Iron oxide thin films MBE growth in low oxygen pressure and electrical and magnetic properties

Author’s Accepted Manuscript
Iron oxide thin Films: MBE growth in low oxygen
pressure and electrical and magnetic properties
Dang Duc Dung, Wuwei Feng, Duong Van Thiet,
Yunki Kim, Sunglae Cho
www.elsevier.com

PII:
DOI:
Reference:

S0167-577X(15)30477-8
/>MLBLUE19483

To appear in: Materials Letters
Received date: 13 June 2015
Revised date: 10 August 2015
Accepted date: 25 August 2015
Cite this article as: Dang Duc Dung, Wuwei Feng, Duong Van Thiet, Yunki Kim
and Sunglae Cho, Iron oxide thin Films: MBE growth in low oxygen pressure
and
electrical
and
magnetic
properties, Materials
Letters,
/>This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which

could affect the content, and all legal disclaimers that apply to the journal pertain.


Iron Oxide Thin Films: MBE Growth in Low Oxygen Pressure and Electrical and
Magnetic Properties

Dang Duc Dung1,2, Wuwei Feng3, Duong Van Thiet1,2, Yunki Kim4, and Sunglae Cho1*
1
2

3

Department of Physics, University of Ulsan, Ulsan 680-749, Republic of Korea

Department of General Physics, School of Engineering Physics, Ha Noi University of Science
and Technology, 1 Dai Co Viet road, Ha Noi, Viet Nam

School of Materials Science and Engineering, China University of Geosciences, Beijing 100083,
China
4

Department of Electrical and Biological Physics, Kwangwoon University, Seoul 139-701,
Republic of Korea

Abstract
Epitaxial iron oxide films were grown on MgO(001) substrate by using molecular beam epitaxy.
The growth modes, magnetism and transport properties of the films were strongly dependent on
the oxygen pressure during film growth. Fe3O4 film was grown in oxygen poor environment
(2.3x10-7 Torr), while -Fe2O3 film was grown in 8.2x10-6 Torr oxygen. The average roughness
decreases from 1.02 to 0.26 nm with oxygen pressure during growth.

Keywords: Fe3O4; thin film; magnetism; magnetotransport

*) Corresponding author e-mail:
I. Introduction
Iron oxides usually exist in three kinds of oxidative states (Fe2+, Fe3+, and Fe4+) in wustite FeO,
magnetite Fe3O4, hematite -Fe2O3, and maghemite -Fe2O3 compounds. -Fe2O3 has a cubic
1


spinel structure (a = 8.35 Å) and is known to be ferrimagnetic. It is a metastable phase at ambient
conditions and transforms to a stable antiferromagnetic -Fe2O3 above 400 °C. -Fe2O3 has a
wide range of applications such as gas sensor, electrochromic and photocatalytic applications.
Magnetite Fe3O4 has ferrrimagnetic spinel structure. It is predicted to possess as half-metallic
nature, P  100% spin polarization, and has a high Curie temperature (TC ~ 850 K) [1, 2].
Experiments demonstrated that the P ~ (80 ± 5) %, ~ (60 ± 5) %, and ~40-55% for epitaxial
(111), (110) and (001)-oriented Fe3O4 thin films, respectively [3-5]. The state of Fe valence has
been controlled by substrate, growth rate and oxygen partial pressure, etc.
Here we report that the growth modes, magnetism and transport properties of iron oxide films on
MgO(001) were strongly dependent on the oxygen partial pressure during film growth. Fe3O4
film was grown in oxygen poor environment (2.3x10-7 Torr), while -Fe2O3 film was grown in
8.2x10-6 Torr oxygen. The average roughness of iron oxide film decreases from 1.021 to 0.263
nm for the oxygen pressure environment increase from 2.3x10-7 to 8.2x10-6 Torr.
II. Experiment
The 800 Å iron oxide thin films have been grown on MgO(100) substrate using
molecular beam epitaxy (VG Semicon. Inc.). The growth temperature and evaporation rate of Fe
was fixed at 400C and 0.2 Å/s, respectively, while the oxygen pressure was changed; 2.3x10-7,
2.2x10-6, and 8.2x10-6 Torr. For Fe evaporation, we used high temperature effusion cell and, for
oxygen, we used the thermal oxygen cracking source. The growth quality of films was monitored
by in-situ reflection high energy electron diffraction (RHEED). The surface morphology and
roughness were characterized by field emission scan electron microscopy (FE-SEM) and atomic

force microscopy (AFM), respectively. The crystal structure of the samples was characterized by
2


