Computational Materials Science 39 (2007) 580–586

www.elsevier.com/locate/commatsci

Ab initio study of structural and electronic properties of

Received 22 April 2006; received in revised form 17 August 2006; accepted 17 August 2006

Abstract

A density-functional theory study of structural and electronic properties of III-arsenide compounds BAs, AlAs, GaAs and InAs, is
presented. For the exchange-correlation potential, local-density approximation (LDA) and generalized gradient approximation (GGA)
have been used to calculate theoretical lattice parameters, bulk modulus, its pressure derivative, and cohesive energy. The electronic band
structure of these compounds have been calculated by using the above two approximations as well as another form of GGA, proposed by
Engel and Vosko (GGA-EV). The results obtained for structural and electronic properties are compared with experimental data and
other computational work. It has been found that the energy bands with all these approximations are similar except band gap values.
It has also been found that our results with LDA, GGA and GGA-EV are in good agreement with other computational work wherever
these are available. Our results in general are also in agreement with the experimental values.
Ó 2006 Elsevier B.V. All rights reserved.

Keywords: WIEN2k; Density functional theory; Electronic band structure; Semiconductor compounds; III–V Compounds; FP-LAPW; Bandgap

1. Introduction gap partner for alloying with AlAs and GaAs [4]. Boron
arsenide due to strong covalent character possesses a pecu-
III–V Compound semiconductors provide good basis for liar behavior compared to other III–V compounds [5]. In
many established commercial and new technological appli- literature, very little experimental and theoretical work has
cations in electronic and optoelectronic devices, such as light been reported about the structural and electronic properties
emitting diodes, photo detectors, lasers, modulators, inte- of this compound. Similarly Gallium arsenide, another
grated circuits and filters. III-Arsenide compounds belong member of this series, due to its direct band gap and having
to a family of common-anion III–V semiconductors. These a higher saturated electron velocity as well as higher electron
compounds have the widest range of energy gaps apart from mobility, has wide applications, especially where the reliabil-
III-nitrides [1]. Under normal conditions, these compounds ity is a major concern. AlAs, GaAs and InAs are fundamen-
crystallize in the zincblende structure [2]. Boron arsenide and tal to a wide variety of optoelectronic heterojunction
Aluminum arsenide are indirect band gap semiconductors, systems. This includes short period super-lattices (GaAs)m/
while Gallium arsenide and Indium arsenide are direct band (AlAs)n and (InAs)n/(GaAs)m [6], bulk alloys such as
gap semiconductors. Boron arsenide has resemblance with AlxGa1xAs and InxGa1xAs [7], epitaxial alloy films on
silicon electronically [3] and is an alternate to GaN as a wide GaAs, InxGa1xAs/GaAs quantum dots and GaAs-embed-
ded InAs quantum dots [8].
*
Corresponding author. Tel.: +92 42 9231137; fax: +92 42 9231253. The operating characteristics of these electronic
E-mail address: rasofi@hotmail.com (R. Ahmed). and optoelectronic devices depend not only on material

0927-0256/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.commatsci.2006.08.014
 R. Ahmed et al. / Computational Materials Science 39 (2007) 580–586 581

engineering at a practical level, it also demands clear (FP-LAPW) method within local-density approximation
understanding of the properties of materials, and the (LDA) or generalized gradient approximation (GGA)
fundamental science behind these properties. Owing to and pseudo-potential methods [9]. The electronic structure
their technological significance, it is important to know is rarely predicted theoretically with GGA-EV. Compre-
different properties of these semiconductor compounds. hensive study of structural and electronic properties of
Theoretical investigations as well as experimental research III-arsenide series is therefore elusive.
is therefore of vital interest to all those working in this area In this work, calculations of various structural and
of research. electronic properties, like cohesive energy, bulk modulus,
Theoretically, III-arsenide compounds have been pressure derivative etc., have been performed using
studied by employing different approaches; from pheno- ground-state total energy calculation approach of the sys-
menological methods such as k Æ p theory or empirical tem. These calculations provide a one stop shop for funda-
pseudo-potentials methods [8] to atomistic ab initio meth- mental understanding of the structural and electronic
ods, such as the full-potential linear augmented plane wave properties of the III-arsenide compounds. An in-depth

