Computational Materials Science 144 (2018) 147–153

Contents lists available at ScienceDirect

Computational Materials Science

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

DFT calculation for stability and strength of iron borides

H.H. Huang a, Xiaofeng Fan a,⇑, G.M. Yang a,d, David J. Singh c, W.T. Zheng a,b
a
Key Laboratory of Automobile Materials (Jilin University), Ministry of Education, and College of Materials Science and Engineering, Jilin University, Changchun 130012, China
b
State Key Laboratory of Superhard Materials, Jilin University, Changchun 130012, China
c
Department of Physics and Astronomy, University of Missouri, Columbia, MO 65211-7010, USA
d
College of Physics, Changchun Normal University, Changchun 130032, China

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

Article history: The structural stability and mechanical properties of FeBx have been investigated by first principles meth-
Received 1 November 2017 ods with semiempirical model. With the analysis of the mechanical properties in both elastic and plastic
Received in revised form 12 December 2017 regions, it is found that the hardness of FeBx becomes higher by following the increase of B content up to
Accepted 14 December 2017
FeB2. From the stress-strain curves with the semiempirical model about hardness, the hardness of FeB2 is
about 31 GPa and similar to that of FeB4. It is expected that these results can give some light on the appli-
cations of FeBx with high B content, such as the boronizing of iron surface.
Keywords:
Ó 2017 Elsevier B.V. All rights reserved.
First-principles methods
Superhard materials
Hardness
Stress-strain curves
Ideal strength

1. Introduction rior electronic properties, such as the typical superconductive

Superhard materials are of tremendous interest in both techno-
logical applications and fundamental science. Due to the superior
mechanical and physical property, they can be used in a variety
of industrial applications such as cutting tools, hard coatings, and
abrasives and are synthesized by the design [1–4]. Among the
superhard materials, the borides are expected to have the excellent
properties [5,6]. For example, boron can combine with other light
elements, such as C and N to form the superhard alloys BC5 [7–
9], BC3 [10–12], BC2N [13–15], and so on. These materials with
the strong covalent bond network have fine mechanical properties.
However, their preparation condition is harsh and the increased
production costs prevent them from being used in industry [6].
Recently, another category of borides with the combination of
boron and transition metals, such as Cr, Re, W, Os, Fe, etc were
found to have the ultra-incompressible and expected to have good
mechanical properties [6,16–24]. For example, ReB2 was synthe-
sized under ambient pressure and with a hardness of 48 GPa
[16], which was comparable to that of cubic BN [25]. OsB2 was
shown to have an ultra-incompressible hard behavior [6]. The
strength of WB4 was reported to reach up to 46 GPa [17,18]. In
addition to high hardness, these borides also possibly exhibit supe-
⇑ Corresponding author.
E-mail address: xffan@jlu.edu.cn (X. Fan).
MgB2 with a Tc of 39 K [19].
With the structure search, iron-based borides FeB4 (op10-FeB4)
were considered to thermodynamically stable and predicted to
have the phonon-mediated superconducting characteristic,
recently [26]. Subsequently, the route for the FeB4 synthesis with
pressure was suggested [27] and it was synthesized experimen-
tally at a pressure above 8 GPa and high temperature by Gou
et al. [28] The results of magnetic susceptibility and heat capacity
demonstrated the superconductivity below 2.9 K. In addition, it
was found that the new phase FeB4 had the highly incompressible
with a nanoindentation hardness of about 62 GPa [28]. The
reported anomalous high hardness more than 60 GPa is considered
to be a strong deviation from that of the well-known transition
metal borides. The stability of FeB4 with different crystal structure
was studied [29] and the hardness derived by the semiempirical
method was shown to be lower than the experimental result, not
exceeding the superhard threshold of 40 GPa. Moreover, Li et al.
[30] investigated the ideal strength of FeB4 using first principles
methods to analyze the contribution of the stress response. The
stress-strain relations under different loading conditions showed
the indentation strength of FeB4 was about 17 GPa [30] and its
Vickers hardness was estimated to be about 25 GPa [31]. As we
know, the detailed understanding of the stability and mechanical
properties of Fe-B alloy system is still limited, though that of
FeB4 may be analyzed well [32–36].

