Anda di halaman 1dari 8

Materials and Design 89 (2016) 815–822

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier.com/locate/jmad

Wear and friction behavior of austempered ductile iron as railway


wheel material
Ning Zhang a, Jiwang Zhang a,b,⁎, Liantao Lu a, Mintang Zhang c, Dongfang Zeng a, Qingpeng Song a
a
State Key Laboratory of Traction Power, Southwest Jiaotong University, Chengdu 610031, China
b
CSR Qishuyan Institute Co. Ltd., Changzhou 213000, China
c
Henan Aoudi Co. Ltd, Hebi 456750, China

a r t i c l e i n f o a b s t r a c t

Article history: Austempered ductile irons (ADIs) with three strength grades and one kind of wheel steel were matched with
Received 23 January 2015 conventional rail steel and rolling-sliding wear tests were conducted. The results show that the wear rate de-
Received in revised form 24 April 2015 creases while the friction coefficient increases with the increase of matrix hardness. The increase of subsurface
Accepted 10 October 2015
hardness is due to work-hardening and stain-induced transformation of retained austenite to martensite. The
Available online 22 October 2015
main wear mechanism is delamination and becomes mild with the increase of matrix hardness. ADI austempered
Keywords:
at 340 °C shows the reasonable friction and wear behavior as well as relatively superior mechanical properties.
Austempered ductile iron © 2015 Elsevier Ltd. All rights reserved.
Wear
Coefficient of friction
Delamination

1. Introduction considered that ADIs are suitable as an alternative material for railway
wheel [16]. In fact, ADI has been applied to railway wheel successfully
Austempered ductile iron (ADI), with an ausferrtic matrix micro- in Europe [17]. Though the wear tests on ADI have been conducted a
structure, characterized by the presence of bainitic ferrite and retained lot, investigations on tribological performance of ADI matched with rail
austenite can be obtained by austempering treatments on ductile iron. steel are not enough, while it is important for the designers to consider
This special microstructure provides a good combination in mechanical it as wheel material. Especially, research on wear and friction behavior
properties, covering high tensile strength and good ductility, high fa- of ADIs with different strengths is limited, that has kept the field of ADI
tigue strength and fracture toughness, and superior wear resistance in railway applications because of the fact that reasonable hardness
[1–5]. Because of these advantages, ADI has been emerged as an impor- matching plays a significant role for increasing service life of wheel/rail
tant engineering materials in recent years and used extensively in many materials. Meanwhile, the wear and friction properties are not only attrib-
structural applications in automotive industry, defense and earth mov- uted to the relevant material itself, but also strongly depend on the envi-
ing machineries, etc. [6–8]. Many of these are subjected to rolling and ronment and other experimental conditions, such as temperature and
sliding wear. Therefore lots of investigations on the friction and wear humidity. [18–20] when the railcar runs at a high speed, strong wind
behavior of ADI have been reported [9–14]. will be caused and it will induce cooling effect on the wheel. Though
With the raising speed and weight in rail traffic, the property optimi- the results of dry and uncooling tribological tests show excellent wear re-
zation of wheel material draws more and more researchers' attention. The sistance performance, the tribological properties of ADI under cooling
optimization is based on balancing the cost, weight, wear resistance, noise condition have not been investigated.
reduction and rolling contact strength [15]. Compared with traditional In the present investigation, the dry rolling-sliding wear tests of ADIs
steel, ADI exhibits high strength, toughness and three times higher with three strength grades were conducted. Considering the air cooling
damping which can substantially lower the traveling noise. In addition, in the actual operation, the testing procedure incorporated a blast of dry
ADI has 10% lower density because of the graphite nodules dispersed in compressed air to simulate actual running contact conditions and pre-
the matrix, which promises a decrease in components' weight. So it is vent the oxidation at high temperature. For comparison, a widely used
railcar wheel steel ER8 was also investigated under the same wear con-
⁎ Corresponding author at: State Key Laboratory of Traction Power, Southwest Jiaotong
ditions. Meanwhile, the microstructure, friction and wear behavior as
University, Chengdu 610031, China. well as wear mechanism of three strength grades ADIs were discussed
E-mail address: zhangjiwang@swjtu.cn (J. Zhang). in detail.

http://dx.doi.org/10.1016/j.matdes.2015.10.037
0264-1275/© 2015 Elsevier Ltd. All rights reserved.
816 N. Zhang et al. / Materials and Design 89 (2016) 815–822

Table 2
Heat treatment and mechanical properties of ADIs specimen.

