- Email: [email protected]
Study of the effects of austempering temperature and time on scuffing behavior of austempered Ni–Mo–Cu ductile iron
Wear 290–291 (2012) 99–105
Contents lists available at SciVerse ScienceDirect
Wear journal homepage: www.elsevier.com/locate/wear
Study of the effects of austempering temperature and time on scuffing behavior of austempered Ni–Mo–Cu ductile iron J.M. Han a,b,n, Q. Zou a, G.C. Barber a, T. Nasir c, D.O. Northwood c, X.C. Sun d, P. Seaton d a
Oakland University, Rochester Hills, MI, USA University of Turabo, Gurabo, PR, USA c University of Windsor, Ontario, Canada d Chrysler LLC Group, Auburn Hills, MI, USA b
a r t i c l e i n f o
abstract
Article history: Received 29 September 2011 Received in revised form 22 May 2012 Accepted 23 May 2012 Available online 5 June 2012
Scuffing can occur in various engineering components, including engine cylinders and liners, camshafts, crankshafts, and gears. Austempered ductile iron (ADI) is finding increasing application in these components due to its self-lubricating characteristics and excellent mechanical properties. The objective of this research is to study the scuffing behavior of an austempered ductile iron material austempered at different temperatures and for varying periods of time. Rotational ball-on-disk tests were run with white mineral oil as the lubricant at two sliding speeds. A step load was applied until scuffing occurred. The scuffed specimens were studied using optical and scanning electron microscopy to determine their scuffing mechanisms. Improved scuffing resistance, as evidenced by a higher scuffing load, is related to a decreased hardness and higher level of retained austenite which produce a more ductile material. & 2012 Elsevier B.V. All rights reserved.
Keywords: Scuffing test Scuffing mechanism Plastic deformation Austempered ductile iron
1. Introduction As noted by Qu et al. [1]: ‘‘The term ‘scuffing’ has been used to describe surface damage in various contexts throughout the field of engineering’’. Scuffing is associated with a sharp rise in friction and surface temperature, usually accompanied by a rise in noise and vibration [2,3]. There has been no general agreement on a definition for scuffing. This has, to a large extent, been due to the complexity of the process. However, one definition that has captured many of the features of scuffing is: ‘‘Scuffing is a form of sliding-induced contact damage to a bearing surface, usually associated with asperity-scale plastic deformation that results in localized and perceptible changes in roughness or appearance without significantly altering the geometric form of the part on which the damage occurs [1].’’ Usually scuffing damage is catastrophic and not self-healing so that the scuffed part must be replaced. Scuffing may be delayed or prevented by selecting materials with appropriate microstructure and hardness. However due to the complexity of the scuffing process, there is a need to conduct a variety of experiments to better understand the scuffing mechanism, and to evaluate the influence of material microstructure and hardness on scuffing.
n Corresponding author at: University of Turabo, Gurabo, PR, USA. Tel.: þ1 787 743 7979; fax: þ 1 787 744 5476. E-mail address: [email protected] (J.M. Han).
0043-1648/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.wear.2012.05.003
Austempered ductile iron (ADI) has recently appeared as a significant engineering material owing to its exceptional combination of high strength, ductility, toughness, machinability and wear and fatigue resistance [4]. The attractive properties are related to its unique microstructure that includes ferrite (a) and high carbon austenite (gHC ), called ausferrite. This is different from the austempered steels where the microstructure involves ferrite and carbide. The products of ADI can be molded, which allows cost reduction compared to conventional steels. Castability removes unnecessary forging and assembly requirements saving cost and weight [5,6]. Therefore, it appears that ADI can be substituted for forged and cast steels in many engineering applications such as camshafts, crankshafts, and piston rings, as well as other applications in the rail and heavy engineering industries [5–7]. A two stage heat treatment is used for ADI, austenitization (815–950 1C) and austempering (230–400 1C) [4,7–9]. Mechanical properties of ADI vary over a wide range of values, mostly controlled by the microstructure which depends on the heat treatment parameters, such as the austenitizing and austempering temperature/times [4,9–11]. A significant number of studies have been carried out on the tribological behavior of ADI [12–19], but few studies have examined scuffing. Magalha~ es and Seabra [20] found that the properties of ADI may help it resist scuffing. However there is little published research dealing with the effect of both microstructure and hardness on the scuffing behavior of ADI.
100
J.M. Han et al. / Wear 290–291 (2012) 99–105
The objectives of the present investigation were to examine the effect of heat treatment schedules, and resulting microstructure and hardness of ADI, on its scuffing performance and to determine the mechanisms involved in the scuffing process.
2. Experiment details 2.1. ADI material and heat treatment The ADI was an alloyed nodular ductile cast iron with a composition of 1.61% Ni, 0.11% Mo, 0.78% Cu, 3.76% C, 0.24% Mn, 2.51% Si, 0.057% Mg, and traces of S and P. This Ni–Mo–Cu ductile cast iron was initially heat treated in a salt bath at 890 1C for 20 min for austenitization, then quenched directly in another salt bath down to the austempering temperature of either 275 1C, 300 1C, 325 1C, 350 1C, or 375 1C for different periods of time, namely 10 min, 60 min, or 150 min. After austempering, the samples were immediately cooled in air to ambient temperature. The detailed heat treatment process is shown schematically in Fig. 1. To determine the volume fraction of retained austenite for each heat treatment process, an X-ray diffraction (XRD) method was used with monochromatic Cr–Ka radiation (wave length ˚ at 20 kV and 20 mA. The recorded profiles were l ¼2.29 A) analyzed to obtain the precise diffraction peak positions and integrated intensities. The volume fraction of retained austenite was determined by the direct comparison method using the integrated intensities of the (200)aand (211)a peaks of ferrite, and the (200)g and (220)g peaks of austenite [21,22]. 2.2. Scuffing test A ball-on-disk tribometer (shown in Fig. 2) was used to carry out the tests at room temperature. The ball sample was made of 52100 steel with a diameter of 7.94 mm and a hardness of 66 HRC. The ADI disk specimens had a diameter of 75 mm and a thickness of 10 mm. The disk surface was finished by grinding after the heat treatment, producing an average roughness of Ra ¼ 0:399 mm. During the test, the rotational speed of disk is 700 rpm. The ball was located at two different radii, which
890ºC/20min
10min 375ºC 350ºC 325ºC 300ºC 275ºC
60min
150min
Fig. 1. Ni–Mo–Cu ductile cast iron austenitization and austempering process chart.
