57Fe emission Mössbauer spectroscopy following dilute implantation of 57Mn into In 2O3

Abstract

Emission Mössbauer spectroscopy has been utilised to characterize dilute 57Fe impurities in In 2O3 following implantation of 57Mn (T
1/2 = 1.5 min.) at the ISOLDE facility at CERN. From stoichiometry considerations, one would expect Fe to adopt the valence state 3 + , substituting In 3+, however the spectra are dominated by spectral lines due to paramagnetic Fe2+. Using first principle calculations in the framework of density functional theory (DFT), the density of states of dilute Fe and the hyperfine parameters have been determined. The hybridization between the 3d-band of Fe and the 2p band of oxygen induces a spin-polarized hole on the O site close to the Fe site, which is found to be the cause of the Fe2+ state in In 2O3. Comparison of experimental data to calculated hyperfine parameters suggests that Fe predominantly enters the 8b site rather than the 24d site of the cation site in the Bixbyite structure of In 2O3. A gradual transition from an amorphous to a crystalline state is observed with increasing implantation/annealing temperature.

Full text

Hyperfine Interact (2016) 237:75
DOI 10.1007/s10751-016-1258-9
57Fe emission M¨
ossbauer spectroscopy following dilute
implantation of 57Mn into In2O3
A. Mokhles Gerami1,2 ·K. Johnston1,3 ·H. P. Gunnlaugsson1·K. Nomura4·
R. Mantovan5·H. Masenda6·Y. A . M atv eye v7·T. E. Mølholt1·M. Ncube6·
S. Shayestehaminzadeh8·I. Unzueta9·H. P. Gislason8·P. B. Krastev10 ·
G. Langouche11 ·D. Naidoo6·S. ´
Olafsson8·the ISOLDE collaboration1
© Springer International Publishing Switzerland 2016
Abstract Emission M¨
ossbauer spectroscopy has been utilised to characterize dilute 57Fe
impurities in In2O3following implantation of 57Mn (T1/2=1.5 min.) at the ISOLDE
This article is part of the Topical Collection on Proceedings of the International Conference on the
Applications of the M¨
ossbauer Effect (ICAME 2015), Hamburg, Germany, 13–18 September 2015
H. P. Gunnlaugsson
Haraldur.p.gunnlaugsson@cern.ch
1CERN, PH Div, 1211 Geneve 23, Switzerland
2Department of physics, K.N.Toosi University of Technology, P.O.Box 15875-4416, Tehran, Iran
3Universit¨
at des Saarlandes, Experimentalphysik, 66123 Saarbrucken, Germany
4Photocatalysis International Research Center, Tokyo University of Science, Yamazaki 2641, Noda,
278-8501, Japan
5Laboratorio MDM, IMM-CNR, Via Olivetti 2, 20864 Agrate Brianza (MB), Italy
6School of Physics, University of the Witwatersrand, Witwatersrand, South Africa
7Moscow Institute of Physics and Technology, 9 Institutskiy per., Dolgoprudny, Moscow Region,
141700, Russian Federation
8Science Institute, University of Iceland, Dunhaga 3, IS-107 Reykjav´
ık, Iceland
9BCMaterials & Elektrizitate eta Elektronika Saila, Euskal Herriko Unibertsitatea (UPV/EHU),
48048 Bilbao, Spain
10 Institute for Nuclear Research and Nuclear Energy, Bulgarian Academy of Sciences, 72
Tsarigradsko Chaussee Boulevard, Sofia, 1784, Bulgaria
11 KU Leuven, Instituut voor Kern-en Stralings Fysika, Leuven, Belgium

75 Page 2 of 9 Hyperfine Interact (2016) 237:75
facility at CERN. From stoichiometry considerations, one would expect Fe to adopt the
valence state 3+, substituting In3+, however the spectra are dominated by spectral lines
due to paramagnetic Fe2+. Using first principle calculations in the framework of density
functional theory (DFT), the density of states of dilute Fe and the hyperfine parameters have
been determined. The hybridization between the 3d-band of Fe and the 2p band of oxygen
induces a spin-polarized hole on the O site close to the Fe site, which is found to be the
cause of the Fe2+state in In2O3. Comparison of experimental data to calculated hyperfine
parameters suggests that Fe predominantly enters the 8b site rather than the 24d site of the
cation site in the Bixbyite structure of In2O3. A gradual transition from an amorphous to a
crystalline state is observed with increasing implantation/annealing temperature.