https://doi.org/10.1016/j.commatsci.2017.12.021
0927-0256/Ó 2017 Elsevier B.V. All rights reserved.
148 H.H. Huang et al. / Computational Materials Science 144 (2018) 147–153
In this work, we investigated the structural stability and
mechanical properties of Fe-B alloy by considering the different
phases including Fe2B and FeB from the previous Fe-B phase dia-
gram, and both FeB2 and FeB4 from the structural research with
first-principles methods. By following the change of B content,
the relative stability of these structures under pressure was ana-
lyzed by considering the formation enthalpy. With the calculations
of elastic constants and the analysis of shear stress-strain relations,
the mechanical properties of FeBx with different B contents were
explored. The charge distribution in real space was analyzed in
order to get a deep insight into the tendency in mechanical
strength with the content of boron in the Fe-B compounds.
2. Computational method
We carried out the first-principle calculations with the accurate
frozen-core full-potential projector augmented wave (PAW)
method as implemented in the Vienna ab initio simulation package
(VASP) [37,38]. The calculations performed were based on density
functional theory with the exchange and correlation effects
described by the generalized gradient approximation (GGA) of Per-
dew, Burke, and Ernzerhof (PBE) [39]. A place wave basis with an
energy cutoff of 500 eV was used. The Monkhorst-Pack grid [40]
with a k-point spacing of 0.02 Å were chosen in the Brillouin zone
for the total energy convergence at the 1 meV/atom level. The
energy convergence criterion was 10
6 eV and the Hellmann-
Feynman forces were relaxed to below 0.01 eV Å
1. The spin-
polarization effect is considered due to the partial occupations of
3d-orbitals of Fe. In addition, to consider the strong correlations
from 3d_Fe, the GGA + U was tested and the Hubbard parameter
U didn’t result in the obvious effect on mechanical properties of
FexB.
From the structural research about binary alloy FeBx [26], there
are four possible single-phases for the different contents of B
including x = 0.5, 1, 2 and 4. For the four phases Fe2B, FeB, FeB2
and FeB4 (op10-FeB4), the corresponding possible stable structures
are with space groups I4/mcm, Pnma, Pnma, and Pnnm [26], respec-
tively. In addition, the energy-favorable structure (tp10-FeB4) of
FeB4 under high pressure is found to be with P42/nmc. These struc-
tures are plotted in Fig. 1.
Fig. 1. Crystal structures and coordination numbers for different single-phase structures of FeBx including Fe2B (a), FeB (b), FeB2 (c), ambient pressure phase FeB4 (op10-FeB4)
(d) and high pressure phase FeB4 (tp10-FeB4) (e).
The elastic constants were obtained by calculating the stress
under small special strain with first-principles methods. For tetrag-
onal phases, the bulk modulus and shear modulus with the Voigt
model can be expressed by the equation [42],
BV ¼ ð1=9Þ½2ðC11 þ C12 Þ þ C33 þ 4C13 ; and GV
¼ ð1=30ÞðM þ 3C11
3C12 þ 12C44 þ 6C66 Þ:
For the Reuss model, they are expressed as,
BR ¼ C2 =M and GR
n o
1
¼ 15 ð18BV =C2 Þ þ ½6=ðC11
C12 Þ þ ð6=C44 Þ þ ð3=C66 Þ :
In these formulas, M and C2 are denoted by,
M ¼ C11 þ C12 þ 2C33
4C13 ; and C2 ¼ ðC11 þ C12 ÞC33
2C213 :
For orthorhombic phases, the bulk and shear modulus with the
Voigt model can be expressed by the equation,
BV ¼ ð1=9Þ½C11 þ C22 þ C33 þ 2ðC12 þ C13 þ C23 Þ; and
GV ¼ ð1=15Þ½C11 þ C22 þ C33 þ 3ðC44 þ C55 þ C66 Þ
ðC12 þ C13 þ C23 Þ:
For the Reuss approximation, they are expressed as,
BR ¼ D½C11 ðC22 þ C33
2C23 Þ þ C22 ðC33
2C13 Þ
2C33 C12
þC12 ð2C23
C12 Þ þ C13 ð2C12
C13 Þ þ C23 ð2C13
C23 Þ
1 ; and
GR ¼ 15f4½C11 ðC22 þ C33 þ C23 Þ þ C22 ðC33 þ C13 Þ þ C33 C12

