Physics Express Phys.

Express 2014, 4: 16

The influence of P substitution for As on magnetic anisotropy in epitaxial

Ga Mn As P films grown on GaAs substrates

M. Yahyaouia,*, K. Boujdariaa, M. Cubukcub,c, C. Testelinc, C. Gourdonc

a
Laboratoire de Physique des Matériaux: Structures et Propriétés, Faculté des Sciences de Bizerte,
Université de Carthage, 7021 Zarzouna, Bizerte, Tunisia
b
[Current adress: ] SPINTEC, 17 rue des Martyrs, 38054 GRENOBLE Cedex 9, France
c
Institut des Nanosciences de Paris, Université Pierre et Marie Curie, CNRS-UMR7588, Paris, F-75005 France
*
Author for Correspondence: M. Yahyaoui, email: yahya.makrem@gmail.com
Received 4 Apr 2013; Accepted 17 May 2013; Available Online 17 May 2013

Abstract

We present a theoretical study on the influence of phosphorus fraction on the magnetic anisotropies in ferromagnetic
(Ga,Mn)(As,P) layers. Our calculations are based on the 40-band k.p model and treats the antiferromagnetic exchange coupling within the
mean field approximation. We show that the present k.p model reproduces accurately the overall band diagram as well as the band-gap
energy and the Luttinger parameters in the strained (Ga,Mn)(As,P)/GaAs system. Our simulations clearly show that the experimental results
for the magnetic anisotropy are quantitatively compared with microscopic calculations based on a mean-field model. Almost perfect
agreement between experiment and theory is found for the uniaxial out-of-plane and cubic-in-plane anisotropy parameter.

Keywords: Band structure; Diluted magnetic semiconductor; Magnetic anisotropy; k.p approach

important advantages over (Ga Mn )As. For example, it

1. Introduction has been predicted that the decrease in the lattice constant of
(Ga Mn )(As P ) should lead to an increase in the
For the past two decades, magnetic anisotropy (MA) Curie temperature, the exchange integral and the exchange
has focused much attention because it is essential not only for constant [8]. In a recent article [9], we have investigated the
the fundamental understanding of the microscopic origins of effect of epitaxial strain on the MA in a ferromagnetic
ferromagnetism in the diluted magnetic semiconductor (DMS), (Ga,Mn)(As,P) thin film with a typical composition of Mn
but also it is of central importance for applications of these and P . We have reported that the uniaxial and cubic MA
materials, since it determines the easy axis of magnetization, parameters reveal a pronounced dependence on the vertical
the anisotropy fields and the magnetic domain wall width and strain. We also showed that the microscopic calculations are in
specific energy. As a consequence of the semiconductor band good agreement with the theoretical and experimental results
structure (BS) of DMS, the MA is strongly dependent on Mn especially in the description of the fourth-order in-plane and
and carrier concentrations, temperature, and strain [1-3]. This out-of-plane components of the MA. The anisotropy constants
dependence allows the use of various static or quasi-static obtained from the microscopic calculations allow to construct
methods to intentionally manipulate the magnetic easy axes. a three-dimensional magnetic free energy surface, which
Strain control of MA in (Ga Mn )(As P ), which we clearly indicates that the easy axis of magnetization depends
focus on in this paper, has so far been explored less on the strain conditions. One of the remaining open questions
extensively. It has been reported that introducing a few percent is the influence of phosphorus fraction on the electronic and
of phosphorus at the arsenic sites in (Ga Mn )As layers, magnetic properties of (Ga,Mn)(As,P).
can also modify the epitaxial strain in the (Ga,Mn)(As,P) As far as we know, although several works were
layers continuously from tensile to compressive independently devoted to the study of Ga Mn As P [5-8,10-12], the
from the Mn doping level [4-7]. Thus phosphorus composition dependence of the
(Ga Mn )(As P ) compounds have been attracting Ga Mn As P band parameters and the link between
considerable attention due to the ability to produce layers with MA and the band structure has not been theoretically
a perpendicular to plane anisotropy such that the layers are investigated. Let us note that one cannot simply map onto
under tensile strain when grown on a GaAs substrate. Also, Ga Mn As P the calculations performed for
these layers are expected to contain much less defects than in
Ga Mn As. These generally use the band parameters of
other layer structures designed to produce normal to plane
GaAs. Here one needs to input the band structure of the
magnetic easy axis, such as (Ga Mn )As grown on a
GaAs P alloy.
relaxed (Ga,In)As buffer layer. Consequently, the self-
organization of magnetic domains in periodic patterns,
2. Band Structure
impossible to reach in (Ga,Mn)As/(Ga,In)As, can be achieved
in (Ga,Mn)(As,P)/GaAs owing to the low density of domain
2.1. Methodology
wall pinning defects. These quaternary layers have many other

