Journal of Alloys and Compounds 606 (2014) 55–60

Contents lists available at ScienceDirect

Journal of Alloys and Compounds

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

Structural, magnetic and nanomechanical properties in Ni-doped AlN

films
Juan Xiong a, Peng Guo a, Yaxuan Cai a, Bohumir Stradel b, Jan Brumek b, Yunbin He c, Haoshuang Gu a,⇑
a
Hubei Collaborative Innovation Center for Advanced Organic Chemical Materials, Faculty of Physics & Electronic Science, Hubei University, Wuhan 430062, China
b
Department of Materials Engineering, Technical University of Ostrava, Ostrava, Czech Republic
c
Faculty of Materials Science & Engineering, Hubei University, Wuhan 430062, China

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

Article history: Synthesis and characterization of structural, magnetic and nanomechanical properties of Ni-doped AlN
Received 6 October 2013 films deposited by magnetron sputtering are reported. The films exhibited ferromagnetism with a Curie
Received in revised form 26 March 2014 temperature above 400 K. The observed room temperature saturation magnetization tended to decrease
Accepted 28 March 2014
with increasing Ni concentration, which has the maximum 4.56 emu/cm3. The nanomechanical analysis
Available online 5 April 2014
testified that the degradation of crystalline quality in the films with high Ni content results in the
decreasing hardness, Young’s modulus and slight decrease of friction coefficient. Al1xNixN films for
Keywords:
x = 0.021 exhibited favorable nanomechanical properties and outstanding ferromagnetism which confirm
Ni-doped AlN films
Ferromagnetic
the possible application in spin electronics.
Nanomechanical Ó 2014 Elsevier B.V. All rights reserved.
Magnetron sputtering

1. Introduction in Ni-doped AlN nanostructure and polycrystalline powders have

Dilute magnetic semiconductor (DMS) materials have been
attracting considerable interest for their potential applications in
the field of spintronic and nonvolatile storage [1,2]. Ferromagnetic
semiconductors with Curie temperature (Tc) above room tempera-
ture are of special interest from the viewpoints of both fundamen-
tal research and practical applications [3,4]. Following theoretical
predictions [5], high temperature ferromagnetism (FM) has
recently been reported in doped oxides and nitrides, such as ZnO
[6,7], TiO2 [8], SnO2 [9], GaN [10], and AlN [11]. Among them,
AlN possesses attractive advantages such as direct-wide band
gap (6.2 eV), high thermal conductivity (2.85 W/cm K), and high
breakdown field (1.2–1.8  106 V/cm) [12].
Recently high-temperature FM has also been observed in
transition metals doped AlN host semiconductors. Cr-doped AlN
films with a Tc higher than 900 K was reported by Kumar et al.
[13]. Nonmagnetic element, Cu, doped AlN film with Tc  360 K
has been experimentally observed by Ran et al. [14]. In previous
studies, ferromagnetic Ni-doped GaN and ZnO films with Tc above
room temperature have been successfully synthesized both by
reactive sputtering and by pulsed-laser deposition, which suggest-
ing Ni dopant can induce ferromagnetic in non-ferromagnetic
semiconductors. And room-temperature ferromagnetic behaviors
⇑ Corresponding author. Tel.: +86 027 88665568.
E-mail address: guhsh@hubu.edu.cn (H. Gu).
been reported [15,16].
Moreover, the successful fabrication of devices based on the
epitaxial AlN thin films requires better understanding of the
mechanical characteristics in addition to its electrical perfor-
mances, since the contact loading during processing or packaging
can significantly degrade the performance of these devices
[17,18]. Consequently, there is a growing demand of investigating
the mechanical characteristics of materials, especially in the nano-
scale regime, for device applications.
In the present investigation, we report on the magnetic and
nanomechanical characteristics of room temperature ferromagne-
tism (RTFM) in Al1xNixN films synthesized by reactive sputtering.
