This paper studies the effect of adding a small amount of zinc on the microstructure, hardness and high temperature creep resistance of SnSb8Cu4 babbitt alloy. An indentation creep test device with a flat end indenter was designed to characterize the creep deformation behavior of babbitt alloy at 100°C ambient temperature and different loads. After adding 0.83wt% Zn to SnSb8Cu4 alloy, a large number of fine and dispersed SnSb particles precipitated in the matrix, and these particles tended to precipitate along the grain boundaries of the Sn matrix. The total volume fraction of intermetallic compounds, namely Cu6Sn5 and SnSb particles, increased from 14.9% to 21.2%. Although the increase in Brinell hardness at room temperature was not obvious, its creep resistance was significantly improved. The addition of Zn reduced the solid solubility of Sb in the Sn matrix, resulting in more SnSb particles precipitating along the grain boundaries, which played a pinning role in the grain boundary sliding during creep deformation, thereby leading to an increase in creep resistance. In the steady-state creep stage, the indentation creep rate is exponentially related to the indentation stress. According to the measured data, the indentation stress index of SnSb8Cu4 and SnSb8Cu4Zn is 2.95 and 2.73 respectively.

SnSb8Cu4 is a typical tin-based babbitt alloy, with a tin-based solid solution as the matrix, in which hard phase particles such as Cu6Sn5 and SnSb are distributed. It has excellent compliance, embedding and anti-seizure properties, and is often used as a bearing lining alloy material for sliding bearings. The melting point of babbitt alloy is low, and the normal service temperature is 60~80℃, which is equivalent to 0.6~0.7Tm (melting point of babbitt alloy).
Long-term service at this temperature is prone to creep deformation. As modern industry places higher and higher demands on production efficiency, the continuous service time of babbitt alloy bearings is getting longer and longer, and the accumulation of creep deformation is getting larger and larger. When the creep deformation exceeds the gap between the bearing shell and the shaft diameter, the bearing shell may directly contact the shaft diameter, resulting in dry friction and burning of the bearing shell. This failure mode of creep deformation tolerance accounts for an increasing proportion in the failure mode of the bearing shell. Therefore, enhancing the creep resistance of babbitt alloy has become an important topic for this alloy. Generally speaking, the methods to improve the high-temperature creep resistance of materials include increasing the grain size, alloying, solid solution strengthening or precipitation strengthening, etc., but there are few reports on research work to improve the creep resistance of babbitt alloy.

On the other hand, the creep performance test of metal materials is generally carried out using rod-shaped specimens in accordance with GB/T-2039 “Metal Tensile Creep and Endurance Test Methods”, but for babbitt alloy, its actual application state is to make a thin layer of bearing lining on the back of steel shells. Its molding method and organizational structure are completely different from those of rod-shaped specimens used for creep testing, so the test results of conventional rod-shaped specimens cannot be directly applied to actual working conditions. Qian Kangle et al. reported an indentation creep test method for babbitt alloy bearing linings, that is, a small-diameter flat indenter is used to directly apply a certain load on the surface of the babbitt alloy bearing lining and maintain it for a period of time. By measuring the relationship between the indentation depth and the load and the holding time, the creep resistance of the babbitt alloy bearing lining can be quantitatively studied. Since this test method directly uses the actual bearing shell for testing, its results are more valuable for reference to bearing design.

In order to study the method of improving the creep resistance of babbitt alloy, this paper intends to introduce a small amount of zinc into SnSb8Cu4 babbitt alloy, and use the indentation creep test method to study its creep behavior. At the same time, in order to deepen the understanding of the creep resistance mechanism of babbitt alloy, the metallographic structure and phase composition of the alloy are studied in detail using scanning electron microscopy and X-ray diffractometer.