Steel Ball
3. Results 3.1. Microstructure of austempered ADI The microstructures of the ADI austempered at 5 different temperatures and 3 time periods are shown in Table 1. Distinct differences in microstructure were observed for different austempering temperatures and times. For short austempering times (10 min), the microstructure consists of nodular graphite and a very small amount of ausferrite (ferrite and austenite) in a martensitic matrix. Martensite was detected for all 5 austempering temperatures. With an increase in the austempering temperature, more ausferrite is developed and the fine needle ausferrite becomes coarser. For long austempering times (60 min, 150 min), the microstructure consists of nodular graphite in an ausferrite matrix without martensite. Long austempering times result in the martensite being transformed into high carbon austenite. The fine ausferrite needles grow larger with an increase in austempering temperature. A coarse feathery ausferrite is produced at the higher austempering temperatures, as shown in Table 1 (350 1C/60 min, 150 min and 375 1C/60 min, 150 min). Increasing the austempering temperature results in higher quantities and coarser ausferrite. The measured volume fractions of retained austenite are shown in Table 2. It can be seen that the volume fraction of retained austenite is lower for the 10 min austempering time than for the 60 min and 150 min austempering times. The volume fraction of retained austenite is slightly higher for the 60 min austempering process than in the 150 min austempering time. This can be explained by a 2 stage transformation during the austempering process. In the first stage during the austempering process, austenite which was developed during the austenitization process, decomposes into ferrite (a) and carbon enriched austenite ðgHC Þ:
g-a þ gHC Air cooling
Normal Load
resulted in two wear tracks and two linear velocities. One wear track had a diameter of 0.045 m and a linear velocity of 1.649 m/s, and the other one had a diameter of 0.037 m and a linear velocity of 1.356 m/s. The applied normal load was increased by 22 N every 120 s and the test was terminated when there was a sudden increase of the coefficient of friction, noise level and severe vibration, which usually -occurred simultaneously with the onset of scuffing. The load at this time was defined as the scuffing load. The friction force was measured with a strain gage mounted on the sample holder. The disk specimen was lubricated by white mineral oil with a viscosity of 53 cP at room temperature. All tests were repeated 4 times and the average coefficient of friction (COF) and scuffing load were recorded. Typically the coefficient of friction was approximately 0.1 before scuffing and increased rapidly to approximately 0.25 when scuffing occurred.
Strain Gauge Wear Track
ð1Þ
If the material is austempered for longer times, then the carbon enriched austenite ðgHC Þ further decomposes into ferrite and carbide:
gHC -a þ carbide
ð2Þ
It can be assumed that the 60 min duration had produced the highest amount of stable enriched austenite. When kept for a longer period (150 min), the 2nd stage process occurred, reducing the amount of stable austenite [4,7,10]. 3.2. Hardness of austempered ADI
Rotation Direction Disc Specimen Fig. 2. Schematic view of the ball-on-disk test rig.
The influence of austempering temperature and time on the hardness of ADI is shown in Table 3. It is found that low austempering temperature and short time cause ADI to have high
J.M. Han et al. / Wear 290–291 (2012) 99–105
101
Table 1 Microstructure of austempered ADI.
Time Austempering Temperature
10(min)
60(min)
150(min)
275°C
AF (fine needle)
300°C
325°C
350°C
375°C AF (feathery-shape)
hardness value, with the highest hardness of HRC 55.8 for 275 1C/ 10 min. Hardness decreases with increasing austempering temperature and time: the lowest hardness of HRC 27.6 occurs at 375 1C/60 min and 150 min. It can be seen that hardness values at 350 1C/10 min and 375 1C/10 min are similar to those at 275 1C/ 60 min and 275 1C/150 min. All five principal hardness grades (I–V) of ADI can be developed by choice of austempering temperature and time [23].
3.3. Scuffing resistance of ADI Fig. 3 shows the scuffing load at two different sliding speeds for all the austempered ADI samples. From the comparison of scuffing load, it can be seen that the scuffing load decreases with the increase of sliding speed, except for two cases at 325 1C/ 60 min and 350 1C/150 min. There is larger variation in scuffing loads at a sliding speed of 1.649 m/s for all austempered ADIs,
102
J.M. Han et al. / Wear 290–291 (2012) 99–105
Table 2 Volume fraction (vol %) of retained austenite in austempered ductile iron. Austempering temperature (1C)
Time (min)
275 300 325 350 375
10
60
150
12 18 23 27 29
15 28 33 35 36
14 27 33 34 35
Table 3 Hardness (HRC) of austempered ADI. Austempering temperature (1C)
Time (min) 10
275 300 325 350 375
55.8 51.0 45.7 45.4 45.2
60 (V) (V) (III) (III) (III)
46.0 42.1 35.2 32.0 27.6
150 (IV) (III) (II) (II) (I)
45.1 42.4 36.7 32.0 27.6
(III) (III) (III) (II) (I)
Fig. 4. Relationship between (a) % retained austenite and scuffing resistance, (b) hardness and scuffing resistance, (c) hardness and retained austenite %.
Table 4 Coefficient of determination R2.
Fig. 3. Comparison of average scuffing load at two sliding speeds.
than at a sliding speed of 1.356 m/s. This suggests that scuffing resistance at the higher sliding speed is more sensitive to the microstructure of the austempered ADI. The maximum scuffing load was obtained at the austempering temperature of 375 1C for 60 min with a scuffing load of 176 N at 1.649 m/s and 183 N at 1.356 m/s. The main reason for this is thought to be the presence of a large volume fraction of feathery ausferrite that has high carbon content (see Table 1). The lowest scuffing load was observed for the sample austempered at 275 1C/10 min and tested at 1.649 m/s (88 N) due to the lowest fraction of ausferrite in the martensitic matrix. The values of the scuffing load for the austempered ADI indicate that a large volume fraction of feathery ausferrite has greater scuffing resistance than a small volume fraction of needle-like ausferrite.