Keywords In2O3·Fe doping ·57Mn implantation ·Emission M ¨
ossbauer spectroscopy ·
Annealing ·Density functional theory
1 Introduction
Doping non-magnetic semiconductors with 3d-elements to obtain dilute magnetic semi-
conductors (DMSs) has motivated a large amount of research following the theoretical
prediction of room temperature (RT) ferromagnetism in ZnO-based DMSs [1]. Here Fe
doped In2O3is among the systems where DMS has been reported [2], and where the charge
state of Fe has been suggested to play an important role in the magnetic properties [3,4].
In2O3is a wide bandgap semiconductor, and is of interest due to its good electrical
conductivity and high optical transparency [5]. Crystalline In2O3has a body-centred cubic
structure known as Bixbyite (space group Ia-3 (206)). There are two different kinds of
cation sites, denoted as 24d and 8b, and the ratio of 24d and 8b sites in a unit cell is 3:1
[6–8]. The 8b cation site has octahedral symmetry and is surrounded by 6 equivalent nearest
neighbour oxygen atoms. The 24d cation site has asymmetric structure and is surrounded
by 6 neighbour oxygen atoms with three pairs distances (d1<d
2<d
3)(Fig. 1b). These
sites have been proposed to have different magnetic states [9–12].
In order to investigate the site occupancy and chemical or magnetic properties of dilute
Fe in ion-implanted In2O3, we have performed 57Fe emission M ¨
ossbauer Spectroscopy
(eMS) following implantation of radioactive 57Mn (T1/2=1.5 min.) at the ISOLDE facility
at CERN. The results are compared to calculations using density functional theory (DFT)
within the Wien2K package [13,14].
2 Experimental methods and samples
Beams of 57 Mn (T1/2=1,5 min.) were produced at the ISOLDE/CERN by 1.4 GeV
proton induced fission in a heated UCxtarget. Element selective laser ionization was used
to ionize manganese, which was then accelerated to 50 keV. After magnet mass separa-
tion, a clean beam with intensity of few times 10857
Mn+/s was obtained. The 57Mn ions
were implanted into In2O3polycrystalline films with columnar structure heated from the
backside with a halogen lamp inside an implantation chamber. Implantation took place at
30◦relative to the sample surface normal. Emission M¨
ossbauer spectra were measured at
60◦relative to the sample surface normal with a Parallel Plate Avalanche Detector (PPAD)
equipped with 57Fe-enriched stainless steel mounted on a conventional drive system out-
side the implantation chamber. The implantation fluence rate was ∼10957
Mn/(cm2·s) and

Hyperfine Interact (2016) 237:75 Page 3 of 9 75
OIn8b In24d
d1
d1
d2
d2
d3
d3
Fig. 1 Crystalline structure of In2O3.Left: Big spheres are the two different In sites and small spheres are
O anions. Right: the crystallographic view of the two In sites
the maximum fluence per sample was ∼1012 57Mn/cm2, corresponding to a maximum con-
centration of 4 ×10−4at.%. Isomer shifts and velocities are given relative to the centre of
the α-Fe spectrum at RT.
3 Computational method
Using the WIEN2K code [13], the isomer shift (δ) and quadrupole splitting (EQ) param-
eters have been calculated within density functional theory by considering the generalized
gradient approximation (GGA) [14]. The Pedrew Burke Ernzerhof (PBE) of GGA func-
tional was used for all of the DFT calculations [15]. In the calculations performed, the radii
of the muffin-tin atomic spheres of In, O, and Fe are set to 2.15, 1.8, and 2.10 a.u., respec-
tively. In addition, the energy value of -6 Ry is set as the boundary separating the core
electron states and valence electron states. The cut-off parameter RMTKMAX, which controls
the size of the basis set, is set to 7.0, a mesh of (4×4×4)k-points in the irreducible part
of the first Brillouin zone was applied to the self-consistent total energy calculation. Within
this approach, the calculations of the Fe-doped In2O3in different configurations were per-
formed with the lattice constant of virgin In2O3at room temperature, based on a periodic
supercell of 80 atoms.
