Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
Ferroelastic twin structures in epitaxial WO3 thin filmsShinhee Yun, Chang-Su Woo, Gi-Yeop Kim, Pankaj Sharma, Jin Hong Lee, Kanghyun Chu, Jong Hyun Song,Sung-Yoon Chung, Jan Seidel, Si-Young Choi, and Chan-Ho Yang Citation: Applied Physics Letters 107, 252904 (2015); doi: 10.1063/1.4938396 View online: http://dx.doi.org/10.1063/1.4938396 View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/107/25?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Epitaxial growth of high quality WO3 thin films APL Mater. 3, 096102 (2015); 10.1063/1.4930214 Structural and magnetic properties of a series of low-doped Zn 1−x Co x O thin films deposited from Zn and Cometal targets on (0001) Al 2 O 3 substrates J. Appl. Phys. 95, 7187 (2004); 10.1063/1.1667805 Dielectric properties of strained ( Ba, Sr)TiO 3 thin films epitaxially grown on Si with thin yttria-stabilized zirconiabuffer layer Appl. Phys. Lett. 78, 2542 (2001); 10.1063/1.1367309 Effect of finite domain-wall width on the domain structures of epitaxial ferroelectric and ferroelastic thin films J. Appl. Phys. 89, 1355 (2001); 10.1063/1.1332086 Model of crystal lattice strained along the preferential direction by anisotropic stress for GaAs heteroepitaxialfilms grown on vicinal Si(001) and Si(110) substrates by molecular-beam epitaxy J. Vac. Sci. Technol. A 19, 287 (2001); 10.1116/1.1323971
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions. IP: 143.248.11.147 On: Wed, 17 Feb 2016 02:28:03
http://scitation.aip.org/content/aip/journal/apl?ver=pdfcovhttp://oasc12039.247realmedia.com/RealMedia/ads/click_lx.ads/www.aip.org/pt/adcenter/pdfcover_test/L-37/1734683719/x01/AIP-PT/APL_ArticleDL_121615/APR_1640x440BannerAd11-15.jpg/434f71374e315a556e61414141774c75?xhttp://scitation.aip.org/search?value1=Shinhee+Yun&option1=authorhttp://scitation.aip.org/search?value1=Chang-Su+Woo&option1=authorhttp://scitation.aip.org/search?value1=Gi-Yeop+Kim&option1=authorhttp://scitation.aip.org/search?value1=Pankaj+Sharma&option1=authorhttp://scitation.aip.org/search?value1=Jin+Hong+Lee&option1=authorhttp://scitation.aip.org/search?value1=Kanghyun+Chu&option1=authorhttp://scitation.aip.org/search?value1=Jong+Hyun+Song&option1=authorhttp://scitation.aip.org/search?value1=Sung-Yoon+Chung&option1=authorhttp://scitation.aip.org/search?value1=Jan+Seidel&option1=authorhttp://scitation.aip.org/search?value1=Si-Young+Choi&option1=authorhttp://scitation.aip.org/search?value1=Chan-Ho+Yang&option1=authorhttp://scitation.aip.org/content/aip/journal/apl?ver=pdfcovhttp://dx.doi.org/10.1063/1.4938396http://scitation.aip.org/content/aip/journal/apl/107/25?ver=pdfcovhttp://scitation.aip.org/content/aip?ver=pdfcovhttp://scitation.aip.org/content/aip/journal/aplmater/3/9/10.1063/1.4930214?ver=pdfcovhttp://scitation.aip.org/content/aip/journal/jap/95/11/10.1063/1.1667805?ver=pdfcovhttp://scitation.aip.org/content/aip/journal/jap/95/11/10.1063/1.1667805?ver=pdfcovhttp://scitation.aip.org/content/aip/journal/apl/78/17/10.1063/1.1367309?ver=pdfcovhttp://scitation.aip.org/content/aip/journal/apl/78/17/10.1063/1.1367309?ver=pdfcovhttp://scitation.aip.org/content/aip/journal/jap/89/2/10.1063/1.1332086?ver=pdfcovhttp://scitation.aip.org/content/avs/journal/jvsta/19/1/10.1116/1.1323971?ver=pdfcovhttp://scitation.aip.org/content/avs/journal/jvsta/19/1/10.1116/1.1323971?ver=pdfcov
Ferroelastic twin structures in epitaxial WO3 thin films
Shinhee Yun,1 Chang-Su Woo,1 Gi-Yeop Kim,2 Pankaj Sharma,3 Jin Hong Lee,1
Kanghyun Chu,1 Jong Hyun Song,4 Sung-Yoon Chung,5,6 Jan Seidel,3 Si-Young Choi,2