Table 1
0
The lattice parameter (a), Volume (V), bulk modulus (B), its pressure derivative (B ) and cohesive energy per unit cell (Ecoh) of BAs, AlAs, GaAs and InAs
compounds
0
Compounds Method a (Å) V (a.u.)3 B (GPa) B ECoh (eV)
BAs
Present work FP-LDA 4.741 179.795 147.5 4.216 11.303
FP-GGA 4.817 188.563 131.2 4.179 09.463
Experiment 4.777 [25] 3.984 [29]
Other calculations FP-LDA 4.743 [39] 152 [39] 3.65 [39]
FP-GGA 4.812 [39] 133 [39] 3.75 [39]
FP-GGA 4.784 [9] 137 [9] 3.49 [9]
HF 4.828 [30] 162 [30] – 5.497 [36]
HF 4.83 [41] 149 [41] – 5.6 [41]
AlAs
Present work FP-LDA 5.633 301.526 75.1 4.512 08.962
FP-GGA 5.734 317.988 66.5 4.184 07.494
Experiment 5.620 [25] 82.0 [29] 4.182 [29] 07.70 [35]
5.661 [26] 77.3 [1]
Other calculations PP-LDA 5.614 [28] 74.7 [28] 4.182 [28]
PP-GGA 5.690 [37] 71.0 [37] 7.25 [37]
FP-LDA 5.664 [32] 71.0 [32] 08.8 [32]
FP-LDA 5.644 [40] 75.4 [40] 4.4 [40]
HF 5.741 [30] 90.0 [30] 4.7 [36]
HF 5.76 [41] 79.0 [41] 4.9 [41]
GaAs
Present work FP-LDA 5.608 297.606 75.2 4.814 8.044
FP-GGA 5.748 320.390 60.8 4.800 6.367
Experiment 5.654 [25] 77.0 [29] 4.487 [29]
5.653 [35] 76.0 [35] 6.52 [35]
Other calculations PP-LDA 5.530 [28] 75.7 [28] 4.487 [28]
PP-GGA 5.700 [37] 65.0 [37] 6.18 [37]
PP-GGA 8.30 [42]
FP-LDA 5.651 [32] 63.0 [32] 7.80 [32]
FP-LDA 5.508 [33] 77.1 [33] 4.8 [40]
FP-LDA 5.649 [40] 74.2 [40] 4.46 [34]
FP-LDA 5.592 [34] 81.1 [34] 4.45 [34]
FP-GGA 5.726 [34] 68.0 [34]
HF 5.755 [30] 87.0 [30] 3.537 [36]
HF 5.760 [41] 77.0 [41] 3.50 [41]
InAs
Present work FP-LDA 6.030 370.879 60.9 4.691 7.586
FP-GGA 6.195 401.104 48.1 4.683 5.944
Experiment 6.036 [25] 58.0 [29] 4.79 [38] 4.487 [29]
Other calculations PP-LDA 5.921 [28] 61.7 [28] 4.545[28]
PP-LDA 5.902 [31] 61.9 [31] 4.488[31]
FP-LDA 6.015 [40] 60.3 [40] 4.9 [40]
HF 6.152 [30] 74.0 [30] 3.51 [36]
HF 6.15 [41] 64.0 [41] 3.50 [41]
582 R. Ahmed et al. / Computational Materials Science 39 (2007) 580–586