X-ray diffraction (XRD) studies. The valence states of Fe were characterized using X-ray
photoelectron spectroscopy (XPS). Transport measurements were carried out with a physical
property measurement system (PPMS). Magnetic properties were characterized by a
superconducting quantum interference device (SQUID) magnetometer (Quantum Design Inc.).
III. Result and discussions
Figure 1 (a)-(c) shows the RHEED patterns of iron oxide thin films grown under oxygen
pressure of 2.3x10-7, 2.2x10-6 and 8.2x10-6 Torr, respectively. The RHEED patterns indicated the
epitaxial growth of iron oxide thin films on the MgO(001) substrate. However, the growth modes
were strongly dependent on the oxygen pressure during growth. The spotty patterns were
observed in the sample grown at low oxygen pressure, 2.3x10-7 Torr, as shown in the Fig. 1(a),
indicating that the film was grown under the 3D island growth mode. The streaky-line patterns
were observed in the samples grown at high oxygen pressures, 2.2x10-6 and 8.2x10-6 Torr, as
shown in the Fig. 1(b) and (c), indicating that the films were grown under the layer-layer growth
mode. The growth mode is strongly affected by surface and interface during growth [6]. Figure 1
(c) -(i) show an SEI (secondary electron imaging) and BEI-COMPO (composition mode in
backscattered electron imaging) images of the samples grown at different oxygen pressures,
compared with the RHEED data. The SEI and BEI-COMPO images of the sample grown at the
oxygen pressure of 2.3x10-7 Torr showed the rough surface, as shown in Fig. 1 (d) and (g),
whose average island sizes were approximately 30-40 nm. However, the homogeneous surface
was obtained when sample was grown at higher oxygen pressure, as shown in SEI of Fig. 1(e)
and (f) and BEI-COMPO of Fig. 1(h) and (i). The roughness of the samples was increased with
the oxygen pressure, as shown in AFM images (Fig. 1 (k)-(m)). The averages roughness was

3

1.021, 0.309 and 0.263 nm for oxygen pressure of 2.3x10-7, 2.2x10-6 and 8.2x10-6 Torr,
respectively.
The XRD patterns of iron oxide thin films grown at different oxygen pressures were shown in
Fig. 2 (a), which was compared to the pure Fe pattern. The sample grown under low oxygen
pressure, 2.3x10-7 Torr, has several phases, Fe, FeO, and Fe3O4. However, in oxygen rich
environment, 8.2x10-6 Torr, a single phase, -Fe2O3 was observed. The peaks positions of iron
oxide were magnified in range of 41-45, as shown in Fig. 2 (b). Since the lattice parameter of
Fe3O4 (0.839 nm) and -Fe2O3 (0.834 nm) are very close to each other, therefore, using -2
XRD pattern it is hard to distinguish these two phases of iron oxide. Furthermore, in order to
clarify the chemical state of the iron oxide, the Fe2p core level XPS spectra were recorded for
iron oxide films grown under oxygen pressures of 2.2x10-6 and 8.2x10-6 Torr, as shown in Fig. 2
(c) and (d), respectively. The observed binding energies centered at about 711 eV and 725 eV in
the XPS spectra correspond to the Fe 2p3/2 and Fe 2p1/2, respectively, as shown in Fig. 2 (c). The
values match very well to the literature values, indicating that the Fe3O4 phases were existed [7].
However, the satellite peaks were clearly obtained around 720 eV for growth samples at high
oxygen pressure of 8.2x10-6 Torr, as marked in Fig. 2 (d), indicating that fully oxidation state of
iron were obtained as -Fe2O3 phase. Note that FeO has a clear satellite feature at 715.5 eV,
while -Fe2O3 shows a satellite feature at 719.1 eV [7-10]. Such satellite structures are frequently
used as fingerprints to identify the other iron oxide phases.
Figure 3 (a) shows the temperature (T) dependent electrical resistivity (). The inset of Fig. 3 (a)
shows the temperature dependent electrical resistivity of pure Fe on MgO(001) substrates,
indicating that samples exhibited the metallic properties. The TV, metal-insulator transition in
4


