scispace - formally typeset
Open AccessJournal ArticleDOI

β phase and γ-β metal-insulator transition in multiferroic BiFeO3

TLDR
In this paper, the authors report on extensive experimental studies on thin film, single crystal, and ceramics of multiferroic bismuth ferrite BiFeO3 using differential thermal analysis, high-temperature polarized light microscopy, hightemperature x-ray diffraction, dc conductivity, optical absorption and reflectivity, and domain imaging.
Abstract
We report on extensive experimental studies on thin film, single crystal, and ceramics of multiferroic bismuth ferrite BiFeO3 using differential thermal analysis, high-temperature polarized light microscopy, high-temperature and polarized Raman spectroscopy, high-temperature x-ray diffraction, dc conductivity, optical absorption and reflectivity, and domain imaging, and show that epitaxial (001) thin films of BiFeO3 are clearly monoclinic at room temperature, in agreement with recent synchrotron studies but in disagreement with all other earlier reported results. We report an orthorhombic order-disorder beta phase between 820 and 925 (±5) °C, and establish the existence range of the cubic gamma phase between 925 (±5) and 933 (±5) °C, contrary to all recent reports. We also report the refined Bi2O3-Fe2O3 phase diagram. The phase transition sequence rhombohedral-orthorhombic-cubic in bulk [monoclinic-orthorhombic-cubic in (001)BiFeO3 thin film] differs distinctly from that of BaTiO3. The transition to the cubic gamma phase causes an abrupt collapse of the band gap toward zero (insulator-metal transition) at the orthorhombic-cubic beta-gamma transition around 930 °C. Our band structure models, high-temperature dc resistivity, and light absorption and reflectivity measurements are consistent with this metal-insulator transition.

read more

Content maybe subject to copyright    Report

Durham Research Online
Deposited in DRO:
27 November 2012
Version of attached le:
Published Version
Peer-review status of attached le:
Peer-reviewed
Citation for published item:
Palai, R. and Katiyar, R.S. and Schmid, H. and Tissot, P. and Clark, S.J. and Robertson, J. and Redfern,
S.A.T. and Catalan, G. and Scott, J.F. (2008) ' phase and - metal-insulator transition in multiferroic
BiFeO3.', Physical review B., 77 (1). 014110.
Further information on publisher's website:
http://dx.doi.org/10.1103/PhysRevB.77.014110
Publisher's copyright statement:
c
2008 The American Physical Society
Additional information:
Use policy
The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for
personal research or study, educational, or not-for-prot purposes provided that:
a full bibliographic reference is made to the original source
a link is made to the metadata record in DRO
the full-text is not changed in any way
The full-text must not be sold in any format or medium without the formal permission of the copyright holders.
Please consult the full DRO policy for further details.
Durham University Library, Stockton Road, Durham DH1 3LY, United Kingdom
Tel : +44 (0)191 334 3042 | Fax : +44 (0)191 334 2971
https://dro.dur.ac.uk

