erosion and corrosion performance of low-carbon Stellite alloys

Materials and Design 78 (2015) 95–106

Contents lists available at ScienceDirect

Materials and Design journal homepage: www.elsevier.com/locate/matdes

Effects of molybdenum content on the wear/erosion and corrosion performance of low-carbon Stellite alloys Rong Liu a,b,⇑, Jianhua Yao b, Qunli Zhang b, Matthew X. Yao c, Rachel Collier c a

Department of Mechanical and Aerospace Engineering, Carleton University, 1125 Colonel By Drive, Ottawa, Ontario K1S 5B6, Canada Research Center of Laser Processing Technology and Engineering, Zhejiang University of Technology, Hangzhou, Zhejiang 310014, PR China c Kennametal Stellite Inc., 471 Dundas St E, Belleville, Ontario K8N 5C4, Canada b

a r t i c l e

i n f o

Article history: Received 5 January 2015 Revised 16 April 2015 Accepted 18 April 2015 Available online 18 April 2015 Keywords: Stellite alloy Intermetallic compound Carbide Wear/erosion and corrosion

a b s t r a c t The strengthening agents of Stellite alloys are commonly various carbides, but intermetallic compounds may play a similar role to the carbides. In this research two low-carbon Stellite alloys with high molybdenum content are developed and studied, which are modified version of Stellite 21. This particular elemental content combination results in large amounts of Co3Mo intermetallic compound precipitated in these alloys. The microstructures of the alloys are analyzed using SEM/EDX/XRD and DSC. The dry sliding wear resistance and solid-particle erosion resistance of the alloys are evaluated experimentally. The corrosion performance of the alloys in 3.5 wt.% sodium chloride (NaCl) aqueous solution is investigated under electrochemical tests. It is shown that the intermetallic compounds enhance hardness and wear resistance as the carbides do in Stellite alloys, but do not favor solid-particle erosion resistance due to their brittleness. The presence of the intermetallic compounds does not worsen corrosion resistance, compared to Stellite 21. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Stellite system is cobalt-based alloys, which contain mainly alloying elements chromium (Cr) and tungsten (W) or molybdenum (Mo) as well as a certain amount of carbon (C), addition to cobalt (Co) [1]. With unique chemical compositions, these alloys display exceptional properties including high-temperature strength, excellent wear/erosion and corrosion/oxidation resistance. Chromium is the main alloying element of Stellite alloys; it is added for resistance to corrosion and oxidation; it is also the predominant carbide former, that is, most of the carbides in Stellite alloys are Cr-rich, in the meanwhile, it provides strengthening to the solution matrix. Tungsten and molybdenum provide additional strength to the solution matrix as solute atoms in Stellite alloys because of their large atomic size that impedes dislocation flow [1], and also promote formation of W-rich carbides [2–5], for example, in Stellite 12 and Stellite 3, and formation of Mo-rich carbides [6], for instance, in Stellite 712 and Stellite 720, when present in large quantities (>5 wt.%). Molybdenum can also improve general corrosion resistance of Stellite alloys [1].

⇑ Corresponding author at: Department of Mechanical and Aerospace Engineering, Carleton University, 1125 Colonel By Drive, Ottawa, Ontario K1S 5B6, Canada. E-mail address: [email protected] (R. Liu). http://dx.doi.org/10.1016/j.matdes.2015.04.030 0261-3069/Ó 2015 Elsevier Ltd. All rights reserved.

In the CoCrMo Stellite alloy system, Stellite 21, which contains very small amount of C (0.25 wt.%), has a wide range of application mainly involving high temperature and corrosion, owing to its unique properties, such as creep resistance and mechanical strength at elevated temperatures, good resistance to corrosion [1,6–9], for example, valve trims for high pressure steam, oil and petrochemical processes, forging or hot stamping dies. It is also used on a large scale in medical implants and prosthetics [10– 14] because of its good mechanical and corrosion properties, excellent compatibility with human body environments, and better wear resistance than stainless steels. However, in some applications, for example, hip implants, which require metal-on-metal bearing, Stellite 21 is found deficient in wear resistance. Therefore, a modification on Stellite 21 with improved wear resistance and meanwhile maintaining good corrosion resistance was proposed, which motivated this research. As mentioned above, Mo is an important alloying element of Stellite alloys. It provides strengthening to the solution matrix as a solute and can also form Mo-rich carbides when present in large quantities (>5 up to 11 wt.%). In addition to these functions, Mo was reported to form intermetallic compounds Co3Mo in low-C Stellite alloy, which improved the wear resistance of the alloy [8]. The effects of Mo on the performance of Stellite alloys have been studied by some researchers [15–20]. For example, the wear and corrosion behavior of Stellite 6 hardfacing with addition of