Sample Heat treatment Ultimate Elongation Impact


tensile (%) energy
Austenitized Austempered
strength (J)
(MPa)

ADI1 910 °C/120 min 380 °C/60 min 963 11.2 125
ADI2 900 °C/110 min 340 °C/60 min 1140 10.2 115
ADI3 910 °C/150 min 300 °C/120 min 1290 7.5 98

This twin-disk rolling-sliding testing machine can provide a load


range of 0–2000 N and promise a line contact between the two cylindri-
cal test disks which simulates the normal load and slip presenting at
rail/wheel contact area. Fig. 2 shows the dimensions and shapes of
rolling-sliding wear test specimens and schematic representation of
the wear apparatus. The upper disk is made of railway steel U75V
with a diameter of 40 mm, while the lower disks are made of three
grades strength ADIs and one kind of wheel steel ER8 with a same diam-
eter of 38 mm. The contact width between the test disks is 5 mm. The
rotation speeds of the upper disk (rail material) was 180 rev/min com-
Fig. 1. Positions from where the test blanks were taken. (a) ADIs disks were taken from the
bottom of ductile iron Y-blocks, (unit: mm); (b) wheel and rail steel disks were taken from pared to 200 rev/min for the lower disk (wheel material). This results in
railway wheel tread and railhead. a 0.0209 m/s sliding speed in the tests. To simulate the wearing condi-
tion in actual running operation, all wear tests were carried out under
a high contact load of 870 N. The maximum Hertzian contact pressure,
2. Experimental procedure P0, and the contact length, a, between the test disks can be obtained
by the following equations quoted by Johnson [23]:
2.1. Material and processing
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P0 ¼ P=πbRE ð1Þ
Wheel specimens of ADIs used for dry rolling-sliding wear testing
were machined from ductile iron keel blocks (Fig. 1(a)). For compari- qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
son, specimens codenamed ER8 cut from railway wheel tread were a¼ 4PRE =πb ð2Þ
also used to carry out the study. Specimens for rail codenamed U75V
were gained from the top surface of railhead showed in Fig. 1(b). The    
main chemical compositions and hardness of ER8 and U75V are given E ¼ 1  v21 =E1 þ 1  v22 =E2 ð3Þ
in Table 1. By controlling the heat treatment, three strength grades
ADIs (shortened as ADI1, ADI2 and ADI3) with different matrix struc- where P is the contact load (N), b is the line contact width (mm), E1 and
tures were obtained. Table 2 shows the mechanical properties of ADIs E2 is the modulus of elasticity of rail (upper) and wheel (lower) disk in
and heat treatments employed on the ductile irons. GPa, respectively, R is given by:

2.2. Test equipment and conditions R ¼ R1 R2 =ðR1 þ R2 Þ ð4Þ

2.2.1. Tensile and impact testing where R1 and R2 are the radii (mm) of the rail (upper) and wheel
After machining to final dimensions, tensile testing was carried out (lower) disk, respectively.
based on ASTM standard E-8 [21]; five samples were tested in each
heat treated conditions. The tests were performed on a servo-
hydraulic MTS (Material Test System) test machine. All of the samples
were tested at room temperature and ambient atmosphere. Load and
displacement plots were obtained on an X-Y recorder; from these
load–displacement diagrams, the ultimate tensile strength and elonga-
tion values were calculated. Impact toughness of the ADI samples was
measured by the Charpy impact test. Tests were performed at room
temperature with unnotched specimens, three samples were tested in
each grade of ADI.

2.2.2. Wear testing


Dry rolling-sliding wear tests were carried out on an Amsler type
tribotester on the basis of the Chinese standard GB 12444. 1–90 [22].

Table 1
Main chemical composition of the test materials.

