Tribological behavior of NiTi alloy against 52100 steel and WC at elevated temperatures

Tribological behavior of NiTi alloy against 52100 steel and WC at elevated temperatures

M A TE RI A L S CH A RACT ER IZ A TI O N 61 ( 20 1 0 ) 6 8 9 –6 9 5 available at www.sciencedirect.com www.elsevier.com/locate/matchar Tribological...

2MB Sizes 2 Downloads 37 Views

M A TE RI A L S CH A RACT ER IZ A TI O N 61 ( 20 1 0 ) 6 8 9 –6 9 5

available at www.sciencedirect.com

www.elsevier.com/locate/matchar

Tribological behavior of NiTi alloy against 52100 steel and WC at elevated temperatures M. Abedini, H.M. Ghasemi⁎, M. Nili Ahmadabadi School of Metallurgy and Materials Engineering, University of Tehran, Tehran, Iran

AR TIC LE D ATA

ABSTR ACT

Article history:

The dry tribological behavior of a Ti–50.3 at.% Ni alloy at temperatures of 25 °C, 50 °C and

Received 28 July 2009

200 °C was studied. The wear tests were performed on a high temperature pin-on-disk

Received in revised form

tribometer using 52100 steel and tungsten carbide pins. The worn surfaces of the NiTi alloy

27 March 2010

were examined by scanning electron microscope. The results showed that in the wear tests

Accepted 30 March 2010

involving steel pins, the wear rate of the NiTi decreased as the wear testing temperature was increased. However, for the NiTi/WC contact, a reverse trend was observed. There was also a

Keywords:

large decrease in the coefficient of friction for the NiTi/steel contact with increasing wear

NiTi alloy

testing temperature. The formation of compact tribological layers could be the main reason

High temperature wear

for the reduction of the wear rate and coefficient of friction of the NiTi/steel contact at

Friction

higher wear testing temperatures.

WC

© 2010 Elsevier Inc. All rights reserved.

Steel

1.

Introduction

NiTi alloys are well known for the superelastic and shape memory effects. Since their first discovery in 1960s, NiTi alloys have been extensively studied. Both shape memory effect and superelasticity have been exploited to design functional and smart structures in mechanical and biomedical engineering [1–3]. Many more potential applications and mechanical behaviors of shape memory alloys have been investigated. For instance, shape memory and superelastic nickel–titanium alloys have been increasingly used in medical surgery and identified as possible materials for micro-electromechanical systems (MEMS) [4,5]. In these applications, the wear performance of the material plays a critical role [5]. Another field in which the wear behavior of NiTi alloys may have a major importance is that of mechanical machining [6]. The wear resistance of conventional tribo-materials strongly depends on their mechanical properties such as hardness, toughness, and work-hardening [7,8]. Ball [9] attributed the wear properties of NiTi simply to the work-

hardening. Singh et al. [10] showed that during dry sliding wear tests, the wear rate of Ti50Ni47Fe3 alloy against SAE 52100 bearing steel was only 2–5% of that of the steel counterface. This was mainly attributed to the work-hardening of surface layers and formation of iron oxide layer at the worn surfaces. For traditional materials, both theory and experiments show that the wear resistance increases with the hardness [7]. The relationship between wear and hardness of five annealed NiTi shape memory alloys was studied by Arciniegas et al. [11]. They showed that the wear resistance of austenitic NiTi alloys is consistent with the hardness values of each alloy, higher as hardness increases. Most of the previous work was concerned with the effect of superelasticity on the wear behavior of NiTi alloys. It is believed that the super wear resistance of shape memory NiTi is mainly due to the recovery of the superelastic deformation [5,12–22]. However, the positive effect of the superelasticity on the wear behavior of NiTi alloy was observed at working temperatures near austenitic transformation finish temperature [22]. No studies have been found at higher working

⁎ Corresponding author. Tel.: +98 21 88012999; fax: + 98 21 88006076. E-mail address: [email protected] (H.M. Ghasemi). 1044-5803/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.matchar.2010.03.017

690

MA TE RI A L S CH A R A CT ER IZ A TI O N 61 ( 20 1 0 ) 6 8 9–6 9 5

temperatures where superelasticity and shape memory effects may not exist. Therefore, characterization of high temperature wear behavior of NiTi alloy could be highly beneficial for better understanding of the wear mechanism of NiTi alloys. The wear mechanisms involved and the tribological layers formed may depend on the counterface material during sliding. The main aim of the present study is to characterize the dry sliding wear behavior of an austenitic NiTi alloy against 52100 steel and tungsten carbide (WC) pins at wear testing temperatures of 25 °C, 50 °C and 200 °C.