R. Liu et al. / Materials and Design 78 (2015) 95–106

different contents of Mo was investigated experimentally and the results demonstrated that with increasing the Mo content the wear and corrosion resistance of the hardfacing was enhanced, which was attributed to the formation of Mo6C carbides [17]. The comparative studies between W-containing Stellite alloys and Mo-containing Stellite alloys found that the latter exhibited unusual combination of excellent abrasive, adhesive and erosive wear resistance and corrosion resistance in reducing environments. The Mocontaining Stellite alloys also had adequate cracking resistance compared with the W-containing Stellite alloys [18]. Because of these advantages of Mo, in this research, attempt was made to add large amounts (>11 wt.%) of Mo in low-C Stellite alloys to create new alloys which were expected to have better wear resistance than and comparable corrosion resistance to Stellite 21. These alloys were investigated under dry-sliding wear, solid-particle erosion, electrochemical corrosion tests, Stellite 21 and Stellite 6 were also tested under the same conditions for comparison. 2. Materials and methods 2.1. Microstructure characterization The chemical compositions of the new alloys, together with those of Stellite 21 and Stellite 6, are given in Table 1. The new alloys have similar Cr and Mo contents, and contain very small C content. Alloy 2 contains slightly high C content than alloy 1 and additional element of niobium (Nb). Addition of Nb is to enhance the strength of the alloy by forming niobium carbide [21]. The alloy specimens are cast products. The microstructural analyses of the alloys were conducted on the IGMA HV-01-043 (Carl Zeiss SMT Pte Ltd, Germany) SEM, equipped with the Bruker Nano Xflash Detector 5010 EDX system (Bruker, Germany) for elemental analysis and quantitative mapping. The phases in the microstructures were identified using the X’Pert PRO X-ray photoelectron spectroscopy (PNAlytical Ltd, Holland) with Cu Ka radiation, which was operated at 40 kV and 40 mA to generate monochromatic Cu Ka radiation with a wavelength of 0.154 nm. To help understand the microstructures, DSC analysis was conducted on the new alloys with a DSC 404C instrument which can detect and qualify almost all calorimetric effects occurring in materials. The maximum heating temperature used in this experiment was 1500 °C which was determined based on the melting point of Co. The alloy samples were heated to 100 °C at a rate of 10 °C/ min and kept for 20 min, then heated again up to the ultimate temperature, kept for 10 min, and finally cooled down to room temperature with the same rate. The phase transformation behavior of the material during the heating and cooling process is characterized by the DSC curve that can be obtained in the test.

under a normal force of 10 N against a static disk that was the specimen with dimensions 12  12  4 mm. The pin used was a ball having a radius of 2.5 mm and was made of a composite containing 94 wt.% WC and 6 wt.% Co, with the hardness of HV1534. During the test the specimen disk was placed horizontally with the center at a distance of 3 mm away from the vertical axis of the pin shaft. The pin (ball) was spinning at a speed of 350 rpm, corresponding to a linear speed of 110 mm s1. As the result of friction/wear, a 6 mm diameter circular wear track was generated on the specimen surface, which represents the wear loss of the specimen material. The test duration of each specimen was set to 2.5 h, leading to a total sliding length of 990 m. Three specimens were tested for each alloy to verify the wear loss results. For each alloy, the wear loss was evaluated by calculating the volume of the wear track. Four locations were selected uniformly along the wear track to calculate the average cross-sectional area. The cross-sectional profiles of the wear tracks were simulated utilizing a D150 Surface Profile Measuring System. The plots in Fig. 1 are the examples of the cross-sectional profiles of wear track for the new alloys. The cross-sectional areas of the wear tracks at the selected locations were calculated automatically by the associated software of the D150 Surface Profile Measuring System. The volume of each wear track was then calculated using the average cross-sectional area multiplied by the periphery length of the wear track, pD (D = 6 mm). The worn counterparts (pins) were also examined. The unused pin had a sphere shape with a radius (R) of 2.5 mm. The wear in the test resulted in an approximately circular area on the pin surface. By measuring the diameter (d) of the circle, as illustrated in Fig. 2, using the DEKTAK 150 Surface Profile Measuring System, the volume or wear loss of the pin can be estimated by the following equations: 2

V ¼ ph

(a)

  h R 3

5000 0

Wear track depth (nm)

96

-5000

0

500

Element Cr

Alloy 1 Alloy 2 Stellite 21 Stellite 6

27 24.2 27 28.5

W

Mo

C

Fe

Ni

Si

Mn

4.5

11 11.8 5.5 1.5

0.25 0.35 0.25 1.2

3 1 3 3

2.75 3.8 2.75 3

1 0.45 1 1.5

1 0.52 1 1

Nb

2500

2000

2500

-20000 -25000 -30000 -35000

Wear track width (µm)

(b) 5000 0 -5000 0

Wear track depth (nm)

Alloy

2000

-15000

-40000

Table 1 Chemical compositions (wt.%, Co in balance) of the new alloys and existing Stellite alloys.

1500

-10000

2.2. Pin-on-disk wear test Wear tests of the alloys were performed on a Neoplus pin-ondisk tribometer in dry sliding condition, according to ASTM: G99 – 05(2010). This apparatus used a rotating pin that was pressed

1000

500

1000

1500

-10000 -15000 -20000 -25000 -30000 -35000 -40000 -45000

2.07

-50000

Wear track width (µm)

Fig. 1. Cross-sectional profiles of wear track: (a) alloy 1 and (b) alloy 2.

