An Exploration of SPS Fabrication and the Sliding Wear Properties of γ-TiAl-Ag Self-Lubrication Materials

Abstract

To promote the optimization of the anti-friction and anti-wear behavior of lightweight TiAl alloys, γ-TiAl-10 wt.% Ag self-lubricating composites were fabricated, and their mechanical and tribological properties were tested. The results showed that the silver in TiAl-10 wt.% Ag slightly reduced its mechanical properties compared with those of pure TiAl alloys. A silver-enriched lubrication film formed on a wear scar, which was helpful in improving the friction and wear behavior. It was found that a large amount of silver gathered at a wear scar, gradually spread out under the action of the sliding friction force, and then increased the silver distribution areas on the wear scar, leading to the good formation of a silver-rich film. Furthermore, an identification model was established to calculate the specific area η of the silver film. A quantitative relationship indicated that an increase in the Ag distribution area improved the tribological behavior of γ-TiAl-10 wt.% Ag. When the specific area η of a silver-rich film was maintained at 44–51%, the small friction coefficient (almost 0.28) and wear rate (about 2.25 × 10⁻⁴ mm³·N⁻¹·m⁻¹) were well stabilized. This provides a new research method to improve the tribological performance of TiAl-Ag samples.

1. Introduction

TiAl alloys possess a number of unique physical and mechanical properties, such as low density, large specific strength, high oxidation resistance, etc. [1,2,3]. They have been used to prepare air release valves, turbine blades, and seal stents. However, their poor friction and wear behavior poses an unexpected obstacle in increasing the service life and usage precision of TiAl-based components [4,5,6,7]. Especially in the dry sliding wear process, the friction coefficient and wear rate are relatively large when compared to those under other wear conditions, and they are not sufficient to satisfy the requirements of industrial lubrication.

This is due to TiAl alloys’ low shearing strength of 140–170 MPa, small yield strength of 20–25 MPa, and eminent ductility of 55% at 450 °C [8,9,10]. In the friction and wear process, a large amount of silver on a wear scar was gradually distributed to form a lubrication film, thus resulting in low friction and wear. Hence, silver can be chosen to optimize the tribological properties of TiAl alloys. Xu et al. [8] studied the migration behavior of the silver contained in TiAl-based materials. It was found that the silver content of the wear scar was continually improved with an increase in the applied load. This might be because the large loads caused the substrate to produce deformation, which led to a large amount of silver migrating to the wear scar from the TiAl substrate, thus increasing the silver content on the wear surface. This was beneficial for the formation of a lubrication film and facilitated a reduction in sliding resistance and material damage. Yang et al. [9] explored the friction and wear behavior of a TiAl/Ag composite and found that, during the process of dry wear sliding, the silver was squeezed out of the TiAl base material and enriched in the wear scar. This resulted in the formation of a silver-enriched film for lubrication, thus improving the tribological behavior. Subsequently, Shi et al. [10] reported the tribological behavior of a TiAl/Ag material sliding against Si₃N₄ balls. They found that its excellent friction and wear properties were mainly attributable to the plastic deformation of the silver-rich film. According to the aforementioned studies of Xu et al. [8], Yang et al. [9], and Shi et al. [10], a silver-rich film on the wear scar is beneficial for the ideal lowering of the friction coefficient and wear rate. Namely, the aforementioned discussions mainly focus on the silver-enriched film formation and plastic deformation, and they confirm the important influence of the silver lubrication film on the tribological behavior. However, during the friction and wear process, it is impossible to form a complete film that is entirely distributed on the wear surface for good lubrication. Hence, it is difficult to determine the distribution area or specific area of the lubrication film on the wear surface, which seriously hinders the quantitative analysis of the silver-rich lubrication film. To date, it has been difficult to determine the quantitative effects of the lubrication film area on the alloy’s tribological properties.