Material Compositions(%) Hardness


(HV0.2)
C Si Mn P S

Ductile iron 3.7 2.6 0.19 0.025 0.013 –


ER8 0.51 0.93 0.93 0.009 0.001 325
U75V 0.75 0.7 0.95 ≤0.03 ≤0.03 350
Fig. 2. Schematic diagram of wear tests (unit: mm).
N. Zhang et al. / Materials and Design 89 (2016) 815–822 817

To prevent the surface oxidation caused by the tribological heating,


the contact surface between the testing disks was continuously cooled
by a blast of dry compressed air of 0.6 MPa during the tests. Under con-
tinuous cooling condition, the temperature and humidity was main-
tained at 10 °C and 5% during the wear and friction process. The
comparison wear test without the air cooling was carried out on the
wear pairing of ADI2/U75V. The tests were performed for 100,000 rev-
olutions on each wear pairing. During wear tests, the disks were sug-
gested to remove the surface hardening layer at the first 20,000
revolutions. Then, the disks were taken down and cleaned in an acetone
ultrasonic bath for 10 min, dried in a jet of hot air, and the weight of
them was obtained by using a digital balance with an accuracy of
0.1 mg (FA2003A). After that, the samples (wear pairings) were
remounted in the testing machine at the same location to conduct an-
other 80,000 revolutions and weight loss were gained by the digital bal-
ance. The weight loss was used to calculate the wear rate which was
defined as a function of sliding distance. The friction coefficient during
steady state was also recorded during each test.

2.3. Characterization

The microstructure of ADIs was evaluated under an optical micro-


scope after polishing and etching with 4% Nital solution. The morpholo-
gy of the graphite nodules were determined by means of image analysis
system. The hardness was examined using a HVS-1000Z microhardness
tester. Phase analysis was carried out on Philips X'pert PRO SUPER X-ray
diffractometer (employing Cu-Kα radiation, λ = 1.544 Å) before and
after wear tests. The tube acceleration voltage was 40 kV, the current
was 40 mA and 2θ values were between 0 and 120°. Computer software
X'pert High Score Plus, was used to determine the volume fractions level
of retained austenite before and after the tests by utilizing the integrat-
ed areas of both austenite (220) and (311) planes.
After the wear tests, the worn surfaces of the wheel and rail disks
were examined using scanning electron microscopy (SEM) to analyze
the wear behavior and mechanism. Characteristics of the subsurface
was also investigated by using microstructural investigation and mea-
suring microhardness on the cross-section wear test samples.

3. Results and discussions

3.1. ADIs microstructure and mechanical properties

In Fig. 3(a)–(c), the microstructures of ADI conducted under differ-


ent heat treatments are depicted. It shows that the matrix of all ADI
samples generally consists of a two-phase mixture of bainitic ferrite
and retained austenite. The morphology of the microstructure in ADIs
alters with the austempering temperature. With the reduction in
austempering temperature, the microstructure becomes finer and the
bainitic ferrite turns more acicular. The graphite nodule characteristics
in all testing ADIs samples have the same following measured data: av-
erage nodule diameter of about 33 μm, nodularity of 91%, and nodular
count of about 126 nod/mm2. The mechanical properties of ADIs treated
under different austempering temperature are shown in Table 2. The ul-
timate tensile strengths of them increase from 963 MPa to 1290 MPa,
the elongations decrease from 11.2% to 7.5% and the impact energies de-
Fig. 3. Microstructure of ADIs observed by optical microscope. (a) ADI1; (b) ADI2;
crease from 125 J to 98 J with the decrease of austempering tempera- (c) ADI3.
ture. It can be seen that the ultimate strength, the elongation and the
impact toughness of ADI1, ADI2 and ADI3 satisfy the requirement of
ASTM 897-06 grade 900/650/09, grade 1050/750/07 and grade 1200/ when austempered at 340 °C and 300 °C. The microhardness is 294,
850/04, respectively. 359 and 403 HV0.05, respectively for the ADI1, ADI2 and ADI3. Austenite
Fig. 4 shows the amount of retained austenite obtained by XRD is a softer phase to bainitic ferrite. Therefore, as the volume fraction of
analysis and microhardness values as a function of the austempering austenite increases the hardness is decreasing. The coarse grain gained
temperature. The volume fraction of austenite increases with the at high austempering temperature also causes the reduction in
decrease of austempering temperature [1,2,24] while the values of hardness [25]. It concludes that austempering treatment has a profound
hardness decrease. Samples of ADI austempered at 380 °C are found influence on the amount of retained austenite and the microhardness
containing 37% retained austenite, which decreases to 35% and 30% of ADI.
818 N. Zhang et al. / Materials and Design 89 (2016) 815–822