Cognizure 1
www.cognizure.com/pubs © Cognizure. All rights reserved.
Physics Express Phys. Express 2014, 4: 16

Table 1. (Ga Mn )(As P ) sample parameters; The lattice mismatch lm, the relaxed lattice parameter, the corresponding strain
component, the saturation magnetization at T=4 K, the effective Mn concentration, the splitting parameter of the valence subbands BG and
the p-d exchange energy
Sample A B C D E F G
[P] (%) 0 2.6 3.4 4.3 5.6 7 11.3
lm (ppm) 4990 280 0 -1200 -2320 -3580 -6620
0.23 0.013 0 -0.055 -0.11 -0.16 -0.31
5.6661 5.6539 5.6532 5.6501 5.6472 5.6439 5.6359
1.20 1.207 1.208 1.21 1.212 1.214 1.22
M (emu cm 37.6 37.7 37.3 40.75 42.6 39.2 35
3.69 3.67 3.63 3.96 4.14 3.80 3.49
-26.0 -26.3 -26.0 -28.5 -30.0 -27.5 -24.7

same Mn (7%) concentration, same thickness of the films (50

nm) and different P concentrations. It is clear from this Table
that the strain in the Ga Mn As P film changes from
compressive to tensile i.e., the sign of changes from
negative to positive value as the P composition increases in the
Ga Mn As P epilayer.
The presence of Mn ions in Ga Mn As P is
taken into account by including the exchange
interactions. Hence, describes the
exchange coupling between the localized d-electrons and the
delocalized holes and electrons . The exchange
interaction can be further simplified by the mean-field
approximation where the virtual crystal approximation is made
[16, 1]. The conduction and valence bands exchange splittings
are obtained from the exchange energies and where
is the concentration of cation sites and and are the
, and exchange integrals, respectively. In order to
obtain the values of we assume and
Figure 1. Band structure of Ga Mn As/GaAs meV nm . While the p-d exchange energy is taken
compound semiconductor along the high symmetry lines of the for the computations, there is no report concerning
first Brillouin zone.
experimental or theoretical measurement of exchange
In order to obtain the dependance of the MA as a energy . Its value eV, as observed for II-VI diluted
function of phosphorus fraction, the initial step is the band- magnetic semiconductors [17], has been adopted for our
structure calculation of (Ga Mn )(As P ) bulk computations. One should also note that, in the Zener model,
semiconductors using the k.p theory. In this model, the energy the exchange splitting of the valence subbands
band structure is obtained by a set of parameters which are important parameters for the band structure
represent the energy gaps at , the momentum matrix and the magnetic anisotropy. The parameter describing the
elements, and the strength of the spin-orbit coupling. The enhancement of spin density of states by the carrier-carrier
number of energy bands (or levels) that are effectively interactions is taken as 1.4 (close to the value 1.2 used in [1]).
involved is related to the precision of the results. In this According to the k.p effective Hamiltonian theory, the analysis
section, we propose to extend the 40-band k.p model that was of structure of epitaxial Ga Mn As P layers grown on
first introduced by Saïdi et al [13]. The accuracy of several GaAs substrates require the solution of the total Hamiltonian
bulk semiconductor band structure calculations with this .
approach is discussed in Ref. [13]. Here, we do not reproduce
all the complicated basis orbitals, energy level structures, band 2.2. Results
edge energies, and matrix elements in a 40-band k.p model. In this section we present and discuss the results of
For detailed description instead referring the reader to Ref. the application of our k.p model to the electronic structure
[14]. calculations for bulk Ga Mn As P SC . All these
The strain effect on the electronic structure has been calculations were carried out at the relaxed lattice constant. An
included through the strain Bir-Pikus Hamiltonian model example of the BS calculated with the full Hamiltonian for
[15]. We assumed that our materials only have the epitaxial and is shown in Figure 1. The electronic
biaxial tensile strain, which breaks the crystallographic cubic BS of this compounds shows features similar to that of bulk
symmetry due to the lattice mismatch between the bulk GaAs. The band diagram is well reproduced on a width of
Ga Mn As P epilayer and the relaxed bulk GaAs about eV. The extended k.p method is known to be valid
buffer layer. The lattice mismatch lm was determined by high up to eV above and eV below the top of the valence band
resolution x-ray diffraction. The characteristics of the samples (VB) in four directions namely , , and . It is
which we study in this paper is summarized in Table 1. The important to note that our k.p approach respects the continuity
seven samples are selected in order to compare layers with the between and , equivalents points
2 Cognizure
© Cognizure. All rights reserved. www.cognizure.com/pubs
Physics Express Phys. Express 2014, 4: 16