R. Liu et al. / Materials and Design 78 (2015) 95–106

97

Fig. 2. Schematic showing of wear loss of a ball pin.

and

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  2 d h ¼ R  R2  : 2

2.3. Solid-particle erosion test The specimens for solid-particle erosion test were machined to rectangular blocks in 75  25  3 mm. The specimen surfaces were polished with silicon carbide (SiC) abrasive paper having a size of 320# grit. The erosion test was conducted in ambient temperature environment according to ASTM: G76 – 13. The distance between the nozzle head and specimen surface was set to 5 mm, which was determined based on the hardness of the tested alloys (HRC40), so that reasonable mass losses could be identified on the specimens for the set test duration of time. The specimen holder contains a screw that allows for the specimen to be adjusted and rotated at various angles, in order to achieve different particle impingement angles, a, in the erosion test, as schematically shown in Fig. 3. Two typical angles, a = 30° and 90°, were used in this experiment to investigate the erosion behavior of the alloys under acute and normal particle impact. The duration time of each test was set to 10 min and the particle impact velocity was controlled at 84 m s1. The erosion test unit containing erodent particles (grit) was placed on a scale throughout the test process; the mass of this unit was recorded before and after running each test. Thus the amount of grit used in each test could be quantified by taking the difference between these two mass values. The grit (erodent particles) used for this test was alumina and angular in shape, as shown by the SEM image in Fig. 4, having the hardness about HRC70. Before the test each specimen was weighed and after each test the specimen was cleaned and weighed again, thus the weight loss (erosion

Fig. 3. Schematic diagram depicting the impingement angle of solid-particle erosion.

Fig. 4. SEM morphology of erodent particles – angular alumina powder.

loss) of the specimen material due to the erosion could be obtained. Three specimens were tested for each alloy in each test condition (impingement angle). The final results of erosion loss for each alloy under each test condition are the averages of the three test results. 2.4. Hardness test Hardness is used quite often in the field of wear resistance as a qualitative indicator for materials. To better understand the wear/ erosion behavior of the new alloys, the overall hardness of these alloys, together with Stellite 21 and Stellite 6, was investigated on a Wilson Rockwell machine. Five tests were performed on each alloy specimen and the average results were taken as the hardness values of the alloys. The hardness of individual phases of the alloys was also measured using a Microhardness Tester Unit, Model SMTX7 Dual Indenter. 2.5. Electrochemical corrosion test The electrochemical corrosion tests, including potentiodynamic and cyclic polarization tests, were conducted on the alloys in 3.5 wt.% sodium chloride (NaCl) aqueous solution using a Gamry™ PC14™ potentiostat system, according to ASTM: G59 – 97(2014), to investigate the general corrosion resistance of the alloys. This solution is a common corrosive medium that is used to rank materials for corrosion resistance in industry. The alloy specimens were machined to round plates with a diameter of 16 mm and thickness of 1.6 mm and lightly polished with1 lm diamond suspension on specific polishing cloth, and then thoroughly cleaned in an ultrasonic bath. Potentiodynamic polarization test is one of electrochemical methods, which has been adopted as a common testing method to investigate the corrosion properties of metals. Polarization resistance of metals can be related to the rate of general corrosion for metals at or near their corrosion potential Ecorr [22]. Polarization resistance measurement provides an accurate and rapid approach to obtaining the general corrosion rate of the metal under test, in terms of ASTM: G102 – 89(2010). Cyclic potentiodynamic polarization test was performed on the alloys to investigate their localized corrosion susceptibility, that is, pitting and crevice corrosion in the corrosive solution. The difference between potentiodynamic polarization test and cyclic potentiodynamic polarization test is that the latter has an additional reverse scan which can provide a hysteresis loop. This hysteresis loop is an evidence of revealing the susceptibility of the tested material to localized corrosion.

98

R. Liu et al. / Materials and Design 78 (2015) 95–106

Fig. 5. Electrochemical cell setup for potentiodynamic polarization test.

Fig. 6. SEM microstructure of alloy 1: (a) at low magnification and (b) at high magnification.

The Gamry™ PC14™ potentiostat system consists of an electrochemical cell and a control system that includes a data acquisition and analysis system. The electrochemical cell provides the corrosive environment for the measurement of electrochemical parameters, and consists of four primary components: cell body, electrolyte, electrodes, and devices monitoring and controlling the environment. Electrodes include a working electrode (testing specimen), auxiliary or counter electrodes (two carbon rods), and a reference electrode, as shown in Fig. 5. The specimen is mounted on a Teflon holder and the testing surface is pressed against an Oring, which results in an exposed area of the surface to the electrolyte through the aperture about 10 mm in diameter. In order to measure and control the potential across the interface between the specimen surface and electrolyte, termed as interfacial potential, a counter electrode is introduced, which is electrically connected to the working electrode through the electrolyte. To minimize the variation in the interfacial potential, a reference electrode is introduced between the working and counter electrodes. In this research a saturated mercury-mercury chloride (calomel) electrode (SCE) was used as the reference electrode and all potential values obtained were against SCE. During the potentiodynamic polarization test, the specimen was immersed in the surrounding solution (corrosive medium) for a few minutes until the solution reached equilibrium to establish the steady state open-circuit potential Eocp [22]. The polarization scan started at 300 mV versus open circuit potential and ended at 1500 mV versus SCE with a scan rate of 0.1667 mV/s. Experimental data were

Fig. 7. SEM microstructure of alloy 2: (a) at low magnification and (b) at high magnification.

99

R. Liu et al. / Materials and Design 78 (2015) 95–106

(a)

cps/eV

50 40 30

Mo C Cr Co

Mo

Cr

Co

20 10 0 2

(b)

4

6

8

10

12

14

8

10

12

14

8

10

12

14

keV

cps/eV

50 40 30

Mo C Cr Co

Cr

Mo

Co

20 10 0 2

4

6

keV

(c)

cps/eV

50 40 30

Mo C Cr Co

Mo

Si

Cr

Co

20 10 0 2

(d)

4

6

keV

cps/eV

50 40 30

Mo Nb C Cr Co

Mo Si Nb

Cr

Co

20 10 0 2

4

6

8

10

12

14

keV Fig. 8. EDX spectra: (a) gray phase, (b) black phase of alloy 1, (c) white phase of alloy 1and (d) white phase of alloy 2.

