Vacuum 107 (2014) 132e140

Contents lists available at ScienceDirect

Vacuum
journal homepage: www.elsevier.com/locate/vacuum

Microstructure and tribological properties of iron-based metallic glass

coatings prepared by atmospheric plasma spraying
Yulong An a, Guoliang Hou a, Jie Chen a, b, Xiaoqin Zhao a, *, Guang Liu c, Huidi Zhou a, *,
Jianmin Chen a
a
State Key Laboratory of Solid Lubrication, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, China
b
Graduate School, Chinese Academy of Sciences, Beijing 100049, China
c
Ningbo Branch of China Academy of Ordnance Science, Ningbo 315103, China

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

Article history: Iron-based metallic glass coatings (denoted as FeWCrNiMoBSiC) were prepared on 1Cr18Ni9Ti stainless
Received 22 January 2014 steel cylinders by atmospheric plasma spraying at different parameters. The morphology, microstructure,
Received in revised form and crystalline structure of as-prepared Fe-based metallic glass coatings were analyzed by scanning
16 April 2014
electron microscopy, transmission electron microscopy, and X-ray diffraction. A Pycnometer and a
Accepted 17 April 2014
Vickers hardness tester were adopted to measure the porosity and microhardness of iron-based metallic
glass coatings. Moreover, differential scanning calorimetry analysis was conducted to investigate the
Keywords:
crystallization behavior of various iron-based metallic glass coatings, and a ball-on-disk tribometer was
Atmospheric plasma spraying
Iron-based metallic glass coating
performed to evaluate the tribological properties of the coatings coupled with silicon nitride ceramic
Microstructure balls under unlubricated conditions. It has been found that the microhardness of iron-based metallic
Tribological properties glass coatings increases with increasing plasma arc power, which is related to the degree of melting of
feedstock powders and the compactness of as-prepared coatings. Besides, the phase compositions of as-
sprayed coatings consist of amorphous structure and limited crystalline structure, and the contents of the
amorphous structure and crystalline structure vary with plasma arc power. Moreover, iron-based
metallic glass coatings deposited at different plasma arc powers show similar steady-state friction co-
efficients (0.8e0.9), but their wear rate varies with varying plasma arc power. Particularly, iron-based
metallic glass coating with next to the highest hardness exhibits the best anti-wear ability, which is
the outcome of the compromise between the hardness and brittle fracture as well as abrasive wear of the
coatings during sliding process.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction eliminated by making use of thermal spraying method, since

Amorphous alloy, a kind of materials without characteristics of
long-range ordered crystal, possesses many excellent properties
such as high hardness, superior corrosion resistance, and good
tribological properties thanks to no dislocation and grain boundary
in them [1e7]. To date, several methods including melt spinning
[8], high-pressure die casting [9], water quenching [10] and suction
casting [11] are available for preparing amorphous metals. How-
ever, these methods are only suitable for fabricating amorphous
alloys with a thickness of centimeter scale, which largely restricts
their wide application. This drawback, fortunately, could be
* Corresponding authors.
E-mail addresses: zhaoxiaoqin@licp.cas.cn (X. Zhao),
(H. Zhou).
thermal spraying has a rapid quenching rate of molten particles
(around 105e107 K s
1) and is suitable for preparing amorphous
coatings on the surfaces of numerous components with different
shapes [12,13].
Among various amorphous metals, Fe-based bulk metallic
glasses (BMGs) are perhaps the most important antiwear materials,
due to the combination of rather low material cost, ultrahigh
hardness and strength, and outstanding corrosion-resistance [14e
19]. In the field of tribology, Fe-based BMGs are of particular sig-
nificance, since they can be introduced onto the surface of various
mechanical components so as to reduce friction and wear and
alleviate deteriorations caused by progressive and undesirable loss
or degradation of surface materials. To name a few, Kishitake and
co-workers claimed that plasma-sprayed amorphous Fee10Cre
hdzhou@lzb.ac.cn 13Pe7C alloy coatings exhibit high hardness and excellent corro-
sion resistance [20]. Cherigui et al. reported that Fe3Si and FeNb