and Chan-Ho Yang1,6,a)1Department of Physics, KAIST, Daejeon 34141, South Korea2Korea Institute of Materials Science, Changwon 51508, South Korea3School of Materials Science and Engineering, University of New South Wales, Sydney,New South Wales 2052, Australia4Department of Physics, Chungnam National University, Daejeon 34134, South Korea5Graduate School of EEWS, KAIST, Daejeon 34141, South Korea6KAIST Institute for the NanoCentury, KAIST, Daejeon 34141, South Korea
(Received 24 October 2015; accepted 8 December 2015; published online 22 December 2015)
Tungsten trioxide is a binary oxide that has potential applications in electrochromic windows, gas
sensors, photo-catalysts, and superconductivity. Here, we analyze the crystal structure of atomically
flat epitaxial layers on YAlO3 single crystal substrates and perform nanoscale investigations of the
ferroelastic twins revealing a hierarchical structure at multiple length scales. We have found that the
finest stripe ferroelastic twin walls along pseudocubic h100i axes are associated with cooperativemosaic rotations of the monoclinic films and the larger stripe domains along pseudocubic h110i axesare created to reduce the misfit strain through a commensurate matching of an effective in-plane lat-
tice parameter between film and substrate. The typical widths of the two fine and larger stripe
domains increase with film thickness following a power law with scaling exponents of �0.6 and�0.4, respectively. We have also found that the twin structure can be readily influenced by illumina-tion with an electron beam or a tip-based mechanical compression. VC 2015 AIP Publishing LLC.[http://dx.doi.org/10.1063/1.4938396]
In recent years, much attention has been paid to twin/do-
main walls as a low dimensional mesoscopic entity exhibiting
exotic physical properties distinct from bulk properties.1–13
One of the noticeable materials to illustrate the significance of
twin walls is tungsten trioxide (WO3). Many interesting phe-
nomena at ferroelastic twin walls have been discovered such
as electronic conduction in contrast to the insulating bulk,14
superconductivity at low temperatures,15 and ionic migra-
tion.16,17 Moreover, alkali-metal doped or oxygen reduced
WO3 occupies 5d orbitals of tungsten ions as an n-type con-ductor18–22 that might further expand the versatile features of
twin wall. However, until now, the usability has been mainly
studied based on bulk crystals or nano-structures,23–27 and few
studies have been made on epitaxial films.28–31 Furthermore,
creation of the twin wall structure in an epitaxial WO3 film has
not been achieved yet. This is a clear limitation to further
advance our understanding of twin wall functionalities and
explore new possibilities in this material. Advantages of epi-
taxial films associated with controllable thickness, effects of
heteroepitaxial interface including misfit strain, tunability of
the orientation, and periodicity of domain walls can be broadly
beneficial to the manipulation of the inherent functionalities of
WO3.
In this paper, we report on the crystal structure of epitax-
ial WO3 layers on a nearly matched (110)o (hereafter the
subscript “o” represents the orthorhombic index) surface of
YAlO3 substrate. We have found that atomically flat epitax-
ial films show an exotic ferroelastic twin domain arrange-
ment with a hierarchical structure at multiple length scales
made of monoclinic building blocks. These ferroelastic
domains sensitively respond to electron beam irradiation and
mechanical pressure.