The rapid development of image recognition has provided a possible method for the identification of the silver distribution area on a wear scar. The analysis of the main component using K-means clustering, enabling the effective identification of similar primary elements [11,12], could enable the identification of the silver on a wear scar. Hence, the silver on a wear scar could be segmented to identify the specified distribution areas; then, the specific area η of silver distribution could be calculated. This is possible because the gray-color values of the silver and TiAl are observably different under EPMA backscattering conditions. This makes it possible to determine the area distribution of Ag on the friction surface. Consequently, the K-means method for main-component clustering has been introduced for the calculation of the distribution area of the silver on a wear scar, and the effects of silver areas on the tribological properties of the γ-TiAl base material can be comprehensively understood.

In this study, using an Instron 1341 tester and a nanomechanical instrument, the mechanical behaviors of TiAl-based materials were characterized. With reference to the ASTM No. G99-95 Standard [13], the MFT-5000 tribometer, with a high temperature and ball-on-disk setup, was used to measure the friction and wear properties of a γ-TiAl-Ag material. The silver distribution on a wear scar was examined under EPMA backscattering conditions. Subsequently, the K-means method for main-component clustering was used to calculate the silver distribution area of the wear scar. Finally, the main element content in the wear scar cross-section was tested by energy-dispersive spectroscopy (EDS). The cross-sectional morphology of the wear scar was observed with the assistance of field emission scanning electron microscopy (FESEM). Thus, this study provides an important reference toward improving the anti-friction and wear-resistant behavior of a TiAl alloy base component.

2. Material Preparation and Characterization

The γ-TiAl-10 wt.% Ag composites were prepared using an in situ technique via spark plasma sintering (SPS) using the D.R. Sinter^® SPS3.20 device. A starting powder with an atom ratio (at.%) of 48:47:2:2:1 was used to prepare the γ-TiAl matrix material. Starting powders with a mean size of 10–20 μm and more than 99.5 wt.% purity were used in this study. The fabrication process of the silver lubricant has been well described in the reported literature [14,15,16]. The magnified morphology of the silver, obtained using FESEM, is shown in Figure 1a. Before the SPS process, the raw powders were mixed by vibration milling at a 50 Hz frequency and then loaded in a graphite mold with a 25 mm inner diameter. After loading, a γ-TiAl-10 wt.% Ag composite could be fabricated for 10–15 min under an atmosphere of pure argon. The operational parameters chosen to prepare the TiAl base samples mainly included a heating rate of 95–110 °C/min heating rate, sintering temperature of 900–1150 °C, and applied pressure of 30–35 MPa.

With reference to the thermodynamic second laws described by Yue et al. [17], Xu et al. [18], and Yang et al. [19], the main synthetic reactions of γ-TiAl-based materials can be expressed using the following formulas:

$$\left\{\begin{array}{cc}\Delta {\mathrm{G}}^0>0\hfill & \hfill \mathrm{No}\mathrm{synthesis}\mathrm{reaction}\\ \Delta {\mathrm{G}}^0=0\hfill & \hfill \mathrm{Equilibrium}\mathrm{synthesis}\mathrm{reaction}\\ \Delta {\mathrm{G}}^0<0\hfill & \hfill \mathrm{Spontaneous}\mathrm{synthesis}\mathrm{reaction}\end{array}\right.$$

(1)

Ti + Al→TiAl

(2)

∆G₁⁰ = −152.91 + 44.44 + 42.72 = −65.75 < 0
Ti + 3Al→TiAl₃

(3)

∆G₂⁰ = −291.06 + (42.72 × 3) + 44.44 = −118.46 < 0
3Ti + Al→Ti₃Al

(4)

∆G₃⁰ =−266.5 + (44.44 × 3) + 42.72 =−90.46 < 0
TiAl + 2Ti→Ti₃Al

(5)

∆G₄⁰ = −266.5 + (44.44 × 2) + 152.91 = −24.71 < 0
TiAl + 2Al→TiAl₃

(6)

∆G₅⁰ = −291.06 + (42.72 × 2) + 152.91 = −52.71 < 0
TiAl₃ + 2Ti→3TiAl

(7)

∆G₆⁰ = −152.91 × 3 + (44.44 × 2) + 291.06 =−103.35 < 0
Ti₃Al + 2Al→3TiAl

(8)