100

R. Liu et al. / Materials and Design 78 (2015) 95–106

Table 2 Element contents (at.%) of the EDX spectra. Spectrum

Fig. Fig. Fig. Fig.

8a 8b 8c 8d

Element Co

Cr

Mo

C

Si

Nb

53.17 17.62 29.82 32.59

27.66 51.27 21.90 21.77

4.92 6.59 22.32 14.30

14.25 24.53 23.67 23.97

2.30 2.77

4.59

collected and then analyzed with the Gamry Electrochemistry DC105™ software.

3. Experimental results discussion 3.1. Microstructure and phases The SEM microstructures of the new alloys are shown in Figs. 6 and 7. The microstructure of alloy 1 has three distinct phases in gray, black and white, respectively, see Fig. 6, but alloy 2 only has two distinct phases in gray and white, respectively, and no black phase is observed in this alloy, see Fig. 7. To identify the phases in the microstructures, EDX was performed on each of the phases; the corresponding spectra are presented in Fig. 8 and the element contents are summarized in Table 2. It should be pointed out that the EDX system is unable to accurately quantify carbon content but the relative contents of metallic elements are correct. It can be seen that the gray phase is Co-rich, the black phase is Cr-rich, the white phase of alloy 1 is Mo-rich, and the white phase of alloy 2 is also Mo-rich but contains additional element Nb. To further investigate the phases of the microstructures, XRD was conducted on the alloys; the XRD patterns are presented in Fig. 9. It is shown that the gray phases of both alloys are fcc Co solid solution; the Cr-rich phase of alloy 1 is Cr23C6 carbide and the Mo-rich phase is Co3Mo intermetallic compound. NbC carbide phase was detected in alloy 2. However, it is interesting to notice that there are no NbC carbides observed in the microstructure of alloy 2, even in the image at very high magnification (20k), see Fig. 10, although the XRD analysis has detected this phase in the alloy, see Fig. 9. To further investigate the presence of NbC in alloy 2, EDX line-scan analysis for Nb and C elements in the microstructure of this alloy was conducted and the results are presented in Fig. 10. It is shown that high Nb is present in the white phase and C is distributed uniformly within the white and gray phases. To confirm the presence of NbC in the white phase, EDX elemental mapping of Nb was performed on alloy 2 and the results are reported in Fig. 11, along with the mapping of Mo and Co for comparison. It is evident that Nb is

Co3Mo

Cr23C6

fcc

3100

NbC

2900 2700 Alloy 2

Counts

2500 2300 2100 1900 1700

3.2. Wear loss

1500 1300

rich and scattered within the entire white phase where Mo is also rich while Co is rich in the gray phase which is solid solution matrix. Although NbC carbides are not observed in the microstructure of alloy 2, the EDX analyses imply that these carbides are tiny and mixed with Co3Mo intermetallic compounds, dispersing within the white phase. The carbides of Stellite alloys are generally Cr-rich, but the presence of Nb can alter the carbides in these alloys. Element Nb has stronger affinity to C than element Cr [23] so that the very small amount of C (0.35 wt.%) in alloy 2 forms carbides easier with Nb than with Cr, which results in lack of Cr23C6 carbide in this alloy. Youdelis and Kwon [24] studied the carbide phases in Stellite 21 and they found that addition of small amount (1.5 at.%) of Nb in this alloy resulted in formation of very fine NbC which replaced coarse Cr23C6. Their findings are consistent with the results of the present research on alloy 2. For comparison, the SEM microstructures of Stellite 21 and Stellite 6 are presented in Fig. 12. Different magnifications were used in the SEM images for these two alloys because the former consists of nearly entire solid solution with minor intermetallic compounds and carbides due to both low C content and low Mo content; at lower magnification the intermetallic compounds and carbides are not clearly visible. The SEM image at higher magnification shows that Stellite 21 contains minor Co3Mo intermetallic compounds (white) and Cr23C6 carbides (black) embedded in the solution matrix. Stellite 6 is composed of primary solid solution matrix with moderate amounts of Cr7C3/Cr3C2 carbides (black) and minor (Co, W)6C carbides (white) embedded, because of the higher C content. Furthermore, the volume fractions of each phase in the alloys were estimated using the SEM image analysis software; the results are presented in Table 3. It is shown that the volume fractions of solid solution (gray phase) in the two new alloys are close, but it is not known if the amounts of Co3Mo intermetallic compound in the two alloys are the same because the amount of NbC carbide within the intermetallic compounds of alloy 2 cannot be identified due to the same color in SEM image. Nevertheless, it is evident that the amount of intermetallic compounds is much larger than that of carbides in the two alloys, since the C contents of the alloys are all very low. The volume fraction of carbides in Stellite 6 is slightly larger than that of intermetallic compounds and carbides in the new alloys, but Stellite 21 contains very small amounts of intermetallic compounds and carbides due to its low C and Mo contents. The DSC curves of the new alloys are plotted in Fig. 13. It is shown that the melting point of alloy 1 (1368.9 °C) is slightly higher than that of alloy 2 (1328.2 °C) and the former has higher solidification temperature (1334.5 °C) than the latter (1260.6 °C), but the latter has much larger solidification range (176.2 °C) than the former (67.6 °C). According to the C contents, these two alloys are all hypoeutectic, with Co solid solution being the primary phase. On the curves of cooling, alloy 1 has two exothermic peaks; the first (1379.6 °C) represents the solidification of the primary Co solid solution phase and the second (1330.4 °C) indicates the formation of the eutectic phase that includes Co solid solution, Co3Mo intermetallic compounds and Cr23C6 carbides. Alloy 2 has three exothermic peaks; the first small peak (1407 °C) represents the segregation of Co solid solution; the second (1348.6 °C) stands for the solidification of Co solid solution and the last peak denotes the precipitation of the eutectic phase that contains Co solid solution, Co3Mo intermetallic compounds and NbC carbides.