Relations
Speed 1 (1.649 m/s)
Speed 2 (1.356 m/s)
Load vs. % g Load vs. hardness
0.749 0.728
0.586 0.540
For both sliding speeds, the trend shows a higher scuffing load with higher percentage retained austenite and lower hardness. The experimental data were fitted using polynomial, logarithmic, exponential, power law and linear relationships. Generally, the polynomial relationships gave the best fit in terms of R2 values with R2 values in the range 0.540–0.749 (Table 4). The equations relating scuffing load (L) to either hardness (HRC) or % retained austenite (RA) were as follows: L ¼ 2352:561H0:0H2 ð1:649 m=sÞ
ð3Þ
L ¼ 294:65:685H0:054H2 ð1:356 m=sÞ
ð4Þ
L ¼ 1384:496RA0:144ðRAÞ2 ð1:649 m=sÞ
ð5Þ
L ¼ 1384:496RA0:110ðRAÞ2 ð1:356 m=sÞ 3.4. Effect of hardness and percentage of retained austenite on scuffing The scuffing load is plotted against the hardness and the volume percentage of retained austenite in Fig. 4(a) and (b), respectively.
ð6Þ 2
The higher speed data (1.649 m/s) has a higher R compared to the lower speed values. This indicates that the scuffing load failure was more regular and predictable for the higher sliding speed. Fig. 4c illustrates the inverse relationship between hardness and % retained austenite. The relation is best described with a linear curve fitting, with a coefficient determination of 0.7687.
J.M. Han et al. / Wear 290–291 (2012) 99–105
3.5. Subsurface of ADI after scuffing Fig. 5(a)–(c) shows the cross-sectional views of the scuffed track for specimens with an austempering time of 10 min. Cracks are observed at the graphite/martensite matrix interfacial regions in the subsurface and cracks are seen to be propagating from the interfacial regions. The result of this crack propagation will be the generation of a wear particle and the ‘‘pull-out’’ of graphite. This observation of the microprocess of crack initiation and propagation agrees with other research studies. Shelton and Bonner [24] in their study of copper-containing ADI suggested that sub-surface crack initiation and propagation which lead to the surface delamination, always initiated at the graphite nodules of ADI. An in-situ observation of the microprocess of crack initiation and propagation in ADI by Dai et al.[25] also gave similar findings. Microcracks were always found to initiate and propagate along the graphite–matrix interphase. Very little plastic deformation is observed on these samples except for the ADI samples austempered at 375 1C (Fig. 5c), where there is a higher ausferrite content. Moderate plastic deformation is observed for the scuffed specimens which were austempered at temperatures of 275 1C and 300 1C for 60 min and 150 min, as shown in Fig. 5(d)–(f). A few cracks are also observed which are expected to produce a wear particle after propagation. There is also deformation of the graphite due to the scuffing process. A large amount of plastic deformation is found in the ausferrite matrix with high austempering temperatures of 350 1C and 375 1C at 60 min and 150 min, as shown in Fig. 5(h)–(i). A trace of a white layer is observed in Fig. 5(g). This mysterious ‘‘white layer’’ constituent has been mentioned in cast iron research
103
studies dating back to 1940s. Clayton and Jenkins [26] found that a cast iron surface rubbing against steel develops a thin layer of white constituent. The material was assumed to be having two phases—cementite and a high carbon ferritic base developed from austenite. Ludema [27] in his research on scuffing for a lubricated surface mentioned the formation of a white layer. This white layer is termed as W2. XRD and transmission electron microscope (TEM) analysis by Cranshaw and Campany [3] suggested that W2 was a heavily deformed mixture of austenite and martensite. A few cracks are also present in the interfacial region of the white layer and ausferrite matrix rather than in the interfacial region of the graphite and the ausferrite matrix: see Fig. 5(g). Graphite is sheared into an elliptical or even strip shape due to the shear force. The elliptical or strip of graphite causes a protrusion on the contact surface, which is expected to result in a rougher contact surface.
4. Discussion of scuffing mechanism The influence of austempering temperature and time on the microstructure of ADI can be seen in Table 1, which indicates that the microstructures of ADI produced by the austempering processes are strongly dependent on the transformation temperature and time. Due to the slow rate of carbon diffusion at low austempering temperature (below 300 1C) with short austempering time (10 min), only a small amount of fine needle ausferrite forms in a martensitic matrix that has a high hardness ranging from HRC 51.0 to 55.8, as shown in Table 2. Cracks can initiate easily in the harder surface during the sliding contact. This
Fig. 5. SEM image in the subsurface micrograph at sliding speed of 1.649 m/s (the horizontal arrow shows the sliding direction) for ADI austempered at (a) 325 1C/10 min, scuffing load ¼117 N; (b) 350 1C/10 min, scuffing load ¼125 N; (c) 375 1C/10 min, scuffing load¼ 127 N; (d) 275 1C/60 min, scuffing load ¼ 103 N; (e) 275 1C/150 min, scuffing load ¼117 N; (f) 300 1C/60 min, scuffing load ¼ 125 N; (g) 325 1C/150 min, scuffing load¼ 127 N; (h) 350 1C/60 min, scuffing load¼ 169 N; and (i) 375 1C/60 min, scuffing load ¼176 N.
104
J.M. Han et al. / Wear 290–291 (2012) 99–105
Fig. 6. Feathery ausferrite produced by austempering temperature at (a.) 350 1C–60 min (medium feathery), (b.) 350 1C–150 min (highly feathery), (c.) 375 1C–60 min (medium feathery), (d.) 375 1C–150 min (highly feathery) and (e.) higher magnification of (d).
microstructure with a hard martensitic matrix cannot ‘‘protect’’ the graphite nodules. The graphite is prone to pull out during sliding because of its weak strength; consequently, cracks can be propagated (see Fig. 5(a)–(c)). After a low critical number of cycles, a large amount of wear particles are produced. These particles destroy the lubricant film, leading to metal-to-metal contact and eventually, catastrophic failure. Consequently, this leads to a low scuffing resistance. With higher austempering temperature, the carbon diffusion rate is faster. Consequently, a larger volume fraction of ausferrite matrix is developed without martensite. The ausferrite becomes coarser with a lower hardness with increase of austempering temperature and time. The feathery ausferrite produced by 350 1C/60 min, 150 min and 375 1C/60 min, 150 min austempering heat treatments (see Fig. 6(a)–(e)) protects grap hite from being pulled out and exhibits a good combination of high strength and ductility. During the sliding contact under these cases, more plastic deformation occurs due to the shear stress created by frictional force between the contacting surfaces (see Fig. 5 (h)–(i). Plastic deformation results in a large amount of heat which changes the material properties, softens the material and facilitates further plastic flow with a resulting increase in roughness. This roughening creates high localized stress and the onset of scuffing with an abrupt rise in the coefficient of friction and severe adhesion between the contacting surfaces. This is
typically accompanied by sudden noise and vibration. The influence of ausferrite matrix on the tribological properties was also observed by Sahin et al. [18], who found that the wear resistance of ADI was higher with an increase of the ausferrite volume fraction.