The isomer shift is obtained from the contact densities (ρ) as [16]:
δ=α(ρA−ρs)(1)
where, ρAand ρsare the charge densities at the nucleus in the absorber (A) and sample (S)
material, respectively. The calibration constant of α=−0.29 a.u.3mm/sasinRef.[17]was
used. The electric field gradient tensor (EFG), expressed by its principal component (VZZ)
and the asymmetry parameter (η) gives the quadrupole splitting as
EQ=eQVZZ
21+η2/3(2)

75 Page 4 of 9 Hyperfine Interact (2016) 237:75
Fig. 2 57Fe emission
M¨
ossbauer spectra obtained after
implantation of 57 Mn into In2O3
sample held at the temperatures
indicated
-4 -3 -2 -1 0 1 2 3
)stinu.bra(noissimeevitaleR
Velocit
y
(mm/s)
763 K
652 K
537 K
303 K
426 K
where eis the elementary charge and Q=0.16 ·10−28 m2[18] is the nuclear quadrupole
moment of the 14.4 keV M¨
ossbauer state of 57Fe.
4 Experimental results
The eMS spectra obtained (Fig. 2) are dominated by a single asymmetric quadrupole split
component with isomer shift characteristic of high spin Fe2+. The spectra were analysed
using the program Vinda [19] with a quadrupole splitting distribution where the distribu-
tion function was simulated with two linear segments [20]. To account for the asymmetry, a
coupling between isomer shift and quadrupole splitting was assumed as δ=d0+d1·EQ,
where d0and d1are fitting variables. As seen in the fit in Fig. 2, this model adequately
explains the experimental data although the spectrum obtained at 426 K could benefit from a
more complicated (3 segment) distribution function. At room temperature, there are notice-
able signs of a misfit due to a broad Fe3+component. Figure 3shows a zoom-in of the
spectral intensity close to the background in the 300 K measurement.
There is a small, but significant tail around v∼4 and -8 mm/s, which the quadrupole
splitting distribution cannot describe. A plausible explanation for this would be high spin
Fe3+showing slow paramagnetic relaxations, as observed in other oxides in similar exper-
iments [21–23]. Such a feature could account for ∼6 % of the relative spectral area. At
temperatures >426 K, this feature has disappeared completely from the spectra.
The isomer shift (Fig. 4a) clearly identifies the doublet component as due to high-
spin Fe2+. There is a change of trend on how the experimental data follows the second
order Doppler shift, suggesting an annealing stage between 400 and 500 K. The average

Hyperfine Interact (2016) 237:75 Page 5 of 9 75
Fig. 3 Spectral intensity close to
the baseline in the 300 K
spectrum. The peak of the
spectrum is at 220 % above the
background
-12 -8 -4 0 4 8 12
noissimE
Velocity (mm/s)
5%
quadrupole splitting (Fig. 4b) shows a gradual decrease through the temperature range as
expected for a temperature-dependent population of 3d orbitals. The width of the quadrupole
splitting distribution represented by its standard deviation (Fig. 4c) also shows a change
between 400 and 500 K, again suggesting a change in the nature of the Fe sites, as was
observed with the trend of the isomer shift (Fig. 4a). The coupling between isomer shift and
quadrupole splitting (Fig. 4d) shows a complicated dependence. Below 500 K it is negative,
peaks with a positive value at ∼650 K and then decreases again.