Experimental part

The solid solubility of Zn in Sn at room temperature is ~1%. To avoid the formation of low melting point Sn-Zn eutectic phase, the amount of Zn added should be less than 1%. This paper uses 0.9% Zn addition as the material. Pure Sn, Sb, Cu, Zn and other elements are melted into SnSb8Cu4 and SnSb8Cu4Zn according to the designed proportion for comparative test. In order to simulate the working conditions of the actual workpiece, the Babbitt alloy is cast on the steel substrate to achieve the purpose of simulating the actual bearing. The steel substrate uses 20 steel with a thickness of 20mm and a length and width of 100mm×100mm. The actual composition of the two samples is shown in Table 1.

In order to improve the bonding strength between the Babbitt alloy and the steel backing and keep it consistent with the actual bearing manufacturing process, the steel block is pre-placed in 300℃ tin liquid for tinning (hot dip tinning) treatment before casting, and then the Babbitt alloy is cast. In order to obtain a casting layer of a certain thickness, before casting, the 20 steel back bottom is surrounded by steel sheets, which are ~20mm higher than the surface of the steel back bottom; the preheating temperature of the steel back bottom before casting is 260℃, and the temperature of the babbitt alloy is ~400℃ during the casting process, and the casting thickness of the babbitt alloy is ~15mm. Finally, the cast sample is cut into 40mm×40mm×35mm samples, of which the height of 35mm includes a 20mm steel base layer and a babbitt alloy layer of about 15mm. The babbitt alloy layer of the cut SnSb8Cu4 and SnSb8Cu4Zn samples is polished to ~5mm thickness with sandpaper, then cleaned with ethanol, blown dry, and corroded with 4% nitric acid ethanol solution for metallographic observation. Using 10 or more metallographic photos, the volume fraction of the precipitated phase was obtained by calculation and analysis using relevant image analysis software. The Brinell hardness was measured using a DHB-3000 Brinell hardness tester with a load of 15.625 kg, a holding time of 15 seconds, and a head size of 2.5 mm. The indentation creep test was carried out in a resistance furnace resistance heating oven at 100°C, and the creep test device used was shown in Figure 1. The flat head indenter used for the indentation creep test had a diameter of 0.7 mm, and the loads were 3 kg, 4.5 kg, and 6 kg, respectively. During the creep test, the experiment was stopped according to the set time, and the indentation depth was measured using a Keyence VHX-2000E digital microscope.

The physical phase analysis used a Shimadzu XRD-6100 X-ray diffractometer, and the metallographic analysis used a scanning electron microscope (SEM) model of a JEOL JSM7800F field emission scanning electron microscope.

Results and Discussion

Experimental Results

Figure 2 shows the microstructure of SnSb8Cu4 and SnSb8Cu4Zn under an optical metallographic microscope. The microstructure of SnSb8Cu4 has been fully studied at home and abroad, and the main phases include β-Sn matrix phase and Cu6Sn5 and SnSb hard phases. From Figure 2(a), it can be seen that in the metallographic structure of SnSb8Cu4, the slender Cu6Sn5 dendrites precipitated in a star shape are embedded in the Sn matrix, and a small amount of smaller precipitated phase particles can also be observed. From Figure 2(b), it can be seen that compared with SnSb8Cu4, after adding a small amount of Zn element to Babbitt alloy, the proportion of granular precipitated phase particles in the matrix increases significantly.

The metallographic microscope slides were statistically analyzed using image analysis software, and the volume fractions of the hard phase precipitated in SnSb8Cu4 and SnSb8Cu4Zn were obtained, as shown in Figure 3, which were 14.9% and 21.2%, respectively. The introduction of a small amount of zinc resulted in an increase of 42.3% in the volume fraction of the precipitated hard phase.

Figure 4 shows the X-ray diffraction spectra of SnSb8Cu4 and SnSb8Cu4Zn samples, respectively. By comparing with the standard spectrum, it can be seen that the two Babbitt alloy samples with different compositions are composed of three phases, namely, β-Sn matrix phase and Cu6Sn5 and SnSb hard phases.