2.

Experimental Procedure

A NiTi alloy with composition of Ti–50.3 at.% Ni used for the present study was prepared in a Vacuum Induction Melting furnace (VIM). Disks, 5 mm in thickness, were wire-cut from the bar obtained in VIM. The disks were hot forged at 800 °C into a thickness of 3.5 mm (a 30% reduction), solution annealed in a vacuum furnace at 1000 °C for 12 h in a controlled argon atmosphere and quenched in water. Finally, the disks were aged at 400 °C for 60 min, followed by water quenching. The disks were then ground to a surface finish of about 0.9 ± 0.1 μm. The phase transformation temperatures were measured by a Mettler differential scanning calorimeter (DSC). Measurements were carried out at temperatures ranging from − 120 °C to 120 °C under a controlled cooling/heating rate of 10 °C/min. In order to characterize the mechanical behavior of the material, uni-axial compression tests were performed in an MTS universal testing machine at testing temperatures of 25 °C and 200 °C. The diameter of the cylindrical compression specimens was 3 mm with a length of 4.5 mm, i.e., a length/ diameter ratio of 1.5. In these tests, the specimens were loaded to fracture. The wear tests were performed on a high temperature pinon-disk tribometer using the NiTi disk with a diameter of 40 mm. SAE 52100 bearing steel and tungsten carbide (WC– 6 wt.% Co) pins were used as slider. The diameter of the pins was 5 mm with 3.1 mm radius on their tips. The wear tests were performed under a normal load of 20 N at a constant sliding speed of 0.3 m/s for a sliding distance of 1000 m at wear testing temperatures of 25 °C, 50 °C and 200 °C in dry conditions. Prior to each test, all contact surfaces were ultrasonically cleaned in acetone, dried and weighed to a precision of 0.1 mg. Three wear tests were performed for each condition and the average weight loss was calculated. The coefficient of friction was calculated using the measured friction and the normal load. Finally, the worn surfaces were studied with a scanning electron microscope (SEM).

3.

Table 1 – Phase transformation temperatures (°C) of the NiTi alloy. Austenite start As

Austenite finish Af

Martensite start Ms

Martensite finish Mf

25

45

16

−1

sliding and wear could raise the surface temperature of the contact [7] and, therefore, the alloy could be transformed to an austenitic structure. At the higher wear testing temperatures (50 °C and 200 °C, i.e., above Af), however, the material could be completely in the austenitic state. Fig. 1 shows the variation of wear rates of the NiTi samples against steel and WC pins as a function of the wear testing temperature under a normal load of 20 N. The figure shows that as the wear testing temperature was increased the wear rate of the NiTi alloy against 52100 steel pin decreased. However, there was an increase in the wear rate of the NiTi alloy against WC pin with an increase in the wear testing temperature. In Fig. 2 the compressive stress–strain curves of the NiTi alloy at the testing temperatures of 25 °C and 200 °C are shown. At the testing temperature of 25 °C the stress–strain curve of the NiTi alloy showed a plateau region, which could be attributed to stress induced martensitic phase transformation [23]. However, at the testing temperature of 200 °C, which was far above the Af temperature of 45 °C in Table 1, no such plateau was observed. This could indicate that no phase transformation occurred and, therefore, the alloy acted like a non-superelastic material. The yield strengths of the alloy at the temperatures of 25 °C and 200 °C were measured from the stress–strain curves in Fig. 2. The figure shows that with an increase in the testing temperature from 25 °C to 200 °C, the yield strength was decreased from a value of 940 MPa to 560 MPa, resulting in a reduction of about 40%. The lower yield strength of the material at the temperature of 200 °C could result in a higher wear rate of the NiTi against WC at the wear testing temperature of 200 °C as expected [7]. However, the wear results of the NiTi against a steel pin in Fig. 1 showed a reverse trend. The wear rates of the NiTi alloy against steel

Results and Discussion

The phase transformation temperatures, measured by differential scanning calorimetry (DSC), are listed in Table 1. At the wear testing temperature of 25 °C, which was less than the austenitic transformation finish temperature (Af) of 45 °C, the microstructure of the alloy would be a mixture of martensitic and austenitic NiTi [23]. However, the frictional heating during

Fig. 1 – Wear rate of the NiTi alloy against 52100 steel and WC pins as a function of wear testing temperature.