5 Theoretical results
The hyperfine parameters and the local structure around the Fe probe in In2O3were sim-
ulated using first principle calculations for Fe on two different In sites with a dilution of
1:32 which corresponds to concentration of 3.12 %, the octahedrons of Fe at 8b site and the
distorted octahedrons of Fe at 24d site as illustrated in Fig. 1. The density of states (DOSs)
of the two calculated configurations are presented in Fig. 5. The valence band of In2O3is
mainly composed of 2p oxygen states, while the conduction band is mainly composed of 4s
In states. By doping Fe on the In8b site in In2O3, the 3d energy levels are split into three- and
two-fold degenerate t2g(dxy ,d
xz,d
zy )and eg-levels (dz2,d
x2−y2)with the t2g levels lower
in energy (Fig. 5b). Four electrons are in t2g (3t2g ↑,1t2g ↓)level energy and two electrons
are in eg(2eg↑)level energy. From Fig. 5b and d, it is seen that the energy level of 3d
states are merged with the valence band. This causes hybridization between the 3d and 2p
states. This hybridization creates spin polarized hole states on the oxygen sites. Since, the
3d band of Fe ions is more than half filled, the spin on the hole state of the oxygen sites
will be parallel to the spin of Fe ions. These findings are in accordance with the findings of
Huang et al. [7]. Therefore, we expect the charge state of Fe substituting the In3+site to be
close to 2+. This calculation is in agreement with the measured eMS data, which show Fe
in the high spin 2+state. Due to the presence of asymmetry at the Fe24d site, the d levels
of energy for Fe get split to three energy levels (dz2,d
x2−y2+dxy,d
xz+dyz)(Fig. 5d). The
calculated hyperfine parameters are given in Table 1.
6 Discussion
A simple explanation of the temperature-dependence of the hyperfine parameters (Fig. 4)
is in terms of the annealing of local implantation-induced amorphous zones between 400 K

75 Page 6 of 9 Hyperfine Interact (2016) 237:75
0.8
1.0
1.2
1.4
1.6
1.8
2.0
300 400 500 600 700 800
E
Q,av.
(mm/s)
Tem pe rature (K )
0.6
0.7
0.8
0.9
1.0
300 400 500 600 700 800
av.
(mm/s)
Temperature (K)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
300 400 500 600 700 800
P(E
Q
) st. dev. (mm/s)
Temperature (K)
-0.3
-0.2
-0.1
0.0
0.1
300 400 500 600 700 800
d
1
Tem pe rature (K )
b
a
c
d
Fig. 4 Temperature dependence of hyperfine parameters obtained from the analysis of the eMS spectra in
Fig. 2.aIsomer shift compared to the second order Doppler shift (SOD). bAverage quadrupole splitting.
cStandard deviation of the quadrupole splitting distribution dThe coupling parameter between quadrupole
splitting and isomer shift
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7
-50
-40
-30
-20
-10
0
10
20
30
40
50
Energy (eV)
)Ve/setatS(etatsfoytisneD
Total DOS
-4 -3 -2 -1 0 1 2 3 4
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
Energy (eV)
Density of state (States/eV)
2p of O
3d of Fe
eg
t2g
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7
-50
-40
-30
-20
-10
0
10
20
30
40
50
Ener
gy
(eV)
)Ve/setatS(etatsfo
y
tisneD
Total DOS
-4 -3 -2 -1 0 1 2 3 4
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
Ener
gy
(eV)
Density of state (States/eV)
2P of O
3d of Fe
d
z
2
d
xy+d
x2-y2
d
xz
+d
yz
E
F
E
F
a) 8-b site
d)24-d site
b)8-b site
c)24-d site E
F
E
F
Fig. 5 Projected DOSs of the total structure of In1−xFexOfora) 8b site of Fe and c) 24d site of Fe and the
contribution of Fe and O orbitals are plotted of the b) 8b site and d) 24d site

Hyperfine Interact (2016) 237:75 Page 7 of 9 75
Tab l e 1 DFT calculated nuclear hyperfine parameters for Fe on the two different sites of In in In2O3,as
indicated in the symmetry structure
Position of Fe VZZ(×1021 V/m2)δ(mm/s) EQ(mm/s) M ¨
ossbauer spectroscopy Line
Fe8b6.3 0.81 1.06 Fe2+in crystalline phase
Fe24d3.06 0.82 0.51
and 500 K. This would explain the change in isomer shift (Fig. 4a) and width of the
quadrupole splitting distribution (Fig. 4c). The data then suggests that Fe2+in amorphous
zones is characterized by a negative coupling between isomer shift and quadrupole split-
ting, whereas the crystalline environment is characterized by a positive coupling (Fig. 4d).
The two cation sites in the Bixbyite structure suggest that in the crystalline environment,
one should observe two quadrupole doublets (if both sites are populated) and this could
be the source of a positive coupling between the isomer shift and quadrupole splitting
around 650 K. Alternatively, nearby defects can affect both the isomer shift and quadrupole
splitting, giving an alternative explanation of these features.