∆G₇⁰ = −152.91 × 3 + (42.72 × 2) + 266.50 = −106.79 < 0
4Ti + TiAl₃→Ti₃Al + 2TiAl

(9)

∆G₈⁰ = −266.50 − (152.91 × 2) + (44.44 × 4) + 291.06 = −103.50 < 0
4Al + Ti₃Al→TiAl₃ + 2TiAl

(10)

∆G₉⁰ = −291.06 + (−152.91 × 2) + (42.72 × 4) + 266.50 = −159.50 < 0

(11)

In consideration of the abovementioned reactions, a γ-TiAl-based material can be continuously synthesized with a decrease in Ti, Al, Ti₃Al, and TiAl₃.

At the scanning speed of 0.01 s⁻¹, the main phase composition of γ-TiAl-10 wt.% Ag was examined using an X-ray diffractometer (XRD) with Cu Ka radiation at 30 kV and 40 mA. Figure 1b,c show the FESEM cross-sectional microstructure and XRD pattern of the γ-TiAl-10 wt.% Ag sample, respectively. As shown in this figure, the phase composition of γ-TiAl-10 wt.% Ag mainly consisted of γ-TiAl and silver, as indicated by their diffraction peaks of [1 1 1], [2 2 1] and [1 1 1], [2 0 0], [2 2 0].

Before the tests, γ-TiAl-10 wt.% Ag was ground to mechanically remove the surface layer using emery paper with 1000–1200 grit and a polishing diamond paste of 0.02–0.05 µm. The EPMA backscattering pattern and the main elemental distribution on the surface of γ-TiAl-10 wt.% Ag are shown in Figure 2. It can be seen that the silver was mainly dispersed in the white regions, as shown in Figure 2a, in accordance with the Ag element distribution shown in Figure 2b.

3. Results and Discussion

3.1. Analysis of Mechanical Properties

According to the ASTM Standard No. B962-08 and the Archimedes principle [20], silver with a density of 10.94 g/cm³ was used to prepare γ-TiAl-10 wt.% Ag to achieve a mean density of about 4.24 g/cm³. It was found to be higher than that of the TiAl alloy (3.95 ± 0.42 g/cm³). This might be because high-density silver was added into the TiAl-based self-lubricating sample, thus resulting in a significant increase in the mean density. Using an Instron 1341 tester, the mean yield strength of γ-TiAl-10 wt.% Ag (approximately 890 MPa) was smaller than that of the TiAl alloy (approximately 975 MPa). The nanoindentation morphology and indentation load curves of the γ-TiAl-10 wt.% Ag sample are shown in Figure 3. With the assistance of a nanomechanical apparatus, at least five tests were carried out to assess the nanohardness. The nanoindentation morphologies for the two tests of the γ-TiAl-10 wt.% Ag sample can be observed in Figure 3a; the corresponding nanoindentation load curves are indicated in Figure 3b, showing the close relationship between the indentation load and penetration depth. The nanohardness of γ-TiAl-10 wt.% Ag (5.98 ± 0.52 GPa) was found to be lower than that of the TiAl alloy (about 6.18 GPa). The mean elasticity modulus was reduced to 130 ± 1.52 GPa from 150 ± 1.63 GPa. Hence, silver (with almost 0.25 GPa hardness and an elasticity modulus of 68 GPa) was chosen to prepare γ-TiAl-10 wt.% Ag, causing the mechanical properties of the TiAl alloy to be slightly reduced. This was because the soft silver reduced the mechanical properties of the TiAl alloy base samples.