Figure 5 shows the metallographic structure morphology of the two Babbitt alloys under a scanning electron microscope. After polishing, the samples were etched with a 4% nitric acid alcohol solution and imaged using backscattered electron signals. As can be seen from Figure 5(a), in the SnSb8Cu4 sample, there are two types of precipitation phases on the matrix: a larger elongated precipitation phase and an equiaxed fine (about 2-5µm in diameter) granular precipitation phase; after adding 0.83wt% Zn, as shown in Figure 5(b), the proportion of fine granular precipitation phase increases significantly. After a partial enlargement of Figure 5(b), it is found that a large number of fine particles tend to precipitate along the grain boundaries of the Sn matrix, as shown in Figure 5(c). The solid line in Figure 5(c) is a schematic diagram of the grain boundary.

The composition of the precipitated phase in Figure 5 was analyzed using the energy dispersive spectrometer (EDS) on the scanning electron microscope. The results are shown in Figure 6. It can be determined that the elongated precipitated phases at point 1 in Figure 5 (a) and position 3 in (b) are Cu6Sn5 intermetallic compounds; the granular precipitated phases at point 2 in Figure 5 (a) and position 4 in (b) are SnSb intermetallic compounds. It should be noted that, as shown in Figure 6 (c), Cu6Sn5 in the SnSb8Cu4Zn sample contains zinc, indicating that zinc may be enriched in Cu6Sn5 particles.

The Brinell hardness of SnSb8Cu4 and SnSb8Cu4Zn alloys at room temperature was measured, and the results are shown in Figure 7. The Brinell hardness values ​​are 21.7HBW and 25.1HBW, respectively, indicating that the introduction of Zn increases the hardness of SnSb8Cu4 by 15.7%.

The creep behavior of two babbitt alloy samples, SnSb8Cu4 and SnSb8Cu4Zn, was studied at 100℃ using a homemade indentation creep test device. Three constant loads of 3kg, 4.5kg and 6kg were selected for the indentation creep test, and the corresponding stresses were 77, 115 and 153MPa, respectively. The curve of the change of indentation depth versus loading time is shown in Figure 8.

It can be observed from Figure 8 that with the increase of the applied load and loading time, the indentation depth shows a trend of continuous increase. The increase of indentation depth can be divided into two stages: the first stage is the rapid growth stage of indentation depth, the creep rate (i.e., the slope of the curve) is very high, but as time goes on, the creep rate decreases rapidly; in the second stage, the indentation depth increases linearly with the extension of the holding time, which belongs to the stable creep (constant rate creep) stage. In addition, it should be noted that, unlike the traditional tensile creep test, the indentation creep curve does not have the final accelerated creep stage. The abnormally rapid increase in creep rate in the final accelerated creep stage of the tensile creep test is due to the reduction of the effective cross-sectional area of ​​the specimen, while the indentation creep test does not have this problem.

The second stage of indentation creep occupies the vast majority of the creep cycle. Therefore, the creep rate (i.e., the slope of the creep curve) in this stage represents the creep resistance of the alloy. The creep rates of the two alloys in Figure 8 under different loads are summarized in Table 2.

Discussion

For practical engineering applications, the relationship between creep rate and applied load should be obtained first. In this experiment, the compressive stress (σ) is defined as the load (F) divided by the cross-sectional area of ​​the indenter (A), that is, σ=F/A. Combined with the results of Figure 8, the steady-state indentation creep behavior can be expressed by formula (1): 𝑑̇ = 𝑐𝜎‘𝑛 (1)
In formula (1), 𝑑̇ is the creep rate, that is, the sinking speed of the flat-head indenter, c is the creep constant, and n is the indentation stress index. Taking the logarithm of both sides of formula (1), we get formula (2): 𝑙𝑔(𝑑̇) = 𝑙𝑔(𝑐) + 𝑛 ∙ 𝑙𝑔(𝜎) (2)
From formula (2), it can be seen that there is a linear relationship between lg(𝑑̇) and lg(σ). Putting the data in Table 2 into the logarithmic coordinate graph, it can be seen that there is an obvious linear relationship between lg(𝑑̇) and lg(σ), as shown in Figure 9, which conforms to the description of formula (2). Linear fitting is performed on these data points in Figure 9. The slope of the fitting line is the indentation creep stress index n of the corresponding babbitt alloy sample. The creep constant c can be determined by the intercept of the fitting line and the ordinate axis in the logarithmic coordinate graph. The calculation results are listed in Table 3.