C12 ðC23 þ C12 Þ
C13 ðC12 þ C13 Þ
C23 ðC13 þ C23 Þ=D
þ3½ð1=C44 Þ þ ð1=C55 Þ þ ð1=C66 Þg
1 ;
where D = C13(C12C23
C13C22) + C23(C12C13
C23C11) + C33(C11C22

C212).
The shear modulus and bulk modulus are derived from the
Voigt-Reuss-Hill average value. In terms of the Voigt-Reuss-Hill
approximations [41], the shear modulus G and bulk modulus B
are expressed as, G = (GR + GV)/2 and B = (BR + BV)/2 with Young’s
modulus, E = 9BG/(3B + G). The shear strengths are related to the
plastic deformation and obtained from stress-strain curve. The
stress-strain curve along a direction in some planes was obtained
by straining the structure with the way of step-by-step from
 H.H. Huang et al. / Computational Materials Science 144 (2018) 147–153 149

ground state [43,44]. In each step of a given strain, the in-plane lat- the mixing binding per Fe atom is calculated for the four single-
tice is free for relaxation. In order to obtain the enthalpy of com- phases. As shown in Fig. 2(c), with increasing of B composition,
pounds and the term PV of pressure effect, the cell was the mixing binding of Fe is increased since the number of B atoms
optimized under a fixed pressure. near each Fe is increased. However, after the ratio of Fe/B more than
1/2 (such as FeB4), the mixing binding is not increased obviously.
3. Results and discussion This implies the number of B atoms nearby Fe for neighbor should
reach to the maximum with FeB2 phase.
3.1. Structural stability of FeBx

3.2. Mechanical properties of FeBx

The calculated lattice parameters are shown in Table 1. The lat-
tice parameters are in agreement with the reported theoretical and
experimental value [45,46].
The formation energy is calculated based on the ‘mother’ com-
pounds, bulk bcc-Fe and a-B. It is found that all the formation ener-
gies for four phases, Fe2B, FeB, FeB2 and FeB4 (op10-FeB4) are
negative. This implies that B can be dissolved into the lattice of
Fe to form the compounds of FeBx.
In order to consider the effect of pressure, the enthalpy is calcu-
lated by the formula, H = E + PV. The formation enthalpy of FeBx
can be calculated by the formula,
1  x y 
DHFex By ¼ HFex By
HFe
Ha
B12 ;
xþy 2 12
where HFex By , HFe and Ha
B12 are the enthalpies of FexBy, bulk bcc-Fe,
and bulk a-B [47], respectively.
As shown in Fig. 2(a), among these compounds, the formation
enthalpy of FeB is the smallest. This indicates it is the most stable
structure and is agreed well with the experiments that the FeB
phase is the easiest one to be synthesized. It is also found that
the other phases, such as Fe2B and FeB2 with negative formation
enthalpies are possible to be synthesized, especially under the
pressure with suitable thermal conditions.
Since the FeB phase is synthesized easily, we can calculate the
mixing enthalpy based on the FeB and a-B (or bcc-Fe). As shown
in Fig. 2(b), the mixing enthalpy of Fe2B, FeB2 and low-pressure
phase FeB4 (op10-FeB4) is lower than that of FeB. This implies
the phases Fe2B and FeB2 can be synthesized with the proper ratio
of Fe to B without the phase separation into FeB. For op10-FeB4
phase, it has a trend to separate into FeB2 and this implies that it
is difficult to be synthesized. It is also noticed that FeB4 (op10-
FeB4) will become the new phase tp10-FeB4 under high pressure
about 60 GPa, as reported before [33].
By the formula,
1 x y 
DHFex By ¼ HFex By
HFe2
Ha
B12 ;
x 2 12
To better understand the mechanical properties, we calculate
the elastic constants which can reflect the response of the crystal
to external stress in elastic region, as listed in Table 2. Based on
these elastic parameters, these structures of four single-phases
about FeBx are found to be satisfied with the criteria of mechanical
stability [42] and this indicates these structures are mechanically
stable, as reported previously.
From Table 2, it is found that the C11 values of FeBx are very
large with the exception of B-rich phase op10-FeB4. It is indicated
that these compounds are incompressible along a axis. The large
ð1Þ values of C11 and C33 for the B-rich high-pressure phase show
tp10-FeB4 is with the high incompressible. The elastic constant
C44 is an important criterion to determine the shear modulus.
The calculated C44 of tetragonal Fe2B is the lowest among these
compounds which means that the shear modulus of Fe2B is possi-
bly the lowest.
Both bulk modulus B and shear modulus G are two basic mod-
ule to demonstrate the mechanical properties. B describes the
capacity of incompressibility and G presents the ability of lattice
plastic deformation against shear stress. They can be derived on
the basis of the elastic constants with the Voight-Reuss-Hill
approximation. To understand the intrinsic hardness of FeBx in
term of elastic properties, the empirical model, Hv = 2.0(k2G)0.585