M A TE RI A L S CH A RACT ER IZ A TI O N 61 ( 20 1 0 ) 6 8 9 –6 9 5

691

tion at the higher temperatures under high loads and stresses could form a mechanically mixed layer (MML) or a solid compact oxide layer on the surface of the NiTi alloy. Therefore, the formation of this layer on the wear surface of the NiTi alloy in contact with a steel pin could decrease the wear damage by about 80% at the testing temperature of 200 °C compared to 25 °C (Fig. 1). A stable layer could lower the wear damage by

Fig. 2 – Stress–strain curves of the NiTi alloy obtained from uni-axial compression tests.

and WC pins were almost equal at the wear testing temperature of 25 °C; however, at the testing temperature of 200 °C the wear rate of the NiTi alloy against 52100 steel was about an order of magnitude lower than for the NiTi/WC contact. The mechanisms responsible for the observed wear kinetics have been investigated through the analysis of the wear tracks. Fig. 3 shows the micrographs of the worn surface of the NiTi disk against a steel pin at the wear testing temperature of 25 °C. No oxide layers were formed on the wear track at this temperature. Fig. 3a shows that ploughing and abrasive mechanisms could be important in the sliding wear of the NiTi alloy. The SEM micrograph in Fig. 3b also shows surface fatigue cracks on the worn surface of the NiTi alloy along with plastic deformation. Due to repeated cyclic loading at the sliding interface, the surface or subsurface cracks could be formed. These could result in the breakup of the surface into large fragments, leaving large pits on the worn surface [24] as shown in Fig. 3b. Figs. 4 and 5 show the micrographs of the worn surface of the NiTi, which ran against a steel pin at the wear testing temperatures of 50 °C and 200 °C, respectively. Fig. 4a shows that some patches were formed on the worn surface of the NiTi alloy at the wear temperature of 50 °C. Results of the Energy Dispersive X-ray Spectroscopy (EDS) analysis of two different regions of the wear track in Fig. 4b are listed in Table 2. The results revealed that the tribological layer formed in region A mainly contained iron and oxygen, which could indicate formation of iron based oxides on the wear surface at the wear testing temperature of 50 °C. In region B, the EDS analysis could probably suggest a mixture of nickel and titanium oxides and no formation of iron base oxides. Fig. 5a shows that a larger region of the worn surface of the NiTi alloy was covered with tribological compacted layer at the wear testing temperature of 200 °C compared to 50 °C. Results of the EDS analysis of two different regions of the wear track in Fig. 5b are listed in Table 3. The results also revealed that the compacted oxide region was mainly iron oxides. During sliding at higher temperatures, the rate of oxide generation of the steel pin in the interface could increase. The formation of wear debris and transfer of the oxides to the wear surface of the NiTi sample [16] and their subsequent compac-

Fig. 3 – SEM micrographs of worn surfaces of the NiTi alloy against 52100 steel pin under a load of 20 N for sliding distance of 1000 m at wear testing temperature of 25 °C: (a) at a low magnification, and (b) at a higher magnification.

692

MA TE RI A L S CH A R A CT ER IZ A TI O N 61 ( 20 1 0 ) 6 8 9–6 9 5

Fig. 4 – SEM micrographs of worn surfaces of the NiTi alloy against 52100 steel pin under a load of 20 N for sliding distance of 1000 m at wear testing temperature of 50 °C, (a) at a low magnification, and (b) at a higher magnification. acting as a solid lubricant as well as reducing the metallic contact. In fact, the formation of tribological oxide layers could cause a transition from a more severe wear mechanism at the testing temperature of 25 °C to a milder wear at the testing temperature of 200 °C. SEM micrographs of the worn surfaces of the NiTi against WC pins at the wear testing temperatures of 25 °C and 200 °C are shown in Fig. 6. The figure shows that, at both testing temperatures, the main wear mechanisms of the NiTi alloy

Fig. 5 – SEM micrographs of worn surfaces of the NiTi alloy against 52100 steel pin under a load of 20 N for sliding distance of 1000 m at wear testing temperature of 200 °C, (a) at a low magnification, and (b) at a higher magnification.