RT-magnetism, favorable mechanical properties and strong adhe-
sion with the substrate of Ni-doped AlN films are reported. The
effects of Ni concentration on the ferromagnetic and nanomechan-
ical properties were investigated. The origin of RTFM the Ni-doped
AlN films was also discussed in detail.
2. Experimental
The thin film samples of composition Al1xNixN (x = 0.021, 0.047, 0.064, and
0.082) were deposited on p-type (1 0 0) Si substrate by reactive sputtering from Al
and Ni purity 99.999% targets in a sputtering system (GJP 450). Al1xNixN films were
prepared by alternatively depositing AlN and Ni films. A constant partial gas pres-
sure ratio of Ar:N2 = 2:3 with a total pressure of 0.5 Pa and a RF power of 150 W was
applied for depositing AlN film. A sputtering pressure of pure Ar gas with 0.5 Pa was
applied for doping Ni with the RF power of 60 W. The base pressure in the chamber
was below 2.0  104 Pa and the substrate temperature was held on 300 °C. The

http://dx.doi.org/10.1016/j.jallcom.2014.03.178
0925-8388/Ó 2014 Elsevier B.V. All rights reserved.
56 J. Xiong et al. / Journal of Alloys and Compounds 606 (2014) 55–60
thickness of Ni doping and AlN film is about 5–10 nm and 100 nm, respectively, for
each layer. The whole films had a thickness of about 600 nm. The dopant concentra-
tion of Ni was controlled by varying the sputtering time, which was 15 min for AlN
film and 2–8 s for Ni doping, respectively. Both the beginning and the end of the
layers are AlN films to prevent Ni atoms from permeation to substrate and
oxidation.
The dopant concentration of Ni was determined by X-ray fluorescence (XRF,
XRF 1800). Structural properties were characterized by field emission scanning
electron microscopy (SEM, JSM-7100F), X-ray diffraction (XRD, D8 Advance) and
atomic force microscope (AFM, Nanoscope IIIA). Photoluminescence (PL) spectra
of the samples were collected at RT excited by the 250 nm line of a Xe lamp as exci-
tation source at RT. The magnetic property was performed on a superconducting
quantum interference device magnetometer (SQUID, MPMS XL-7) which features
a sensitivity of 108 emu. Nanomechanical properties including nanohardness
and scratch were measured by Nanoindenter (Hysitron TI 950). The hardness (H)
and Young’s modulus (E) of Al1xNixN films were determined adopting Oliver and
Pharr model from the load–displacement curves with a diamond Berkovich inden-
ter tip. During the nanoscratch tests, the load was ramped from 0 to 8 mN in 10 s
and the scratch length was set as 15 lm with spherical indenter. The coefficient
of friction was calculated by dividing the lateral force encountered during the
scratch by the normal force. At least 5 independent indents and scratches were con-
ducted for each sample. An in situ imaging analysis was used to observe the surface
morphologies around the scratch marks after the nanoscratch.
3. Results and discussion
Cross-sectional SEM image of the 2.1 at.% Ni-doped specimen
indicates that the monolayer and homogeneity structure of AlN
films, as shown in Fig. 1(a). Fig. 1(b) shows the typical XRD pattern
for Al1xNixN thin films with different Ni concentrations deposited
on p-type (1 0 0) Si substrate. There is only one sharp peak,
Fig. 1. (a) Cross-sectional SEM image of the 2.1 at.% Ni-doped AlN and (b) X-ray
diffraction patterns of Al1xNixN samples with varying Ni composition deposited on
p-type (1 0 0) Si substrate at 300 °C.