3.0 [48], where k is the ratio of G to B, can be used to estimate
the hardness value. The calculated results are summarized in
Table 2. The calculated hardness of FeB2 (31 GPa) is the largest
one. The hardness value of Fe2B (3.6 GPa) is very small and is
agreement with the small C44. Moreover, the value of B/G ratio
reflects materials brittleness and ductility. If the value is smaller
than 1.75, the material is brittle and has high hardness. As shown
in Table 2, the larger B/G values imply Fe2B and FeB are with metal-
lic ductility, while the smaller B/G for FeB2 and FeB4 means the
possible higher hardness.
In order to analyze the plastic deformation under larger shear
ð2Þ
stress, we can calculate the shear stress-strain relationships in

Table 1
Formation energy per atom DE (eV), lattice constants a, b and c (Å) and the shortest distances of Fe-B and B-B (Å) for different single-phase structures of FeBx with various space
groups.

Phases Space groups DE a b c Fe-B B-B

Fe2B I4/mcm
0.32 4.94 4.21 2.125 2.106
5.11a 4.25a
4.95e 4.20e
FeB Pnma
0.38 5.20 3.09 3.90 2.088 1.850
5.51a 2.95a 4.06a
FeB2 Pnma
0.31 4.82 4.81 3.74 2.033 1.737
4.82d 4.81d 3.74d
FeB4(op10) Pnnm
0.18 4.53 5.29 3.00 2.0 1.709
4.52b 5.28b 3.00b
4.48f 5.25f 3.00f
FeB4(tp10) P42/nmc
0.07 3.63 5.14 2.117 1.813
3.64c 5.16c
a
Ref. [45].
b
Ref. [35].
c
Ref. [33].
d
Ref. [26].
e
Ref. [46].
f
Ref. [29].
150 H.H. Huang et al. / Computational Materials Science 144 (2018) 147–153

Fig. 2. Formation enthalpy per atom (a), mixing enthalpy (b) and mixing binding (c) for different single-phase structures of FeBx, including Fe2B, FeB, FeB2, op10-FeB4 and
 
high-pressure phase tp10-FeB4. Note that the formation enthalpy is calculated based on solid a-B and bcc-Fe with the formula, DHFex By ¼ xþy1 y
HFex By
2x HFe2
12 Ha
B12 , the
 
mixing enthalpy is calculated based on FeB and a-B (or bcc-Fe), and the mixing binding is calculated with the formula, DHFex By ¼ x HFex By
2 HFe2
12 Ha
B12 .
1 x y

Table 2
Elastic constants Cij, bulk modulus B (GPa), shear modulus G (GPa), the ratio of B to G and hardness value (GPa) from the empirical model for different single-phase structures of
FeBx.