a smaller amplitude. The increase in the gap seen in Figure 3

results from the alloying effect (i). The plateau near [P] 4 %
results from the cancellation of strain contributions (iii) and
(iv) at zero strain together with a slightly increased exchange
down-shift (13% larger for [P]=6% than for 0% and 11.3%).
An accurate description of the BS makes our results
reliable to calculate the Luttinger parameters .
One has to keep in mind, however, that within the two-level
k.p model, which is often employed to explain the electronic
band structure of (III Mn )V semiconductor, one needs to
specify the Luttinger parameters to take into account the
interaction of the valence band with remote bands. This can be
achieved for (Ga,Mn)As by taking the parameters of GaAs but
these parameters are not available for (Ga,Mn)(As,P). In the
framework of the 40-band model the coupling between bands
is described in terms of the matrix elements and band edges
that are deduced from the experiments. Once the band
Figure 2. Calculated band-gap energy E of structure is well reproduced with the 40-band model, we then
Ga Mn As P as a function of content for the calculate the Luttinger parameters [13] which can be used
samples of Table 1. within the 6-band model. This is an important advantage of the
40-band model as compared to the 6-band model. The
behavior of Luttinger parameters is a consequence of the
competition between the alloy effect and the strain effect. In
Figure 3, we show the dependence of Luttinger parameters
versus . An important effect of alloy can explain this
variation of as a function of phosphorus fraction. The effect
of strain differs according to regime of strain considered.
Indeed, comparatively to unstrained case, the compressive
strain leads to a decrease of the Luttinger parameters while the
tensile strain leads to an increase of .

3. Magnetic Anisotropy

MA is the dependence of the free energy density of

a magnetic system on the orientation of the magnetization
direction . As given in Ref. [18], can be
Figure 3. Luttinger parameters of
divided in two contributions:
Ga Mn As P /GaAs obtained with 40-band k.p model . The first term, namely , represents an
(symbols). The full lines represent the Luttinger parameters for intrinsic contributions to which are arising from spin-orbit
unstrained GaAsP. coupling in the VB. For a phenomenological description, we
expand in powers of the components , , and
of BZ which is not obtained by construction in contrast with of along the cubic axes , , ,
pseudopotentials and tight-binding methods. This continuity is respectively. Considering terms up to the fourth order in ,
the strongest numerical difficulty of this method. the only intrinsic contribution to can be written as
To investigate the effect of phosphorus fraction on the .
electronic properties of alloys, we performed calculations of The parameters denote the magnetic intrinsic
energy gaps at high-symmetry points in the Brillouin zone. anisotropy parameters. Extrinsic contributions to , namely
The band gap energy at point, , is defined as the difference , are the shape anisotropy, caused by the
between the lowest conduction band and the highest valence demagnetization field perpendicular to the layer, and an
band in the presence of strain and exchange. The variation of uniaxial in-plane contribution, whose microscopic origin is
as a function of the phosphorus content in the quaternary under debate [3]. In our theoretical approach, we have
alloy layers Ga Mn As P /GaAs is displayed in Figure considered only the contribution. The numerical
2. This variation results from four contributions: (i) the procedure to determine the MA is described in Refs. [9, 19].
increase of the gap of the bulk Ga(As,P) with P concentration We use the sample parameters specified in Table 1. Due to the
(alloying effect), (ii) the down-shift of the gap resulting from presence of charge compensating defects depending on growth
the exchange splitting of the valence and conduction bands conditions in Ga Mn As P samples, the exact
(almost constant here since and have small variations determination of the Fermi level in ferromagnetic
with [P]), (iii) the effect of the hydrostatic part of the strain Ga Mn As P is difficult. Hence, for reasons of
that increases the gap for compressive strain and decreases it simplicity and lack of data on the hole concentration, we have
for tensile one thus opposing the alloying effect, and finally assumed that the Fermi level has a constant value equal to -110
(iv) the effect of the biaxial strain on the valence band that meV from the top of the VB for all samples except for sample
gives a variation similar to the hydrostatic component but with G, where is taken equal to -81 meV.
Cognizure 3
www.cognizure.com/pubs © Cognizure. All rights reserved.
Physics Express Phys. Express 2014, 4: 16