The calculated hyperfine parameters for Fe on the 8b site in In2O3(Table 1) are in reason-
able agreement with the experimental hyperfine parameters (Fig. 4), as extrapolated from
the crystalline phase. By comparing the experimental data with the DFT calculations, we
can conclude that the fraction of Fe atoms at the 8b site is larger than at the 24d site.
It is noteworthy to add that a spectral component related to interstitial Fe, which could
be expected because of the ER≈40 eV recoil imparted on the 57Fe daughter in the
β−decay of 57 Mn, is not observed. A possible explanation is that the threshold displace-
ment energy is too high (∼90 eV), but this seems to be at variance with the results of
Walsh et al. [24] which based on theoretical calculations predicted threshold displacement
energy of 14.2 eV. Alternatively, the Debye-Waller factor of interstitial Fe is too low to
allow for the detection of interstitial Fe or that interstitial Fe is incorporated on In sites much
faster than the lifetime of the M¨
ossbauer state (140 ns).
The hyperfine parameters obtained here differ significantly from the parameters reported
by Yan et al. [3] who used sol-gel In2O3material with a ∼15 % Fe doping, showing two
Fe3+doublets assigned to Fe on 8b sites and 24d sites. After annealing in vacuum, the Fe3+
doublet due to Fe on 24d sites disappeared, and a magnetic sextet due to Fe2+appeared
in the spectrum assigned to Fe2+on 24d sites stabilized by the introduction of oxygen
vacancies in the annealing process [3].
It is difficult to see any correlation between our results and the results of [3], and the
different concentration of Fe and/or sample material must be the explanation for the very
different findings. Yang et al. [4] found both Fe2+and Fe3+in <16 % Fe doped In2O3
samples prepared by magnetron sputtering, suggesting a transformation from the dilute case
where Fe2+dominates to a Fe3+dominating state, depending on the Fe concentration. This
seems though to be at variance with the results of Nomura et al. which saw only doublets
due to Fe3+in sol-gel indium tin oxide material with 6 % Fe [25].
It should be noted, that the results of the calculations presented here differ from those of
Ref. [26]. One possible reason is that in the calculations of Ref. [26] one Fe per half unit
cell is considered, while in the present calculations one Fe per unit cell is considered. The
splitting observed in our calculations, is closer to predictions of crystal field theory of the
local symmetry (twofold splitting in 8b symmetry and threefold splitting in 24d symmetry),
giving us confidence to the current calculations.

75 Page 8 of 9 Hyperfine Interact (2016) 237:75
7 Conclusions
57Fe Emission M ¨
ossbauer spectroscopy following dilute implantation of 57Mn shows an
annealing stage at 400–500 K. Above this annealing stage, the Fe probe atoms are incorpo-
rated on regular substitutional In lattice sites. The experimental data show that Fe2+replaces
In3+. With the spin of Fe parallel with the spin of the hole state on oxygen orbital which is
due to strong hybridization between the 3d orbitals of Fe and the 2p orbitals of O, the den-
sity functional theory calculations suggest Fe2+state in In 2O3. Furthermore, the calculated
hyperfine parameters are in reasonable agreement with Fe2+predominantly incorporated
on the 8b site of the Byxibite In2O3structure.
Acknowledgments This work was supported by the European Union Seventh Framework through
ENSAR (Contract No. 262010). R. Mantovan acknowledges support from MIUR through the FIRB Project
RBAP115AYN “Oxides at the nanoscale: multifunctionality and applications.” H. Masenda, D. Naidoo, and
M. Ncube acknowledge support from the South African National Research Foundation and the Department
of Science and Technology. T. E. Mølholt, H. P. Gislason, and S. ´
Olafsson acknowledge support from the
Icelandic Research Fund (Grant No. 110017021-23). Unzueta acknowledge financial support from Basque
Government Grants nos. IT-443-10 and PRE 2014 214.