3.2. Friction and Wear Behavior of γ-TiAl-10 wt.% Ag

Figure 4a shows a schematic of the matching sample/ball tribopairs. As shown in Figure 4a, on the MFT-5000 tribometer, the friction and wear behavior of the as-prepared samples at 12 N and 450 °C was examined. The TiAl alloy and γ-TiAl-10 wt.% Ag sample were driven to slide against Si₃N₄ balls with 15.2 GPa hardness. At the relative humidity of 45–65%, the friction coefficients at 0–240 min were recorded in succession using the MFT-5000 computer-controlled system. The measured values are shown in Figure 4b. The sliding wear rate W can be calculated using W = (A·D)/(F·L) [21]. The calculated results are exhibited in Figure 4c. Herein, L and F are, respectively, the sliding distance and applied load; the parameters A and D are, respectively, the cross-section area and perimeter of the wear scar. As shown in Figure 4b,c, compared to those of the TiAl alloy, the γ-TiAl-10 wt.% Ag sample obtained smaller friction coefficients and wear rates. In accordance with the tested values, the friction coefficients and wear rates were rapidly reduced at 0–60 min and then approached small and stable values at the sliding time of 60–240 min.

Figure 5a shows the FESEM cross-sectional morphology of a wear scar of γ-TiAl-10 wt.% Ag. After 240 min, the stratified structure mainly consisted of a lubrication film, grain-refined layer, and microdeformation layer. Figure 5b,c show the silver intensity and Ag content under line scanning AA (LS-AA) (see Figure 5a) after 240 min. As shown in Figure 5b,c, under the wear scar subsurface, the Ag element’s intensity was gradually reduced from the surface (0 μm) to 3.5 μm, and the Ag content was lowered from 55.31 wt.% to 19.28 wt.%. This indicates that a silver-rich film formed on the wear scar, which was helpful in lowering the sliding friction resistance to result in low friction. It effectively prevented the wear surface from being destroyed to realize a low wear rate. Hence, a silver-enriched lubrication film appeared on the grain-refined layer, which was beneficial in improving the tribological behavior of γ-TiAl-10 wt.% Ag compared to that of the pure TiAl alloy.

Figure 6 shows the EBSD characterization of a wear scar cross-section of γ-TiAl-10 wt.% Ag. As shown in Figure 6a, layer A was identified as a lubrication film. Layer B mainly consisted of low submicron grains and was identified as the grain-refined layer. Layer C referred to the γ-TiAl-based composite, demonstrating the existence of large grains. As can be seen from Figure 6b, small grain strain existed in the lubrication film. High strain appeared in the grain-refined layer. The phase distribution in Figure 6c indicates that a large amount of silver mainly appeared in the blue regions. A small amount of γ-TiAl alloy was mainly dispersed in the white regions. The grain orientations are exhibited in Figure 6d. A misorientation angle at line D (see Figure 6d) is shown in Figure 6e. As shown in Figure 6d,e, the smallest misorientation angles appeared in the lubrication film compared to the grain refinement layer, and higher grain misorientation angles mainly existed in the γ-TiAl-based composite (layer C).

With an increase in the sliding time, which was limited to 0–240 min, the silver was squeezed out of the γ-TiAl-based composite and enriched on the wear scar, thus resulting in the formation of a silver-rich lubrication film (see Figure 6c). As a result of the approximately 25 MPa yield strength, massive amounts of silver led to plastic deformation, which resulted in low strain (see Figure 6b) and a small angle (see Figure 6e) in the formed lubrication film. Additionally, with the formation of the lubrication film, the wear debris on the wear scar was continually refined to form the grain refinement layer, which can be seen in Figure 6a. A silver-rich lubrication film with low strain and a small angle existed on the grain-refined layer, which was helpful in achieving small friction coefficients and low wear rates; this further facilitated the excellent behavior of γ-TiAl-10 wt.% Ag under friction and wear.