angular dependences by using , , and as fit

parameters.
The results of the microscopic calculations are shown
by full symbols in Figures 4(a) and 4(b). The full lines show
least-squares fits to the calculated data using Eqs. 9 and 10 in
Ref. [19]. We observe that the fit curves are in very good
agreement with the microscopic results. This good fit clearly
demonstrates that the microscopic theory can be well described
by , , and . Therefore a realistic description of the
part of the free energy only needs terms up to the fourth
order in . We proceed to understand the role of the
phosphorus fraction on the magnetic properties of
Ga Mn As P quaternary alloys. For that purpose we
calculate and for three representative P
concentrations: (tensile strain), ( zero
strain), and (compressive strain), respectively.
Figure 4(a) shows that in Ga Mn As P there is a
stronger fourfold cubic in-plane anisotropy constant under
tensile strain than under compressive strain. Such a behavior
Figure 4. calculated as a function of the magnetization
orientation and phosphorus concentration for M (a) in the (010)
has also been shown in the case of Ga Mn As
plane and (b) in the (100) plane. The results of the microscopic ferromagnetic SC [19, 20].
model are represented by the solid symbols. The solid lines are Applying the phenomenological procedure described
least-square fit curves. above to the whole set of Ga Mn As P /GaAs layers
under study, the anisotropy parameters , , and
were determined as a function of phosphorus fraction.
We have taken the experimental data at in
order to compare with our calculations performed at zero
temperature. Firstly, we focus on the behavior of
parameter. In Figure 5(a), is plotted versus phosphorus
fraction . The microscopic calculations are in very good
agreement with experimental data. However, the results of
deduced from the microscopic model are found to be slightly
different from the experimental data for and
. The microscopic model predicts that is linearly
proportional to the uniaxial deformation . Indeed, a zero
value of the uniaxial out-plane parameter is found for zero
strain ( ). However a non-zero value is obtained
experimentally. This discrepancy might originate from the
experimental difficulty to determine the strain in this range
giving a large error bar at zero strain: 600 ppm. Additionally,
it might be necessary to include higher order terms in for
the analysis of experimental results when , , and
are close to zero. It can also be caused by the presence of
residual strain in the experimentally fabricated film. For the
range an almost linear decrease of is
evidenced. For larger concentrations ,
depends less strongly on . This result can be explained by the
behavior of Luttinger parameters in the range
and the lower Fermi energy for .
Figure 5. Phosphorus concentration dependence of (a) the The fourth-order parameters and are
uniaxial out-of-plane magnetic anisotropy constant , (b) the presented in Figures 5(b) and 5(c), respectively. We can see
in-plane cubic anisotropy constant and (c) the out-of-plane that the experimental data of are very well reproduced by
cubic anisotropy constant . The magnetic anisotropy
the microscopic calculations. The values obtained using
constants obtained with microscopic calculation (circles) are
compared with the experimental data obtained from FMR the the 40-band k.p model in (Ga,Mn)(As,P) are in much
experiments at T=4.2 K (squares). better agreement with experimental values than in ternary
(Ga,Mn)As using the 6-band k.p model [19]. We believe that
Once is calculated numerically in the the more accurate description of the MA originates from a
microscopic model as a function of the crystalline orientations better description of the bands with our 40-band k.p model in
with three strain conditions , , the range used for the calculation of the MA , in particular
. The phenomenological description of the MA from a good determination of the Luttinger parameters.
(Eqs 9 and 10 in Ref. [19]) can be fitted to the resulting Besides, the assumption that consists in taking the Fermi level