References
1. Dietl, T., Ohno, H., Matsukura, F., Cibert, J., Ferrand, D.: Zener model description of ferromag-
netism in Zinc-Blende magnetic semiconductors. Science 287, 1019–1022 (2000). doi:10.1126/sci
ence.287.5455.1019
2. He, J., Xu, S., Yoo, Y.K., Xue, Q., Lee, H.-C., Cheng, S., Xiang, X.D., Dionne, G.F., Takeuchi, I.: Room
temperature ferromagnetic n -type semiconductor in (In1−xFex)2O3−σ. Appl. Phys. Lett. 86, 052503
(2005). doi:10.1063/1.1851618
3. Yan, S., Qiao, W., Zhong, W., Au, C.-T., Dou, Y.: Effects of site occupancy and valence state of Fe ions
on ferromagnetism in Fe-doped In2O3diluted magnetic semiconductor. Appl. Phys. Lett. 104, 062404
(2014). doi:10.1063/1.4865102
4. Yang, D., Feng, D., Wu, Z., Ma, G., Liu, J., An, Y.: Structure, optical and magnetic proper-
ties of (In1−xFex)2O3films by magnetron sputtering. J. Alloy. Compd. 619, 869–875 (2015).
doi:10.1016/j.jallcom.2014.09.094
5. Bierwagen, O.: Indium oxide–a transparent, wide-band gap semiconductor for (opto)electronic applica-
tions. Semicond. Sci. Technol. 30, 024001 (2015). doi:10.1088/0268-1242/30/2/024001
6. Hu, S.-j., Yan, S.-s., Lin, X.-l., Yao, X.-x., Chen, Y.-x., Liu, G.-l., Mei, L.-m.: Electronic structure of Fe-
doped In2O3magnetic semiconductor with oxygen vacancies: Evidence for F-center mediated exchange
interaction. Appl. Phys. Lett. 91, 262514 (2007). doi:10.1063/1.2828041
7. Huang, L.M., Moys´
es, C., Ara´
ujo, Ahuja, R.: Magnetic and electronic properties of 3d transition-metal-
doped In2O3: an ab initio study. Europhys. Lett. 87, 27013 (2009). doi:10.1209/0295-5075/87/27013
8. Qi, S., Jiang, F., Fan, J., Wu, H., Zhang, S.B., Gehring, G.A., Zhang, Z., Xu, X.: Carrier-mediated
nonlocal ferromagnetic coupling between local magnetic polarons in Fe-doped In2O3and Co-doped
ZnO. Phys. Rev. B 205204, 84 (2011). doi:10.1103/PhysRevB.84.205204
9. Kohiki, S., Sasaki, M., Murakawa, Y., Hori, K., Okada, K., Shimooka, H., Tajiri, T., Deguchi, H.,
Matsushima, S., Oku, M., Shishido, T., Arai, M., Mitome, M., Bando, Y.: Doping of Fe to In2O3.Thin
Solid Films 505, 122–125 (2006). doi:10.1016/j.tsf.2005.10.022
10. Xu, X.-H., Jiang, F.-X., Zhang, J., Fan, X.-C., Wu, H.-S., Gehring, G.A.: Magnetic and transport
properties of n-type Fe-doped In2O3ferromagnetic thin films. Appl. Phys. Lett. 212510, 94 (2009).
doi:10.1063/1.3147190
11. Jiang, F.-X., Xu, X.-H., Zhang, J., Fan, X.-C., Wu, H.-S., Alshammari, M., Feng, Q., Blythe, H.J., Score,
D.S., Addison, K., Al-Qahtani, M., Gehring, G.A.: Room temperature ferromagnetism in metallic and
insulating (In1−xFex)2O3thin films. J. Appl. Phys. 109, 053907 (2011). doi:10.1063/1.3559298
12. Tandon, B., Shanker, G.S., Nag, A.: Multifunctional Sn-, Fe-codoped In2O3nanocrystals plasmonics
and magnetism. J. Phys. Chems. Lett. 5, 2306–2311 (2014). doi:10.1021/jz500949g

Hyperfine Interact (2016) 237:75 Page 9 of 9 75
13. Blaha, P., Schwarz, K., Madsen, G.K.H., Kvasnicka, D., Luitz, J.: WIEN2k, an augmented plane wave
+local orbitals program for calculating crystal properties, ISBN 3-9501031-1-2, Karlheinz Schwarz,
Techn. Universit¨
at Wien, Austria (2001)
14. Perdew, J.P., Burke, K., Ernzerhof, M.: Generalized gradient approximation made simple. Physical
Review Letters 77, 3865 (1996). 10.1103/PhysRevLett.77.3865
15. Perdew, J.P., Chevary, J.A., Vosko, S.H., Pederson, M.R., Singh, D.J., Fiolhais C.: Atoms, molecules,
solids, and surfaces: Applications of the generalized gradient approximation for exchange and correla-
tion. Phys. Rev. B 46, 6671 (1992). doi:10.1103/PhysRevB.46.6671
16. Neese, F.: Prediction and interpretation of the 57Fe isomer shift in M¨
ossbauer spectra by density
functional theory. Inorg. Chim. Acta 337, 181–192 (2002). doi:10.1016/S0020-1693(02)01031-9
17. Wdowik, U.D., Ruebenbauer, K.: Calibration of the isomer shift for the 14.4 keV transition in 57Fe
using the full-potential linearized augmented plane-wave method. Phys. Rev. B 76, 155118 (2007).