resulting in a decline in wear rate of ADI3 lower than that of ADI1 and
ADI2.

3.3. Coefficient of friction (CoF) of ADIs samples

Fig. 6 shows the steady-state coefficient of friction versus rolling cy-


cles at constant load(870 N)and sliding speed (0.0209 m/s) for three
strength grades ADIs. Although under the same condition, dry and con-
tinuous cooling, the average values of coefficient of friction differ from
each other. Austempered at 300 °C, ADI3 shows more stable but higher
coefficient of about 0.73than that of ADI1 and ADI2. The sample of ADI2
austempered at 340 °C gives lower friction coefficient lever of about
0.69, but shows relatively stable friction behavior. Finally, further in-
crease in temperature to 380 °C, friction behavior of ADI1 reveals the
most unstable and the lowest friction coefficient values of about 0.61.
This difference is caused by the converting in hardness. Previous inves-
tigation reported that the values of steady-state friction coefficient will
Fig. 4. Variation of hardness and retained austenite with austempering temperature. increase with the increase of samples' hardness [28]. Generally, graphite
present in the matrix structure can serve as lubricant when the disks roll
against to each other. As expected, ADI1 with the lowest bulk hardness
3.2. Wear rate of ADIs samples might undergo the most severe plastic deformation leads to the largest
amount of graphite smeared on the contact surface, which results in the
The wear performances of the three strength grades ADIs were eval- lowest coefficient of friction. In Fig. 6 the friction coefficient of ADI2
uated using an Amsler twin-disk rolling-sliding tester. Fig. 5 shows the gained from the wear test without the air cooling shows much lower
cumulative wear rate of ADI disks after 100,000 revolutions pairing value than that of those obtained under dry and cooling condition.
with rail steel disks. As shown, ADI1 with the softest matrix reveals This demonstrates that the graphite attributes to the self-lubricant ef-
the highest wear rate which means the lowest wear resistance. On the fect during the friction and wear process. Meanwhile, the coefficient
other hand, ADI3 with the hardest matrix reveals the lowest wear of friction in this study is much higher compared with the results in pre-
rate, reflecting its highest wear resistance. Wear rate of ADI2 is between vious report [11] due to the removal of graphite coveraged on the sur-
those of the other two and stands much closer to the ADI3, indicating face by the compressed air.
relatively good wear resistance as compared to the ADI1. Taking the ul-
timate tensile strength as the function of microhardness for three ADI 3.4. Study of the subsurface of ADIs samples
materials, it can be found the values of hardness increase as UTS in-
creases. Previous report has shown that the propagation of cracks in Fig. 7 shows the microhardness of the ADIs disks in the subsurface
fine microstructure tends to be split and deflected, which can optimize region measured on the cross section versus the distance below worn
the fracture toughness of the materials [25,26]. That means the wear re- surface. It can be generally found that the microhardness near the
sistance monotonically increases with the increases of hardness, worn surface significant improved compared to that in the bulk. It is
strength and fracture toughness and that results in the relatively good noted that the microhardness gradually decreases from the surface
wear resistance of ADI2 for its comparatively high microhardness and and thereafter, it reduces to the original value. The microhardness
UTS as well as good toughness. Additionally, wear rate level may also re- near the worn surface increases from 290 to 440HV0.05 for the ADI1.
late to the volume fraction of retained austenite which indicates that the Additionally, from 360 to 460HV0.05 for ADI2, and from 400 to
wear resistance depends on the amount of retained austenite [27]. The 500HV0.05 for ADI3. The variation of the microhardness becomes
reduction in austempering temperatures leads to the increase of bulk wider with the decrease in bulk hardness of the ADI samples. Measure-
hardness and UTS due to the decrease the amount of retained austenite, ment results on the values of thickness of the hardened layers in the