4 Cognizure
© Cognizure. All rights reserved. www.cognizure.com/pubs
Physics Express
Figure 6. Three-dimensional plot of intrinsic contribution of the
free-energy densities for Ga Mn As P at 4.2K
with different phosphorus fraction.
as a constant can be considered as relevant because of the
almost perfect agreement between experiment and theory.
Unlike the results of , those of (Figure 5(c))
show large variations between the microscopic values and the
experimental ones. This discrepancy becomes less important
for samples which present close experimental values of
and (samples B, D, E, and G). However, the predictions of
microscopic calculations could not explain the large variation
observed in samples A, C and F. Moreover, we can notice that
the change of sign for is not reproduced by microscopic
calculations. The same problem has also been reported in Ref.
[19] for Ga Mn As. At the moment, the reason for this
discrepancy is not yet understood. We cannot exclude the
possibility that the change of sign of is an artifact of the
experimental method for determining [19, 20]. In samples
with high coefficient the experimental determination of
might be unreliable [3].
From the numerical values of the anisotropy constants
we can calculate the free-energy density as a function of the
crystalline orientation for several phosphorus fractions. The
three-dimensional (3D) free energy density diagrams can be
constructed using the magnetic anisotropy parameters given in
Figure 5. Figure 6 presents the 3D plots of at 4.2K for
different phosphorus fraction, namely , ,
and , respectively. These surfaces show the energy
barriers for the magnetization reorientation and define the
preferential axes of the magnetization. The free energy for the
magnetization oriented along a specific direction is given by
the distance from the center of the coordinate system to the
surface. We can notice that the free energy in the case of
Ga Mn As P /GaAs is dominated by the out-of-plane
components of the magnetic anisotropy parameters, namely
and , as shown in Figure 5. These 3D plots given in
Figure 6 illustrate clearly that in the regime of tensile strain
exhibits a minimum for oriented out-
of-plane. For compressive strain , the minima
occur for oriented in-plane. Thus, the theoretical model
correctly describes the well-known experimental fact that the
magnetically hard axis along in compressively strained
layers turns into an easy axis in tensily strained layers as
observed in Ref. [7]. Note, however, that for the regime of
zero strain , the easy axis is not aligned along any
high direction. This result is due to the values approximately
equal taken by the cubic anisotropy parameters . Phys. Rev. B. 75 (2007) 045202.
These observations clearly point out that the easy axis of
magnetization is sensitive to the variation of the phosphorus
fraction or, in other words, the preferential axes of the
magnetization are closely related to the strain conditions.
Cognizure
www.cognizure.com/pubs © Cognizure. All rights reserved.
Phys. Express 2014, 4: 16
4. Conclusions
We have presented a theoretical survey of the
magnetic properties of ferromagnetic
Ga Mn As P /GaAs epilayers using the mean-field
approximation and the 40-band k.p model, in which the band
structure and band parameters are calculated. The present k.p
description accurately reproduces the overall band diagram, as
well as the band shifts and the Luttinger parameters versus
phosphorus content. We have considered the effect of the
phosphorus composition on the MA. The present microscopic
model has been validated through an accurate set of
comparisons with experimental results. In particular, the
values of and anisotropy parameters are very well
reproduced. The systematic dependence of the easy-axis of the
magnetization on the phosphorus fraction observed in this
study can be readily understood by referring to the 3D free
energy surface diagrams.
In the next step, it would be instructive to calculate
the optical properties, in particular the dielectric permittivity
and the spectral dependence of the magneto-optical Kerr
effect, which is of great importance for the optimization of
optical pump-probe experiments and magneto-optical imaging
of magnetic domains in these ferromagnetic semiconductors.
Supplementary Information
The explicit form of the different Hamiltonian
matrices used in our calculations is given in the supplementary
information.
Acknowledgements
The authors thank Laura Thevenard from INSP for
useful discussions. This work was performed in the framework
of the MANGAS project (ANR 2010-BLANC-0424-02).
References
1. T. Dietl, H. Ohno, and F. Matsukura, Phys. Rev. B. 63 (2001)
195205.
2. M. Abolfath, T. Jungwirth, J. Brum, and A. H. MacDonald,
Phys. Rev. B. 63 (2001) 054418.
3. J. Zemen, J. Kucera, K. Olejnik, and T. Jungwirth, Phys. Rev.
B. 80 (2009) 155203.
4. P. R. Stone, J. W. Beerman, K. M. Yu, and O. D. Dubon,
Physica B 401-402 (2007) 455.
5. P. R. Stone, K. Alberi, S. K. Z. Tardif, J. W. Beerman, K. M.
Yu, W. Walukiewicz, and O. D. Dubon, Phys. Rev. Lett. 101
(2008) 087203.
6. A. Lemaître, A. Miard, L. Travers, O. Mauguin, L. Largeau, C.
Gourdon, V. Jeudy, M. Tran, and J. -M. George, Appl. Phys.
Lett. 93 (2008) 021123.
7. M. Cubukcu, H. J. von Bardeleben, Kh. Khazen, J. L. Cantin,
O. Mauguin, L. Largeau, and A. Lematre, Phys. Rev. B. 81
(2010) 041202.
8. J. Masek, J. Kudronovsky, F. Maca, J. Sinova, A. H.
Macdonald, R. P. Campion, B. L. Gallagher, and T. Jungwirth,
9. M. Yahyaoui, C. Testelin, C. Gourdon, and K. Boujdaria, J.
Appl. Phys. 111 (2012) 033902.
10. S. Haghgoo, M. Cubukcu, H. J. von Bardeleben, L. Thevenard,
A. Lemaître, and C. Gourdon, Phys. Rev. B. 82 (2010) 041301.
5
 Physics Express Phys. Express 2014, 4: 16