Table 2 – Chemical compositions (at.%) of two regions in the worn surface of NiTi alloy against steel pin at wear testing temperature of 50 °C (Fig. 4b).

Region A Region B

Ti

Ni

Fe

O

5.88 36.54

5.94 37.06

25.63 –

62.56 26.40

693

M A TE RI A L S CH A RACT ER IZ A TI O N 61 ( 20 1 0 ) 6 8 9 –6 9 5

Table 3 – Chemical compositions (at.%) of two regions in the worn surface of NiTi alloy against steel pin at wear testing temperature of 200 °C (Fig. 5b).

Region A Region B

Ti

Ni

Fe

Cr

O

5.48 36.54

5.33 37.06

26.75 –

0.47 –

61.97 26.40

Table 4 – Chemical compositions (at.%) of two regions in the worn surface of NiTi alloy against WC pin at wear testing temperature of 200 °C (Fig. 6b).

Region A Region B

Ti

Ni

O

35.26 51.03

36.44 48.97

28.29 –

against WC were ploughing and abrasive wear. The higher wear rate of the alloy at the wear testing temperature of 200 °C in Fig. 1 could be the result of lower yield strength of the alloy at this temperature as inferred from Fig. 2. The EDS analysis of the wear track in Fig. 6b is listed in Table 4. The results revealed that the patches formed on the worn surface of the NiTi alloy in region A contained a certain amount of nickel, titanium and oxygen. This might indicate formation of a mixed nickel and titanium oxide layer on the worn surface of the NiTi at the wear testing temperature of 200 °C. However, the layer was not compact enough to help decrease the wear of the alloy in contact with WC pin at the higher temperature. The average coefficient of friction for the NiTi/steel and the NiTi/WC contacts at different wear testing temperatures is shown in Fig. 7. A small decrease in the average coefficient of friction was observed for the NiTi/WC contact. This reduction could be attributed to the formation of tribolayer (oxide patches in Fig. 6b) on the worn surface of the NiTi alloy at higher temperatures. However, there was more than 35% decrease in the average coefficient of friction for the NiTi/steel as the testing temperature increased from 25 °C to 200 °C. At the wear testing temperature of 25 °C the average coefficient of friction for the NiTi/steel contact is higher than that for the NiTi/WC contact, which could indicate greater adhesion between the steel pin and the NiTi disk. However, at the wear testing temperature of 200 °C the average coefficient of friction for the NiTi/steel contact became lower than that for the NiTi/WC contact. This could be the result of the formation of compact iron oxide based tribological layer as shown in Fig. 5. Variations of coefficient of friction for the NiTi/steel and the WC/steel contacts with sliding distance at the wear testing

Fig. 6 – SEM micrographs of worn surfaces of the NiTi alloy against WC pin under a load of 20 N for sliding distance of 1000 m at wear testing temperature of: (a) 25 °C, and (b) 200 °C.

Fig. 7 – Variation of average coefficient of friction as a function of wear testing temperature for the NiTi/steel and the NiTi/WC contacts under a normal load of 20 N for sliding distance of 1000 m.

694

MA TE RI A L S CH A R A CT ER IZ A TI O N 61 ( 20 1 0 ) 6 8 9–6 9 5

wear testing temperature of 200 °C (Figs. 8b and 9b). At the wear testing temperature of 200 °C, the formation of tribological oxide layer could lower the adhesion between mating surfaces, resulting in a lower instability in the coefficient of friction.

4.

Conclusions

1. The yield strength of the NiTi alloy was decreased about 40% as the testing temperature increased from 25 °C to 200 °C. 2. The wear rate of the NiTi alloy in contact against WC pin increased with the wear testing temperature. However, the wear rate of the alloy in contact against steel pin decreased with increasing in the wear testing temperature. 3. The reduction of the wear rate and the average coefficient of friction of the NiTi/steel contact as the wear testing temperature increased were attributed to the formation of compact tribological layer on the wear surface of the NiTi alloy. The layer mostly contained iron oxides, which could limit adhesion between the mating surfaces. Fig. 8 – Variation of coefficient of friction of the NiTi/steel contact with sliding distance under a load of 20 N at the wear testing temperature of: (a) 25 °C, and (b) 200 °C.

temperatures of 25 °C and 200 °C are shown in Figs. 8 and 9, respectively. At the wear testing temperature of 25 °C, Figs. 8a and 9a show a higher instability in the coefficient of friction of both the NiTi/steel and the WC/steel contacts compared to the

Fig. 9 – Variation of coefficient of friction of the NiTi/WC contact with sliding distance under a load of 20 N at the wear testing temperature of: (a) 25 °C, and (b) 200 °C.