http://dx.doi.org/10.1016/j.vacuum.2014.04.021
0042-207X/Ó 2014 Elsevier Ltd. All rights reserved.
 Y. An et al. / Vacuum 107 (2014) 132e140
Table 1
Atmospheric plasma spraying parameters for depositing iron-based metallic glass coatings on stainless steel cylinders.
Coatings Plasma spray parameters
Current Voltage (V) Power Argon flow
(A) (kW) rate (L/min)
Coating 1 500 50 25 60
Coating 2 500 60 30 60
Coating 3 500 70 35 60
Coating 4 500 80 40 60
metallic glass coatings deposited by high velocity oxy-fuel spraying
(HVOF) and atmospheric plasma spraying (APS) exhibit good
magnetic properties [21,22]. Chokethawai and co-workers pre-
pared highly amorphous HVOF coatings from partial amorphous
FeCrMoWBCSi powders; and they found that the content of
amorphous phases in as-sprayed coating is as much as 50% (mass
fraction, the same hereafter), much higher than the amorphous
phase content in sprayed powders (18%) [23]. More importantly,
the above-mentioned iron-based metallic glass coatings deposited
by thermal spraying technology can be well used to ameliorate
wear resistance of substrate materials [24,25]. In addition, APS is an
effective and widely recognized technique for preparing metallic
glass coatings with high quality [26], together with it is more
convenient and economical than other thermal spray technologies
(such as HVOF, low pressure plasma spraying (LPPS) [27] and vac-
uum plasma spraying (VPS)), and thus investigating the properties
of Fe-based metallic glass coatings prepared by APS has important
scientific significance and engineering value.
Bearing those perspectives in mind, APS technique was
employed in the present research to prepare Fe-based metallic
glass coatings on cast 1Cr18Ni9Ti stainless steel substrate with the
amorphous powder of FeWCrNiMoBSiC as the feedstock powder;
owing to the spraying parameters strongly affect the structure,
phase composition and other properties of APS-spraying coatings
[28e30], the influence of spraying parameters on the microstruc-
ture and amorphous content as well as tribological properties of as-
sprayed Fe-based metallic glass coatings was also investigated. This
paper reports the relationship among the spraying parameters and
the microstructure, phase composition and tribological properties
of as-prepared Fe-based metallic glass coatings. Meanwhile, the
structure of as-sprayed coating is compared with iron-based
metallic glass coating that was deposited by LPPS in our previous
researches [12].
2. Experimental procedure
2.1. Feedstock powders
Commercially available Fe-based amorphous powders (FeW-
CrNiMoBSiC) provided by Beijing Sangyao Technology development
Co., Ltd. (China) with a size of about 5w50 mm were prepared by
alloy water atomization. The composition of the feedstock powders
is Fee10We4Cre3Nie2Moe4Be4Sie1C by mass ratio.
2.2. Preparation of coatings
An atmospheric plasma spraying system equipped with a PQ-1S
plasma torch (Beijing Institute of Aeronautical Manufacturing
Technology; Beijing, China) manipulated with an IRB 2400/16 robot
(ABB, Switzerland) was employed to spray the feedstock powders
onto 1Cr18Ni9Ti stainless steel cylinders (dimensions: 24 mm in
diameter and 8 mm in height) to form iron-based metallic glass
coatings with a thickness of about 180e220 mm (measured with a
digital micrometer at a resolution of 1 mm). Ultra pure argon (purity,
133
Hydrogen flow Powder feed rate Spray distance Gun speed Injector
rate (L/min) (g/min) (mm) (mm/s) angle ( )
11 40 100 800 90
16 40 100 800 90
22 40 100 800 90
30 40 100 800 90
99.9999%) was used as the primary gas, and highly pure hydrogen
(purity, 99.999%) was adopted as the secondary gas. The plasma arc
powers were adjusted properly so as to obtain desired FeW-
CrNiMoBSiC metallic glass coatings under different spraying pa-
rameters. The details about the sets of spraying parameters are
presented in Table 1. For convenience, coatings prepared under
different plasma spraying conditions are denoted as coating 1,
coating 2, coating 3 and coating 4, respectively. During the plasma
spraying process, no accelerated cooling of the coating and steel
substrate was conducted, which helps to eliminate the effect of
accelerated cooling on formation of amorphous phase from molten
feedstock powders.
2.3. Characterization of feedstock powders and coatings
The morphology of feedstock powders and as-sprayed iron-
based metallic glass coatings was analyzed with a JSM-5600LV
scanning electron microscope (SEM; JEOL, Japan), with which sec-
ondary and backscattered electron images were recorded at a
voltage of 20 kV.
The porosity of the iron-based coatings was measured based on
that the true volume of one material with some pores is smaller
than its geometric volume, and the ratio of the difference between
them to the geometric volume is the porosity of this material.
Therefore, the porosity of the iron-based coatings could be calcu-
lated as
P ¼ ð1
VcT =Vc Þ  100% ¼ ð1
Vs;cT
V =V
s s;c
Vs Þ  100%, where
Vs was the geometric volume of stainless steel substrate, Vs,c was
the geometric volume of the substrate coated with the iron-based
coating, Vc was the geometric volume of the iron-based coating
T was the true volume of the substrate
and equal to (Vs,c
Vs); Vs;c
coated with the iron-based coating, and VcT was the true volume of
the iron-based coating and equal to ðVs;c T
V Þ. V and V
s s s,c in this
paper was 12 p  8 mm and 12 p  8.2 mm3, respectively. The Vs;c
2 3 2 T
was determined with Micromeritics AccuPyc 1330 Pycnometer
(Micrometerics Instrument Corporation, Norcross, GA, USA) in high
purity helium (99.999%). The Micromeritics AccuPyc 1330 Pyc-
nometer is a fast, non destructive, fully automatic true volume
analyzer that provides high-precision volume. This pycnometer is a
device that determines the porosity of samples by actually
measuring their true volume very precisely. The AccuPyc works by
measuring the amount of displaced gas (helium). The pressures
observed upon filling the sample chamber and then discharging it
into a second empty chamber allow computation of the sample
solid phase volume. Gas molecules rapidly fill the tiniest pores of
the sample; only the truly solid phase of the sample displaces the
gas [31e33].
A JEM2010 transmission electron microscope (TEM; JEOL, Japan)
was used to analyze the microstructure of as-sprayed iron-based
metallic glass coatings. The phase composition of the starting
powders and as-sprayed coatings was analyzed with a D/Max-2400
X-ray diffractometer (XRD, Rigaku, Japan; Cu-Ka radiation, poten-
tial 40 kV, current 100 mA), with which the diffraction data were
collected at a step size of 0.02 in the range of 10  2q  90 .
134 Y. An et al. / Vacuum 107 (2014) 132e140