5. Conclusions In order to better understand the scuffing mechanism, ADI samples were austempered at five austempering temperatures for three austempering time periods and tested in a ball-on-disk rig lubricated by white mineral oil. The different microstructures and hardnesses give rise to different scuffing behaviors. The main conclusions were: 1) Scuffing is initiated by plastic deformation for ADI austempered at high austempering temperatures ( Z350 1C) and the long austempering times (60 min and 150 min), and by cracks for ADI austempered at low austempering temperatures ( r325 1C) and short austempering times (10 min). 2) A feathery-ausferrite produced at high austempering temperatures and long austempering times (350 1C/60 min, 150 min and 375 1C/60 min, 150 min) exhibits higher scuffing resistance than
J.M. Han et al. / Wear 290–291 (2012) 99–105
the fine needle-like ausferrite produced at low austempering temperatures (o325 1C) and short austempering times (10 min). 3) The improvement in scuffing resistance of austempered ADI with the increase of austempering temperature and time is related to an increase in % retained austenite, a decrease in hardness and hence improved ductility. 4) The samples austempered at 3751/60 min had the highest scuffing resistance due to the feathery ausferrite microstructure.
References [1] J. Qu, J.J. Truhan, P.J. Blau, R. Ott, The development of a ‘‘pin-on-twin’’ scuffing test to evaluate materials for heavy-duty diesel fuel injectors, Tribology Transactions 50 (2007) 50–57. [2] T.E. Cranshaw, R.G. Campany, The study of scoring and scuffing on lubricated sliding surfaces by Mossbauer spectrography, Journal De Physique C2 (1979) 589–591. [3] P.J. Blau, J. Qu, Transient Scuffing of Candidate Diesel Engine Materials at Temperatures up to 6001 C, Project Milestone Report, March 31st, Oak Ridge National Library, Oak Ridge, TN, 2003. [4] V. Kilicli, M. Erdogan, Effect of ausferrite volume fraction and morphology on tensile properties of partially austenitized and austempered ductile irons with dual matrix structures, International Journal of Cast Metals Research 20 (2007) 202–214. [5] A design study in austempered ductile iron, Smith Foundry Drive Wheel, Copyright 2005 by the American Foundry Society. /http://www.sfsa.org/ tutorials/drive_wheel/Drive%20Wheel.pdfS (D.O.A: 1/18/2010). [6] /http://www.gearshub.comS, Leading supplier, manufacturer and exporter of industrial gears and custom made gears, /http://www.gearshub.com/ ductile-iron-gears.htmlS (D.O.A: 1/17/2010). [7] B. Radulovic, B. Bosnjak, The influence of austenitising temperature on the microstructure and mechanical properties of low-alloyed Ni–Mo–Cu austempered ductile iron, Materiali in Tehnologije 35 (5) (2000) 207–212. [8] Y.J. Kim, H. Shin, Investigation into mechanical properties of austempered ductile cast iron (ADI) in accordance with austempering temperature, Materials Letters 62 (2008) 357–360. [9] D. Myszka, Austenite–martensite transformation in austempred ductile iron, Archives of Metallurgy and Materials 52 (3) (2007) 475–480.
105
[10] B. Radulovic, B. Bosnjak, The influence of austenitising temperature on austempering kinetics of Ni–Mo alloyed ductile iron, Materiali in Tehnologije 28 (6) (2004) 307–312. [11] O. Eric, M. Jovanovic, Microstructure and mechanical properties of Cu–Ni–Mo austempered ductile iron, Journal of Mining and Metallurgy 40B (1) (2004) 11–19. [12] J. Yang, S.K. Putatunda, Effect of microstructure on abrasion wear behavior of austempered ductile cast iron (ADI) processed by a novel two-step austempering process, Materials Science and Engineering A406 (2005) 217–228. [13] L. Magalha~ es, J. Seabra, C. Sa´, Experimental observations of contact fatigue crack mechanisms for austempered ductile iron (ADI) discs, Wear 246 (2000) 134–148. [14] U.R. Kumari, P.P. Rao, Study of wear behavior of austempered ductile iron, Journal of Material Science 44 (2009) 1082–1093. [15] R.C. Dommarco, P.C. Bastias, H.A. Dall’O, G.T. Hahn, C.A. Rubin, Rolling contact fatigue (RCF) resistance of austempered ductile iron (ADI), Wear 221 (1998) 69–74. [16] M.C. Cakir, A. Bayram, Y. Isik, B. Salar, The effect of austempering temperature and time onto the machinability of austempered ductile iron, Materials Science and Engineering A 407 (2005) 147–153. [17] A.S.M.A. Haseeb, M.A. Islam, M.M.A. Bepari, Tribological behavior of quenched and tempered, and austempered ductile iron at the same hardness level, Wear 244 (2000) 15–19. [18] Y. Sahin, M. Erdogan, V. Kilicli, Wear behavior of austempered ductile iron with dual matrix structures, Materials Science and Engineering A 444 (2007) 31–38. [19] M.J. Perez, M.M. Cisneros, H.F. Lopez, Wear resistance of Cu–Ni–Mo austempered ductile iron, Wear 260 (2006) 879–885. [20] L. Magalha~ es, J. Seabra, Wear and scuffing of austempered ductile iron gears, Wear 215 (1998) 237–246. [21] Diffraction Notes, Lambda Technologies, No. 33, Fall 2006, pp. 1–3. [22] ASTM Standard (E975-03), Standard Practice for X-Ray Determination of Retained Austenite in Steel with Near Random Crystallographic Orientation, American Society for Testing and Materials, November 1, 2003. [23] 10th edition,Metals Handbook, 1, ASM International, 1990. [24] P.W. Shelton, A.A. Bonner, The effect of copper additions to the mechanical properties of austempered ductile iron (ADI), Journal of Materials Processing Technology 173 (2006) 269–274. [25] P.Q. Dai, Z.R. He, C.M. Zheng, Z.Y. Mao, In-situ SEM observation of the fracture of austempered ductile iron, Material Science and Engineering A319–321 (2001) 531–534. [26] D. Clayton, C.H.M. Jenkins, Physical changes in rubbing surfaces on scuffing, British Journal of Applied Physics 2 (1951) 69–78. [27] K.C. Ludema, A review of scuffing and running-in of lubricated surfaces, with asperities and oxides in perspective, Wear 100 (1984) 315–331.