corresponding to (1 0 0) of the AlN structure in the pattern, which
indicates all the films has the hexagonal wurtzite structure with-
out any other impurity phases such as NiN, Ni2N3, NiO, Ni2O3 or
Ni clusters. The lattice constants a as a function of Ni contents
are shown in Table 1. It can be seen from the table that the
(1 0 0) peak position shifts to higher angles as Ni concentration
from 2.1 to 8.2 at.% and the lattice parameter, a, slightly shrinks
with the increase of Ni doped concentration, probably due to the
Ni substitution for Al. This tendency has been found in Si-doped
AlN films and provides the clear evidence that the doping ions
were incorporated into the substitutional sites in the AlN structure
[19]. However, the tendency is in the opposite direction of Cr-
doped AlN films, in which the c-axis lattice constant decreases as
a function of Cr doping concentration [20]. A possible reason for
the different tendency could be the orientation of the film, since
the AlN (0 0 2) oriented in the Cr-doped films but AlN (1 0 0) in
our present work. It also can be observed that the intensity of
(1 0 0) peaks was decreased with increasing Ni concentration. The
decreasing crystalline property may be due to the Ni doping
induced defects in the AlN films.
Moreover, the Scherrer’s formula shown below was employed
to estimate the average grain size of Ni-doped AlN (1 0 0) films
[21]:
0:9k
D¼ ð1Þ
B cos h
where D, k; B and h are denoted as the average grain size, the wave-
length of the X-ray (1.5406 Å), the full-width at half-maximum of
(1 0 0) peak, and the corresponding diffraction angle, respectively.
Thus, the estimated average grain sizes of 2.1, 4.7, 6.4 and 8.2 at.%
Ni-doped AlN (1 0 0) films are 43.2, 35.1, 28.7, and 25.6 nm, respec-
tively, indicating a decreasing average grain size with increasing Ni
concentration.
The AFM measurement technique was employed to discuss the
effect of Ni concentrations on the surface morphology of Ni-doped
AlN films. Fig. 2 shows the three-dimensional surface morpholo-
gies acquired over a 1 lm  1 lm area. It can be seen that the sur-
face of AlN thin film with 2.1 at.% Ni doping is relatively smooth
(Fig. 2(a)). As Ni concentration increasing to 8.2 at.% the grain size
of AlN films reduced and the surface roughness increased obvi-
ously. The result is in agreement with the result of XRD. The root
mean square (RMS) surface roughness of 2.1, 4.7, 6.4 and 8.2 at.%
doped AlN films are 2.48 nm, 3.46 nm, 6.61 nm and 7.46 nm,
respectively. Since the XRD result illustrates no exist of secondary
phase in Ni-doped AlN film, this indicates that more Ni atoms dis-
solving into the AlN lattice brought down the surface activity of the
grains and leaded to uneven nucleation, therefore, the grain
growth was inhibited and the surface roughness was increased. A
similar trend of increasing surface roughness or decreasing grain
size with increasing doping content is reported for Fe-doped ZnO
film [22] and Na-doped CdS film [23].
The photoluminescence (PL) spectra of undoped and Ni doped
samples were carried out to confirm that the Ni incorporated into
the AlN lattice indeed (Fig. 3) for some transition metal doped AlN
materials were expected to have strong luminescence properties
[24]. No emission peak was found in the spectrum range of 300–
460 nm. However, obvious blue emission peaks located at
Table 1
Variation of peak position, and lattice parameter of the samples with Ni doped
concentration.
Ni concentration (at.%) (1 0 0) Peak position, 2h Lattice parameter, a (Å)
2.1 33.24 3.1095
4.7 33.28 3.1055
6.4 33.30 3.1036
8.2 33.36 3.0984
 J. Xiong et al. / Journal of Alloys and Compounds 606 (2014) 55–60
Fig. 2. Three-dimensional AFM images of Al1xNixN thin films for x = (a) 0.021, (b) 0.047, (c) 0.064 and (d) 0.082 in an area of 1 lm  1 lm.
Fig. 3. PL spectra of (a) undoped and (b) 2.1, (c) 4.7, (d) 6.4 and (e) 8.2 at.% Ni-doped
AlN samples measured at room temperature.