Alloy 1 20

30

40

50

60

70

80

Angle (2θ θ) Fig. 9. XRD patterns of the new alloys.

90

100

110

For each alloy, the average wear loss was obtained based on three test results. The average wear losses of the new alloys, Stellite 21 and Stellite 6 are illustrated in Fig. 14 with the standard deviations of statistics analysis provided representing the test

R. Liu et al. / Materials and Design 78 (2015) 95–106

101

Fig. 10. EDX line-scan spectra for Nb and C elements of alloy 2.

Fig. 11. EDX elemental mapping of alloy 2 (light represents the rich region of mapped element).

errors. It is shown that the new alloys exhibit much higher wear resistance than Stellite 21 and comparable wear resistance to Stellite 6. Compared with the wear losses of the tested alloys, the wear losses of the pins against each alloy specimen are trivial.

This may be due to the fact that the pin material (HV1534) is much harder than that of the tested alloys (HV410  HV630). To further understand the wear test results, the worn surfaces of the specimens were examined under SEM. No distinct difference

102

R. Liu et al. / Materials and Design 78 (2015) 95–106

Fig. 13. DSC curves: (a) alloy 1 and (b) alloy 2.

Fig. 12. SEM microstructures: (a) Stellite 21 at 6000 and (b) Stellite 6 at 500.

between the worn surfaces of the new alloys was observed, but the difference between the worn surface of Stellite 21 and that of the new alloys is obvious, as shown in Fig. 15. The wear track of the former is much wider and smoother than that of the latter (alloy 1), indicating that the former is less resistant against wear, in other words, material was easier removed from its surface. On the contrary, the wear track of the latter exhibits typical plough scars, which means that large plastic deformation occurred in this surface against wear. It is also seen that the wear track of the new alloy contains many microcracks, see Fig. 15b. This is because this alloy has large volume fraction of hard but brittle intermetallic compounds; under mechanical attack of wear, these particles were vulnerable, which resulted in cracks. Therefore these two alloys exhibited different wear mechanisms; the wear loss of Stellite 21 was due to the wearing off of the solution matrix, while that of the new alloy was caused mainly by fracture of the intermetallic compounds.

Fig. 14. Wear losses of the new alloys and existing Stellite alloys under sliding wear.

For sliding wear resistance, as demonstrated by the wear test results, the new alloys with large amounts of Co3Mo intermetallic compounds have much better wear resistance than Stellite 21, and

Table 3 Volume fractions of each phase in the new alloys and existing Stellite alloys. Alloy

Alloy 1 Alloy 2 Stellite 21 Stellite 6

Phase White phase (Mo-rich phase or W-rich phase)

Black phase (Cr-rich phase)

Gray phase (Co solid solution)

8.92 11.31 (including NbC) 0.66 0.15

2.49 0 1.28 15.57

88.59 88.69 98.06 84.28

103

R. Liu et al. / Materials and Design 78 (2015) 95–106 45

Rockwell hardness (HRC)

40 35 30 25 20 15 10 5 0

Alloy 1

Alloy 2

Stellite 21

Stellite 6

Fig. 16. Hardness of the new alloys and existing Stellite alloys.

Fig. 17. Erosion weight losses of the new alloys and existing Stellite alloys at particle impact velocity of 84 m s1.

hardness results were not reliable, due to the fact that the intermetallic compound and carbide were all in the eutectic phase, mixed with Co solid solution phase, and also they were small in size. This caused the problem that the hardness tested on the intermetallic compound phase and carbide phase was influenced by the presence of Co solid solution phase. However, the macro hardness results of the alloys imply that the intermetallic compound is harder than the carbide, if comparing hardness between Stellite 6 and the new alloys in Fig. 16, because the solid solution in the new alloys (HV334) is softer than that in Stellite 6 (HV411), which were measured using the Microhardness Tester Unit. 3.3. Erosion rate Fig. 15. SEM morphologies of worn surface: (a) Stellite 21 and (b) alloy 1.

comparable wear resistance to Stellite 6 which has similar volume fraction of carbides to that of intermetallic compounds in the new alloys. It is generally accepted that the harder the material, the higher the wear resistance is, for most of metallic materials. The average results of hardness with the maximum error less than 2% for each alloy are reported in Fig. 16. It is shown that the new alloys have much higher hardness than Stellite 21, which indicates that Co3Mo intermetallic compound can enhance the hardness of Stellite alloys. Alloy 2 is harder than Stellite 6 but alloy 1 is slightly softer. Attempt was made to measure the hardness of Mo-rich intermetallic compound in the new alloys and Cr-rich carbide in Stellite 6 using the Microhardness Tester Unit, but the obtained