Fig. 5. Wear rate for three strength grades ADIs as wheel material at 5.4% slip and 870 N Fig. 6. Steady-state friction coefficient curves of rolling-sliding wear tests for ADI1, ADI2
contact load. and ADI3 under air cooling and ADI2 without air cooling.
N. Zhang et al. / Materials and Design 89 (2016) 815–822 819

Fig. 7. Microhardness profiles of a cross section of worn surface of ADIs.

three ADI samples are 130 μm, 100 μm, and 70 μm, approximately and
respectively. According to the previous reports, two major reasons ac-
count for this increase in hardness. Firstly, this increase is caused by
the subsurface deformation produced by the high load of 870 N applied
on the testing samples in rolling-sliding. Predominating over any fric-
tion heating effect, strain hardening of the matrix at the subsurface re-
gion induces increase in hardness of the samples [29]. Secondly, as
strain-induced transformation of austenite to martensite has been re-
ported [30], phase transformation contributes to the increase of hard-
ness near the worn surface. The X-ray diffraction patterns obtained
from three grades ADIs is shown in Fig. 8. The alteration in austenite
peaks for the samples before and after the wear tests can be analyzed
qualitatively by the diffraction patterns. In the case of the worn samples,
austenite peaks labeled (111), (220) and (311) have been reduced to
nearly background noise levels. This huge reduction in size of the austen-
ite peaks coupled with a little broadening of the ferrite/martensite (110),
(200) and (211) peaks showed in Fig. 8. It suggests that under extreme
conditions retained austenite can transform to martensite [31,32].
Fig. 9(a)–(c) shows the cross section observation of worn disks. It
can be seen that serious shear plastic deformation has occurred due to
friction force during wear tests. The depth of the plastic deformation
layers decreases with the increase of the material hardness and they
are in good accordance with the hardness measurements showed in
Fig. 7. The deformed layer of ADI1 is deeper than those of other two

Fig. 9. Cross section observed with SEM. (a) ADI1; (b) ADI2; (c) ADI3.

types. The friction force produces a plastic flow in the direction of sliding
in the worn samples. Meanwhile, micro-cracks are recognizable near
the surface in the plastic deformed area and they propagate along the
direction of the plastic flow. The graphite nodules close to surface also
turned into the same direction and squeezed to the surface and causes
lubrication on the contact area.
In general, the contact surfaces develop an oxidized layer during dry
sliding wear test under high load and uncooling condition [9,11,33,34].
Fig. 8. X-ray diffraction diagrams for ADIs before and after wear tests. However, oxidation does not occur in present work. As shown in Fig. 8,
820 N. Zhang et al. / Materials and Design 89 (2016) 815–822

the Fe3O4 (magnetite) or F2O3 (hematite) diffraction peaks are not ob-
served in the XRD spectrum at the surface of ADI specimens after
wear test. Besides, reddish-brown oxide areas are not found on the
worn surface shown in Fig. 9 and Fig. 10. So it is considered that the sur-
face oxidation is well prevented due to the cooling of compressed air
and the influence of it on the wear behavior is prevented.