Fe3O4, was observed about 115 and 108 K for the samples grown at oxygen pressures of 2.3x10-7
and 2.2x10-6 Torr, respectively. However, the resistivity monotonically increased with
decreasing temperature for the sample grown at the oxygen pressures of 8.2x10-6 Torr. Herein,
we note that the -Fe2O3 is semiconductor with band gap round 2.0-2.2 eV [11]. It has been
reported that the Verwey transition of 120 K in bulk Fe3O4 is strongly affected by many

parameters such as stoichiometry and stress, etc. [11, 12]. TV linearly decreased from 124 K for
δ = 0 to 81 K for δ = 0.0121 in Fe3-δO4 single crystals [13]. The TV decreased linearly with
increasing applied pressure [14, 15]. Jain et al. reported that TV depends on buffer layers such as
Ta, Ti, SiO2, and Fe2O3, etc, which is maybe due to the impurities at the interface between the
buffer layer and the Fe3O4 films [16, 17]. Furthermore, Fig. 3 (b)-(d) show the logarithmically
resistivity as the function of T-1/2 for the samples grown under oxygen pressure of 2.3x10-7,
2.2x10-6 and 8.2x10-6 Torr, respectively. The linear fitting in the temperature range between 400
and 160 K for the sample grown under oxygen pressure of 8.2x10-6 Torr follows the power 1/2
law, indicating hopping conductivity in granular metals suggested by Sheng et al., which
described the tunneling barrier [18]. Jang et al. reported that in thermally assisted intraparticle
tunneling between single nanoparticles, no Verwey transition is observed [19]. In addition, an inplane magnetoresistance (MR) was measured as a function of magnetic field in order to further
study the magnetic ordering of the films. Figure 3 (e)-(g) shows magnetoresistance (MR) at
selected temperatures for the samples grown under oxygen pressure of 2.3x10-7, 2.2x10-6 and
8.2x10-6 Torr, respectively. Negative MR was observed for all samples when the magnetic field
was perpendicular to the applied current. This is due to the decreased scattering centers caused
by the increase in the magnetic domain with the increased magnetic field [20, 21], which is also
evidence of ferrimagnetic ordering in the iron oxide films. The highest MR ratio, 2.3% at 170 K,
5


was obtained for sample grown at high oxygen pressure. In addition, the unsaturated behavior of
the field-dependent parameters is suggested to attribute to the antiphase boundaries in epitaxial
iron oxide thin films. Since the electron transport across the antiphase boundaries is understood
by electron hopping model [22]. The evidence in transport properties which consisted of the -T
curve where resistivity  as proposal with temperature with power of 1/2, as show in Fig. 3 (c)(d). Therefore, we suggest that the observation of magnetoresistance which resulted from spin
depend scattering between the antiphase boundaries. Furthermore, the high magnetoresistance
ratios were strong affected to the anomalous Hall signal. Figure 3 (h)-(k) shows the Hall
resistance as the function of applied magnetic field at selected temperatures. However, we could
not obtain the AHE at temperature below TV temperature due to asymmetric Hall probe
configuration, i. e. the strong negative magnetoresistance affect to unsymmetrical anomalous

Hall effect curves, as shown in Fig. 3 (e)-(g).
Figure 4 (a) shows the zero-field-cooled and field-cooled magnetizations as a function of
temperature for the samples grown under different oxygen pressures. The TVs were estimated
around 115 and 108 K for sample grown under oxygen of 2.3x10-7 and 2.2x10-6 Torr,
respectively. However, for the sample grown under 8.2x10-6 Torr oxygen pressure, TV was
disappeared because -Fe2O3 is ferromagnetic without metal-insulator transition around 120 K.
We suggested that a shift in Verway transition temperature as functions of selected oxygen
pressure resulted from oxygen nonstoichiometry in Fe3O4 and it disappeared because -Fe2O3
phase was more stable than that of Fe3O4 phase under high growth oxygen pressure. The results
were consisted with Tvs values which were obtained from transport measurement. The net
magnetic moments become smaller with increasing the oxygen pressure during growth, as shown
6