REFERENCES [1] Funakubo H. Shape memory alloys. New York: Gordon and Bleach; 1987. [2] Van Humbeeck J. Non-medical applications of shape memory alloys. Mater Sci Eng A 1999;273–275:134–48. [3] Duerig T, Pelton A, Stoeckel D. An overview of nitinol medical applications. Mater Sci Eng A 1999;273–275:149–60. [4] Kahn H, Huff MA, Heuer AH. The TiNi shape memory alloy and its applications for MEMS. J Micromech Microeng 1998;8: 213–21. [5] Li DY. Exploration of TiNi shape memory alloy for potential application in a new area: tribological engineering. Smart Mater Struct 2000;9:717–26. [6] Gialanella S, Ischia G, Straffelini G. Phase composition and wear behavior of NiTi alloys. J Mater Sci 2008;43:1701–10. [7] Hutchings IM. Tribology; friction and wear of engineering materials. London: Edward Arnold; 1992. [8] Glaser WA. Materials for tribology. Amsterdam, the Netherlands: Elsevier Science Publishers; 1992. [9] Ball A. On the importance of work hardening in the design of wear-resistant materials. Wear 1983;91:201–7. [10] Singh J, Alpas AT. Dry sliding wear mechanisms in a Ti50Ni47Fe3 intermetallic alloy. Wear 1995;181–183:302–11. [11] Arciniegas M, Casals J, Manero JM, Pena J, Gil FJ. Study of hardness and wear behaviour of NiTi shape memory alloys. J Alloys Compod 2008;460:213–9. [12] Li DY. A new type of wear-resistant material: pseudo-elastic TiNi alloy. Wear 1998;221:116–23. [13] Li DY, Liu R. The mechanism responsible for high wear resistance of pseudo-elastic TiNi alloy—a novel tribo-material. Wear 1999;225–229:777–83. [14] Li DY. Development of novel wear-resistant materials: TiNi-based pseudoelastic tribomaterials. Mater Des 2000;21: 551–5. [15] Liu R, Li DY. Indentation behavior of pseudoelastic TiNi alloy. Scr Mater 1999;41(7):691–6. [16] Liu R, Li DY. Experimental studies on tribological properties of pseudoelastic TiNi alloy with comparison to stainless steel 304. Metall Mater Trans A 2000;31:2773–83.

M A TE RI A L S CH A RACT ER IZ A TI O N 61 ( 20 1 0 ) 6 8 9 –6 9 5

[17] Li DY. Wear behaviour of TiNi shape memory alloy. Scr Mater 1996;34(2):195–200. [18] Jin J, Wang H. Wear resistance of Ni–Ti alloy. Acta Metall Sinica 1988;1:76–81. [19] Shida Y, Sugimoto Y. Water jet erosion behaviour of Ti–Ni binary alloys. Wear 1991;146:219–28. [20] Yan W. Mechanics examination on the wear behaviour of shape memory alloys. Fourth Australasian Congress on Applied Mechanics, Melbourne, Australia; 2005. [21] Yan W. Theoretical investigation of wear-resistance mechanism of superelastic shape memory alloy NiTi. Mater Sci Eng A 2006;427:348–55.

695

[22] Qian L, Sun Q, Xiao X. Role of phase transition in the unusual microwear behavior of superelastic NiTi shape memory alloy. Wear 2006;260:509–22. [23] Otsuka K, Ren X. Physical metallurgy of Ti–Ni-based shape memory alloys. Prog Mater Sci 2005;50:511–678. [24] Lin HC, Liao HM, He JL, Chen KC, Lin KM. Wear characteristics of TiNi shape memory alloys. Metall Mater Trans A 1996;28: 1871–7.