phase and
-
metal-insulator transition in multiferroic BiFeO
3
R. Palai,
1
R. S. Katiyar,
1
H. Schmid,
2
P. Tissot,
2
S. J. Clark,
3
J. Robertson,
4
S. A. T. Redfern,
5
G. Catalan,
5
and J. F. Scott
5
1
Department of Physics and Institute for Functional Nanomaterials, University of Puerto Rico,
San Juan, Puerto Rico 00931-3343, USA
2
Department of Inorganic, Analytical and Applied Chemistry, University of Geneva, CH-1211 Geneva 4, Switzerland
3
Department of Physics, Durham University, Durham DH1 3LE, United Kingdom
4
Department of Engineering, University of Cambridge, Cambridge CB2 1PZ, United Kingdom
5
Department of Earth Science, University of Cambridge, Cambridge CB2 3EQ, United Kingdom
Received 17 August 2007; revised manuscript received 17 October 2007; published 28 January 2008
We report on extensive experimental studies on thin film, single crystal, and ceramics of multiferroic
bismuth ferrite BiFeO
3
using differential thermal analysis, high-temperature polarized light microscopy, high-
temperature and polarized Raman spectroscopy, high-temperature x-ray diffraction, dc conductivity, optical
absorption and reflectivity, and domain imaging, and show that epitaxial 001 thin films of BiFeO
3
are clearly
monoclinic at room temperature, in agreement with recent synchrotron studies but in disagreement with all
other earlier reported results. We report an orthorhombic order-disorder
phase between 820 and 925
5 °C, and establish the existence range of the cubic
phase between 925 5 and 933 5 °C, contrary
to all recent reports. We also report the refined Bi
2
O
3
-Fe
2
O
3
phase diagram. The phase transition sequence
rhombohedral-orthorhombic-cubic in bulk monoclinic-orthorhombic-cubic in 001BiFeO
3
thin film differs
distinctly from that of BaTiO
3
. The transition to the cubic
phase causes an abrupt collapse of the band gap
toward zero insulator-metal transition at the orthorhombic-cubic
-
transition around 930 °C. Our band
structure models, high-temperature dc resistivity, and light absorption and reflectivity measurements are con-
sistent with this metal-insulator transition.
DOI: 10.1103/PhysRevB.77.014110 PACS numbers: 64.70.K, 71.30.h, 78.30.j, 77.55.f
I. INTRODUCTION
Magnetoelectric ME multiferroics
1
are technologically
and scientifically promising because of their potential appli-
cations in data storage, spin valves, spintronics, quantum
electromagnets, microelectronic devices, etc.
24
Ferroelec-
tricity originates from off-center structural distortions d
0
electrons and magnetism is involved with local spins d
n
electrons, which limit the presence of off-center structural
distortion.
5
These two are quite complementary phenomena,
but coexist in certain unusual multiferroic materials. BiFeO
3
BFO is one of the most widely studied multiferroic mate-
rial because of its interesting ME properties, i.e., ferroelec-
tricity with high Curie temperature
6
T
C
810830 °C and
antiferromagnetic properties below T