in Fig. 4 (b)-(d). In addition, the sign investigated of magnetic moment were obtained at high
temperature for sample grown under high oxygen. The iron oxide samples grown on MgO(001)
substrates that contributed a strong diamagnetic background to the signal measured in a
conventional magnetometer [23]. The background magnetic properties of some commercial
substrates were well investigated by Wu et al. [24]. The reduction of magnetic moment was
suggested result from oxygen nonstoichiometry in Fe3O4 and/or from antiferromagnetic and
frustrated exchange interaction across the antiphase boundaries [25, 26]. The other possibilities
are from present of -Fe2O3 than that Fe3O4 phase under high oxygen pressure during growth
where saturation magnetization of -Fe2O3 is smaller than that of Fe3O4 [27]. Moreover, Fe3O4
has a cubic spinel structure in which tetrahedral A sites contain one-third of the Fe ions as Fe3+,
while octahedral B sites contain the remaining Fe ions, with equal numbers of Fe2+ and Fe3+ in
B1 and B2 sites, respectively. Below TC, magnetite is ferromagnetic with A-site magnetic
moments aligned antiparallel to the B-site moments. Indeed, most of the exchange interactions
are Fe-O-Fe superexchange ones.
IV. Conclusion
The epitaxial iron oxide thin films were fabricated on MgO(001) by using MBE. The -Fe2O3

thin film with homogeneous and flat surface was grown under high oxygen pressure during
growth, while Fe3O4 film was grown in oxygen poor environment (2.3x10-7 Torr). The Verwey
transition of 110 K lower than 120 K in bulk might be due to oxygen deficiency in Fe3O4 film.
Acknowledgments

7


This work was supported by a grant from Energy Efficiency & Resources program of the Korea
Institute of Energy Technology Evaluation and Planning (KETEP) funded by the Korean
Ministry of Knowledge Economy (20132020000110).

8


REFERENCES
1) Z. Zhang, and S. Satpathy, Phys. Rev. B 44 (1991) 13319.
2) V. I. Anisimov, I. S. Elfimov, N. Hamada, and K. Terakura, Phys. Rev. B 54 (1996) 4387.
3) Y. S. Dedkov, U. Rudiger, and G. Guntherodt, Phys. Rev. B 65 (2002) 064417.
4) M. Fonin, Y. S. Dedkov, J. Mayer, U. Rudiger, and G. Guntherodt, Phys. Rev. B 68 (2003)
45414.
5) M. Fonin, R. Pentcheva, Y. S. Dedkov, M. Sperlich, D. V. Vyalikh, M. Scheffler, U. Rudiger,
and G. Guntherodt, Phys. Rev. B 72 (2005) 104436.
6) X. H. Liu, A. D. Rata, C. F. Chang, A. C. Komarek, and L. H. Tjeng, Phys. Rev. B 90 (2014)
125142.
7) A. Barbieri, W. Weiss, M. A. Van Hove, and G.A.Somorjai, Surf. Sci. 302 (1994) 259.
8) J. B. Moussy, J. Phys. D: Appl. Phys. 46 (2013) 143001.
9) T. Fujii, F. M. F. de Groot, G. A. Sawatzky, F. C. Voogt, T. Hibma, and K. Okada, Phys. Rev.
B 59 (1999) 3195.
10) C. Karunakaran, and S. Senthilvelan, Electrochem. Commun. 8 (2006) 95.

11) E. J. W. Verwey, and P. W. Haayman, Physica 8 (1941) 979.
12) A. R. Muxworthy, and E. McClelland, Geophys. J. Int. 140 (2000) 101.
13) J. P. Shepherd, R. Argon, J. W. Koenitzer, and J. M. Honig, Phys. Rev. B 32 (1985) 1818.