The erosion rate due to solid particle impact can be expressed as the ratio of the mass (weight) loss or volume loss of target material to the mass of erodent particles. The weight losses of target materials and the mass of the erodent particles were obtained after each of the erosion tests. The average erosion rates of the new alloys expressed as the ratio of the mass weight loss of target material to the mass of erodent particles are illustrated in Fig. 17, together

Table 4 Densities (kg/m3) of the new alloys and existing Stellite alloys. Alloy 1

Alloy 2

Stellite 21

Stellite 6

8420

8400

8450

8380

104

R. Liu et al. / Materials and Design 78 (2015) 95–106

Fig. 18. Erosion volume losses of the new alloys and existing Stellite alloys at particle impact velocity of 84 m s1.

on the mechanisms of solid-particle erosion have been reported in literature [25–33], however, the two mechanisms: the cutting and extensive/repeated deformation of target material, which correspond to ductile and brittle erosion, respectively, are mostly accepted. According to these mechanisms, the total erosion loss can be considered as the sum of cutting wear and deformation wear [27]. When the particles impact at acute angles, cutting wear dominates the total erosion loss, while with increasing the particle impingement angle, deformation wear prevails [28]. The resistance of a material to erosion wear depends on ductility and hardness. Cutting wear induces plastic deformation at the sub-surface; when the accumulated plastic deformation reaches the material limit with continuous particle impact, fracture occurs at the sub-surface, leading to the material removal. The ductility of the target material determines its capacity of accommodating plastic deformation, while the hardness provides the resistance to deformation of the target material. Therefore the synergetic effect of the hardness and ductility of target material determines the erosion resistance of the target surface. The new alloys are harder but more brittle than Stellite 21 due to the large volume fraction of intermetallic compounds. This is why the new alloys are not better than Stellite 21 in erosion resistance, which differs from sliding wear. Between the new alloys and Stellite 6, although the volume fraction of intermetallic compounds in the former is close to that of carbides in the latter, the latter is better in erosion resistance. This implies that the carbides are stronger than the intermetallic compounds, in other words, the latter is more brittle than the former, so that the intermetallic compounds are more vulnerable under particle impact, resulting in more erosion loss. 3.4. Polarization curve

Fig. 19. otentiodynamic polarization curves of the new alloys and existing Stellite alloys in 3.5 wt.% NaCl solution.

Table 5 Summary of polarization test results of the new alloys and existing Stellite alloys in 3.5 wt.% NaCl solution. Alloy

Ecorr (VSCE)

Icorr (lA/cm2)

Rp (kX cm2)

Alloy 1 Alloy 2 Stellite 21 Stellite 6

0.325 0.215 0.210 0.320

0.004 0.009 0.004 0.014

3.418  105 1.408  105 3.398  105 8.978  104

with the data of Stellite 21 and Stellite 6 which were obtained under the same test conditions for comparison. As a statistics analysis measurement, the values of standard deviation were computed for error estimation and provided with the average erosion rates. The experimental data of weight losses were converted to the volume losses through the target materials’ densities in Table 4, and the results are presented in Fig. 18. Different from sliding wear, the increased Mo content did not improve the erosion resistance of low-C Stellite alloy, in particular, at normal particle impingement angle, the new alloys were worse than Stellite 21 and Stellite 6. This reveals that the intermetallic compounds in the new alloys do not favor the alloys in resisting erosion and they also do not have the beneficial effect as the carbides in Stellite 6 in regard of erosion resistance. The mechanisms of solid-particle erosion are much more complex than those of sliding wear, in particular, when particle impingement angle is concerned. Various studies and discussions

The potentiodynamic polarization curves of the new alloys tested in 3.5 wt.% NaCl solution are plotted in Fig. 19, along with those of Stellite 21 and Stellite 6 for comparison. These curves describe the response of the current to the applied potential on the electrochemical system. It can be seen that all the alloys have a passive region where the current remains unchanged as the potential is increased, indicating that protective oxide films formed on the alloy surfaces against further corrosion during the electrochemical tests. However, with the potential increased, the oxide films broke, corrosion occurred again on the alloys. In this case, with the raise of the potential, the current increased, characterized by the transpassive regions on the polarization curves in Fig. 19. Nevertheless, all the alloys can form second passive region, as shown on the polarization curves. The low-C Stellite alloys behaved similarly in the electrochemical tests, while Stellite 6 exhibited differently, as its passive region was smaller, but it could quickly form second passive region when the oxide film was broken. Utilizing the Tafel extrapolation approach, the important parameters such as corrosion potential Ecorr, current density Icorr, and polarization resistance Rp, which characterize the corrosion behavior of the tested alloys, can be determined through analysis of the polarization curves [34]. Corrosion potential represents starting of corrosion occurrence on the material in the tested solution. The larger the value of corrosion potential, the better of the material in corrosion resistance is. Current density is a measure of corrosion intensity of the material in the tested solution. The higher the current density, the severe the corrosion is. Polarization resistance means the resistance of the material to corrosion in the tested solution. The larger the value of polarization resistance, the more resistant the material to the corrosive environment is. The calculated values of these parameters from the polarization curves of the alloys are summarized in Table 5. The original data of current were collected in the unit ampere, and they

R. Liu et al. / Materials and Design 78 (2015) 95–106

105

Fig. 20. Cyclic polarization curves of tested alloys in 3.5 wt.% NaCl solution: (a) alloy 1, (b) alloy 2, (c) Stellite 21 and (d) Stellite 6.