N
370 °C.
6,7
The bulk
BFO single crystal shows rhombohedral a =5.58 Å and
=89.5 ° crystal structure at room temperature RT with the
space group R3c and G-type antiferromagnetism.
7,8
If BFO
were an ordinary antiferromagnet, the magnetic space group
would be expected to be R3c Heesch-Shubnikov point
group 3m, however, the antiferromagnetic structure is in-
commensurate IC with a cycloidal modulation.
9
Whereas
this modulation suppresses the linear ME effect, the bilinear
ME effect survives, simulating a paramagnetic point group
3m1
for the IC phase.
10
The structure and properties of bulk
BFO have been studied extensively.
6 8,11,12
Early values of polarization were very low due to sample
quality, but, recently, P
r
=40–50
C/ cm
2
was found in bulk
by several different groups.
13,14
However, recently even,
about P
r
=100
C/ cm
2
was measured on single crystals,
matching the high values of spontaneous polarization
achieved in thin films see below.
15
Thus, these well insu-
lating single crystals, although still relatively small, yielding
a high, filmlike spontaneous polarization, represent a certain
progress relative to the fernlike dendritic single crystals
16
Fig. 1 and to the dendritic polar single domain crystals with
100-pseudo-cubic habitus.
8,17
Thus, the growth of large,
high quality, and defect-free single crystals with low leakage
still remains a challenge.
It has been found that thin films of BFO grown on SrTiO
3
STO substrates show very high values of P
r
关⬃55, 86, and
98
C/ cm
2
for the 001, 101, and 111 BFO films,
respectively and magnetization M
r
150 emu/ cc .
2,18,19
Very recently, Yun et al.
20
have claimed P
r
of 250
C/ cm
2
for the polycrystalline BFO films grown on
Pt/ TiO
2
/ SiO
2
/ Si substrates, but this is probably an artifact
due to leakage and charge injection. This makes BFO as one
of the potential materials for novel device applications, al-
though the mechanisms behind the huge polarization
claimed by some groups is not yet fully understood. Some
experimental results
2,18
and theoretical reports
21
suggest that
the epitaxial strain might be the cause of such high value of
P
r
and M
r
. However, a recent study showed that the epitaxial
strain does not enhance M
r
in BFO thin films.
22
It is believed
that the heteroepitaxy induces significant and important
structural changes in BFO thin films, which may lead to
higher values for P
r
than found in single crystals.
There are some controversies in literature about the crys-
tal structure of 001 epitaxial thin films. There have been
several reports claiming tetragonal,
2,23
rhombohedral,
19,24
and monoclinic
25
structure of 001 BFO films on STO sub-
strates. Therefore, the sequence of transitions in layers is
poorly understood, and there is no understanding of the over-
all physics of the phase transitions involved; more structural
PHYSICAL REVIEW B 77, 014110 2008
1098-0121/2008/771/01411011 ©2008 The American Physical Society014110-1