Figure 7 shows the EPMA morphology, FESEM morphology, backscattered morphology, texture structure, and surface profile of the wear scar at 120 min. As can be seen in Figure 7a, the main wear mechanism of γ-TiAl-10 wt.% Ag was plastic deformation at 120 min, resulting in the existence of a wear scar consisting of plastic deformation. As shown in Figure 7b,c, massive amounts of silver were spread across the wear scar. As shown in Figure 7d,e, during the sliding wear process, the silver showed excellent plastic deformation and good ductility to form a silver-enriched lubrication film as an effect of the sliding friction force. It improved the silver distribution area and repaired the wear scar morphology, resulting in a wear scar with a smooth texture, indicated by Sa:0.04, Sq:0.057, Sku:14.57, and Ssk:1.72 in Figure 7d,e. This smooth wear scar effectively lowered the friction resistance to obtain small friction coefficients at 60–240 min. The silver exhibited good plastic deformation to lower the material loss of the wear scar, resulting in a lower wear rate at 60–240 min. The silver-enriched lubrication film showed outstanding plastic ductility during the process of sliding wear, and this caused the lubrication film to be well maintained at 60–240 min. It resulted in small friction coefficients and lower wear rates for TiAl-10wt.%Ag and facilitated an enhancement in lubrication.

Figure 8a,b exhibit the backscattering morphologies of the wear scars of γ-TiAl-10 wt.% Ag at 10 min and 90 min, respectively. As shown in the figure, in accordance with the foregoing discussion of Figure 4 and Figure 5, after 90 min of sliding wear, a silver-rich film was distributed on the wear scar. The magnified morphology obtained by the FESEM of the wear scar marked by rectangle A in Figure 8b is shown in Figure 8c, which enables a better understanding of the microstructure of the lubrication film. Figure 8d shows the texture structure of the wear scar marked by rectangle A in Figure 8b. As shown in Figure 8d, the height parameters—namely Sa: 0.09 µm, Sq: 0.11 µm, Sku: 2.13, and Ssk: 0.23—were small when sliding wear was performed for up to 90 min, thus ensuring a smooth wear scar. Massive amounts of silver appeared on the wear scar (Figure 5 and Figure 8b,c), showing significant plastic deformation due to sliding friction forces. This effectively repaired the wear scar to create a well-textured structure (see Figure 8d), subsequently causing the wear scar to be smooth. This facilitated an ideal reduction in the sliding resistance and material loss of the wear scars and improved the tribological behavior of the γ-TiAl-10 wt.% Ag sample.

Based on the above discussion, it could be concluded that, in the process of dry wear, massive amounts of silver were gradually squeezed out of γ-TiAl-10 wt.% Ag to gather on the wear scar, which then resulted in an increase in the silver content. A large amount of silver was spread out across the wear scar as an effect of friction forces, increasing the silver distribution and forming a silver-enriched film for lubrication. The low grain strain and small misorientation angle enabled the silver-rich film to effectively resist the sliding friction resistance and reduce the material loss. Because of the low shearing strength and excellent plastic deformation, the silver-rich film was helpful in reducing the sliding friction force to obtain a low friction coefficient. It could prevent the wear scar from being destroyed to realize the small wear rate of γ-TiAl-10 wt.% Ag.

In addition to the findings shown in Figure 4, Figure 5, Figure 6, Figure 7 and Figure 8, the friction and wear behavior was mainly determined by the silver-rich film. However, during the friction process, it was impossible to form a complete lubrication film to cover the wear scar and enable lubrication. Hence, it was necessary to explore the effects of the distribution area of the silver-enriched film on the tribological behavior.

3.3. Tribology Calculation Model

With particular reference to previous reports [22,23,24], the principal component K-means clustering analysis method enables the clustering analysis of similar principal elements using the K-means method. Compared to traditional segmentation algorithms, which are based on image pixel thresholds, the principal component K-means clustering method not only considers the pixel value information of an image but also introduces the geometric structure information of the image. This significantly improves the accuracy and robustness in the labeling of the main elements. Hence, the principal component K-means clustering analysis method was used to perform sliding block processing on the backscattered images of the wear scars during wear to obtain a data set with N image blocks. Subsequently, a principal component K-means clustering analysis was conducted on the data set. The data set was divided into image blocks with and without silver, and then they were labeled. Finally, the mean values of these image blocks, labeled via cluster analysis, were combined to form labeled images. Based on the marked silver areas on the wear surface, the specific area η of the enriched silver region could be calculated to quantify the distribution area.