The microhardness of the coatings was measured with an MH-5-

VM microhardness tester at a load of 3 N and for a dwelling time of
10 s. Microhardness values were obtained from the average of 10
individual measurements performed on polished coating surfaces.

2.4. Differential scanning calorimetry analysis

Differential scanning calorimetry (DSC) analysis of the feedstock

powders and as-prepared coatings was conducted with a Netzsch
STA 449C simultaneous thermal analyzer (Netzsch-Gerätebau
GmbH, Selb, Germany; fixed heating rate 10 K min
1, nitrogen at-
mosphere, from room temperature to 1075 K) equipped with a
DSC-sample carrier. The enthalpy of crystallization was calculated
with an accuracy of 3% by taking into account the DSC data. The
activation energy of crystallization was calculated from the
exothermic peak corresponding to the crystallization process of the
amorphous phase based on Kissinger equation. Then the relative
content of amorphous phase (Pcoating) of the coatings was deter-
mined based on the exothermic enthalpies of the coatings
(DHcoating) and feedstock powders (DHpowder):
Pcoating ¼ DHcoating/DHpowder  100% (1)
2.5. Friction and wear tests
A ball-on-disc tribometer (CSM, Tribo-S-D-0000) was adopted to
evaluate the tribological properties of iron-based metallic glass
coatings sliding against commercially obtained Si3N4 balls (6 mm in
diameter) in air. Before tests, the coatings were sequentially ground
with 320, 600, 800, 1000, and 1200 mesh SiC papers, and polished
with 3 mm and 0.5 mm diamond pastes. The polished coatings and
Si3N4 balls were ultrasonically cleaned in acetone for 5 min and then
dried in air. Then the sliding tests were conducted at a room tem-
perature of about 20  C, a relative humidity of 20%e30%, a sliding
distance of 400 m, a load of 6 N (corresponding to a contact pressure
of 722.8 MPa), and a constant speed of 20 cm s
1. The friction coef-
ficient curves were automatically recorded with a computer con-
nected to the tribometer. The wear rate is calculated as W ¼ V/SF,
where V is the wear volume loss in m3, S the total sliding distance in
m, and F the applied load in N. Three repeat tests were conducted for
each coating sample under pre-set test conditions, and the average
values of the repeat tests are reported in this article. Moreover, upon
completion of friction and wear tests, the volume loss of the coatings
was evaluated with an ADE MicroXAM three-dimensional (3D) non-
contact surface mapping profiler (Massachusetts, USA); and the
morphology of worn coating surfaces was examined also by SEM.
Fig. 1. SEM image of water atomized iron-based amorphous feedstock powders.
particles and pores on the surface Fig. 3(a). This is because such a
low plasma arc power is inadequate to completely melt the feed-
stock powders, thereby leading to loosely combined splats in the
amorphous coating. When the plasma arc power rises to 30 kW and
35 kW, the amount of un-melted partially molten particles declines
greatly, and the pores with large sizes nearly disappear in resultant
coating 2 and coating 3 Fig. 3(b and c). At the highest plasma arc
power of 40 kW, no spherical particles are visible on the surface of
coating 4, and many fine particles emerge owing to the splash of
well melted splats Fig. 3(d), which indicates that a plasma arc po-
wer of 40 kW is high enough for fully melting the feedstock
powders.
It is well known that the microstructures of thermally sprayed
coatings are very complex and always possess some process-
dependent defects such as globular pores, interlamellar pores,
etc. More importantly, porosity influences a wide range of coating
properties [35]. Thus, the inner microstructure and porosities of the
four coatings are also characterized. Fig. 4 gives BSE images of
cross-sections of various as-sprayed coatings. It can be seen that the
number of partially molten particles in the coatings decreases with
increasing plasma arc power, which is considered to be help to
prevent the formation of pores in the coatings. This is because pores
in thermal sprayed coatings are generally derived from poorly