In our experiments (for the details of experimental meth-
ods, see Ref. 32), WO3 thin films were grown on (110)o-ori-
ented YAlO3 substrates (orthorhombic, a¼ 5.18 Å, b¼ 5.31 Å,c¼ 7.35 Å,�
ffiffiffi
2p
apc� �ffiffiffi
2p
apc� � 2apc) by pulsed laser dep-osition. WO3 has a crystal structure similar to the ABO3 perov-
skite except for an empty A-site, which undergoes various
structural transitions with temperature variation (monoclinic
! triclinic! monoclinic! orthorhombic! tetragonal withincreasing temperature).33,34 Near room temperature, there are
two competing phases called c-WO3 and d-WO3. The c-WO3has a monoclinic crystal structure (space group P21/n,
a¼ 7.306 Å, b¼ 7.540 Å, c¼ 7.692 Å, b¼ 90.88�) with a vol-ume of �2apc� �2apc� �2apc, where apc is a pseudocubiclattice parameter (hereafter the subscript “pc” represents the
pseudocubic cell),33 exhibiting its peculiar twin structure.35
The d-WO3 is triclinic with lattice parameters (P�1,a¼ 7.309 Å, b¼ 7.522 Å, c¼ 7.678 Å, a¼ 88.81�, b¼ 90.92�,c¼ 90.93�).36
A representative surface topographic image of a 610-
nm-thick WO3 film exhibits a well aligned ferroelastic twin
structure with a herringbone pattern (Fig. 1(a)). The twin
domains have a width of �30 nm, which is more clearly rec-ognized in the AFM deflection contrast representing the
slope of height (Fig. 1(b)). Since the fast scan axis of the
AFM tip was parallel to the [001]o axis, the contrast for only
the [1�10]o-axis-parallel fine stripe domains (fine-domains) isclearly identified in the deflection image. A schematic of the
herringbone twin structure in Fig. 1(c) shows the fine stripea)E-mail: [email protected]
0003-6951/2015/107(25)/252904/5/$30.00 VC 2015 AIP Publishing LLC107, 252904-1
APPLIED PHYSICS LETTERS 107, 252904 (2015)
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions. IP: 143.248.11.147 On: Wed, 17 Feb 2016 02:28:03
http://dx.doi.org/10.1063/1.4938396http://dx.doi.org/10.1063/1.4938396http://dx.doi.org/10.1063/1.4938396mailto:[email protected]://crossmark.crossref.org/dialog/?doi=10.1063/1.4938396&domain=pdf&date_stamp=2015-12-22
domains parallel to [001]o or [1�10]o axes at a length scale ofa few 10 nm. Larger stripe macro-domains, which are rotated
by 645� relative to the fine domain walls, are present at alarger length scale (a few 100 nm), and they can be classified
into two variants in terms of the aligning axes. Furthermore,
orthogonal bundles of the macro-domains, each of which
contains a single variant of stripe macro-domains, are found
at a few micron length scales. The bumpy twin walls were
observed at the boundaries between two adjacent bundles
creating the irregular shaped super-macro-domain structure
(Fig. 1(d)).
To investigate the microscopic origin for the hierarchi-
cal twin structure, we characterized the crystal structure
through x-ray h-2h scans and reciprocal space maps (RSMs)for a 200-nm-thick film. In Fig. 2(a), the h-2h scan exhibitsthe (001)pc and (002)pc diffraction peaks of the WO3 film.
The out-of-plane lattice parameter of the WO3 film was
determined to be 3.84(8) Å, which is almost the same as the
bulk c-axis pseudocubic lattice parameter of the monoclinicphase (3.84(6) Å).33 Besides, the full-width-at-half-maxi-
mum (FWHM) of the small range x-rocking curve measuredat the (001)pc WO3 peak (inset of Fig. 2(a)) was 0.061
�. Inaddition to this sharp central peak, a larger range scan
revealed the existence of two satellite peaks at off-axis
x-angles of 60.82� (see Fig. S2(b) in Ref. 32). This can beclearer in a RSM for the (002)pc reflection in the HL-plane
(Fig. 2(b)). Accordingly, we know that the ab-plane of thefilm is not exactly parallel to the substrate but tilted by v�0.82� and there are four variants in terms of the tilt direc-tions. It is also worthwhile mentioning that this mosaic tilt
angle of �0.82� detected in this thick-thickness regime(>100 nm) is gradually reduced with film thickness andeventually the tilt becomes negligible when films are thinner
than �50 nm.To further clarify the crystal structure including in-plane
lattice parameters, RSMs for two asymmetric peaks (103)pcand (113)pc were measured for the 200-nm-thick film (Figs.