study of these compounds based on first principle therefore GGA-EV and obtained good results. Ahmed et al. [22]
provides very useful information. obtained good results for band gaps of III-nitrides using
In this paper, we use single theoretical approach to give GGA-EV.
a uniform explanation of the electronic structure of BAs, The RMT values for B and As in BAs compound are 1.7
AlAs, GaAs and InAs. Using the density-functional theory and 2.1 a.u., for Al and As in AlAs compound are 2.0 and
[10,11] and employing different form of exchange-correla- 2.1 a.u., for Ga and As in GaAs compound are 2.0 and 2.1
tion approximation, we have studied structural properties, a.u., and for In and As in InAs compound are 2.4 and 2.1
electronic band structure and band gap properties of these a.u., respectively. For the wave function expansion, maxi-
compounds. The paper is organized as follows. Section 2 mum value of ‘l’ was confined to lmax = 10 inside the
explains the method of calculations. The structural and atomic spheres. For plane wave cut off, Kmax = 8.0/RMT
electronic properties are given in Section 3. Section 4 con- is used to expand the wave functions in the interstitial
cludes our work. region and for the charge density Fourier series was trun-
cated at Gmax = 14. A mesh of 30 special k-points was used
2. The method of calculation in the irreducible wedge of the Brillion zone. Relativistic
effects are included within the scalar-relativistic approxima-
Electronic configuration of BAs, AlAs, GaAs and InAs tion. The effect of spin orbit coupling is neglected because it
is B: He 2s22p1, Al: Ne 3s23p1, Ga: Ar 3d104s24p1, In: Kr does not produce appreciable effect on the results.
4d105s25p1 and As: Ar 3d104s24p3. In our computational
work, we differentiate between the inner-shell electrons of
B (1s2), Al (1s22s22p6), Ga (1s22s22p63s23p6), In (1s22s22p 3. Results and discussion
6 2
3s 3p63d104s24p6), As (1s22s22p63s23p6) and the valence
band electrons of the B (2s22p1), Al (3s23p1), Ga 3.1. Structural parameters
(3d104s24p1), In (4d105s25p1) and As (3d104s24p3). Atoms
in zincblende structure are in FCC positions as R (0, 0, 0); In order to study the structural properties, we used
X (1/4, 1/4, 1/4) where R („B, Al, Ga, In); and X („As). experimental lattice parameter [25,26]. By calculating the
In order to compute equilibrium structural parame- total energy of primitive unit cell at different volumes and
ters and electronic band structure of III-arsenides, we fitting the data with Murnaghan equation of state [27],
have employed WIEN2k code [12]. This includes full- we obtained the equilibrium lattice parameter, bulk modu-
potential (linear) augmented plane wave plus local orbitals lus, its pressure derivative and cohesive energy, which are
((L)APW + lo) method within the density-functional shown in Table 1 along with the experimental data and
theory. In this method, the primitive unit cell is divided other computations. A prototype energy–volume curve
into non-overlapped spheres around atoms and remaining for BAs structure is shown in Fig. 1. As is clear from the
interstitial regions, where the Kohn–Sham wave functions, figure, GGA-EV theoretical lattice parameter is far from
charge density and potential are treated differently in these the reality that, as mentioned in the last section, is due to
regions of the unit cell. Inside the atomic spheres of radius the fact that this functional is adjusted for better reproduc-
RMT around each atom, radial solution of the Schrödinger tion of the exchange-correlation potential at the expense of
equation times the spherical harmonic are used for expan- less accuracy in exchange-correlation energy. Therefore
sion and the plane wave basis set is used in the interstitial
region. Details of this method are available in literature
[12,13]. For the exchange-correlation potentials, we use
local-density approximation [14], and generalized gradient
approximation [15].
For band structure calculations, another form of gener-
alized gradient approximation suggested by Engel and
Vosko (GGA-EV) [16] is also used. It is well known that
GGA underestimates the energy band gap [17–24] which
is due to its simple functional form that is not flexible
enough to simultaneously reproduce accurately exchange-
correlation energy and its charge derivative. The GGA-
EV has a new functional form of GGA which compensates
this deficiency and produces better results for the band gap
values. It has been successfully used by several authors to
explore electronic properties of solids. Dufek et al. [17]
successfully applied GGA-EV to a wide range of solids.
Persson et al. [18] calculated the band structure of gallium Fig. 1. Energy volume curves of BAs binary compound using different
arsenide with GGA-EV. Akbarzadeh and Mokhatari [19– exchange-correlation energy approximations. Arrow indicates experimen-
21] calculated the band structure of II–V nitrides using tal volume.
 R. Ahmed et al. / Computational Materials Science 39 (2007) 580–586 583