3. Results and discussion

3.1. Morphology and phase structure of feedstock powders

Fig. 1 shows a typical SEM image of feedstock powders. It can be

seen that most of the feedstock particles have irregular or near-
spherical shape, which conforms to the morphological features of
water atomized powders [34]. Fig. 2 presents the XRD pattern of the
feedstock powders. The presence of the broad protuberance and
the absence of any obvious XRD peaks of crystalline indicate that
as-obtained feedstock powders are amorphous.

3.2. Microstructure of as-sprayed coatings

Fig. 3 gives SEM images of iron-based metallic glass coatings

obtained under different plasma arc powers. Coating 1 obtained at
a low plasma arc power of 25 kW contains many partially molten Fig. 2. XRD pattern of water atomized iron-based amorphous feedstock powders.
 Y. An et al. / Vacuum 107 (2014) 132e140 135

Fig. 3. SEM images of as-sprayed iron-based metallic glass coatings (a) coating 1, (b) coating 2, (c) coating 3, and (d) coating 4.

Fig. 4. BSE images of cross-sections of as-sprayed iron-based metallic glass coatings (a) coating 1, (b) coating 2, (c) coating 3, and (d) coating 4.
136 Y. An et al. / Vacuum 107 (2014) 132e140

stacked flat particles [30], and the state of the flat particles is highly
dependent on the melted extent of the feedstock powders and the
velocity of plasma jet. Naturally, elevating plasma arc power favors
to increase the melted degree of the feedstock powders and the
velocity of plasma jet as well. As a result, well melted feedstock
powders can easily form flat splats, and these flat splats can inte-
grate tightly, resulting in fewer pores existing at the junctions of the
splats. This is further confirmed by the porosity of various coatings
listed in Table 2. Namely, the porosity of as-sprayed coatings tends
to decline from coating 1 to coating 4.
Compared to LPPS sprayed iron-based metallic glass coating in
Refs. [12], it can be found that both APS and LPPS coatings contain
many partially molten particles under the low plasma arc power,
because the spraying powders can not be well melted in the low
temperature plasma flame. Besides, APS coating exhibit higher
porosity; while the porosity of LPPS coating can be controlled
below 0.5%. This may lie in that LPPS has higher velocity of plasma
jet than APS, and thus the spraying powders can be obtained higher
flight speed in LPPS flame and then spread out more adequately Fig. 5. XRD patterns of various iron-based metallic glass coatings obtained under
upon impact. different spraying parameters.