Phases C11 C22 C33 C44 C55 C66 C12 C13 C23 B G B/G Hv
Fe2B 463 266 123 97 242 258 283 85 3.33 3.6
228a 143a
Fe2Bb 455 302 96 129 231 274 301 87
FeB 556 459 434 180 154 115 172 249 249 309 137 2.26 10.7
287a 138a
286f 140f 2.05f 12.5f
FeB2 584 725 574 271 200 222 181 154 137 312 231 1.35 30.9
311a 231a
FeB2f 590 719 570 273 202 222 179 154 131 310 232 1.34 31.4
FeB4(op10) 402 751 445 217 138 225 164 159 151 274 184 1.49 23.4
274a 187a 24.4f
FeB4(op10)c 408 754 448 219 141 229 165 160 154 277 186 1.49
FeB4(op10)e 379 761 466 226 152 233 136 143 144 263 194 1.36
FeB4(tp10) 658 764 126 379 245 50 307 216 1.42 27.9
FeB4(tp10)d 683 782 153 394 248 46 290 203
a
Ref. [27].
b
Ref. [46].
c
Ref. [35].
d
Ref. [33].
e
Ref. [29].
f
Ref. [32].

the relevant slip planes for these single-phase structures to obtain
the ideal shear strength. As shown in Fig. 3, the shear strengths
which is the maximum of stress in shear stress-strain curve are cal-
culated for the sliding plane (0 0 1) in the direction of [1 0 0] and
[1 1 0], (1 1 1) in the direction of [
1 1 0] and [0
1 1] and (1 1
0) in the direction of [0 0 1] and [
1 1 0], respectively.
Then the ideal shear strength which is the smallest one among
these shear strengths along different directions on the different
slip planes can be chosen. It is clear that the obtained shear
strengths for Fe2B and FeB are lower than 40 GPa in Fig. 3
(a) and (b). The ideal shear strengths for Fe2B and FeB are found
to be about 15 GPa in the (1 1 1) sliding plane along [
1 1 0] direc-
tion and (1 1 0) sliding plane along [
1 1 0] direction, respectively.
It is in agreement with the theoretical hardness of 3.6 GPa and 10.7
GPa from the empirical model for Fe2B and FeB which is much
lower than that for the rich-B phases FeB2 and FeB4.
For FeB2 and op10-FeB4, there is obvious anisotropy for shear
strengths. Both phases have the high shear strength (40 GPa) in
the (1 1 0) planes along the [
1 1 0] and [0 1 1] direction, respec-
tively. However, the shear strength of FeB2 in the (0 0 1)[1 0 0]
shear direction is decreased to 25 GPa and that of op10-FeB4 in
(1 1 0)[
1 1 0] is decreased to 26.8 GPa. The high anisotropy of
shear strengths may be one important reason that the high hard-
ness is observed in some experiments [28], while FeB4 is impossi-
ble to be superhard. For the high-pressure phase (tp10-FeB4) of
FeB4, the anisotropy of shear strength is not very obvious in differ-
 H.H. Huang et al. / Computational Materials Science 144 (2018) 147–153 151

Fig. 3. Shear stress-strain curves for different single-phase structures of FeBx including Fe2B (a), FeB (b), FeB2 (c), low-pressure phase op10-FeB4 (d) and high-pressure phase
tp10-FeB4 (e) on various shear sliding planes in different directions.