total energies obtained by this approximation are inaccu- of crystal, has also been computed. In order to obtain pre-
rate and accordingly the obtained structural properties cise value for the cohesive energy, the total energy calcula-
are far from measured values. Hence we do not use tions for isolated atoms and crystal have been performed at
GGA-EV for calculating structural properties and instead the same level of accuracy. Energy of an isolated atom was
we apply it to other relevant lattice parameters for energy calculated by considering a super cell containing an isolated
band gap studies (Section 3.2). atom. This is achieved by using spin dependent form of
It is observed that our calculated lattice parameter is in functional, with atoms in ground-state spin configuration.
agreement with the experimental data (within 2%) and Energy associated with the crystal is calculated at the opti-
other theoretical work. Table 1 also indicates that in mized lattice constant. Accuracy of DFT calculations is
general GGA slightly overestimates while LDA underesti- also based on the best choice of the exchange-correlation
mates the theoretical lattice parameter compared to the energy functional. In more advanced exchange-correlation
experimental results. It is further observed that our results energy functionals, the Kohn–Sham orbitals or higher
with GGA are in agreement with lattice parameters calcu- derivatives of the density are involved. In Table 1, we pres-
lations using HF method. The calculated Bulk moduli, B ent computed results for the cohesive energy using LDA
(Table 1), agree with other first principle calculations and and GGA. Very little reported work on III-arsenides
are in better agreement with the experiments. It is observed with FP-LAPW is available in literature. Our computed
that LDA gives bulk moduli closer to experiment while values with GGA are in good agreement with other cal-
GGA approximation consistently tends to underestimate culations [32]. It is observed that LDA generally gives
experiment. The obtained pressure derivatives of bulk higher values compared to experiment while GGA provides
modulus, given in Table 1, are in good agreement also with a good agreement [35]. It is also observed that cohesive
the experimental measurements. energy calculated by HF method is significantly lower than
The cohesive energy, defined as the difference between the experimental data and results obtained by DFT
the total energy of the isolated atoms and the total energy approach.

Fig. 2. Electronic band structures of BAs (upper row) and AlAs (lower row) at their experimental lattice parameters using LDA, GGA, and GGA-EV.
584 R. Ahmed et al. / Computational Materials Science 39 (2007) 580–586

3.2. Electronic structure band structure while the GGA-EV conduction band is
shifted with respect to GGA conduction band. This shift
Although the LDA and GGA theoretical lattice con- is the main reason for appropriateness of the GGA-EV
stants are close to the corresponding measured values but for band gap studies.
due to the rather high sensitivity of electronic band gap Using the calculated electronic structure (Figs. 2 and 3)
to the lattice parameter, we have chosen experimental lat- we have obtained the corresponding band gap and listed
tice parameter for consistent study of electronic structure them in Table 2 together with other available theoretical
using different exchange-correlation approximation. There- and experimental data. It is noticeable that, except for
fore at the experimental lattice parameters, eigenvalues of GaAs, no theoretical work using GGA-EV for band struc-
Kohn–Sham equations were calculated along some high ture of these binary compounds is available. Similarly we
symmetry directions in the first Brillion Zone using LDA, did not find any reported GGA band gap value for InAs.
GGA and GGA-EV exchange-correlation approximations Our results show a general agreement and improvement
and the obtained band structures are presented and com- over the earlier theoretical studies where ever these are
pared together in Figs. 2 and 3. An overall similarity is available. A good agreement is also observed between our
clearly visible in all band structures. For all binary com- full-potential calculations and the reported pseudo-poten-
pounds, the semicore state below 9 eV is mainly s orbital tial plane wave (PP-PW) results. Additionally it may be
of As while the valence band is a hybridization of s and p noted that the results obtained with LDA and GGA are
electrons of III elements and p electrons of As. The bottom close to each other for BAs and begin to differ for heavier
of valence band is dominated by the s electrons and metals. It might be attributed to the uniform electron gas
remaining parts are mainly p orbital of As. The bottom approximation used in LDA that is probably less valid for
of conduction band has a hybridized character composed heavy elements having large density variations.
of s and p states of both elements. Obviously, for all cases, Clearly for all compounds, except BAs, GGA-EV
below Fermi energy, GGA and GGA-EV leads to similar leads to much better band gap values compared with the