425 nm corresponding to the blue region were observed in the PL
spectra for the Ni doped samples. A similar emission band was also
observed by Li, in which it was reported that the luminescence
spectra centered at 426 nm in ferromagnetic AlN:Cu polycrystal-
line powders were due to the intracenter transitions and suggested
Cu ions substitute Al ions in AlN lattice [25]. More importantly, the
intensities of the blue emission band increase as the Ni doping con-
centration in our case, which suggests the incorporation of Ni into
the AlN lattice. It is likely to be related to increase of defects result-
ing from the Ni substitution of Al in the AlN lattice. Liu also
57
observed the higher intensity emission band with doping concen-
tration in ferromagnetic AlN:Mn powders [26]. Above all, the PL
spectra demonstrated that the incorporation of Ni into the AlN lat-
tice and the presence of defects in Ni-doped AlN films, although the
mechanism of PL in Ni-doped AlN films needs further investigation.
The magnetic properties of Ni-doped AlN films was studied by
applying magnetic field of 5000 Oe parallel (||) to the sample plane
at 300 K which is presented in Fig. 4(a). The diamagnetic contribu-
tion from pure AlN film has been subtracted. The distinct hystere-
sis loops indicate that the films are ferromagnetic at room
temperature. The saturation magnetization moment (Ms) and coer-
civity (Hc) values as a function of Ni concentration are also shown
in the inset. The data show that the Hc reduces slightly, while the
Ms reduces progressively with the increasing Ni concentration.
The Hc values of Al0.979Ni0.021N film reaches 111 Oe, which is higher
than that in the earlier report for Ni-doped ZnO nanorods [27]. The
Ms value is deduced from 4.56 to 1.19 emu/cm3 with Ni concentra-
tion increasing from 2.1 to 8.2 at.%. The dependence of Ms and Hc
on the Ni concentration has also been observed in the ZnNiO sys-
tem [28]. Fig. 4(b) shows magnetic properties of 2.1 at.% Ni-doped
AlN film by applying magnetic field parallel (k) and perpendicular
(\) to the sample plane, respectively. The film exhibited a stronger
parallel FM, 4.56 emu/cm3, than perpendicular FM, 3.05 emu/cm3.
The magnetic anisotropy is a further proof of intrinsic FM in Ni-
doped AlN film, since cluster-related FM is generally isotropic.
Temperature dependence of magnetic properties for Al0.979Ni0.021N
film under the field-cooled (FC) mode, which was measured by
applying a magnetic field of 100 Oe parallel (||) to the sample plane
as shown in Fig. 5. The magnetization decreases slowly and contin-
uously in the range of 5–400 K, indicating that the Curie tempera-
ture is higher than 400 K.
58 J. Xiong et al. / Journal of Alloys and Compounds 606 (2014) 55–60
Fig. 4. (a) Magnetization versus magnetic field of AlN films with various Ni
concentrations at 300 K. The inset shows the saturation magnetic moment and
coercive field as a function of Ni concentration. (b) Magnetic anisotropy of AlN film
with 2.1 at.% Ni.
The ferromagnetism of the obtained Ni-doped AlN films arise
from a number of possible sources. In our study, a blue emission
band was observed in the PL spectra of the samples. And the
strength of emission band is probably related to the increasing
defects resulted from more Ni substitution of Al ions. XRD results
indicated no impurity phases exhibiting in the films. In addition,
the Hc of samples with different Ni concentrations has different
values indicating that the ferromagnetism of the films does not
originate from a ferromagnetic secondary phase. And the different
magnetic behavior of Al0.979Ni0.021N film in parallel and perpendic-
ular magnetic field provides a further proof of the intrinsic magne-
tism. At present, ferromagnetism based on carrier mediated
Fig. 6. (a) Load–displacement curves and (b) hardness and Young’s modulus of Al1xNixN (1 0 0) films for x = 0.021, 0.047, 0.064 and 0.082 at 8 mN load.