3.3. Phase composition of as-sprayed coatings

Refs. [12], the relative content of amorphous phase of LPPS coating
can be improved to 98%, which is higher than that in APS coating.
Fig. 5 gives XRD patterns of various iron-based metallic glass
This is mainly due to that more serious oxidation of spraying
coatings obtained under different spraying parameters. It is seen
powders in atmospherical environment prompts the crystalline
that coating 1 and coating 2 show only broad protuberances but
process of amorphous phase.
have no obvious signals of crystalline, which indicates that, within
TEM was undertaken to obtain more detailed information on
the resolution limit of XRD analysis, coating 1 and coating 2 are
the microstructure of coating 2 and coating 4 as typical examples
amorphous. Contrary to the above, coating 3 and coating 4 show an
(selected area electron diffraction (SAED) patterns are also given in
obvious crystalline peak at 2q ¼ 44.6 . This implies that high
Fig. 8. In terms of relevant TEM analyses (compare Fig. 8(a) with
plasma arc power could facilitate the formation of supersaturated
Fig. 8(c)), coating 2 and coating 4 seem to have similar micro-
a-Fe solid solution phase in iron-based metallic glass coating [36].
structure. Besides, the diffused halo ring in the SAED pattern of
Fig. 6 gives the DSC curves of various iron-based metallic glass
coating 2 (Fig. 8(b)) confirms that it mainly consists of amorphous
coatings prepared under different spraying parameters in order to
phases, while the diffraction ring in the SAED pattern implies that it
estimate the content of amorphous phases. Fig. 7 shows XRD
also contains a small amount of polycrystalline phases. Similarly,
pattern of the feedstock powders after DSC test. Feedstock powder
the diffused halo ring in the SAED pattern of coating 4 (Fig. 8(d))
and as-prepared iron-based metallic glass coatings show similar
confirms that it also has amorphous structure; however, the
thermodynamic behavior. Namely, they all show three sequential
diffraction spots validate the existence of crystals in coating 4, as
exothermic peaks in the DSC curves, and these peaks are almost
evidence by relevant XRD analysis (see Fig. 5).
located at the same position. Moreover, feedstock powder and iron-
Considering the above analyses, we can conclude that all as-
based metallic glass coatings experience crystallization around
prepared iron-based metallic glass coatings prepared under
873 K and the amorphous phase in them has been completely
different plasma spraying parameters contain amorphous phase
transformed to crystalline phase after DSC test; and the exothermic
and limited crystalline phase, and the contents of the two phases
enthalpy values of these peaks around 873 K of feedstock powder
and coating 1, coating 2, coating 3 and coating 4 are calculated to
be
65.04 J g
1,
52.13 J g
1,
58.01 J g
1,
51.87 J g
1
and
48.55 J g
1, respectively. Accordingly, the relative content of
amorphous phase of coatings 1, 2, 3 and 4 is determined to be 80.2%
(mass fraction; the same hereafter), 89.2%, 79.8% and 74.6%. This
indicates that the content of amorphous phases in as-sprayed iron-
based metallic glass coatings rises first and declines later with the
increase of plasma arc power, but it is always lower than the con-
tent of amorphous phase in the feedstock powder, which is possibly
due to the transformation of a part of amorphous phases into
crystalline phases during plasma spraying process. As reported in

Table 2
Porosity of as-sprayed iron-based metallic glass
coatings.

Coatings Porosity (%)

Coating 1 9.0
Coating 2 5.2
Coating 3 3.9
Fig. 6. DSC curves of various iron-based metallic glass coatings obtained under
Coating 4 2.6
different spraying parameters.
 Y. An et al. / Vacuum 107 (2014) 132e140 137

Fig. 9. Friction coefficient curves of iron-based metallic glass coatings sliding against
Si3N4 balls with sliding distance at 6 N and 20 cm s
1.
Fig. 7. XRD pattern of the feedstock powders after DSC test.

further increased to 35 kW and 40 kW, the oxidation extent of

are strongly dependent on the cooling rate and oxidation degree of
spraying powders as well as heat accumulating inside the as-
sprayed coatings [13]. To be specific, many spraying particles can
not be enough melted at a low plasma arc power of 25 kW owing to
the low flame temperature, and the partially molten zones may
cool down at low cooling rate upon impact, which goes against the
formation of amorphous phase in these zones. Thus, the content of
amorphous phase in coating 1 is only 80.2%. As the plasma arc
power is increased to 30 kW, the amount of partially molten par-
ticles in coating 2 is largely reduced and the cooling rate is greatly
elevated thereby resulting in increased content of amorphous
phase than in coating 1. However, when the plasma arc power is
spraying powders is accelerated, together with the obvious influ-
ence of localized reheating from the latent heat evolutions as
overlapping and connecting layers of melted splats solidify on the
coatings [13], thereby prompting many amorphous phase to be
transformed to crystalline phase. As a result, coating 3 and coating
4 exhibit obvious XRD peaks of crystalline phases (see Fig. 5).
3.4. Tribological properties of as-sprayed coatings
Fig. 9 gives the friction coefficient versus sliding distance curves
of various iron-based metallic glass coatings. The four kinds of iron-
based metallic glass coatings have similar friction coefficients as