ent directions and the ideal shear strength is about 23.5 GPa. From among the shear stress-strain curves of Fe2B, the curve in shear
these results, it is found that op10-FeB4 has the largest ideal shear direction (1 1 1)[0
1 1] has the sdG in the elastic region. As listed
strength among these single-phase structures of FeBx and that of in Table 3, FeB2 has the largest sdG among these structures.
high-pressure phase tp10-FeB4 is not higher than that of op10- This is consistent with the results of shear modulus in Table 2.
FeB4. In addition, the ideal shear strength of FeB2 is similar to that In the elastic region, FeB2 is harder than that of other compositions.
of FeB4. Based on the theoretical hardness from the empirical model, ideal
We also extract the smallest directional shear modulus (sdG) shear strength, and sdG, it suggests that FeB2 may be the hardest
which is obtained under the small stain (e.g. e = 0.04). For example, phase for FeBx and more amount of B content (such as FeB4)
doesn’t enhance the hardness obviously.
To understand the hardness trend of FeBx further, the charge
Table 3
density differences of five structures are calculated. The redistribu-
Ideal strength (the smallest shear strength along all the directions) of FeBx and the tion of charge in real space are analyzed as shown in Fig. 4.
smallest directional shear modulus (sdG) which is obtained under the small stain (e.g. For Fe2B, the electronic charges are found to be localized mostly
e = 0.04) for the shear stress-strain curves in a special direction. near the boron atom. The charge depletion between Fe and B
Phases Ideal strength (GPa) sdG (GPa) results in the long Fe-B bonds length (2.13 Å), as the charge distri-
bution in (1 2 0) plane in Fig. 4(a). In Fig. 4(b), the charge density
Fe2B 15.2 74.5
FeB 15.4 110.8 difference in the (1 3 0) plane of FeB is shown. It is obvious that
FeB2 25.0 199.8 the hybridized states appear between Fe and B. The bond of B-B
op10-FeB4 26.8 139.8 isn’t formed clearly due to the long distant (1.85 Å) between B
Tp10-FeB4 23.5 119.5
atoms. In Fig. 4(c), it is found that the bond network among Fe
152 H.H. Huang et al. / Computational Materials Science 144 (2018) 147–153
Fig. 4. Distribution of charge density difference of tetragonal structure Fe2B in the (0 0 1) and (1 2 0) plane (a), orthorhombic structure FeB in the (1 3 0) plane (b), FeB2 with
orthorhombic structure in the (0 0 1) plane (c), orthorhombic structure op10-FeB4 in the (0 0 1) plane (d), high pressure phase tp10-FeB4 with tetragonal structure in the (1 0
0) plane (e). (Red color represents charge accumulation, and blue color represents charge depletion). (For interpretation of the references to color in this figure legend, the
reader is referred to the web version of this article.)
atoms is formed by the formation of Fe-B bonds. This implies that
the hardness of FeB2 among the phases of FeBx is higher. The
charge density difference of op10-FeB4 and tp10-FeB4 are pre-
sented in Fig. 4(d) and (e) respectively. It is obvious that the charge