Contents lists available at SciVerse ScienceDirect
Wear journal homepage: www.elsevier.com/locate/wear
Study of the effects of austempering temperature and time on scuffing behavior of austempered Ni–Mo–Cu ductile iron J.M. Han a,b,n, Q. Zou a, G.C. Barber a, T. Nasir c, D.O. Northwood c, X.C. Sun d, P. Seaton d a
Oakland University, Rochester Hills, MI, USA University of Turabo, Gurabo, PR, USA c University of Windsor, Ontario, Canada d Chrysler LLC Group, Auburn Hills, MI, USA b
a r t i c l e i n f o
abstract
Article history: Received 29 September 2011 Received in revised form 22 May 2012 Accepted 23 May 2012 Available online 5 June 2012
Scuffing can occur in various engineering components, including engine cylinders and liners, camshafts, crankshafts, and gears. Austempered ductile iron (ADI) is finding increasing application in these components due to its self-lubricating characteristics and excellent mechanical properties. The objective of this research is to study the scuffing behavior of an austempered ductile iron material austempered at different temperatures and for varying periods of time. Rotational ball-on-disk tests were run with white mineral oil as the lubricant at two sliding speeds. A step load was applied until scuffing occurred. The scuffed specimens were studied using optical and scanning electron microscopy to determine their scuffing mechanisms. Improved scuffing resistance, as evidenced by a higher scuffing load, is related to a decreased hardness and higher level of retained austenite which produce a more ductile material. & 2012 Elsevier B.V. All rights reserved.
Keywords: Scuffing test Scuffing mechanism Plastic deformation Austempered ductile iron
1. Introduction As noted by Qu et al. [1]: ‘‘The term ‘scuffing’ has been used to describe surface damage in various contexts throughout the field of engineering’’. Scuffing is associated with a sharp rise in friction and surface temperature, usually accompanied by a rise in noise and vibration [2,3]. There has been no general agreement on a definition for scuffing. This has, to a large extent, been due to the complexity of the process. However, one definition that has captured many of the features of scuffing is: ‘‘Scuffing is a form of sliding-induced contact damage to a bearing surface, usually associated with asperity-scale plastic deformation that results in localized and perceptible changes in roughness or appearance without significantly altering the geometric form of the part on which the damage occurs [1].’’ Usually scuffing damage is catastrophic and not self-healing so that the scuffed part must be replaced. Scuffing may be delayed or prevented by selecting materials with appropriate microstructure and hardness. However due to the complexity of the scuffing process, there is a need to conduct a variety of experiments to better understand the scuffing mechanism, and to evaluate the influence of material microstructure and hardness on scuffing.
n Corresponding author at: University of Turabo, Gurabo, PR, USA. Tel.: þ1 787 743 7979; fax: þ 1 787 744 5476. E-mail address: [email protected] (J.M. Han).
0043-1648/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.wear.2012.05.003
Austempered ductile iron (ADI) has recently appeared as a significant engineering material owing to its exceptional combination of high strength, ductility, toughness, machinability and wear and fatigue resistance [4]. The attractive properties are related to its unique microstructure that includes ferrite (a) and high carbon austenite (gHC ), called ausferrite. This is different from the austempered steels where the microstructure involves ferrite and carbide. The products of ADI can be molded, which allows cost reduction compared to conventional steels. Castability removes unnecessary forging and assembly requirements saving cost and weight [5,6]. Therefore, it appears that ADI can be substituted for forged and cast steels in many engineering applications such as camshafts, crankshafts, and piston rings, as well as other applications in the rail and heavy engineering industries [5–7]. A two stage heat treatment is used for ADI, austenitization (815–950 1C) and austempering (230–400 1C) [4,7–9]. Mechanical properties of ADI vary over a wide range of values, mostly controlled by the microstructure which depends on the heat treatment parameters, such as the austenitizing and austempering temperature/times [4,9–11]. A significant number of studies have been carried out on the tribological behavior of ADI [12–19], but few studies have examined scuffing. Magalha~ es and Seabra [20] found that the properties of ADI may help it resist scuffing. However there is little published research dealing with the effect of both microstructure and hardness on the scuffing behavior of ADI.
100
J.M. Han et al. / Wear 290–291 (2012) 99–105
The objectives of the present investigation were to examine the effect of heat treatment schedules, and resulting microstructure and hardness of ADI, on its scuffing performance and to determine the mechanisms involved in the scuffing process.
2. Experiment details 2.1. ADI material and heat treatment The ADI was an alloyed nodular ductile cast iron with a composition of 1.61% Ni, 0.11% Mo, 0.78% Cu, 3.76% C, 0.24% Mn, 2.51% Si, 0.057% Mg, and traces of S and P. This Ni–Mo–Cu ductile cast iron was initially heat treated in a salt bath at 890 1C for 20 min for austenitization, then quenched directly in another salt bath down to the austempering temperature of either 275 1C, 300 1C, 325 1C, 350 1C, or 375 1C for different periods of time, namely 10 min, 60 min, or 150 min. After austempering, the samples were immediately cooled in air to ambient temperature. The detailed heat treatment process is shown schematically in Fig. 1. To determine the volume fraction of retained austenite for each heat treatment process, an X-ray diffraction (XRD) method was used with monochromatic Cr–Ka radiation (wave length ˚ at 20 kV and 20 mA. The recorded profiles were l ¼2.29 A) analyzed to obtain the precise diffraction peak positions and integrated intensities. The volume fraction of retained austenite was determined by the direct comparison method using the integrated intensities of the (200)aand (211)a peaks of ferrite, and the (200)g and (220)g peaks of austenite [21,22]. 2.2. Scuffing test A ball-on-disk tribometer (shown in Fig. 2) was used to carry out the tests at room temperature. The ball sample was made of 52100 steel with a diameter of 7.94 mm and a hardness of 66 HRC. The ADI disk specimens had a diameter of 75 mm and a thickness of 10 mm. The disk surface was finished by grinding after the heat treatment, producing an average roughness of Ra ¼ 0:399 mm. During the test, the rotational speed of disk is 700 rpm. The ball was located at two different radii, which
890ºC/20min
10min 375ºC 350ºC 325ºC 300ºC 275ºC
60min
150min
Fig. 1. Ni–Mo–Cu ductile cast iron austenitization and austempering process chart.