analyses of thin films are necessary for better understanding
of engineered epitaxial and heterostructure BFO thin films.
Here, we report that the BFO001 thin films are clearly
monoclinic, in agreement with synchrotron data but contrary
to the bulklike rhombohedral and tetragonal structures re-
ported earlier; the
phase is unambiguously orthorhombic,
in disagreement with all other recent publications.
An extremely large and unexplained drop 40% in the
band-gap energy within the monoclinic phase as temperature
is increased from 300 to 800 K implies a strong and un-
known electron-phonon interaction. The concomitant lattice
contraction in the cubic
phase shows the cubic phase is
metallic and the material has large negative thermal expan-
sion, which is very rare e.g., Zr-tungstate and never found
in perovskites. The existence of insulator-metal transition
above magnetic ordered states implies that the metallic tran-
sition could only be due to lattice contraction. Although the
phase sequence is different from that in BaTiO
3
, the succes-
sion of order-disorder transitions resembles that in the eight-
site model of Comes et al.
26
ABO
3
oxide perovskites which are rhombohedral at low
temperatures, such as LaAlO
3
, PrAlO
3
, or NdAlO
3
,
27,28
have
ferroelastic instabilities at the A-ion site that induce displa-
cive phase transitions directly to cubic, but those which have
B-site instabilities instead have order-disorder transitions to
cubic that involve two or more steps. This has been success-
fully described
26,29
by an eight-site model in which the B-ion
displacements are always locally toward a 111 axis, but
thermally average via hopping over 111, 1
¯
11, 11
¯
1, and
111
¯
to give orthorhombic, tetragonal, or cubic time- and
space-global averages. In the present work, we show that this
model may describe BFO, contrary to conventional
wisdom
30
but in agreement with NMR, which shows some
B-site disorder.
31
II. EXPERIMENTAL DETAILS
BFO thin films of 300 nm thickness were grown by
pulsed laser deposition PLD using a 248 nm KrF Lambda
Physik laser. Films were grown on STO100 substrates of
area 5mm
2
with 25 nm thick SrRuO
3
SRO buffer
layer. The orientation, crystal structure, and phase purity of
the films were examined using Siemens D5000 x-ray diffrac-
tometer. The Jobin Yvon T64000 micro-Raman microprobe
system with Ar ion laser =514.5 nm in backscattering
geometry was used for polarized and temperature-dependent
Raman scattering. The laser excitation power was 2.5 mW
and the acquisition time was 10 min per spectrum. The high-
temperature x-ray data were collected from powdered bulk
material prepared by high energy ball milling using a Bruker
D8-ADVANCE diffractometer with Cu radiation, Göbel mir-
ror, and fast Vantec linear position sensitive detector. Scans
from 20° to 80° 2
were obtained every 15° from 40° to the
decomposition point of BFO with an acquisition time of
8 min per pattern and a heating rate of 30 °C/ min between
scans. The differential thermal analysis DTA measurements
on the Bi
2
O
3
-Fe
2
O
3
phase diagram were performed with a
Mettler TA 1 thermoanalyzer in nitrogen atmosphere with a
heating rate of 500 °C/ h up to 750 °C . Above that tempera-
ture, heating-cooling cycles with a rate of 120 °C/ h were
practiced. The sample and reference material weights were
0.2 g, contained in platinum crucibles. Whereas
Bi
2
O
3
/ Fe
2
O
3
powder mixtures were used for determining
the liquidus curve, the peritectic and eutectic horizontals,
crushed fernlike dendritic single crystals Fig. 1 served for
the exact BiFeO
3
composition.
16
The DTA measurements of
target materials of thin film were carried out in air with a
heating rate of 5 °C/min up to 1000 °C using Shimadzu
DTA-50 differential thermal analyzer. The crystal growth of
the fernlike dendrites was performed in platinum crucibles,
using the accelerated crucible rotation technique,
32
with a
crucible content of about 90 g and a platinum crucible cover,
welded tightly to the crucible, leaving only a central hole of
0.1 mm diameter. For example, for a Bi
2
O
3
/ Fe
2
O
3
mixture
of 23 mol % Fe
2
O
3
, the following temperature program was
used: rapid heating to 1000 °C, soaking at 1000 °C for 24 h,
cooling at 3 °C/ h to 960 °C, thereafter linearly at
0.36 °C/ h to 800 °C, cutting off the crucible cover, pouring
off the remaining flux, and final cooling in furnace. Maximal
size of dendritic leaves: 14 51.5 mm
3
.
16
The ferroelastic
domain structure of the
phase of these leaflets is explained
in Ref. 10.
III. RAMAN EFFECT IN
PHASE
The rhombohedral R3c, tetragonal P4mm,
2
and
monoclinic
33
Bb structures of BFO give rise to 13, 8, and
(a)
(b)
FIG. 1. As-grown dendrites of BiFeO
3
: a bunch of leaves
grown in a Pt crucible; b individual leaf, axis parallel to 110pc,
plane of leaf parallel to 110pc. The horizontal axis is 2.5 and
1cmina and b, respectively.
PALAI et al. PHYSICAL REVIEW B 77, 014110 2008
014110-2