Fig. 5. Temperature dependence of the magnetization for the Al0.979Ni0.021N film
under FC mode with a magnetic field of 100 Oe.
models may not be applicable to the observed ferromagnetism in
our Ni-doped AlN films since it is highly resistive at RT. Recently,
Liu et al. and Li et al.‘s work suggested that the introduction of
defects, such as nitrogen deficiency, play an important role in sta-
bilizing and enhancing the ferromagnetism in Cr-doped GaN and
AlN [29,30]. In our case, based on the high resistance nature and
the defects formed in Ni-doped AlN films, the bound magnetic
polaron (BMP) model, which is usually used to explain the ferro-
magnetic ordering in high-resistivity materials, is probably the ori-
gin of the observed ferromagnetism in our samples.
Fig. 6(a) shows the typical load–displacement curve for
Ni-doped AlN (1 0 0) films under a normal load of 8 mN. The corre-
sponding values of hardness (H) and Young’s modulus (E) are plot-
ted in Fig. 6(b). It was found that the hardness of Ni-doped AlN
films show a significant decrease from 34.97 to 30.45 GPa with
the Ni concentration increases from 2.1 to 8.2 at.%. And the values
of Young’s modulus exhibit the similar decrease from 168.0 to
111.6 GPa with increasing Ni concentration. Based on the results
of XRD and AFM, it can be found that both the hardness and
Young’s modulus decrease with the decreasing average grain size
which follows the inverse Hall–Petch effect [31]. It has been pro-
posed by Song et al. that the volume fraction of grain boundaries
increases noticeably as the grain size decreases to tens of nanome-
ters [32]. Consequently, for the small grains the deformation is
dominated by elevated grain boundary phase (via grain boundary
sliding), while the decrease of hardness and strength, since grain
boundary phase is softer than grain due to increased atomic spac-
ing [33]. No pop-in discontinuous dislocation burst behavior in the
load–displacement curves of Ni-doped AlN films illustrates the
inverse Hall–Petch effect, as shown in Fig. 6(a). The hardness and
 J. Xiong et al. / Journal of Alloys and Compounds 606 (2014) 55–60
Fig. 7. Nanoscratch load plotted with respect to the coefficient of friction of
Al1xNixN (1 0 0) films, for x = 0.021, 0.047, 0.064 and 0.082, at a constant normal
load of 8 mN.
elastic modulus evolution with the grain size was also observed in
Fe60Co20B20 films by Chen and Chang [34]. In their case, the
decrease of hardness and elastic modulus with the thermal anneal-
ing time was explained by the increasing grain size and surface
roughness of the film. It should be also noted that the values of
hardness of our Ni-doped AlN (1 0 0) films are higher than those
of the reported (0 0 2) and (1 0 3) oriented AlN films, which are
12 GPa and 21.4 GPa, respectively [35,36].
The calculated coefficient of friction of Al1xNixN (1 0 0) films as
a function of x is plotted in Fig. 7 and the corresponding in situ
Fig. 8. The in situ images of Al1xNixN (1 0 0) films, for x = (a) 0.021, (b) 0.047, (c) 0.064 and (d) 0.082, after nanoscratch test at a constant normal load of 8 mN.
59
images after nanoscratch test at a constant normal load of 8 mN
are shown in Fig. 8. In Fig. 7 the friction coefficients exhibit a slight
decrease with the increasing Ni content as shown. It is also
revealed a significant creep in the AlN film when the Ni content
is 0.82 in Fig. 7. And as shown in Fig. 8(d), in addition to plowing
scratch tracks, pile-up and scratch debris appeared inside or
around the nanoscratch in the Al0.918Ni0.082N films. The depen-
dence of the film grain size and surface roughness on the friction
coefficient has been studied by other authors [37]. Fan et al. have
observed the friction coefficient is proportional to the grain size
in amorphous carbon films deposited by chemical vapor deposition
on copper substrate [38]. They correlated the increase of friction
coefficient with the increase of the grain size and the film rough-
ness. In our case, we could explain the slight decrease of friction
coefficient with the Ni content by a decrease of adhesion forces
in the tip–film interface. The presence of Ni not only enhanced
the roughness of the AlN film but also reduced the grain size.