11. L. Thevenard, E. Peronne, C. Gourdon, C. Testelin, M.

Cubukcu, E. Charron, S. Vincent, A. Lemaître, and B. Perrin,
Phys. Rev. B. 82 (2010) 104422.
12. L. Thevenard, S. A. Hussain, H. J. von Bardeleben, M.
Bernard, A. Lemaître, and C. Gourdon, Phys. Rev. B. 85
(2012) 064419.
13. I. Saïdi, S. Ben Radhia, and K. Boujdaria, J. Appl. Phys. 107
(2010) 043701.
14. N. Rim, I. Saïdi, and K. Boujdaria, J. Appl. Phys. 112 (2012)
053716.
15. G. E. Pikus and G. L. Bir, Fiz. Tverd. Tela (Leningrad) 1
(1959) 1642, Sov. Phys. Solid. State 1 (1960) 1502; G. L. Bir
and G. E. Pikus, Symmetry and Strain-Induced Effects in
Semiconductors, Wiley, New-York (1974).
16. T. Dietl, H. Ohno, F. Matsukura, J. Cibert, and D. Ferrand,
Science 287 (2000) 1019.
17. F. Glas, G. Patriarche, L. Largeau, and A. Lemaître, Phys. Rev.
Lett. 93 (2004) 086107.
18. W. Limmer, M. Glunk, J. Daeubler, T. Hummel, W. Scoch, R.
Sauer, C. Bihler, H. Huebl, M. S. Brandt, and S. T. B.
Goennenwein, Phys. Rev. B. 74 (2006) 205205.
19. M. Glunk, J. Daeubler, L. Dreher, S. Schwaiger, W. Schoch, R.
Sauer, W. Limmer, A. Brandlmaier, S. T. B. Goennenwein, C.
Bihler, and M. S. Brandt, Phys. Rev. B. 79 (2009) 195206.
20. X. Liu and K. Furdyna, J. Phys. Condens. Matter 18 (2006)
R245.

Cite this article as:

M. Yahyaoui et al.: The influence of P substitution for As on magnetic anisotropy in epitaxial Ga Mn As P films
grown on GaAs substrates. Phys. Express 2014, 4: 16

6 Cognizure
© Cognizure. All rights reserved. www.cognizure.com/pubs