Fig. 8. TEM images of coating 2 and coating 4 ((a) and (c)) as well as their SAED patterns ((b) and (d)).
138
Fig. 10. Microhardness and wear rate of as-sprayed iron-based metallic glass coatings
after sliding wear tests at 6 N and 20 cm s
1.
they slide against the ceramic ball under dry sliding conditions; and
the steady-state friction coefficients remain in the range of 0.8e0.9.
This means that altering plasma spraying parameters have few
effects on the friction coefficient of as-prepared iron-based metallic
glass coatings.
Fig. 10 presents the microhardness and wear rate of as-sprayed
iron-based metallic glass coatings. It is seen that the microhardness
of various iron-based metallic glass coatings tends to rise to some
extent with increasing plasma arc power; and in particular, coating
4 obtained under the highest plasma arc power has the highest
hardness, which should be closely related to its dense structure and
strong crystallization as well. Moreover, although the wear rate of
various iron-based metallic glass coatings roughly tends to decline
with increasing plasma arc power, it is coating 3 with next to the
highest hardness rather than coating 4 with the highest hardness
that exhibits the lowest wear rate. This could be explained from
two aspects. On the one hand, as described by Archard wear
equation, the wear rate of materials is inversely proportional to
Fig. 11. SEM images of wear scars of various as-sprayed iron-based metallic glass coatings after sliding wear tests at 6 N and 20 cm s
1: (a) coating 1, (b) coating 2, (c) coating 3, and
(d) coating 4.
Y. An et al. / Vacuum 107 (2014) 132e140
hardness [37,38]. On the other hand, the nanocrystal will take an
effect in strengthening and toughing the as-sprayed coatings,
which is also beneficial to improve the anti-wear property of
coatings [39,40]. From the XRD patterns of the four coatings (see
Fig. 5), together with their TEM images and SAED patterns, it can be
inferred that the polycrystalline in coating 1 and coating 2 has
developed into nanocrystal in coating 3, but has developed into
crystal with big sizes in coating 4. As a result, the coating 3 has the
best toughness and thus shows better anti-wear behavior than
coating 4.
SEM images and 3D profiles of the wear scars of various iron-
based metallic glass coatings are given in Figs. 11 and 12. It is
seen that the worn surfaces of all coatings show signs of brittle
fracture and splat detachment (Fig. 10). This could be rationally
understood, since iron-based metallic glass coatings possess poor
ductility and toughness [41]. In other words, although iron-based
metallic glass coatings possess high microhardness, they are
liable to brittle facture during friction process. Besides, plasma
sprayed coating is built up by splats, and the splats tend to suffer
detachment from the coating under high Hertz load and repeated
shear stress [42]. As a result, wear of various iron-based metallic
glass coatings is dominated by brittle fracture and splat detach-
ment during sliding against the ceramic balls. Particularly, coating 1
and coating 2 deposited at low plasma arc powers exhibit severe
splat detachment Fig. 11(a and b), due to non-complete melting of
the feedstock powders as well as poor inter-splat combination.
Thanks to fully melting of feedstock powders at increased plasma
arc powers of 35 kW and 40 kW, coating 3 and coating 4 possess
increased combination strength of splats and better wear resis-
tance than coating 1 and coating 2. Namely, the worn surfaces of
coating 1 and coating 2 seem to be relatively rough (see Fig. 12(a
and b)), and those of coating 3 and coating 4 seem to be quite
smooth Fig. 12(c and d). In particular, cracks on the worn surface of
coating 3 seems to be less than that of coating 4 (see Fig. 11(c and
d)), which well conforms to the observation that coating 3 has
better toughness and wear resistance than coating 4.
 Y. An et al. / Vacuum 107 (2014) 132e140
Fig. 12. 3D non-contact surface mapping profiles of the wear scars of various iron-based metallic glass coatings after sliding wear tests at 6 N and 20 cm s
1: (a) coating 1, (b)
coating 2, (c) coating 3, and (d) coating 4.
4. Conclusions
Four kinds of FeWCrNiMoBSiC metallic glass coatings were
prepared on stainless steel substrate by atmospheric plasma
spraying under different spraying parameters. The influences of
spraying power plasma arc power on microstructure, porosity,
amorphous phase content as well as tribological properties of as-
prepared iron-based metallic glass coatings were investigated.
The main conclusions are drawn as follows:
(a) The degree of melting of feedstock powders and the
compactness of as-prepared iron-based metallic glass
coatings are ameliorated at elevated plasma arc power. The
microhardness of as-prepared iron-based metallic glass
coatings depends on both compactness and phase
composition. Coating 4 obtained under the highest plasma
arc power of 40 kW has the most compact microstructure
and contains a relative high content of crystalline, and
hence it possesses the highest microhardness.
(b) Phase compositions of as-sprayed iron-based metallic glass
coatings consist of amorphous structure and limited crys-
talline structure, and the contents of the amorphous
structure and crystalline structure are strongly dependent
on the cooling rate and oxidation degree of spraying
powders as well as heat accumulating inside the as-
sprayed coatings. Elevating plasma arc power favors full
melting of feedstock powders and elevated cooling rate,
thus coating 2 obtained at 30 kW contains a higher content
139
of amorphous phase than coating 1. However, coating 3
and coating 4 deposited at 35 kW and 40 kW contain a
lower content of amorphous phase than coating 2, due to
serious oxidation of the feedstock powders and the
obvious influence of localized reheating from the latent
heat evolutions as overlapping and connecting layers of
melted splats solidify on the coatings. In this sense, a
plasma arc power of 30 kW might be suitable for fabri-
cating desired iron-based metallic glass coatings on
stainless steel substrate.
(c) Fe-based metallic glass coatings deposited at different
plasma arc powers show similar steady-state friction co-
efficients of 0.8e0.9 when they slide against Si3N4 balls
under unlubricated conditions. The wear rate of iron-
based metallic glass coatings, however, varies with vary-
ing plasma arc power; and in particular, coating 3 with
next to the highest microhardness rather than coatings 4
with the highest microhardness exhibits the lowest wear
rate, which is a compromise between the microhardness
and brittle fracture as well as abrasive wear of the
coatings.
Acknowledgments
This project is financially supported by the West Light Founda-
tion of the Chinese Academy of Sciences, the National Natural Sci-
ence Foundation of China (Grant No. 51302272). The authors also
appreciate constructive comments of reviewers.
140 Y. An et al. / Vacuum 107 (2014) 132e140