were divided by the exposed specimen surface area thus converted to absolute values. The exposed surface area of the specimen in the solution was calculated to be 78.5 mm2. From Table 5, obviously, the new alloys have comparable corrosion resistance to Stellite 21, and higher corrosion resistance than Stellite 6. According to the Ecorr values of the alloys, alloy 1 easier gets corroded than alloy 2 and Stellite 21, but once corrosion starts, it occurs more severe on alloy 2 because of the higher Icorr value. The Rp values further confirm that alloy 2 is less corrosion-resistant than alloy 1 and Stellite 21. The cyclic potentiodynamic polarization curves of the new alloys, Stellite 21 and Stellite 6, tested in 3.5 wt.% NaCl solution are plotted in Fig. 20. The reverse scan or backward curve is called hysteresis loop. If the hysteresis loop is electropositive, that is, the current density of backward curve is smaller than that of forward curve at the same potential, the tested alloy exhibits localized or pitting corrosion resistance. The more the offset of the electropositive loop, the better the localized corrosion resistance of the alloy is. In terms of this concept, all the alloys in Fig. 20 lack pitting corrosion resistance, because the new alloys and Stellite 21 almost have no hysteresis loop, this is, corrosion current in backward process maintains the same as forward process, and Stellite 6 shows electronegative hysteresis loop, that is, the current density of backward curve is larger than that of forward curve at the same potential. Stellite alloys commonly contain high Cr content for corrosion and oxidation resistance by forming protective Cr2O3 film. This is reflected by the passive region on the polarization curves in Fig. 19. The passive region of Stellite 6 is smaller than that of the other alloys, which implies that the Cr2O3 film on this alloy is

weaker and more vulnerable. This is because Stellite 6 contains high C content which results in large amounts of Cr-rich carbides. Thus the Cr concentration in the solid solution is reduced, leading to less Cr2O3 film, compared with the new alloys and Stellite 21. 3.5. Discussion The Mo-rich intermetallic compounds in low-C Stellite alloys exhibited different effects on sliding wear and on solid-particle erosion. This is because these two wear modes involve different mechanisms. Sliding wear generally occurs by three mechanisms: oxide control, contact stress and subsurface fatigue. Therefore, the Stellite alloys which perform well under sliding conditions do so either by virtue of their oxidation behavior or their ability to resistance deformation and fracture [1]; thus their sliding-wear properties are controlled predominantly by the overall hardness and resistance to deformation. With the point of this view, the Mo-rich intermetallic compounds in low-C Stellite alloys did benefit sliding wear resistance. For solid-particle erosion, with the complex mechanisms, however, the general abrasion mechanism may not be warranted. For example, in solid-particle erosion, ductility of Stellite alloys and brittleness of hard strengthening phases can be a critical factor, hence bulk hardness has very little effect on resistance to the erosion [1]. 4. Conclusions Molybdenum plays an important role in Stellite alloys. When present in large quantities (up to 11 wt.%), it induces large

106

R. Liu et al. / Materials and Design 78 (2015) 95–106

amounts of Co3Mo intermetallic compound in low-C Stellite alloys, which has a laminar structure. These intermetallic compounds enhance the hardness and sliding-wear resistance of Stellite alloys as the carbides in Stellite alloys do, but they do not favor the solidparticle erosion resistance of the alloys due to brittleness. The presence of intermetallic compound Co3Mo does not affect the corrosion resistance of low-C Stellite alloy in 3.5 wt.% NaCl solution, which is better than the carbides. For the applications involving sliding wear and corrosion, the modified version of Stellite 21 with increased Mo content is recommended. However, carbide-strengthened Stellite alloys such as Stellite 6 may be the best option for solid-particle erosion conditions. Acknowledgements The authors are grateful for both financial and in-kind supports of Kennametal Stellite Inc., and also for financial supports from the National Natural Science Foundation of China (51475429) and Youth Foundation Project of Natural Science Foundation of Zhejiang Province (LQ13E050012). References [1] J.R. Davis, Nickel, Cobalt, and Their Alloys, ASM International, Materials Park, 2000, pp. 362–370. [2] S. Kapoor, R. Liu, X.J. Wu, M.X. Yao, Microstructure and wear resistance relations of Stellite alloys, Int. J. Adv. Mater. Sci. 4 (3) (2013) 231–248. [3] R. Liu, J.H. Yao, Q.L. Zhang, M.X. Yao, R. Collier, Sliding wear and solid-particle erosion resistance of a novel high-tungsten Stellite alloy, Wear 322–323 (2015) 41–50. [4] R. Liu, X.J. Wu, S. Kapoor, M.X. Yao, R. Collier, Effects of temperature on the hardness and wear resistance of high-tungsten Stellite alloys, Metall. Mater. Trans. A 46 (2) (2015) 587–599. [5] S. Nsoesie, R. Liu, K.Y. Chen, M.X. Yao, Erosion resistance of Stellite alloys under solid-particle impact, J. Mater. Sci. Eng. B 3 (9) (2013) 555–566. [6] R. Liu, M.X. Yao, CRC Handbook on Aerospace and Aeronautical Materials, CRC Press: Taylor & Francis, 2012, pp. 151–235. [7] A. Aizaz, P. Kumar, Properties of Stellite alloy No. 21 made via pliable powder technology, Met. Powder Rep. 40 (9) (1985) 507–510. [8] P. Huang, R. Liu, X.J. Wu, M.X. Yao, Effects of molybdenum content and heat treatment on mechanical and tribological properties of a low-carbon Stellite alloy, J. Eng. Mater. Technol. 129 (4) (2007) 523–529. [9] R. Liu, M.X. Yao, P.C. Patnaik, X.J. Wu, An improved wear-resistant PTA hardfacing - VWC/Stellite 21, J. Compos. Mater. 40 (24) (2006) 2203–2215. [10] A. Matthies, R. Underwood, P. Cann, K. Ilo, Z. Nawaz, J. Skinner, A.J. Hart, Retrieval analysis of 240 metal-on-metal hip components, comparing modular total hip replacement with hip resurfacing, J. Bone Joint Surg. Br. 93 (3) (2011) 307–314. [11] P.F. Doom, J.M. Mirra, P.A. Campbell, H.C. Amstutz, Tissue reaction to metal on metal total hip prostheses, Clin. Orthop. Relat. Res. 329 (1996) 187–205.