density of Fe in op10-FeB4 is redistributed and the charge density
between Fe and B is not high. In tp10-FeB4, the electron of Fe is
transferred to B and the chemical bonds between B atoms are
formed.
4. Conclusions
The structure stability and mechanical properties of FeBx are
investigated by first-principles methods. Besides the FeB4 synthe-
sized recently in the experiment and the rich-Fe phases Fe2B and
FeB in the previous phase diagram, the rich-B phase FeB2 is found
to be popular in Fe-B phase diagram. From the analysis of structure
stability, FeB2 is easier to synthesize than FeB4.
With the analysis of mechanical properties including elastic
constants, different moduli (bulk and shear), the hardness from
the empirical model, ideal shear strength and directional shear
modulus, it is found that the hardness of FeBx becomes higher by
following the increase of B content up to FeB2. The hardness FeB4
is not obviously higher than that of FeB2. By modulating the occu-
pation of B atoms in Fe lattice to increase the distance between B
atoms and Fe-B, the high-pressure phase tp10-FeB4 is with higher
formation energy compared to op10-FeB4 and there may be phase
transition from op-FeB4 to tp10-FeB4 under high pressure (60 GPa).
Acknowledgments
The research was supported by the National Key R&D Program
of China (Grant No. 2016YFA0200400) and the National Natural
Science Foundation of China (Grant No. 11504123, No. 51627805
and No. 51641202). Work at the University of Missouri was sup-
ported by the Department of Energy, BES through the MAGICS
center.
References
[1] S. Veprek, J. Vac. Sci. Technol. A 17 (1999) 2401–2420.
[2] J.C. Crowhurst, A.F. Goncharov, B. Sadigh, C.L. Evans, P.G. Morrall, J.L. Ferreira,
A.J. Nelson, Science 311 (2006) 1275–1278.
[3] R.B. Kaner, J.J. Gilman, S.H. Tolbert, Science 308 (2005) 1268–1269.
[4] J. Haines, J.M. Leger, G. Bocquillon, Annu. Rev. Mater. Res. 31 (2001) 1–23.
[5] J.B. Levine, S.H. Tolbert, R.B. Kaner, Adv. Funct. Mater. 19 (2009) 3519–3533.
[6] R.W. Cumberland, M.B. Weinberger, J.J. Gilman, S.M. Clark, S.H. Tolbert, R.B.
Kaner, J. Am. Chem. Soc. 127 (2005) 7264.
[7] V.L. Solozhenko, O.O. Kurakevych, D. Andrault, Y. Le Godec, M. Mezouar, Phys.
Rev. Lett. 102 (2009) 015506.
[8] Y. Liang, W. Zhang, J. Zhao, L. Chen, Phys. Rev. B. 80 (2009) 113401.
[9] P. Lazar, R. Podloucky, Appl. Phys. Lett. 94 (2009) 251904.
[10] P.V. Zinin, L.C. Ming, H.A. Ishii, R. Jia, T. Acosta, E. Hellebrand, J. Appl. Phys. 111
(2012) 114905.
[11] A.S. Mikhaylushkin, X. Zhang, A. Zunger, Phys. Rev. B 87 (2013) 094103.
[12] M. Zhang, H. Liu, Q. Li, B. Gao, Y. Wang, H. Li, C. Chen, Y. Ma, Phys. Rev. Lett. 114
(2015) 015502.
[13] V.L. Solozhenko, D. Andrault, G. Fiquet, M. Mezouar, D.C. Rubie, Appl. Phys.
Lett. 78 (2001) 1385.
[14] Y. Zhang, H. Sun, C. Chen, Phys. Rev. Lett. 93 (2004) 195504.
[15] X.F. Fan, H.Y. Wu, Z.X. Shen, J.-L. Kuo, Diamond Relat. Mater. 18 (2009) 1278.
[16] C. Hsiu-Ying, W.M. B, L.J. B., K. Abby, Y. Jenn-Ming, T.S. H, K. R. B, Science 316