Therefore, it could lead to a decrease of the adhesive forces, which
will cause the decrease of the friction coefficient [39].
4. Conclusion
In summary, we reported the structural, magnetic and nanome-
chanical properties in Ni-doped AlN films with various Ni concen-
trations deposited by magnetron sputtering. The magnetic and
nanomechanical properties of Al1xNixN films were observed to
be strongly dependent on the Ni concentration. The maximum RT
Ms and Hc obtained were about 4.56 emu/cm3 and 111 Oe, respec-
tively, for Al0.979Ni0.021N films, which also exhibited the maximum
hardness, Young’s modulus and optimal coefficient of friction. The
magnetization measurement indicated that the Curie temperature
is higher than 400 K. It is expected that the room temperature
60 J. Xiong et al. / Journal of Alloys and Compounds 606 (2014) 55–60
ferromagnetic Ni-doped AlN films with stable mechanical proper-
ties can make it possible to the applications for the spin electric
devices.
Acknowledgments
The author would like to acknowledge financial support
from Nature Science Foundation of China (61106070, 61274010
and 51202062) and Supply-Demand Program of Science and
Technology of Wuhan (No. 201150124019).
References
[1] Y. Kozuka, A. Tsukazaki, D. Maryenko, J. Falson, C. Bell, M. Kim, Y. Hikita, H.Y.
Hwang, M. Kawasaki, Phys. Rev. B 85 (2012) 075302.
[2] P. Wójcik, J. Adamowski, M. Wołoszyn, B.J. Spisak, Appl. Phys. Lett. 102 (2013)
242411.
[3] T.S. Herng, M.F. Wong, D.C. Qi, J.B. Yi, A. Kumar, A. Huang, Adv. Mater. 23
(2011) 1635.
[4] M. Zhang, T.F. Zhou, Y.M. Zhang, B. Li, S.N. Zheng, J. Huang, Y.P. Sun, G.Q. Ren,
J.F. Wang, K. Xu, H. Yang, Appl. Phys. Lett. 100 (2012) 041904.
[5] G.S. Chang, J. Forrest, E.Z. Kurmaev, A.N. Morozovska, M.D. Glinchuk, J.A.
McLeod, Phys. Rev. B 85 (2012) 165319.
[6] T. Dietl, H. Ohno, F. Matsukura, J. Cibert, D. Ferrand, Science 287 (2000) 1019.
[7] J.P. Fan, X.L. Li, Z.Y. Quan, X.H. Xu, Appl. Phys. Lett. 102 (2013) 102407.
[8] T. Kataoka, Y. Yamazaki, V.R. Singh, A. Fujimori, Phys. Rev. B 84 (2011) 153203.
[9] C. Ke, W. Zhu, J.S. Pan, Z. Yang, Z.P. Li, L. Wang, J. Alloys Comp. 509 (2011)
5008–5011.
[10] N.H. Hong, N. Poirot, J. Sakai, Phys. Rev. B 77 (2008) 033205.
[11] M. Roever, J. Malindretos, A. Bedoya-Pinto, A. Rizzi, Phys. Rev. B 84 (2011)
081201(R).
[12] W.W. Lei, P.W. Zhu, X.H. Chen, Q. Zhao, G.H. Wen, Q.L. Cui, G.T. Zou, Appl. Phys.
Lett. 95 (2009) 162501.
[13] D. Kumar, J. Antifakos, M.G. Blamire, Z.H. Barber, Appl. Phys. Lett. 84 (2004)
5004.
[14] F.Y. Ran, M. Subramanian, M. Tanemura, Y. Hayashi, T. Hihara, Appl. Phys. Lett.