3.5. Wear mechanism under dry rolling-sliding

Fig. 10(a)–(d) shows the SEM micrographs of worn surface of three


strength grades ADIs. The worn surfaces of ADI samples in (a)–(c) show
different morphology, spalling layers and deformed graphite holes can
be seen on the worn surface which means ADI disks undergo wear by
delamination due to the subsurface fatigue during rolling and sliding.
Meanwhile, Fig. 10(d) shows the worn morphology of ADI2 after wear
test matched with rail steel without air cooling. Delamination is the
main wear mechanism under no air cooling condition also. Superficial
graphite in ADI2 smeared over the surface can be seen in Fig. 10(d).
The present results of wear characteristic are strongly related to the sur-
Fig. 11. Wear rate for ADIs/steel and conventional wheel/rail steel pairings at 5.4% slip and
face hardness of ADIs, the delamination becomes mild with the increase 870 N contact load.
of surface hardness. As previously mentioned, the shearing friction force
caused plastic flow, in which the matrix and graphite nodules deformed
plastically as cracks and propagated in the sliding direction [28]. Actual- comparison with the conventional wheel/rail steel pairing, which was
ly, with the decrease of the bulk hardness, the area of deformed layer tested under the same condition, should be figured out. Fig. 11 shows
becomes wider, which induced more occurrence of the crack. In addi- the wear rate of three strength grades ADIs matched with rail steel
tion, cracks in the soft sample show longer propagation distance as U75V and a conventional wheel/rail (ER8/U75V) pairing under a normal
shown in Fig. 9. It is concluded that with the increase of the microhard- load of 870 N after 100,000 rotations. As illustrated in Fig. 11, the ADI1/
ness, the spalling caused by delamination becomes easier and larger. U75V pairing shows the largest wear rate on wheel disk and the lowest
This may explain the surface feature changing in ADI samples. wear on rail disk. The cause of this is primarily the lower hardness of
ADI1 relative to U75V. ADI2/U75V and ADI3/U75V pairings show more
3.6. Wear behavior comparison between ADIs and wheel steel ER8 favorable wear characteristics. ADI2 and ADI3 show much less wheel
disks wear than that of ER8, but show almost the same damage to rail
To investigate the favorable wear characteristics of ADIs with differ- disks despite their relatively high hardness. It is reported that the
ent strength grades as a wheel material matched with rail steel, a better wear rate of ADI disk matched with rail steel is lower than that of

Fig. 10. Observation on worn surface. (a) ADI1, (b) ADI2, (c)ADI3 under dry and cooling condition; (d) ADI2 without air cooling.
N. Zhang et al. / Materials and Design 89 (2016) 815–822 821

conventional wheel/rail steel pairings [16]. In this study, the com- reflect the fracture toughness of ADI, decrease with the increasing UTS.
pressed air may account for this difference. During the tests, the com- According to the wear test results, both ADI2 and ADI3 have relative
pressed air of 0.6 MPa may blow away the smeared graphite, which higher hardness and superior friction and wear behavior than those of
can limit the lubricating action in the boundary layer between friction ADI1 and conventional wheel steel ER8. In addition the morphology of
partners. However, in the actual operation, the wind caused by the worn surfaces of rail specimens show that ADI1 and ADI2 causes milder
high speed of the railcar running generally cannot achieve such high damage than that for ADI3 to rail specimens. Thus ADI2 shows reason-
pressure especially under the extreme situation. So in actual running able friction and wear behavior as well as relative superior mechanical
ADI2 and ADI3 may have better wear and friction properties due to properties, which promise to meet the demands of railway wheel fac-
the existence of graphite in the matrix. tors, reflecting the best suitability as an alternative material for railcar
In Fig. 12(a), (b) and (c), top views of rail steel specimens matched wheels.
with ADI1, ADI2, and ADI3 after the tests are shown respectively. In
Fig. 12(a) the worn surface is characterized by plowing grooves along
4. Conclusions
the wear direction, as well as thin spalling marks. As for Fig. 12(b) the
micropitting and mild fatigue spalling due to microcracks is significant.
In this study, the dry rolling-sliding friction and wear behaviors of
Long delamination cracks are clearly observed on the worn surface of
three strength grades austempered ductile irons matched with conven-
rail steel specimen matched with ADI3, in Fig. 12(c). In the wear pairing
tional rail steel were investigated. Their behaviors were also compared
of ADI1/U75V, the appearance of plowing reveals the mildest wear
to that of a conventional wheel steel. The following points can be
damage on the rail specimen. As a result, rail specimen pairing with
drawn:
ADI1 shows the lowest wear rate (shows in Fig. 11). The worn surfaces
in Fig. 12(b) and c show cracks at the surfaces of rail specimens. Propa-
1. With the increase of material hardness of ADIs, the mass loss and
gation of the microcrack leads to top pitting and thin flakes break away
wear rate are getting lower while the friction coefficient becomes
from the surface, while long cracks may cause severe fatigue spalling on
more stable and higher.
the surface resulting in large loss of material and wear rate. So it can be
found that the wear level of rail specimen becomes severe with increase 2. The increase of hardness on subsurface after wear test is due to work-
of the hardness of their matched materials. hardening and stain-induced transformation of retained austenite to
In general, selection of a proper wheel material must satisfy the re- martensite.
quirement of the counteraction of its wear, fatigue and fracture proper- 3. The main wear mechanism is delamination where spalling layers can
ties, which may induce a catastrophic failure [35]. Material with higher be seen on the worn surface. The delamination of ADIs becomes mild
wear resistance, usually possesses higher hardness, leading to a lower with the increase of matrix hardness.
fracture toughness and a higher fatigue crack growth rate. Therefore, 4. ADI austempered at 340 °C shows the reasonable friction and wear
the combination of higher wear resistance and higher fracture tough- behavior as well as relative superior mechanical properties, which
ness should be taken into account in the development of a wheel mate- reflecting the best suitability as an alternative material for railcar
rial. As shown in Table 2, values of impact energy and elongation, which wheel.