Steel Ball
3. Results 3.1. Microstructure of austempered ADI The microstructures of the ADI austempered at 5 different temperatures and 3 time periods are shown in Table 1. Distinct differences in microstructure were observed for different austempering temperatures and times. For short austempering times (10 min), the microstructure consists of nodular graphite and a very small amount of ausferrite (ferrite and austenite) in a martensitic matrix. Martensite was detected for all 5 austempering temperatures. With an increase in the austempering temperature, more ausferrite is developed and the fine needle ausferrite becomes coarser. For long austempering times (60 min, 150 min), the microstructure consists of nodular graphite in an ausferrite matrix without martensite. Long austempering times result in the martensite being transformed into high carbon austenite. The fine ausferrite needles grow larger with an increase in austempering temperature. A coarse feathery ausferrite is produced at the higher austempering temperatures, as shown in Table 1 (350 1C/60 min, 150 min and 375 1C/60 min, 150 min). Increasing the austempering temperature results in higher quantities and coarser ausferrite. The measured volume fractions of retained austenite are shown in Table 2. It can be seen that the volume fraction of retained austenite is lower for the 10 min austempering time than for the 60 min and 150 min austempering times. The volume fraction of retained austenite is slightly higher for the 60 min austempering process than in the 150 min austempering time. This can be explained by a 2 stage transformation during the austempering process. In the first stage during the austempering process, austenite which was developed during the austenitization process, decomposes into ferrite (a) and carbon enriched austenite ðgHC Þ:
g-a þ gHC Air cooling
Normal Load
resulted in two wear tracks and two linear velocities. One wear track had a diameter of 0.045 m and a linear velocity of 1.649 m/s, and the other one had a diameter of 0.037 m and a linear velocity of 1.356 m/s. The applied normal load was increased by 22 N every 120 s and the test was terminated when there was a sudden increase of the coefficient of friction, noise level and severe vibration, which usually -occurred simultaneously with the onset of scuffing. The load at this time was defined as the scuffing load. The friction force was measured with a strain gage mounted on the sample holder. The disk specimen was lubricated by white mineral oil with a viscosity of 53 cP at room temperature. All tests were repeated 4 times and the average coefficient of friction (COF) and scuffing load were recorded. Typically the coefficient of friction was approximately 0.1 before scuffing and increased rapidly to approximately 0.25 when scuffing occurred.
Strain Gauge Wear Track
ð1Þ
If the material is austempered for longer times, then the carbon enriched austenite ðgHC Þ further decomposes into ferrite and carbide:
gHC -a þ carbide
ð2Þ
It can be assumed that the 60 min duration had produced the highest amount of stable enriched austenite. When kept for a longer period (150 min), the 2nd stage process occurred, reducing the amount of stable austenite [4,7,10]. 3.2. Hardness of austempered ADI
Rotation Direction Disc Specimen Fig. 2. Schematic view of the ball-on-disk test rig.
The influence of austempering temperature and time on the hardness of ADI is shown in Table 3. It is found that low austempering temperature and short time cause ADI to have high
J.M. Han et al. / Wear 290–291 (2012) 99–105
101
Table 1 Microstructure of austempered ADI.
Time Austempering Temperature
10(min)
60(min)
150(min)
275°C
AF (fine needle)
300°C
325°C
350°C
375°C AF (feathery-shape)
hardness value, with the highest hardness of HRC 55.8 for 275 1C/ 10 min. Hardness decreases with increasing austempering temperature and time: the lowest hardness of HRC 27.6 occurs at 375 1C/60 min and 150 min. It can be seen that hardness values at 350 1C/10 min and 375 1C/10 min are similar to those at 275 1C/ 60 min and 275 1C/150 min. All five principal hardness grades (I–V) of ADI can be developed by choice of austempering temperature and time [23].
3.3. Scuffing resistance of ADI Fig. 3 shows the scuffing load at two different sliding speeds for all the austempered ADI samples. From the comparison of scuffing load, it can be seen that the scuffing load decreases with the increase of sliding speed, except for two cases at 325 1C/ 60 min and 350 1C/150 min. There is larger variation in scuffing loads at a sliding speed of 1.649 m/s for all austempered ADIs,
102
J.M. Han et al. / Wear 290–291 (2012) 99–105
Table 2 Volume fraction (vol %) of retained austenite in austempered ductile iron. Austempering temperature (1C)
Time (min)
275 300 325 350 375
10
60
150
12 18 23 27 29
15 28 33 35 36
14 27 33 34 35
Table 3 Hardness (HRC) of austempered ADI. Austempering temperature (1C)
Time (min) 10
275 300 325 350 375
55.8 51.0 45.7 45.4 45.2
60 (V) (V) (III) (III) (III)
46.0 42.1 35.2 32.0 27.6
150 (IV) (III) (II) (II) (I)
45.1 42.4 36.7 32.0 27.6
(III) (III) (III) (II) (I)
Fig. 4. Relationship between (a) % retained austenite and scuffing resistance, (b) hardness and scuffing resistance, (c) hardness and retained austenite %.
Table 4 Coefficient of determination R2.
Fig. 3. Comparison of average scuffing load at two sliding speeds.
than at a sliding speed of 1.356 m/s. This suggests that scuffing resistance at the higher sliding speed is more sensitive to the microstructure of the austempered ADI. The maximum scuffing load was obtained at the austempering temperature of 375 1C for 60 min with a scuffing load of 176 N at 1.649 m/s and 183 N at 1.356 m/s. The main reason for this is thought to be the presence of a large volume fraction of feathery ausferrite that has high carbon content (see Table 1). The lowest scuffing load was observed for the sample austempered at 275 1C/10 min and tested at 1.649 m/s (88 N) due to the lowest fraction of ausferrite in the martensitic matrix. The values of the scuffing load for the austempered ADI indicate that a large volume fraction of feathery ausferrite has greater scuffing resistance than a small volume fraction of needle-like ausferrite.