Fig. 3. Electronic band structures of GaAs (upper row) and InAs (lower row) at their experimental lattice parameters using LDA, GGA, and GGA-EV.
 R. Ahmed et al. / Computational Materials Science 39 (2007) 580–586 585

Table 2
The energy band gap (Eg) properties of BAs, AlAs, GaAs and InAs
Compounds Methods XC Eg (eV) Type of band gap
BAs
Present work FP-LAPW LDA 1.13 Indirect (C–Dmin)
GGA 1.18 Indirect (C–Dmin)
GGA-EV 1.79 Indirect (C–Dmin)
Experiment 0.67 [44,49] Indirect (C–Dmin)
Other Calculations FP-LAPW LDA 1.12 [39] Indirect (C–Dmin)
GGA 1.21 [39] Indirect (C–Dmin)
GGA 1.23 [9] Indirect (C–Dmin)
GGA 1.25 [43]
AlAs
Present work FP-LAPW LDA 1.31 Indirect (C–X)
GGA 1.40 Indirect (C–X)
GGA-EV 2.25 Indirect (C–X)
Experiment 2.24 [1] Indirect (C–X)
Other calculations FP-LAPW LDA 1.37 [32] Indirect (C–X)
FP-LAPW LDA 1.25 [48]
PP-PW LDA 1.32 [37] Indirect (C–X)
PP-PW GGA 1.39 [37] Indirect (C–X)
PP-PW LDA 1.35 [45] Indirect (C–X)
PP-PW LDA 1.31 [47] Indirect (C–X)
GaAs
Present work FP-LAPW LDA 0.28 Direct (C–C)
GGA 0.51 Direct (C–C)
GGA-EV 1.03 Direct (C–C)
Experiment 1.42 [46] Direct (C–C)
Other calculations FP-LAPW LDA 0.23 [32] Direct (C–C)
FP-LAPW LDA 0.18 [17] Direct (C–C)
FP-LAPW LDA 0.09 [48]
FP-LAPW GGA-EV 0.97 [19] Direct (C–C)
PP-PW LDA 0.32 [37] Direct (C–C)
PP-PW GGA 0.49 [37] Direct (C–C)
InAs
Present work FP-LAPW LDA 0.00 Direct (C–C)
GGA 0.00 Direct (C–C)
GGA-EV 0.40 Direct (C–C)
Experiment 0.42 [1] Direct (C–C)
Other calculations FP-LAPW LDA 0.64 [48] Direct (C–C)
PP-PW LDA 0.00 [47] Direct (C–C)
XC: exchange-correlation approximation.

experimental results while LDA and GGA considerably InAs have a direct gap in C–C direction. It is not clear if
underestimates it. In the case of InAs this underestimation it is due to the presence of d electrons in Ga and In.
leads to a wrong metallic behavior within LDA and
GGA. For BAs compound we observed that in contrast 4. Conclusion
to the general trend of LDA and GGA, theoretical band
gaps are consistently higher than the experimental value. Our computations with LDA and GGA generally give a
This is not the case for other compounds as discussed good agreement with the experimental data for structural
above. This indicates that either the experimental value properties. We observe that in general, GGA some what
needs to be measured again or some very unusual phenom- overestimates the experimental data while LDA underesti-
ena is taking place and needs a through probe into it. For all mates. However, they agree with the experiment within 5%.
these binary compounds, maximum of the valence band is For electronic band structure, our results with LDA and
in C point while the minimum of the conduction band GGA, although in good agreement with other computa-
appears at different points. In BAs it is a point between C tional work, are lower than the experimental results. This
and X (we call it Dmin), for AlAs it appears in X, and for is due to a simpler form of the exchange-correlation func-
GaAs and InAs the minimum of the conduction band is tional. However, band gap calculated with GGA-EV is
on C point. Therefore BAs and AlAs have an indirect band closer to the experimental values due to a better choice of
gap along C–Dmin and C–X, respectively, while GaAs and the exchange-correlation potential. Our work is the first
586 R. Ahmed et al. / Computational Materials Science 39 (2007) 580–586