Figure 9a shows the derived model of the specific area η of the lubrication film containing massive amounts of silver. As shown in Figure 9a, when the sliding wear time increased to 90 min, the specific silver area was approximately equal to 44.1% when analysis software was used to perform effective calculations, using the theory of the K-means method of main component clustering. Figure 9b shows the variation curve of the specific area η at 60–240 min. With an increase in the sliding time from 60 to 240 min, the specific area η of the silver-rich film was mainly distributed within the range of 44–51%; it was found that the stable distribution area of the silver-enriched lubrication film at 60–240 min sliding wear facilitated consistent lubrication, thus resulting in the low friction coefficient and wear rate of γ-TiAl-10 wt.% Ag.

According to the work by Yamamoto et al. [25], the tribological behavior was mainly determined by the area proportions ɳ of the silver-enriched film distributions, which could be calculated using Formula (11). A model for calculating the friction coefficients is indicated in Formula (12). Similarly, a theoretical model was derived, as expressed in Formula (14), to acquire the sliding wear rate.

$$\eta ={\displaystyle \frac{A_l}{A_m+A_l}}$$

(12)

$${\mu }_c={\mu }_m\cdot (1-\eta )+{\mu }_l\cdot \eta$$

(13)

$$W_c=W_m\cdot (1-\eta )+W_l\cdot \eta$$

(14)

where A_l is the Ag distribution area, A_m is the TiAl-Ag distribution area, μ_l is the friction coefficient between Ag and the Si₄N₃ ball, μ_m is the friction coefficient between γ-TiAl-10 wt.% Ag and the Si₄N₃ ball, W_l is the wear rate between Ag and the Si₄N₃ ball, W_m is the wear rate between the Si₄N₃ ball and γ-TiAl-10 wt.% Ag, and μ_c and W_c are the calculated friction coefficient and wear rate W_c.

In order to understand the quantitative influence of the specific Ag area on the tribological behavior, μ_l and μ_m were chosen as 0.15 and 0.36, and W_l and W_m were selected as 1.05 × 10⁻⁴ mm³·N⁻¹·m⁻¹ and 3.95 × 10⁻⁴ mm³·N⁻¹·m⁻¹. With reference to the computational formulas for the friction coefficient (see Formula (12)) and wear rate (see Formula (13)), the columnar charts of the friction coefficients μ_c and wear rates W_c were obtained, as exhibited in Figure 10. As can be seen from the figure, the calculated friction coefficients at 60–240 min were approximatively equivalent to the tested values. The calculated wear rates were slightly higher than the tested values, and the errors for the sliding tribological behavior were acceptable. Consequently, these models could be used to evaluate the effects of a silver-rich film on the friction and wear behavior of γ-TiAl-10 wt.% Ag. This provides an important reference toward evaluating the tribological behavior of TiAl alloy base components.

4. Conclusions

The main objective of this study was to understand the SPS fabrication and sliding wear properties of γ-TiAl-Ag self-lubrication materials, reaching the following main conclusions:

• A large amount of soft silver used in preparing γ-TiAl-10 wt.% Ag samples caused their mechanical properties to be slightly reduced.
• Because of the low grain strain and small misorientation angle, the silver-rich film was protected from destruction and the lubrication ability was improved, thus leading to the small friction coefficient and wear rate of γ-TiAl-10 wt.% Ag.
• Massive amounts of silver gradually spread out across the wear scar to increase the Ag distribution area. This resulted in the formation of a silver-rich lubrication film, which was helpful in ensuring the excellent tribological behavior of γ-TiAl-10 wt.% Ag.
• The K-means method of main component clustering was utilized to establish a theoretical model, and this facilitated the calculation of the distribution area of the silver-enriched lubrication film. It was found that an increase in the silver distribution area improved the friction and wear behavior of γ-TiAl-10 wt.% Ag; subsequently, the specific area η of the silver-rich film was maintained at 44–51%, and a small friction coefficient (almost 0.28) and wear rate (about 2.25 × 10⁻⁴ mm³·N⁻¹·m⁻¹) were observed for γ-TiAl-10 wt.% Ag.