Relations
Speed 1 (1.649 m/s)
Speed 2 (1.356 m/s)
Load vs. % g Load vs. hardness
0.749 0.728
0.586 0.540
For both sliding speeds, the trend shows a higher scuffing load with higher percentage retained austenite and lower hardness. The experimental data were fitted using polynomial, logarithmic, exponential, power law and linear relationships. Generally, the polynomial relationships gave the best fit in terms of R2 values with R2 values in the range 0.540–0.749 (Table 4). The equations relating scuffing load (L) to either hardness (HRC) or % retained austenite (RA) were as follows: L ¼ 2352:561H0:0H2 ð1:649 m=sÞ
ð3Þ
L ¼ 294:65:685H0:054H2 ð1:356 m=sÞ
ð4Þ
L ¼ 1384:496RA0:144ðRAÞ2 ð1:649 m=sÞ
ð5Þ
L ¼ 1384:496RA0:110ðRAÞ2 ð1:356 m=sÞ 3.4. Effect of hardness and percentage of retained austenite on scuffing The scuffing load is plotted against the hardness and the volume percentage of retained austenite in Fig. 4(a) and (b), respectively.
ð6Þ 2
The higher speed data (1.649 m/s) has a higher R compared to the lower speed values. This indicates that the scuffing load failure was more regular and predictable for the higher sliding speed. Fig. 4c illustrates the inverse relationship between hardness and % retained austenite. The relation is best described with a linear curve fitting, with a coefficient determination of 0.7687.
J.M. Han et al. / Wear 290–291 (2012) 99–105
3.5. Subsurface of ADI after scuffing Fig. 5(a)–(c) shows the cross-sectional views of the scuffed track for specimens with an austempering time of 10 min. Cracks are observed at the graphite/martensite matrix interfacial regions in the subsurface and cracks are seen to be propagating from the interfacial regions. The result of this crack propagation will be the generation of a wear particle and the ‘‘pull-out’’ of graphite. This observation of the microprocess of crack initiation and propagation agrees with other research studies. Shelton and Bonner [24] in their study of copper-containing ADI suggested that sub-surface crack initiation and propagation which lead to the surface delamination, always initiated at the graphite nodules of ADI. An in-situ observation of the microprocess of crack initiation and propagation in ADI by Dai et al.[25] also gave similar findings. Microcracks were always found to initiate and propagate along the graphite–matrix interphase. Very little plastic deformation is observed on these samples except for the ADI samples austempered at 375 1C (Fig. 5c), where there is a higher ausferrite content. Moderate plastic deformation is observed for the scuffed specimens which were austempered at temperatures of 275 1C and 300 1C for 60 min and 150 min, as shown in Fig. 5(d)–(f). A few cracks are also observed which are expected to produce a wear particle after propagation. There is also deformation of the graphite due to the scuffing process. A large amount of plastic deformation is found in the ausferrite matrix with high austempering temperatures of 350 1C and 375 1C at 60 min and 150 min, as shown in Fig. 5(h)–(i). A trace of a white layer is observed in Fig. 5(g). This mysterious ‘‘white layer’’ constituent has been mentioned in cast iron research
103
studies dating back to 1940s. Clayton and Jenkins [26] found that a cast iron surface rubbing against steel develops a thin layer of white constituent. The material was assumed to be having two phases—cementite and a high carbon ferritic base developed from austenite. Ludema [27] in his research on scuffing for a lubricated surface mentioned the formation of a white layer. This white layer is termed as W2. XRD and transmission electron microscope (TEM) analysis by Cranshaw and Campany [3] suggested that W2 was a heavily deformed mixture of austenite and martensite. A few cracks are also present in the interfacial region of the white layer and ausferrite matrix rather than in the interfacial region of the graphite and the ausferrite matrix: see Fig. 5(g). Graphite is sheared into an elliptical or even strip shape due to the shear force. The elliptical or strip of graphite causes a protrusion on the contact surface, which is expected to result in a rougher contact surface.
4. Discussion of scuffing mechanism The influence of austempering temperature and time on the microstructure of ADI can be seen in Table 1, which indicates that the microstructures of ADI produced by the austempering processes are strongly dependent on the transformation temperature and time. Due to the slow rate of carbon diffusion at low austempering temperature (below 300 1C) with short austempering time (10 min), only a small amount of fine needle ausferrite forms in a martensitic matrix that has a high hardness ranging from HRC 51.0 to 55.8, as shown in Table 2. Cracks can initiate easily in the harder surface during the sliding contact. This
Fig. 5. SEM image in the subsurface micrograph at sliding speed of 1.649 m/s (the horizontal arrow shows the sliding direction) for ADI austempered at (a) 325 1C/10 min, scuffing load ¼117 N; (b) 350 1C/10 min, scuffing load ¼125 N; (c) 375 1C/10 min, scuffing load¼ 127 N; (d) 275 1C/60 min, scuffing load ¼ 103 N; (e) 275 1C/150 min, scuffing load ¼117 N; (f) 300 1C/60 min, scuffing load ¼ 125 N; (g) 325 1C/150 min, scuffing load¼ 127 N; (h) 350 1C/60 min, scuffing load¼ 169 N; and (i) 375 1C/60 min, scuffing load ¼176 N.
104
J.M. Han et al. / Wear 290–291 (2012) 99–105
Fig. 6. Feathery ausferrite produced by austempering temperature at (a.) 350 1C–60 min (medium feathery), (b.) 350 1C–150 min (highly feathery), (c.) 375 1C–60 min (medium feathery), (d.) 375 1C–150 min (highly feathery) and (e.) higher magnification of (d).
microstructure with a hard martensitic matrix cannot ‘‘protect’’ the graphite nodules. The graphite is prone to pull out during sliding because of its weak strength; consequently, cracks can be propagated (see Fig. 5(a)–(c)). After a low critical number of cycles, a large amount of wear particles are produced. These particles destroy the lubricant film, leading to metal-to-metal contact and eventually, catastrophic failure. Consequently, this leads to a low scuffing resistance. With higher austempering temperature, the carbon diffusion rate is faster. Consequently, a larger volume fraction of ausferrite matrix is developed without martensite. The ausferrite becomes coarser with a lower hardness with increase of austempering temperature and time. The feathery ausferrite produced by 350 1C/60 min, 150 min and 375 1C/60 min, 150 min austempering heat treatments (see Fig. 6(a)–(e)) protects grap hite from being pulled out and exhibits a good combination of high strength and ductility. During the sliding contact under these cases, more plastic deformation occurs due to the shear stress created by frictional force between the contacting surfaces (see Fig. 5 (h)–(i). Plastic deformation results in a large amount of heat which changes the material properties, softens the material and facilitates further plastic flow with a resulting increase in roughness. This roughening creates high localized stress and the onset of scuffing with an abrupt rise in the coefficient of friction and severe adhesion between the contacting surfaces. This is
typically accompanied by sudden noise and vibration. The influence of ausferrite matrix on the tribological properties was also observed by Sahin et al. [18], who found that the wear resistance of ADI was higher with an increase of the ausferrite volume fraction.