ab initio study of electronic band structure with GGA-EV [23] A. Zaoui, F. El Haj Hassan, J. Phys.: Condens. Matter 13 (2001)
within DFT. 253;
Z. Charifi, H. Baaziz, F. El Haj Hassan, N. Bouarissa, J. Phys.:
Condens. Matter 17 (2005) 4083.
Acknowledgements [24] F. El Haj Hassan, H. Akbarzadeh, Mater. Sci. Eng. B 121 (2005) 170;
F. El haj Hassana, H. Akbarzadeh, S.J. Hashemifar, A. Mokhtari, J.
Part of this work was done at ICTP affiliated Center, Phys. Chem. Solids 65 (2004) 1871;
Isfahan University of Technology, Isfahan, Iran. One of Z. Charifi, F. El Haj Hassan, H. Baaziz, Sh. Khosravizadeh, S.J.
Hashemifar, H. Akbarzadeh, J. Phys.: Condens. Matter 17 (2005)
the authors (R.A.) acknowledges financial support of the 7077.
Isfahan University of Technology, Iran, and University [25] R.W.G. Wyckoff, Crystal Structures, second ed., Krieger, Malabar,
of the Punjab Lahore, Pakistan. 1986.
[26] Sadao Adachi, Properties of Group-IV, III–V and II–VI Semicon-
References ductors, John Wiley & Sons, England, 2005.
[27] F.D. Murnaghan, Proc. Natl. Acad. Sci. U.S.A. 30 (1994) 244.
[28] S.Q. Wang, H.Q. Ye, J. Phys.: Condens. Matter 14 (2002) 9579.
[1] I. Vurgaftman, J.R. Meyer, L.R. Ram-Mohan, J. Appl. Phys. 89
[29] K.H. Hellwege, O. Madelung, Landolt–Börnstein New Series Group
(2001) 5815 and references given therein.
III, vol. 17a, Springer, Berlin, 1982.
[2] A. Mujica, Angel Rubio, A. Munõz, R.J. Needs, Rev. Mod. Phys. 75
[30] Simon Kalvoda, Beate Paulus, Peter Fulde, Hermann Stoll, Phys.
(2003) 864.
Rev. B 55 (1997) 4027.
[3] M. Ferhat, B. Bouhafs, A. Zaoui, H. Aourag, J. Phys.: Condens.
[31] P.E. Van Camp, V.E. Doren, J.T. Devreese, Phys. Rev. B 41 (1990)
Matter 10 (1998) 7995;
1598.
M. Ferhat, B. Bouhafs, A. Zaoui, H. Aourag, J. Phys.: Condens.
[32] B.I. Min, S. Massidda, A.J. Freeman, Phys. Rev. B 38 (1988) 970.
Matter 51 (1998) 285.
[33] Bal K. Agrawal, P.S. Yadav, Sudhir Kumar, S. Agrawal, Phys. Rev.
[4] Gus L.W. Hart, Alex Zunger, Phys. Rev. B 62 (2000) 13522.
B 52 (1995) 4896.
[5] R.M. Wentzcovitch, K.J. Chang, M.L. Cohen, Phys. Rev. B 34 (1986)
[34] Viktor N. Staroverov, Gustavo E. Scuseria, Jianmin Tao, John P.
1071.
Perdew, Phys. Rev. B 69 (2004) 075102.
[6] D.M. Bylander, L. Kleinman, Phys. Rev. B 34 (1986) 5280.
[35] B. Fillipi, D.J. Singh, C.J. Umrigar, Phys. Rev. B 50 (1994) 14947.