27 distinct Raman-active modes, respectively, as listed in
Table I. For the orthorhombic distortion in the
phase, as
discussed below, the tetragonal entries will remain correct,
with only a small splitting of the E modes.
The unpolarized perpendicular to the 001 of the sub-
strate Raman spectra of STO substrate, SRO/STO, and
BFO/SRO/STO at room temperature are given in Fig. 2a.
The comparison of the spectrum of BFO thin film with STO
and SRO/STO spectra precludes any Raman contribution
from the substrate and bottom electrode; to the contrary, we
observed a dip rather than a peak at the STO strongest peak
position. As is evident from the intensity comparison, all of
these peaks were due to BFO normal modes of vibrations
and none of them arose from the substrate. We have verified
our results using target materials and single crystals, and also
by growing 001BFO films on different substrates.
34
The lowest peak around 74 cm
−1
in our spectra looks very
asymmetric and the deconvolution with the phonon fitting
showed Fig. 2b two peaks at 71.6 and 76.3 cm
−1
with R
2
value of 99% the percentage of variation in the observed
and fitted values, while the STO peak appears at 79.8 cm
−1
.
A clear splitting has been observed at cryogenic
temperature.
34
The polarized Raman spectra for 300 nm thick 001BFO
thin film on the SRO/STO substrate and 100STO single-
crystal substrate at room temperature in the parallel,
ZXXZ
¯
, and perpendicular, ZXYZ
¯
, polarization configu-
rations are given in Fig. 3. The comparison of polarized Ra-
man spectra again clearly implies that there is no Raman
contribution from the substrate on the BFO spectra. The
spectra of BFO thin film Fig. 3a revealed strong peaks at
around 72, 76, 140, 172, 219, 261, 269, and 289 cm
−1
, while
weaker peaks were observed at around 352, 370, 406, 478,
529, 547, 609, 808, 946, and 1093 cm
−1
including few
second-order peaks. The existence of 13 well defined and
identical peaks in both the ZXXZ
¯
and ZXYZ
¯
polarization
configurations confirms the Raman selection rules for the
monoclinic structure see Table I instead of tetragonal or
rhombohedral as reported earlier.
19,23,24
This observation
verifies the very recent report of monoclinic structure for the
epitaxial BFO films grown on 001STO substrates by Xu et
al.
25
studied via synchrotron radiation. We also investigated
BFO thin films grown using different deposition techniques
PLD, metallo-organic chemical vapor deposition, and
chemical solution deposition on different single-crystal and
textured substrates, and we found that films show monoclinic
structure at room temperature down to 50 nm thickness.
However, the single crystal clearly satisfies the Raman selec-
tion rules for the rhombohedral structure with R3c symmetry.
For a comparative study between epitaxial thin film and
single crystal, refer to Ref. 34. Note that for ultrathin unre-
laxed films, a tetragonal structure is possible, and our present
work does not address these specimens. As the spectra along
YZZY
¯
and YZXY
¯
were heavily dominated by the scat-
tering from the STO substrate, the contribution from the
BFO film could not be separated.
TABLE I. Selection rules for the Raman active modes for rhom-
bohedral R, tetragonal T, and monoclinic M crystal structures
in different polarization configurations with total number of normal
N Raman modes. The notation 001up” means unpolarized spec-
tra along the pseudocubic 001 direction perpendicular to the sub-
strate. The notation Z along 001 direction and Z
¯
are the direc-
tions of the incident and backscattered light, respectively.
Scattering
geometry
RR3c
C
3
v
TP4mm
C
4
v
MBb
C
s
N Raman 4A
1
+9E 3A
1
+B
1
+4E 13A
+14A
001up 4A
1
+9E 3A
1
+B
1
13A
ZXXZ
¯
A
1
and EA
1
and B
1
A
ZXYZ
¯
E No modes A
YZZY
¯
A
1
A
1
A
YZXY
¯
EE A
200 400 600 800 1000 1200
Int. (arb. units)
Raman shift (cm
-1
)
(100)STO
SRO/STO
BFO/SRO/STO
(a)
50 60 70 80 90
Int. (arb. units)
Raman shift (cm
-1
)
STO
Observed
Fitted
RT
(b)
FIG. 2. a Comparison of unpolarized Raman spectra along
001 direction of 100STO single crystal substrate, SRO/STO, and
001BFO/SRO/STO thin film at room temperature. Lines are
drawn as a guide for the eyes. b Deconvolution of the BFO peak
around 74 cm
−1
at RT shows the evolution of two peaks at 71.6 and
76.3 cm
−1
. The “dotted line with empty circle” shows the fitting
after the deconvolution of these two peaks. The STO100 single-
crystal substrate shows one peak at 79.8 cm
−1
at RT, plotted for
comparison.
PHASE AND
-
METAL-INSULATOR PHYSICAL REVIEW B 77, 014110 2008
014110-3