doi:10.1103/PhysRevB.76.155118
18. Dufek, P., Blaha, P., Schwarz, K.: Determination of the nuclear quadrupole moment of 57Fe. Phys. Rev.
Lett. 75, 3545–3548 (1995). doi:10.1103/PhysRevLett.75.3545
19. Gunnlaugsson, H.P.: Spreadsheet based analysis of M¨
ossbauer spectra hyperfine interact. accepted
(2015)
20. Gunnlaugsson, H.P.: A simple model to extract hyperfine interaction distributions from M¨
ossbauer
spectra. Hyperfine Interact 167, 851–854 (2006). doi:10.1007/978-3-540-49853-724
21. Gunnlaugsson, H.P., Mantovan, R., Mølholt, T.E., Naidoo, D., Johnston, K., Masenda, H., Bharuth-Ram,
K., Langouche, G., ´
Olafsson, S., Sielemann, R., Weyer, G., Kobayashi, Y.: the ISOLDE Collaboration:
M¨
ossbauer spectroscopy of 57Fe in α-Al2O3following implantation of 57Mn∗. Hyperfine Interact 198,
5–14 (2010). doi:10.1007/s10751-010-0184-5
22. Mølholt, T.E., Mantovan, R., Gunnlaugsson, H.P., Naidoo, D., ´
Olafsson, S., Bharuth-Ram, K., Fanciulli,
M., Johnston, K., Kobayashi, Y., Langouche, G., Masenda, H., Sielemann, R., Weyer, G., G´
ıslason, H.P.:
Observation of spin-lattice relaxations of dilute Fe3+in MgO by M¨
ossbauer spectroscopy. Hyperfine
Interact 197, 89–94 (2010). doi:10.1007/s10751-010-0214-3
23. Mantovan, R., Gunnlaugsson, H.P., Johnston, K., Masenda, H., Mølholt, T.E., Naidoo, D., Ncube, M.,
Shayestehaminzadeh, S., Bharuth-Ram, K., Fanciulli, M., Gislason, H.P., Langouche, G., ´
Olafsson, S.,
Pereira, L.M.C., Wahl, U., Torelli, P., Weyer, G.: Atomic-scale magnetic properties of truly 3d-diluted
ZnO. Adv. Electron. Mater. 1, 1400039 (2015). doi:10.1002/aelm.201400039
24. Walsh, A., Woodley, S.M., Richard, C., Catlow, A., Sokol, A.A.: Potential energy landscapes
for anion Frenkel-pair formation in ceria and India. Solid State Ionics 184, 52–56 (2011).
doi:10.1016/j.ssi.2010.08.010
25. Nomura, K., Sakuma, J., Ooki, T., Takeda, M.: M ¨
ossbauer study on indium tin oxides (ITO) doped with
Fe. Hyperfine Interact 184, 117–121 (2008). doi:10.1007/s10751-008-9775-9
26. Sena, C., Costa, M.S., Mu ˜
noz, E.L., Cabrera-Pasca, G.A., Pereira, L.F.D., Mestnik-Filho, J., Carbonari,
A.W., Coaquira, J.A.H.: Charge distribution and hyperfine interactions in the vicinity of impu-
rity sites in In2O3doped with Fe, Co, and Ni. J. Magn. Magn. Mat. 387, 165–178 (2015).
doi:10.1016/j.jmmm.2015.03.092