[7] Herbert Kroemer, Rev. Mod. Phys. 73 (2001) 783.
[36] Beate Paulus, Peter Fulde, Hermann Stoll, Phys. Rev. B 54 (1996)
[8] Kwiseon Kim, P.R.C. Kent, Alex Zunger, C.B. Geller, Phys. Rev. B
2556.
66 (2002) 045208.
[37] Philip P. Rushton, Stewart J. Clark, David J. Tozer, Phys. Rev. B 63
[9] W.E. Pickett, Comput. Phys. Rep. 9 (1989) 117.
(2001) 115206.
[10] P. Hohenberg, W. Kohn, Phys. Rev. 136 (1964) 864.
[38] M. Vubcevich, Phys. Status Solidi B 54 (1972) 219.
[11] W. Kohn, L.J. Sham, Phys. Rev. 140A (1965) 1133.
[39] H. Meradji, S. Drablia, S. Ghemid, H. Belkhir, B. Bouhafs,
[12] G.K.H. Madsen, P. Blaha, K. Schwarz, E. Sjostedt, L. Nordstrom,
A. Tadjer, Phys. Status Solidi B 241 (2004) 2881.
Phys. Rev. B 64 (2001) 195134;
[40] Su-Huai, Alex. Zunger, Phys. Rev. B 60 (1999) 5404.
K. Schwarz, P. Blaha, G.K.H. Madsen, Comput. Phys. Commun. 147
[41] M. Causà, R. Dovesi, C. Roetti, Phys. Rev. B 43 (1991) 11937.
(2002) 71. Available from: <http://www.wien2k.at>.
[42] R.T. Senger, S. Tongey, E. Durgun, S. Ciraci, Phys. Rev. B 72 (2005)
[13] E. Sjostedt, L. Nordstrom, D.J. Singh, Solid State Commun. 114
075419.
(2000) 15.
[43] R.M. Wentzcovitch, M.L. Cohen, J. Phys. C: Solid State Phys. 19
[14] J.P. Perdew, Y. Wang, Phys. Rev. B 45 (1992) 13244.
(1986) 6791.
[15] P. Perdew, K. Burke, M. Emzerhof, Phys. Rev. Lett. 77 (1996) 3865.
[44] H. Bross, R. Bader, Phys. Status Solidi B 191 (1995) 369.
[16] E. Engel, S.H. Vosko, Phys. Rev. B 47 (1993) 13164.
[45] A. Stroppa, M. Peressi, Phys. Rev. B 71 (2005) 205303.
[17] P. Dufek, P. Blaha, K. Schwarz, Phys. Rev. B 50 (1994) 7279.
[46] S. Adachi, J. Appl. Phys. 53 (1982) 8775.
[18] C. Persson, R. Ahuja, B. Johansson, Phys. Rev. B 64 (2001) 033201.
[47] S. Zh. Karazhanov, L.C. Lew Yan Voon, Fizika i Tekhnika
[19] A. Mokhatari, H. Akbarzadeh, Phyisca B 324 (2002) 305.
Poluprovodnikov 39 (2) (2005) 177.
[20] A. Mokhatari, H. Akbarzadeh, Physica B 337 (2003) 122.
[48] Su-Huai, Alex. Zunger, Phys. Rev. B 57 (1998) 8983.
[21] A. Mokhatari, H. Akbarzadeh, J. Phys.: Condens. Matter 16 (2004)
[49] W. Martiensses, H. Warlimont, Springer Handbook of Condensed
6063.
Matter and Materials Data, Springer, Berlin Heidelberg, Germany,
[22] Rashid Ahmed, H. Akbarzadeh, Fazal-e-Aleem, Physica B 370
2005.
(2005) 52.