Fig. 12. Worn surface observations of the rail steel specimens matched with ADI1, ADI2 and ADI3, respectively.
822 N. Zhang et al. / Materials and Design 89 (2016) 815–822

Acknowledgments [16] S. Dündar, Application of austempered ductile iron to rail wheel sets, Journal of En-
gineering Sciences 9 (2003) 283–287.
[17] K. Jokipi, Kymenite in Railway Applications, Conference Group of the ADI-Seminar,
This work was supported by the National Natural Science Founda- Technical University Helsinki 1991, pp. 1–16.
tion of China (No. 51305363) and State Key Laboratory Independent Re- [18] M. Fanling, I. Baker, P. Munroe, The effects of environment on the dry sliding wear of
eutectic Fe 30Ni 20Mn 35Al 15, J. Mater. Sci. 47 (2012) 4827–4837.
search of China (No. 2015TPL_T11). [19] Y. Lyu, Y. Zhu, U. Olofsson, Wear between wheel and rail: a pin-on-disc study of en-
vironmental conditions and iron oxides, Wear 328–329 (2015) 277–285.
References [20] Y. Zhu, X. Chen, W. Wang, H. Yang, A study on iron oxides and surface roughness in
dry and wet wheel–rail contacts, Wear 328–329 (2015) 241–248.
[1] C.-H. Hsu, K.-T. Lin, A study on microstructure and toughness of copper alloyed and [21] ASTM E-8, Annual Book of ASTM Standards, vol. 3.011998 545–557.
austempered ductile irons, Mater. Sci. Eng. A 528 (2011) 5706–5712. [22] GB 12444.1-90, Metallic Materials-MM Mode Wear Test. Standardization Adminis-
[2] J. Zhang, N. Zhang, M. Zhang, L. Lu, D. Zeng, Q.P. Song, Microstructure and mechan- tration of the People's Republic of China, 1990.
ical properties of austempered ductile iron with different strength grades, Mater. [23] K.L. Johnson, Contact Mechanics, Cambirdeg University Press, UK, 1985.
Lett. 119 (2014) 47–50. [24] P.H.S. Cardoso, C.L. Israel, T.R. Strohaecker, Abrasive wear in austempered ductile
[3] M. Tayanç, K. Aztekin, A. Bayram, The effect of matrix structure on the fatigue be- irons: a comparison with white cast irons, Wear 313 (2014) 29–33.
havior of austempered ductile iron, Mater. Des. 28 (2007) 797–803. [25] J. Yang, S.K. Putatunda, Improvement in strength and toughness of austempered
[4] A. Refaey, N. Fatahalla, Effect of microstructure on properties of ADI and low alloyed ductile cast iron by a novel two-step austempering process, Mater. Des. 25 (2004)
ductile iron, J. Mater. Sci. 38 (2003) 351–362. 219–230.
[5] U. Batra, N. Batra, J.D. Sharma, Wear performance of Cu-alloyed austempered ductile [26] Y.P. Wang, D.Y. Li, L. Parent, H. Tian, Improving the wear resistance of white cast iron
iron, J. Mater. Eng. Perform. 22 (2012) 1136–1142. using a new concept — high-entropy microstructure, Wear 271 (2011) 1623–1628.
[6] R. Martins, J. Seabra, L. Magalhães, Austempered ductile iron (ADI) gears: power [27] M.C. Jeng, J.L. Doong, Abrasive wear study of bainitic nodular cast iron, J. Mater. Sci.
