}} = \frac{{M_{{{\text{oxide}}}} \times d_{{{\text{metal}}}} }}{{n \times M_{{{\text{metal}}}} \times d_{{{\text{oxide}}}} }}$$
(1)

where M is the molecular mass, n is the metal atom per one oxide molecule, d is the density.

Fig. 12
figure 12

TGA curves of the samples of Al–10Si–Mg alloy samples

Full size image

Calculating the Pilling–Bedworth ratio for the oxidation of aluminum;

$$2{\text{Al}} + \frac{3}{2}{\text{O}}_{2} = {\text{Al}}_{2} {\text{O}}_{3}$$
(2)

It is known that the molar mass of aluminum is about 26.98 g/mol and molar mass of Al2O3 is about 101.96 g/mol. Density of aluminum is 2.70 g/cm3, and density of Al2O3 is 3.95 g/cm3 [46]. Accordingly, the Pilling–Bedworth ratio for aluminum was calculated using Eq. 3:

$${\text{Pilling}} {-} {\text{Bedworth Ratio of Aluminum}} = \frac{{M_{{{\text{Al}}_{2} {\text{O}}_{3} }} \times d_{{{\text{Al}}}} }}{{n \times M_{{{\text{Al}}}} \times d_{{{\text{Al}}_{2} {\text{O}}_{3} }} }}$$
(3)
$$\begin{aligned} & {\text{Pilling}} {-} {\text{Bedworth Ratio of Aluminum}}\\ & = \frac{{101.96\;{\raise0.7ex\hbox{${\text{g}}$} \!\mathord{\left/ {\vphantom {{\text{g}} {{\text{mol}}}}}\right.\kern-0pt} \!\lower0.7ex\hbox{${{\text{mol}}}$}} \times 2.7\;{\raise0.7ex\hbox{${\text{g}}$} \!\mathord{\left/ {\vphantom {{\text{g}} {{\text{cm}}^{3} }}}\right.\kern-0pt} \!\lower0.7ex\hbox{${{\text{cm}}^{3} }$}}}}{{2 \times 26.98\;{\raise0.7ex\hbox{${\text{g}}$} \!\mathord{\left/ {\vphantom {{\text{g}} {{\text{mol}}}}}\right.\kern-0pt} \!\lower0.7ex\hbox{${{\text{mol}}}$}} \times 3.95\;{\raise0.7ex\hbox{${\text{g}}$} \!\mathord{\left/ {\vphantom {{\text{g}} {{\text{cm}}^{3} }}}\right.\kern-0pt} \!\lower0.7ex\hbox{${{\text{cm}}^{3} }$}}}} = 1.29\end{aligned}$$
(4)

If the Pilling–Bedworth ratio is less than 1, the oxide film formed by the metal is very thin and has no protection. If the Pilling–Bedworth ratio is greater than 2, the oxide film is extremely thick and easily disintegrates. However, if the Pilling–Bedworth ratio is greater than 1 and less than 2, the oxide film formed by the metal is protective. When the Pilling–Bedworth ratio is calculated for the oxide layer formed by aluminum metal, a value of approximately 1.29 is reached. This proves that aluminum forms a protective oxide layer [45]. According to the literature studies, it is understood that the oxidation mechanism for aluminum proceeds very slowly up to a temperature of approximately 550 °C and that there is a not very high oxidation between 550 and 600 °C. This is due to the protective oxide layer formed by aluminum, as explained by the Pilling–Bedworth ratio. In other words, aluminum undergoes oxidation with increasing temperature, but the progress of this oxidation is prevented by the amorphous oxide layer (Al2O3) formed on the aluminum surface. Thus, aluminum can remain resistant to oxidation almost up to its melting temperature [47, 48].

In the present study, the effects of the production of AlSiMg10 alloy by casting method and subsequent secondary treatments on oxidation resistance were investigated. In this context, it is understood that all test specimens have a very low oxidation rate, i.e., very high oxidation resistance, in accordance with the literature studies (Fig. 12). In addition, the relative effects of microstructural changes due to casting and subsequent secondary processing can also be understood from the TGA results. When the wt% change is examined up to a temperature value of approximately 400 °C, there is a small mass decrease for all samples. This is a similar situation to that encountered in the literature and in previous studies. Researchers have interpreted it as a situation encountered due to the removal of water or other residues from the structure. Park et al. observed a decrease in the wt% values of Al up to about 600 °C in their study, which they interpreted as the removal of moisture and impurities absorbed by the sample during the experiment by thermal experiment and thus a decrease in mass [49, 50].

The cross-sectional investigation of the Al–10Si–Mg (as-cast) and Al–10Si–Mg*** coded sample is shown in Fig. 13. Figure 13 shows that the oxidized surface contains nearly three times more oxide than the polished surface. The reason for the presence of oxide on the polished surface is that Al can easily form a protective oxide layer on its surface even at room temperature in an open atmosphere. In addition, when the EDS spectrum analysis of the surfaces and the wt% values of the elements are examined, it is clear that the amount of oxide is still lower than the amount of Al even on the oxidized surface in parallel with the TGA results. This confirms that the oxide layer formed by Al with O is a protective oxide layer.

Fig. 13
figure 13

Cross-sectional investigation and EDS analysis of the Al–10Si–Mg (as-cast) (a) and Al–10Si–Mg*** coded samples (b)

Full size image

4 Conclusions

1. The microstructure of the cast Al–10Si–Mg alloy consists of α(Al), eutectic silicon, primary silicon, and β-Mg2Si phases. Due to the aging heat treatment, the silicon particles in the Al–10Si–Mg alloy microstructure become comparatively spherical. Additionally, heat treatment leads to the removal of β-Mg2Si's Chinese script morphology.

2. Heat treatment led to an approximate 60% increase in hardness, a 70% increase in yield strength, and a 115% increase in tensile strength for the as-cast Al–10Si–Mg alloy.

3. The fracture surfaces of the cast and heat-treated alloy consist of cleavage planes and tear ridges. There are relatively more cleavage planes in the heat-treated state of the alloy. This reveals that the heat-treated alloy exhibits a more brittle behavior.

4. Adhesion, delamination, and cracks were observed on all wear surfaces of the Al–10Si–Mg alloy, both as-cast and heat-treated. It was determined that heat treatment increased the alloy's wear resistance.

5. In the friction and wear tests, it was observed that the variation of the friction coefficient with the sliding distance showed a sudden increase at the beginning of the test and reached a steady state.

6. It was determined that the volume loss values of the alloy increased linearly with the sliding distance and pressure values and the highest wear resistance was reached in the Al–10Si–Mg alloy with aging heat treatment.

5. No significant weight change was observed with increasing temperature for all alloy types until temperatures close to the melting temperature. This result is due to the high oxidation resistance of Al-based alloys. On the other hand, as the melting temperature approached, the oxidation resistance decreased due to the loss of effectiveness of the protective surface layer and the beginning of the transition of the material to the liquid phase.