95 (2009) 112111.
[15] H. Li, H.Q. Bao, B. Song, W.J. Wang, X.L. Chen, Solid State Commun. 148 (2008)
406–409.
[16] X.H. Ji, S.P. Lau, S.F. Yu, H.Y. Yang, Nanotechnology 18 (2007) 105601.
[17] S.R. Jian, G.J. Chen, T.C. Li, Nanoscale Res. Lett. 5 (2010) 935.
[18] J.H. Song, J.L. Huang, T. Omori, J.C. Sung, S. Wu, Thin Solid Films 520 (2012)
2247–2250.
[19] D. Pan, J.K. Jian, Y.F. Sun, R. Wu, J. Alloys Comp. 519 (2012) 41–46.
[20] J. Zhang, S.H. Liou, D.J. Sellmyer, J. Phys.: Condens. Matter. 17 (2005) 3137–
3142.
[21] Z. Ullah, S. Atiq, S. Naseem, J. Alloys Comp. 555 (2013) 263–267.
[22] F. Gao, X.Y. Liu, L.Y. Zheng, M.X. Li, Y.M. Bai, J. Xie, J. Cryst. Growth 371 (2013)
126–129.
[23] S. Mageswari, L. Dhivya, B. Palanivel, R. Murugan, J. Alloys Comp. 545 (2012)
41–45.
[24] H.R. Hu, X. Ji, Z.G. Wu, P.X. Yan, H.G. Zhou, S.S. Du, X.Y. Wu, G. Gong, C. Li,
Mater. Lett. 70 (2012) 34–36.
[25] H. Li, X.L. Chen, B. Song, H.Q. Bao, W.J. Wang, Solid State Commun. 151 (2011)
499–502.
[26] H. Li, H.Q. Bao, B. Song, W.J. Wang, X.L. Chen, L.J. He, W.X. Yuan, Physica B 403
(2008) 4096–4099.
[27] C.W. Cheng, G.Y. Xu, H.Q. Zhang, Y. Luo, Mater. Lett. 62 (2008) 1617–1620.
[28] X.X. Liu, F.T. Lin, L.L. Sun, W.J. Cheng, X.M. Ma, W.Z. Shi, Appl. Phys. Lett. 88
(2006) 062508.
[29] L.J. Shi, L.F. Zhu, Y.H. Zhao, B.G. Liu, Phys. Rev. B 78 (2008) 195206.
[30] J. Zhang, X.Z. Li, B. Xu, D.J. Sellmyer, Appl. Phys. Lett. 86 (2005) 212504.
[31] H. Van Swygenhoven, Science 296 (2002) 66–67.
[32] Z. Song, H.X. Liu, S.J. Zhang, Z.J. Wang, Y.T. Shi, J. Eur. Ceram. Soc. 34 (2014)
1209–1217.
[33] Z.B. Qi, P. Sun, F.P. Zhu, Z.C. Wang, D.L. Peng, C.H. Wu, Surf. Coat. Technol. 205
(2011) 3692–3697.
[34] Y.T. Chen, C.C. Chang, J. Alloys Comp. 498 (2010) 113–117.
[35] S.R. Jian, G.J. Chen, J.S. Jang, Y.S. Lai, J. Alloys Comp. 494 (2010) 219–222.
[36] E.P. Koumoulos, C.A. Charitidis, D.P. Papageorgiou, A.G. Papathanasiou, Surf.
Coat. Technol. 206 (2012). 3823-3813.
[37] W. Yue, X.C. Gao, C.B. Wang, Z.Q. Fu, X. Yu a, J.J. Liu, Mater. Lett. 73 (2012) 202–
205.
[38] Q.H. Fan, E. Peiriera, P. Davim, J. Gracio, C.J. Tavares, Surf. Coat. Technol. 126
(2000) 110–115.
[39] A. Jellad, S. Labdi, C. Malibert, G. Renou, Wear 264 (2008) 893–898.