9


14) S. K. Ramasesha, M. Mohan, A. K. Singh, J. M. Honig, and C. N. R. Rao, Phys. Rev. B 50
(1994) 13789.
15) G. A. Samara, Phys. Rev. Lett. 21 (1968) 795.
16) S. Klotz, G. S. Neumann, T. Strassle, J. Philippe, T. Hansen, and M. J. Wenzel, Phys. Rev. B
77 (2008) 012411.
17) S. Jain, and A. O. Adeyeye, J. Appl. Phys. 97 (2005) 093713.
18) P. Sheng, B. Abeles, and Y. Arie, Phys. Rev. Lett. 31 (1973) 44.
19) S. Jang, W. Kong, and H. Zeng, Phys. Rev. B 76 (2007) 212403.
20) C. L. Chien and C. R. Westgate, Hall Effect and Its Applications (Plenum, New York, 1980).
21) Y. M. Kang, H. J. Kim, and S. I. Yoo, J. Magnetics 17 (2012) 265.
22) W. Eerenstein, T.T.M. Plastra, S.S. Saxena, T. Hibma, Phys. Rev. Lett. 88 (2002) 247204.
23) Y. Zhu, Modern Techniques for Characterizing Magnetic Materials (Springer Science &
Business Media, 2005).
24) X. Wu, Z. Zhang, Y. Yu, and J. Meng, J. Magn. Magn. Mater. 346 (2013) 87.
25) J. P. Shepherd, R. Argon, J. W. Koenitzer, and J. M. Honig, Phys. Rev. B 32, 1818 (1985).
26) Y. Zhou, X. Jin, and I. V. Shvets, J. Appl. Phys. 95, 7357 (2004).
27) H. M. Lu, W. T. Zheng, and Q. Jiang, J. Phys. D: Appl. Phys. 40, 320 (2007)

10


Figure captions
Fig. 1. (color online) The RHEED patterns of iron oxide thin films grown at oxygen pressures of

(b) 2.3x10-7 Torr, (c) 2.2x10-6 Torr, and (d) 8.2x10-6 Torr. (d)-(f) An SEI images and (g)-(i) a
BEI-COMPO images of the samples grown at oxygen pressure of 2.3x10-7 Torr, 2.2x10-6 Torr,
and 8.2x10-6 Torr, respectively. (k)-(m) AFM images of the samples grown at oxygen pressures
of 2.3x10-7, 2.2x10-6, and

8.2x10-6 Torr, respectively.

Fig. 2. (color online) (a) The -2 X-ray diffraction patterns of the samples grown at selected
oxygen pressures, and (b) the magnified XRD patterns in the 2θ ranges of 41 – 45. The XPS
Fe2p core-level spectra for iron oxide thin films grown at the oxygen pressures of (c) 2.2x10-6
Torr and (d) 8.2x10-6 Torr.
Fig. 3. (color online) (a) Temperature-dependent electrical resistivity and ln() versus 1/T2 for
the samples grown at oxygen pressures of (b) 2.3x10-7, (c) 2.2x10-6, and (d) 8.2x10-6 Torr. The
inset of Fig. 3 (a) is the temperature-dependent electrical resistivity for Fe/MgO(001) samples.
(e)-(g) Magnetoresistance and (h)-(k) the magnetic field dependent Hall resistances at selected
temperatures for the samples grown under oxygen pressure of 2.3x10-7 Torr, 2.2x10-6 Torr and
8.2x10-6 Torr, respectively.
Fig. 4. (color online) (a) The magnetization as a function of temperature under the applied
magnetic field of 500 Oe for the as-grown samples at selected oxygen pressure. The magnetic
hysteresis curves at selected temperatures for the samples grown under oxygen pressures of (b)
2.3x10-7 Torr, (c) 2.2x10-6 Torr, and (d) 8.2x10-6 Torr.

Highlights
11


- Epitaxial iron oxide films were grown on MgO(001) substrate by using
MBE.
- Fe3O4 film was grown in oxygen poor environment (2.3x10-7 Torr).
- -Fe2O3 film was grown in 8.2x10-6 Torr oxygen.

- The Verwey transition of Fe3O4 film 110 K lower than 120 K in bulk.

fig.1

12


fig.2

13


fig.3

14


fig.4

15


Graphical abstract

16