[12] J. Fisher, X.Q. Hu, T.D. Stewart, S. Williams, J.L. Tipper, E. Ingham, M.H. Stone, C. Davies, P. Hatto, J. Bolton, M. Riley, C. Hardaker, G.H. Isaac, G. Berry, Wear of surface engineered metal-on-metal hip prosthese, J. Mater. Sci. Mater. Med. 15 (3) (2004) 225–235. [13] S. Williams, J.L. Tipper, E. Ingham, M.H. Stone, J. Fisher, In vitro analysis of the wear, wear debris and biological activity of surface-engineered coatings for use in metal-on-metal total hip replacements, Proc. Inst. Mech. Eng. H 217 (3) (2003) 155–163. [14] A.W. Hodgson, S. Mischler, V. Rechenberg, S. Virtanen, An analysis of the in vivo deterioration of Co–Cr–Mo implants through wear and corrosion, Proc. Inst. Mech. Eng. H 221 (3) (2007) 291–303. [15] T. Sawada, K. Yanagimoto, Effects of chromium and molybdenum contents on properties of Ni–Cr–Mo–B alloy with high corrosion and wear resistance consolidated by HIP, J. Jpn. Soc. Powder Powder Metall. 59 (4) (2012) 167–172. [16] K.T. Oh, K.N. Kim, M. Lee, Y.S. Park, Corrosion wear of high molybdenum and nitrogen stainless steel for biomedical applications, J. Electrochem. Soc. B 149 (4) (2002) 146–153. [17] J.C. Shin, J.S. Kim, D.Y. Lee, J.W. Yang, J.K. Yoon, D.H. Rho, J.K. Lee, Wear and corrosion properties of the Stellite 6 alloys added with molybdenum, J. Korean Inst. Metal Mater. 40 (10) (2002) 1104–1110. [18] M.X. Yao, J.B.C. Wu, Y.S. Xie, Wear, corrosion and cracking resistance of some W- or Mo-containing Stellite hardfacing alloys, Mater. Sci. Eng. A 407 (1–2) (2005) 234–244. [19] M. Zhang, Y. Ma, X.Y. Zhang, A.L. Fan, B. Tang, Corrosive-wear behavior of Mo modified Ti6Al4V alloy by wear-electrochemical noise method, Mocaxue Xuebao/Tribol. 32 (3) (2012) 301–305. [20] B. Uyulgan, H. Cetinel, I. Ozdemir, C. Tekmen, S.C. Okumus, E. Celik, Friction and wear properties of Mo coatings on cast-iron substrates, Surf. Coat. Technol. 174–175 (2003) 1082–1088. [21] M.C. Chaturvedi, Precipitation behavior of NbC in a CoNiCr alloy, Metallography 13 (2) (1980) 167–183. [22] N. Perez, Electrochemistry and Corrosion Science, Kluwer Academic Publishers, Assinippi Park, 2004, pp. 20–70. [23] W.D. Callister Jr., D.G. Rethwisch, Materials Science and Engineering—An Introduction, John Wiley & Sons, Inc., New York, 2010, pp. 18–43. [24] W.V. Youdelis, O. Kwon, Carbide phases in cobalt base superalloy: role of nucleation entropy in refinement, Metal Sci. 17 (1983) 379–384. [25] G.P. Tilly, A two stage mechanism of ductile erosion, Wear 23 (1) (1973) 87– 96. [26] J.G.A. Bitter, A study of erosion phenomena. Part I, Wear 6 (1) (1963) 5–12. [27] H. Rayleigh, The sand blast, Nature 93 (1912) 188–190. [28] J.H. Neilson, A. Gilchrist, Erosion by stream of solid particles, Wear 11 (1968) 111–122. [29] I.M. Hutchings, Introduction to the microscopy of erosion, J. Microsc. 130 (3) (1983) 331–338. [30] G.W. Stachowiak, A.W. Batchelor, D.G. Rethwisch, Engineering Tribology, Elsevier Butterworth-Heinemann, Oxford, 2005, pp. 501–551. [31] E. Akbarzadeh, E. Elsaadawy, A.M. Sherik, J.K. Spelt, M. Papini, The solid particle erosion of 12 metals using magnetite erodent, Wear 282–283 (2012) 40–51. [32] D.R. Andrews, An analysis of solid particle erosion mechanisms, J. Phys. D 14 (11) (1981) 1979–1991. [33] S.G. Sapate, A.V.R. Rao, N.K. Gary, Solid particle erosion studies of weld hardfacing deposits, Mater. Manuf. Process. 15 (5) (2000) 747–759. [34] D.A. Jones, Principles and Prevention of Corrosion, Macmillan Publishing Company, New York, 1992, pp. 27–52.