(2007) 4Please check and provide author names in Ref. [16].36.
[17] Q. Gu, G. Krauss, W. Steurer, Adv. Mater. 20 (2008) 3620.
[18] M. Wang, Y. Li, T. Cui, Y. Ma, G. Zou, Appl. Phys. Lett. 93 (2008) 101905.
[19] J. Nagamatsu, N. Nakagawa, T. Muranaka, Y. Zenitani, J. Akimitsu, Nature 410
(2001) 63.
[20] H. Niu, J. Wang, X.-Q. Chen, D. Li, Y. Li, P. Lazar, R. Podloucky, A.N. Kolmogorov,
Phys. Rev. B 85 (2012) 144116.
[21] X. Chong, Y. Jiang, R. Zhou, J. Feng, J. Alloys Compd. 610 (2014) 684.
[22] A. Knappschneider, C. Litterscheid, D. Dzivenko, J.A. Kurzman, R. Seshadri, N.
Wagner, J. Beck, R. Riedel, B. Albert, Inorg. Chem. 52 (2013) 540.
[23] P. Lazar, X.-Q. Chen, R. Podloucky, Phys. Rev. B 80 (2009) 012103.
[24] X. Fan, M.M. Li, D.J. Singh, Q. Jiang, W.T. Zheng, J. Alloys Compd. 631 (2015)
321–327.
[25] P.B. Mirkarimi, D.L. Medlin, K.F. McCarty, D.C. Dibble, W.M. Clift, J.A. Knapp, J.C.
Barbour, J. Appl. Phys. 82 (1997) 1617.
[26] A.N. Kolmogorov, S. Shah, E.R. Margine, A.F. Bialon, T. Hammerschmidt, R.
Drautz, Phys. Rev. Lett. 105 (2010) 217003.
[27] A.F. Bialon, T. Hammerschmidt, R. Drautz, S. Shah, E.R. Margine, A.N.
Kolmogorov, Appl. Phys. Lett. 98 (2011) 081901.
[28] H. Gou, N. Dubrovinskaia, E. Bykova, A.A. Tsirlin, D. Kasinathan, W. Schnelle, A.
Richter, M. Merlini, M. Hanfland, A.M. Abakumov, D. Batuk, G. Van Tendeloo, Y.
Nakajima, A.N. Kolmogorov, L. Dubrovinsky, Phys. Rev. Lett. 111 (2013)
157002.
[29] L.P. Ding, X.Y. Kuang, P. Shao, X.F. Huang, Inorg. Chem. 53 (2014) 3471.
[30] B. Li, H. Sun, C. Chen, Phys. Rev. B 90 (2014) 014106.
 H.H. Huang et al. / Computational Materials Science 144 (2018) 147–153
[31] M. Zhang, M. Lu, Y. Du, L. Gao, C. Lu, H. Liu, J. Chem. Phys. 140 (2014) 174505.
[32] L.-H. Li, W.-L. Wang, L. Hu, B.-B. Wei, Intermetallics 46 (2014) 211.
[33] Q. Wang, Q. Zhang, M. Hu, M. Ma, B. Xu, J. He, Phys. Chem. Chem. Phys. 16
(2014) 22008.
[34] Y. Gou, Z. Fu, Y. Liang, Z. Zhong, S. Wang, Solid State Commun. 187 (2014) 28.
[35] X. Zhang, J. Qin, J. Ning, X. Sun, X. Li, M. Ma, R. Liu, J. Appl. Phys. 114 (2013)
183517.
[36] M. Yang, Y. Wang, J. Yao, Z. Li, J. Zhang, L. Wu, H. Li, J. Zhang, H. Gou, J. Solid
State Chem. 213 (2014) 52.
[37] G. Kresse, J. Furthmuller, Phys. Rev. B 54 (1996) 11169.
[38] G. Kresse, J. Furthmiiller, Comput. Mater. Sci. 6 (1996) 15.
[39] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 4.
153
[40] H.J. Monkhorst, J.D. Pack, Phys. Rev. B 13 (1976) 5188–5192.
[41] R. Hill, Proc. Phys. Soc. London, Sec. A 65 (1952) 349.
[42] Z.-J. Wu, E.-J. Zhao, H.-P. Xiang, X.-F. Hao, X.-J. Liu, J. Meng, Phys. Rev. B 76
(2007) 054115.
[43] D. Roundy, C.R. Krenn, M.L. Cohen, J.W. Morris, Phys. Rev. Lett. 82 (1999) 2713.
[44] Y. Zhang, H. Sun, C. Chen, Phys. Rev. B 73 (2006) 144115.
[45] K. Balani, A. Agarwal, N.B. Dahotre, J. Appl. Phys. 99 (2006) 044904.
[46] S. Deng, J. Zhao, S. Wei, C. Zhu, J. Lv, Q. Li, W. Zheng, RSC Adv. 6 (2016) 73576–
73580.
[47] G. Will, B. Kiefer, Z. Anorg, Allg. Chem. 627 (2001) 2100.
[48] X.-Q. Chen, H. Niu, C. Franchini, D. Li, Y. Li, Phys. Rev. B 84 (2011) 121405.