If the theoretically derived, polar monoclinic space
group
33
Bb describes, in fact, the ferromagnetism in mono-
clinically distorted thin films,
18
then the linear magnetoelec-
tric effect is possible in principle, because all 13 ferromag-
netic and/or ferroelectric Heesch-Shubnikov point groups are
permitting the linear magnetoelectric effect.
35
The possible
ferromagnetic subgroups of Bb of the nuclear structure are
the ferromagnetic space groups Bb and Bb
magnetic point
groups m and m
, respectively.
33,36
Figure 4 shows temperature-dependent Raman spectra of
BFO thin film. A closer inspection of Raman spectra near the
phase transition shows two noticeable changes: disappear-
ance of all stronger peaks 72, 76, 140, 172, and 219 cm
−1
at 820 °C with the appearance of a few new peaks, and
complete disappearance of spectra around 950 °C. This tem-
perature behavior implies that the BFO maintains its room-
temperature structure up to 820 °C, indicating the
ferroelectric-ferroelectric FE-FE phase transition, in agree-
ment with the earlier investigations on BFO bulk single-
crystal and polycrystalline
6,30
samples. No evidence of soft
phonon modes implies that the BFO has an order-disorder,
first-order ferroelectric transition, unlike PbTiO
3
. No decom-
position of the BFO films was observed until 950 °C, sug-
gesting that our film samples had few defects and disloca-
tions. According to the experience of one of the authors
H.S., bulk BFO is thermodynamically unstable in air with-
out being in equilibrium contact with the Bi
2
O
3
/ Fe
2
O
3
flux.
Since the decomposition is a first-order transition, it needs
nucleation, and this is abundant in the case of the large sur-
face of a very fine powder, starting to decompose at about
400 °C. This widely ignored fact was leading in the past, and
is still leading to the preparation and description of impure
powders or ceramic samples. The surface of single crystals,
e.g., polished with a 1
m diamond paste, proved to remain
observable under the polarized light microscope up to the
decomposition temperature 共⬃933 °C. However, with in-
creasing time, a multitude of dark spots developed on the
surface, indicating nucleation of decomposition on polishing
scratches Fig. 8c. It is expected that a defect-free surface
of as-grown single crystals and of high quality films will
withstand decomposition for a longer time.
IV. RAMAN EFFECT IN
PHASE
The high-temperature Raman spectra Fig. 4 show that
four lines 关⬃ 213, 272, 820, and 918 cm
−1
possibly second-
order peak兲兴 persist above the well known ferroelectric-phase
transition temperature 820 °C. In the cubic perovskite
phase, no first-order Raman lines are allowed because all the
ions sit at the inversion centers and all long wavelength
phonons are of odd parity. The data show that the
phase
above 820 °C cannot be cubic as reported earlier.
30
Since
our backscattering geometry with incidence radiation along
the Z axis favors A
1
and B
1
phonons, four Raman modes
3A
1
+B
1
, normally silent are predicted in the tetragonal or
orthorhombic perovskite phase Table I, in agreement with
the experiment. Small orthorhombic splitting of four unob-
100 300 500 700 900 1100
Z(XX)Z
Int. (arb. units)
Raman shift (cm
-1
)
(a)
400 600 800 1000 1200
Int. (arb. units)
Raman shift (cm
-1
)
Z(XY)Z
100 300 500 700 900 1100
Int. (arb. units)
Raman shift (cm
-1
)
Z(XX)Z
STO (100) single crystal
Z(XY)Z
(b)
FIG. 3. a Polarized Raman spectra of 001BFO thin film on
SRO/STO in ZXXZ
¯
and ZXYZ
¯
scattering configurations. b
Polarized Raman spectra of 100STO substrate given for
comparison.
300 600 900 1200
22
350
Int. (arb. units)
Raman shift (cm
-1
)
1000
950
900
850
840
830
820
810
800
750
700
650
550
BFO(001)/SRO/STO(100)
22-1000
o
C
FIG. 4. Color online Temperature-dependent Raman spectra of
001BFO thin film on SRO/STO substrates.
PALAI et al. PHYSICAL REVIEW B 77, 014110 2008
014110-4