5. Conclusions In order to better understand the scuffing mechanism, ADI samples were austempered at five austempering temperatures for three austempering time periods and tested in a ball-on-disk rig lubricated by white mineral oil. The different microstructures and hardnesses give rise to different scuffing behaviors. The main conclusions were: 1) Scuffing is initiated by plastic deformation for ADI austempered at high austempering temperatures ( Z350 1C) and the long austempering times (60 min and 150 min), and by cracks for ADI austempered at low austempering temperatures ( r325 1C) and short austempering times (10 min). 2) A feathery-ausferrite produced at high austempering temperatures and long austempering times (350 1C/60 min, 150 min and 375 1C/60 min, 150 min) exhibits higher scuffing resistance than
J.M. Han et al. / Wear 290–291 (2012) 99–105
the fine needle-like ausferrite produced at low austempering temperatures (o325 1C) and short austempering times (10 min). 3) The improvement in scuffing resistance of austempered ADI with the increase of austempering temperature and time is related to an increase in % retained austenite, a decrease in hardness and hence improved ductility. 4) The samples austempered at 3751/60 min had the highest scuffing resistance due to the feathery ausferrite microstructure.
References [1] J. Qu, J.J. Truhan, P.J. Blau, R. Ott, The development of a ‘‘pin-on-twin’’ scuffing test to evaluate materials for heavy-duty diesel fuel injectors, Tribology Transactions 50 (2007) 50–57. [2] T.E. Cranshaw, R.G. Campany, The study of scoring and scuffing on lubricated sliding surfaces by Mossbauer spectrography, Journal De Physique C2 (1979) 589–591. [3] P.J. Blau, J. Qu, Transient Scuffing of Candidate Diesel Engine Materials at Temperatures up to 6001 C, Project Milestone Report, March 31st, Oak Ridge National Library, Oak Ridge, TN, 2003. [4] V. Kilicli, M. Erdogan, Effect of ausferrite volume fraction and morphology on tensile properties of partially austenitized and austempered ductile irons with dual matrix structures, International Journal of Cast Metals Research 20 (2007) 202–214. [5] A design study in austempered ductile iron, Smith Foundry Drive Wheel, Copyright 2005 by the American Foundry Society. /http://www.sfsa.org/ tutorials/drive_wheel/Drive%20Wheel.pdfS (D.O.A: 1/18/2010). [6] /http://www.gearshub.comS, Leading supplier, manufacturer and exporter of industrial gears and custom made gears, /http://www.gearshub.com/ ductile-iron-gears.htmlS (D.O.A: 1/17/2010). [7] B. Radulovic, B. Bosnjak, The influence of austenitising temperature on the microstructure and mechanical properties of low-alloyed Ni–Mo–Cu austempered ductile iron, Materiali in Tehnologije 35 (5) (2000) 207–212. [8] Y.J. Kim, H. Shin, Investigation into mechanical properties of austempered ductile cast iron (ADI) in accordance with austempering temperature, Materials Letters 62 (2008) 357–360. [9] D. Myszka, Austenite–martensite transformation in austempred ductile iron, Archives of Metallurgy and Materials 52 (3) (2007) 475–480.
105
[10] B. Radulovic, B. Bosnjak, The influence of austenitising temperature on austempering kinetics of Ni–Mo alloyed ductile iron, Materiali in Tehnologije 28 (6) (2004) 307–312. [11] O. Eric, M. Jovanovic, Microstructure and mechanical properties of Cu–Ni–Mo austempered ductile iron, Journal of Mining and Metallurgy 40B (1) (2004) 11–19. [12] J. Yang, S.K. Putatunda, Effect of microstructure on abrasion wear behavior of austempered ductile cast iron (ADI) processed by a novel two-step austempering process, Materials Science and Engineering A406 (2005) 217–228. [13] L. Magalha~ es, J. Seabra, C. Sa´, Experimental observations of contact fatigue crack mechanisms for austempered ductile iron (ADI) discs, Wear 246 (2000) 134–148. [14] U.R. Kumari, P.P. Rao, Study of wear behavior of austempered ductile iron, Journal of Material Science 44 (2009) 1082–1093. [15] R.C. Dommarco, P.C. Bastias, H.A. Dall’O, G.T. Hahn, C.A. Rubin, Rolling contact fatigue (RCF) resistance of austempered ductile iron (ADI), Wear 221 (1998) 69–74. [16] M.C. Cakir, A. Bayram, Y. Isik, B. Salar, The effect of austempering temperature and time onto the machinability of austempered ductile iron, Materials Science and Engineering A 407 (2005) 147–153. [17] A.S.M.A. Haseeb, M.A. Islam, M.M.A. Bepari, Tribological behavior of quenched and tempered, and austempered ductile iron at the same hardness level, Wear 244 (2000) 15–19. [18] Y. Sahin, M. Erdogan, V. Kilicli, Wear behavior of austempered ductile iron with dual matrix structures, Materials Science and Engineering A 444 (2007) 31–38. [19] M.J. Perez, M.M. Cisneros, H.F. Lopez, Wear resistance of Cu–Ni–Mo austempered ductile iron, Wear 260 (2006) 879–885. [20] L. Magalha~ es, J. Seabra, Wear and scuffing of austempered ductile iron gears, Wear 215 (1998) 237–246. [21] Diffraction Notes, Lambda Technologies, No. 33, Fall 2006, pp. 1–3. [22] ASTM Standard (E975-03), Standard Practice for X-Ray Determination of Retained Austenite in Steel with Near Random Crystallographic Orientation, American Society for Testing and Materials, November 1, 2003. [23] 10th edition,Metals Handbook, 1, ASM International, 1990. [24] P.W. Shelton, A.A. Bonner, The effect of copper additions to the mechanical properties of austempered ductile iron (ADI), Journal of Materials Processing Technology 173 (2006) 269–274. [25] P.Q. Dai, Z.R. He, C.M. Zheng, Z.Y. Mao, In-situ SEM observation of the fracture of austempered ductile iron, Material Science and Engineering A319–321 (2001) 531–534. [26] D. Clayton, C.H.M. Jenkins, Physical changes in rubbing surfaces on scuffing, British Journal of Applied Physics 2 (1951) 69–78. [27] K.C. Ludema, A review of scuffing and running-in of lubricated surfaces, with asperities and oxides in perspective, Wear 100 (1984) 315–331.