loss, pitting and micropitting, Wear 264 (2008) 838–849. 28 (1993) 6555–6561.
[7] L. Magalhaes, R. Martins, J. Seabra, Low-loss austempered ductile iron gears: exper- [28] J. Zhang, N. Zhang, M. Zhang, L. Lu, D. Zeng, Q.P. Song, Rolling-sliding wear of
imental evaluation comparing materials and lubricants, Tribol. Int. 46 (2012) austempered ductile iron with different strength grades. Improvement of ductile
97–105. iron wear resistance through local surface reinforcement, Wear 318 (2014) 62–67.
[8] J. Lefevre, K.L. Hayrynen, Austempered materials for powertrain applications, J. [29] M.A. Islam, A. Haseeb, A.S.W. Kurny, Study of wear of as-cast and heat-treated sphe-
Mater. Eng. Perform. 22 (2013) 1914–1922. roidal graphite cast-iron under dry sliding conditions, Wear 188 (1995) 61–65.
[9] J. Zimba, M. Samandi, D. Yu, T. Chandra, E. Navara, D.J. Simbi, Un-lubricated sliding [30] S. Panneerselvam, C.J. Martis, S.K. Putatunda, J.M. Boileau, An investigation on the
wear performance of unalloyed austempered ductile iron under high contact stress- stability of austenite in austempered ductile cast iron (ADI), Mater. Sci. Eng. A 626
es, Mater. Des. 25 (2004) 431–438. (2015) 237–246.
[10] A. Zammit, S. Abela, L. Wagner, M. Mhaede, M. Grech, Tribological behaviour of shot [31] J. Zimba, D.J. Simbi, E. Navara, Austempered ductile iron: an alternative material for
peened Cu–Ni austempered ductile iron, Wear 302 (2013) 829–836. earth moving components, Cem. Concr. Compos. 25 (2003) 643–649.
[11] G. Straffelini, C. Giuliari, M. Pellizzari, E. Veneri, M. Bronzato, Dry rolling-sliding wear [32] R. Dommarco IG, H. Oriz, P. David, G. Maglieri, The use of ductile iron for wheel
of austempered cast iron, Wear 271 (2011) 1602–1608. loader bucket tips, Wear 249 (2001) 101–108.
[12] G. Balachandran, A. Vadiraj, M. Kamaraj, E. Kazuya, Mechanical and wear behavior of [33] G. Straffelini, M. Pellizzari, L. Maines, Effect of sliding speed and contact pressure on
alloyed gray cast iron in the quenched and tempered and austempered conditions, the oxidative wear of austempered ductile iron, Wear 270 (2011) 714–719.
Mater. Des. 32 (2011) 4042–4049. [34] U.R. Kumari, P.P. Rao, Study of wear behaviour of austempered ductile iron, J. Mater.
[13] M.J. Pérez, M.M. Cisneros, H.F. López, Wear resistance of Cu–Ni–Mo austempered Sci. 44 (2009) 1082–1093.
ductile iron, Wear 260 (2006) 879–885. [35] U.P. Singh, A.M. Popli, D.K. Jain, B. Roy, S. Jha, Influence of microalloying on mechan-
[14] J. Hemanth, Wear characteristics of austempered chilled ductile iron, Mater. Des. 21 ical and metallurgical properties of wear resistant coach and wagon wheel steel, J.
(2000) 139–148. Mater. Eng. Perform. 12 (2003) 573–580.
[15] J. Perez, J. Sinclair, J. Tunna, A review of wheel wear and rolling contact fatigue, Pro-
ceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid
Transit, 221 2007, pp. 271–289.