Figures
Citations
More filters
Journal ArticleDOI

Physics and Applications of Bismuth Ferrite

TL;DR: In this paper, the authors summarize both the basic physics and unresolved aspects of BiFeO3 and device applications, which center on spintronics and memory devices that can be addressed both electrically and magnetically.
Journal ArticleDOI

Domain wall nanoelectronics

TL;DR: In this paper, a review of magnetoelectric domain walls is presented, focusing on magneto-electrics and multiferroics but making comparisons where possible with magnetic domains and domain walls.
Journal ArticleDOI

Advances in magnetoelectric multiferroics.

TL;DR: Progress in the fundamental understanding and design of new multiferroic materials, advances in characterization and modelling tools to describe them, and usage in applications are reviewed.
Journal ArticleDOI

Resistive switching materials for information processing

TL;DR: This Review surveys the four physical mechanisms that lead to resistive switching materials enable novel, in-memory information processing, which may resolve the von Neumann bottleneck and examines the device requirements for systems based on RSMs.
References
More filters
Journal ArticleDOI

Epitaxial BiFeO3 multiferroic thin film heterostructures.

TL;DR: Enhanced polarization and related properties in heteroepitaxially constrained thin films of the ferroelectromagnet, BiFeO3, and combined functional responses in thin film form present an opportunity to create and implement thin film devices that actively couple the magnetic and ferroelectric order parameters.
Journal ArticleDOI

Why Are There so Few Magnetic Ferroelectrics

TL;DR: In this paper, the fundamental physics behind the scarcity of ferromagnetic ferroelectric coexistence was explored and the properties of known magnetically ordered ferro-electric materials were examined.
Journal ArticleDOI

Spiral magnetic ordering in bismuth ferrite

TL;DR: In this article, the magnetic ordering of the iron ions in bismuth ferrite BiFeO3 was obtained by a study with a high-resolution time-of-flight neutron diffractometer.
Journal ArticleDOI

Observation of coupled magnetic and electric domains

TL;DR: Spatial maps of coupled antiferromagnetic and ferroelectric domains in YMnO3 are obtained by imaging with optical second harmonic generation and lead to a configuration that is dominated by the ferroelectromagnetic product of the order parameters.
Journal ArticleDOI

Weak ferromagnetism and magnetoelectric coupling in bismuth ferrite

TL;DR: In this paper, the coupling between the ferroelectric and magnetic order parameters in the magnetoelectric multiferroic was analyzed using density functional theory within the local spin density approximation (LSDA) and the $\mathrm{LSDA}+\mathm{U}$ method.
Related Papers (5)
Frequently Asked Questions (13)
Q1. What are the contributions in this paper?

The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-pro t purposes provided that: • a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders. Please consult the full DRO policy for further details. 

The postulation of a cubic phase in a very narrow temperature interval was initially based on the reflected polarized light observation that the ferroelastic domains of the phase disappear at about 925 °C, leaving the sample optically isotropic up to the decomposition point of the cubic BFO phase at 933 °C. 

It is conjectured that cycling the crystal through the - firstorder transition, as well as ferroelectric switching, will greatly increase the number of antiphase domains and their walls. 

Ferroelastic domains of the rhombohedral phase of BFO have repeatedly been studied by polarized light microscopy, by transmission and reflection in Ref. 16, by transmission alone in Refs. 54–56, and by reflection alone in Refs. 8, 14, 17, 49, and 57. 

The resistivity as a function of temperature shows a change of slope from negative to positive around 920 °C Fig. 11 which the authors interpret as an indication of the metal-insulator M-I transition, although the authors note that the positive slope is anomalously high for a normal metal. 

According to the experience of one of the authors H.S. , bulk BFO is thermodynamically unstable in air without being in equilibrium contact with the Bi2O3 /Fe2O3 flux. 

High-temperature x-ray study of BFO powder Fig. 6 showed that the rhombohedral bulk structure has a strongly first-order transition near 825 5 °C to an orthorhombic structure. 

The sequence of phases is monoclinicorthorhombic-cubic in film and rhombohedral-orthorhombiccubic in bulk; the rhombohedral and orthorhombic phases of BiFeO3 differ notably from those in BaTiO3. 

For the orientation of the orthorhom-bic cell as in BaTiO3, discrimination is possible, because in that case, in addition to 110 pc walls, also 100 pc walls are allowed as in the rhombohedral case. 

In some tiny regions, also lines perpendicular to the former ones can be seen difficult to see in print , which are due to 100 pc walls. 

This work was supported by the DoD W911NF-06-0030 and W911NF-05-1-0340 grants and by a EU-funded project “Multiceral” NMP3-CT2006-032616 at Cambridge. 

The shortened bond length in the orthorhombic and cubic phases, compared with those in the rhombohedral phase, favors a metallic state. 

The data graphed in Fig. 10 a were obtained in two different ways: by conventional absorption spectroscopy at a fixed temperature; and with fixed wavelength 632.8 nm He-Ne by slowly varying temperature and using the Urbach equation to relate absorption coefficient a to band gap Eg T : log a T = E−Eg /Eg +constant.