References [23] Chokethawai K, McCartney DG, Shipway PH. Microstructure evolution and
thermal stability of an Fe-based amorphous alloy powder and thermally
sprayed coatings. J Alloys Compd 2009;480:351e9.
[1] Wang WH, Dong C, Shek CH. Bulk metallic glasses. Mater Sci Eng R 2004;44:
[24] Bolelli G, Bonferroni B, Lauril J, Lusvarghi L, Milanti A, Niemi K, et al. Micro-
45e89.
mechanical properties and sliding wear behaviour of HVOF-sprayed Fe-based
[2] Greer AL. Metallic glasses. Science 1995;267(5206):1947e53.
alloy coatings. Wear 2012;276e277:29e47.
[3] Suryanarayana C, Inoue A. Bulk metallic glasses. New York: CRC Press; 2011.
[25] Zhang C, Liu L, Chan KC, Chen Q, Tang CY. Wear behavior of HVOF-sprayed Fe-
[4] Kovalenko NP, Krasny YP, Krey U. Physics of amorphous metals. Weinheim:
based amorphous coatings. Intermetallics 2012;29:80e5.
Wiley-VCH; 2001.
[26] Suresh K, Yugeswaran S, Rao KP, Kobayashi A, Shum PW. Sliding wear
[5] Vara G, Pierna AR, García JA, Jimenez JA, Delamar M. Influence of nickel
behavior of gas tunnel type plasma sprayed Ni-based metallic glass composite
content on the electrochemical behavior of Finemet type amorphous and
coatings. Vacuum 2013;88:114e7.
nanocrystalline alloys. J Non Cryst Solids 2007;353:1008e10.
[27] Takeda K, Ito M, Takeuchi S. Properties of coatings and application of low-
[6] Baron A, Szewieczek D, Nawrat G. Corrosion of amorphous and nanocrystal-
pressure plasma spray. Pure Appl Chem 1990;62(9):1773e82.
line Fe-based alloys and its influence on their magnetic behavior. Electrochim
[28] Ramachandran CS, Balasubramanian V. On the cyclic hot corrosion behavior of
Acta 2007;52:5690e5.
atmospheric plasma sprayed Lanthanum Zirconate based coatings in contact
[7] Baranowska J, Franklin SE. Characterization of gas-nitrided austenitic steel
with a mixture of sodium sulphate and vanadate salts: a comparison with the
with an amorphous/nanocrystalline top layer. Wear 2008;264:899e903.
traditional YSZ duplex and NiCrAlY coated samples. Vacuum 2013;97:81e95.
[8] Inoue A, Ohtera K, Masumoto T. New amorphous AleY, AleLa and AleCe al-
[29] Wang L, Wang Y, Sun XG. Microstructure and indentation mechanical prop-
loys prepared by melt spinning. Jpn J Appl Phys 1988;27:L736e9.
erties of plasma sprayed nano-bimodal and conventional ZrO2e8 wt%Y2O3
[9] Inoue A, Nakamura T, Nishiyama N, Masumoto T. MgeCueY bulk amorphous
thermal barrier coatings. Vacuum 2012;86:1174e85.
alloys with high tensile strength produced by a high-pressure die casting
[30] Hou GL, An YL, Liu G, Zhou HD, Chen JM, Chen ZJ. Effect of atmospheric plasma
method. Mater Trans JIM 1992;33(10):937e45.
spraying power on microstructure and properties of WCe(W,Cr)2CeNi coat-
[10] Inoue A, Nishiyama N, Matsuda T. Preparation of bulk glassy Pd40Ni10Cu30P20
ings. J Therm Spray Technol 2011;20(6):1150e60.
alloy of 40 mm in diameter by water quenching. Mater Trans JIM 1996;37(2):
[31] Zhang C, Chaudhary U, Das S, Godavarty A, Agarwal A. Effect of porosity on
181e4.
photocatalytic activity of plasma-sprayed TiO2 coating. J Therm Spray Technol
[11] Inoue A, Zhang T. Fabrication of bulk glassy Zr55Al10Ni5Cu30 alloy of 30 mm in
2013;22(7):1193e200.
diameter by a suction casting method. Mater Trans JIM 1996;37(2):185e7.
[32] Drajewicz M, Goral M, Pytel M, Sieniawski J. Thermophysical properties of
[12] Liu G, An YL, Guo ZH, Chen JM, Hou GL, Chen J. Structure and corrosion
selected powders for thermal barrier coatings. J Achi Mater Manuf Eng