Generally, there are three creep mechanisms, namely, diffusion mechanism, grain boundary sliding and dislocation movement. At higher temperatures, grain boundary sliding is the main creep mechanism. On the other hand, the creep deformation mechanism can also be evaluated by the n value. When the n value is around 1, it is a diffusion mechanism. When the n value is 2~3, the grain boundary sliding plays a major role. When the n value is in the range of 4~6, the dislocation climb is dominant. When n＞6, the mechanism related to the dislocation movement is the main mechanism of creep. In this experiment, according to the calculation, the result of n is in the range of 2~3, which can be inferred that grain boundary sliding is the main creep deformation mechanism. Therefore, for SnSb8Cu4 alloy, inhibiting grain boundary sliding is an effective measure to improve creep performance.

In addition, grain size plays an important role in the creep resistance of tin-based babbitt alloys because a large number of grain boundaries can become obstacles to dislocation movement and grain boundary sliding. However, there is still some controversy about the effect of grain size on creep performance. Wu et al. found that the reduction of grain size reduces the stress concentration at the grain boundary, thereby delaying the nucleation of voids. Therefore, the fine dispersed β-Sn particles are the main reason for improving creep resistance. However, the results of Mahmudi et al. show that the reduction of grain size should not be regarded as the main reason for creep enhancement. They believe that the strengthening of Cu6Sn5 particles in the β-Sn matrix leads to the improvement of creep resistance. In this study, since there is no significant difference in the size of β-Sn grains and Cu6Sn5 grains in SnSb8Cu4 and SnSb8Cu4Zn alloys, it can be considered that the SnSb precipitation phase is the main factor causing the change in the creep resistance of the alloy.

As can be seen from Figure 5, in both alloys, the size of SnSb particles precipitated on the Sn matrix is ​​2~5µm, which is much smaller than the SnSb particles (about 80μm) formed in the melt when the SnSb11Cu6 melt is cooled. Therefore, these small-sized SnSb particles are likely to be precipitated from the supersaturated Sn matrix during the cooling process after solidification. During the cooling process after solidification, the solid solubility of Sb in the Sn matrix decreases rapidly, and the excess Sb tends to diffuse to the grain boundaries of the Sn matrix to reduce the system energy, and finally forms the SnSb hard phase at the grain boundaries. Since the solid phase diffusion capacity of the element is much smaller than the liquid phase diffusion, the SnSb hard phase observed in this paper is much smaller than the SnSb hard phase directly formed in the liquid phase. In addition, the introduction of a small amount of Zn may reduce the solid solubility of Sb in the Sn matrix, resulting in the precipitation of a large number of fine and dispersed SnSb particles at the grain boundaries. Because the energy state at the grain boundary is high and the atomic arrangement is chaotic, diffusion becomes easier, and finally forms the structure shown in Figure 5(c). The melting point of SnSb phase is ~240°C, which is much higher than 100°C. That is, at the indentation creep test temperature, these small-sized SnSb particles can still exist at the grain boundary and play a role in pinning the grain boundary during creep deformation, hindering the grain boundary sliding, which is manifested as an improvement in creep resistance on a macro scale.

Conclusion

(1) After adding 0.83wt% Zn to SnSb8Cu4, a large number of fine and dispersed SnSb particles tend to precipitate along the tin matrix grain boundary, and the overall hard phase volume fraction increases by 42.3%, and the hardness increases by 15.7%. At the same time, the introduction of Zn element significantly improves the creep resistance of Babbitt alloy, which can be attributed to the pinning effect of a large number of fine SnSb particles precipitated along the tin matrix grain boundary on the grain boundary, hindering the grain boundary sliding during creep.

(2) The indentation creep test can be used to quantitatively evaluate the creep resistance of Babbitt alloy bearings. Compared with the traditional tensile creep test method, the indentation creep test is convenient and easy to perform, and its samples and test process are closer to the actual operating conditions of the bearing.