behavior of iron-based metallic glass coatings prepared by LPPS. Appl Surf Sci
2012;55(2):902e6.
2012;258:5380e6.
[33] Liu F, Yi GW, Wang WZ, Shan Y, Jia JH. The influence of SrSO4 on the tribo-
[13] Zhou Z, Wang L, He DY, Wang FC, Liu YB. Microstructure and electrochemical
logical properties of NiCreAl2O3 cermet at elevated temperatures. Ceram Int
behavior of Fe-based amorphous metallic coatings fabricated by atmospheric
2014;40:2799e807.
plasma spraying. J Therm Spray Technol 2011;20(1e2):344e50.
[34] Miller SA, Murphy RJ. A gasewater atomization process for producing
[14] Klinger R, Feller HG. Sliding friction and wear resistance of the metallic glass
amorphous powders. Scr Metall 1979;13(8):673e6.
Fe40Ni40B20. Wear 1983;86:287e97.
[35] Kulkarni A, Sampath S, Goland A, Herman H, Dowd B. Computed micro-
[15] Naka M, Hashimoto K, Masumoto T. Change in corrosion behavior of amor-
tomography studies to characterize microstructure-property correlations in
phous FeePeC alloys by alloying with various metallic elements. J Non-Cryst
thermal sprayed alumina deposits. Scr Mater 2004;43:471e6.
Solids 1979;31(3):355e65.
[36] Guo RQ, Zhang C, Chen Q, Yang Y, Li N, Liu L. Study of structure and corrosion
[16] Pang SJ, Zhang T, Asami K, Inoue A. Bulk glassy FeeCreMoeCeB alloys with
resistance of Fe-based amorphous coatings prepared by HVAF and HVOF.
high corrosion resistance. Corros Sci 2002;44:1847e56.
Corros Sci 2011;53(7):2351e6.
[17] Pang SJ, Zhang T, Asami K, Inoue A. Synthesis of FeeCreMoeCeBeP bulk
[37] Rupert TJ, Schuh CA. Sliding wear of nanocrystalline NieW: structural evo-
metallic glasses with high corrosion resistance. Acta Mater 2002;50:489e97.
lution and the apparent breakdown of Archard scaling. Acta Mater 2010;58:
[18] Pardo A, Otero E, Merino MC, López MD, Vázquez M, Agudo P. The influence of
4137e48.
Cr addition on the corrosion resistance of Fe73.5Si13.5B9Nb3Cu1 metallic glass
[38] Leonard BD, Sadeghi F, Shinde S, Mittelbach M. Rough surface and damage
in marine environments. Corros Sci 2002;44:1193e211.
mechanics wear modeling using the combined finite-discrete element
[19] Kishitake K, Era H, Otsubo F. Characterization of plasma sprayed Fee17Cre
method. Wear 2013;305:312e21.
38Moe4C amorphous coatings crystallizing at extremely high temperature.
[39] Yang K, Feng JW, Zhou XM, Tao SY. In-situ formed g-Al2O3 nanocrystals
J Therm Spray Technol 1996;5(3):283e8.
repaired and toughened Al2O3 coating prepared by plasma spraying. Surf Coat
[20] Kishitake K, Era H, Otsubo F. Thermal sprayed Fee10Cre 13Pe7C amorphous
Technol 2012;206(13):3082e7.
coatings possessing excellent corrosion resistance. J Therm Spray Technol
[40] Tian W, Wang Y, Yang Y, Li CG. Toughening and strengthening mechanism of
1996;5(4):476e82.
plasma sprayed nanostructured Al2O3e13 wt.%TiO2 coatings. Surf Coat
[21] Cherigui M, Fenineche NE, Coddet C. Structural study of iron-based micro-
Technol 2009;204(5):642e9.
structured and nanostructured powders sprayed by HVOF thermal spraying.
[41] Ashby MF, Greer AL. Metallic glasses as structural materials. Scr Mater
Surf Coat Technol 2005;192(1):19e26.
2006;54(3):321e6.
[22] Fenineche NE, Cherigui M, Aourag H, Coddet C. Structure and magnetic
[42] Li CJ, Ohmori A. Relationships between the microstructure and properties of
properties study of iron-based thermally sprayed alloys. Mater Lett 2004;58:
thermally sprayed deposits. J Therm Spray Technol 2002;11(3):365e74.
1797e801.