2(c) and 2(d)). To interpret the RSMs, we first consider a
model of pseudocubic unit cell imposed by a monoclinic dis-
tortion similar to the monoclinic unit cell of bulk WO3 (M0in Fig. 2(e)). Given a monoclinic angle b and four-foldtwins, RSMs for {H0L}pc or {K0L}pc reflections of M0 result
in three split peaks (red crosses in Fig. 2(c)). Two of them
((H0L)pc and (-H0L)pc) are shifted toward �L or þL direc-tions by an equal k-space distance away from the expected
pseudo-tetragonal position deduced from (00 L)pc peaks; the
FIG. 1. AFM images of a 610-nm-
thick WO3 film on (110)o YAlO3 sub-
strate. (a) The topographic image and
(b) tip deflection image representing
the slope of height. (c) Schematic of
the herringbone twin structure consist-
ing of macro- and fine-domains. (d)
Large scale unflatten AFM image
showing super-macro-domains sur-
rounded by bumpy boundaries. Terrace
structure with a single unit cell step
height along ½110�subpc is seen superim-posed on the twin wall structure.
FIG. 2. X-ray diffraction study on a 200-nm-thick WO3 film on (110)oYAlO3 substrate. (a) A h-2h scan along [00L]pc showing (001)pc and (002)pcWO3 peaks. The inset exhibits a x-rocking curve measured at the (001)pcWO3 peak. (b) RSM for (002)pc reflection. (c) RSM for (103)pc reflection.
(d) RSM for (113)pc reflection. The reciprocal lattice unit (r.l.u.) was defined
to be the reciprocal value of d-spacing of the YAlO3 (110)o planes. (e) Aschematic displaying the monoclinic unit cell of WO3 rotated from M0 to M
by the mosaic tilt angle around the local b-axis. Four-fold twin structure wasconsidered when the peak indices were assigned in the RSMs.
252904-2 Yun et al. Appl. Phys. Lett. 107, 252904 (2015)
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions. IP: 143.248.11.147 On: Wed, 17 Feb 2016 02:28:03
in-plane H position of them gives the reciprocal of the a-axislattice parameter and the extent of splitting is related to the
monoclinic angle b. In addition, the third peak detected atthe expected L without any shift corresponds to {0KL}pcreflections. The in-plane k-space position of the peak gives
information of b-axis lattice parameter. To ensure the modelof unit cell, we would also check the peak positions of
{HHL}pc reflections. If the model is correct, the peaks of
{HHL}pc should be split into two variant peaks and the
amount of split along L-axis should harmonize with the
aforementioned L-split between (H0L)pc and (�H0L)pcpeaks.
With this in mind, we next extract the lattice parameters
of the WO3 thin film from the measured RSMs. Different
from the usual peak split tendency, both the RSMs in this case
seemed to produce the peak split along the horizontal H-axis
rather than the L-axis. It arose from the fact that the normal of
ab-plane of film was tilted by 60.82� with respect to the nor-mal of substrate creating four-fold mosaic domains. Thus, the
experimental results, which at first seemed to be different
from the expectation, can now be understood by introducing
mosaic tilts into the interpretation. The (103)pc and (�103)pcpeaks were rotated around the K-axis by �0.82� and þ0.82�,respectively. Being traced along powder arcs to recover the
case without the mosaic tilt, the original k-space positions
could be found and marked by red crosses on the maps (Fig.
2). With the monoclinic structural model, we could determine
the pseudocubic lattice parameters of the WO3 thin film to be
a¼ 3.65(6) Å, b¼ 3.75(9) Å, c¼ 3.84(8) Å, and b¼ 89.1(8)�.We note that the deviation of monoclinic angle (b) from 90�
has a very similar value to the tilt angle (v� 0.82�). In fact,the appearance of split peaks coincidentally along the horizon-
tal axis stems from the following situation. As described in
the schematic (Fig. 2(e)), the monoclinic unit cell of the WO3film is rotated around the local b-axis so that the c-direction isperpendicular to substrate. This happens because sharing the
same c-axis of neighboring fine domains is favorable for mini-mizing the elastic deformation at the twin walls. It was also
found that the (103)pc and (�103)pc peaks subject to the mosaicrotation were of diffuse shape along the horizontal reciprocal
axis in accordance with the vertical twin walls separated by
the short width of fine domains (a few 10 nm). On these
grounds, we can conclude that the origin of the fine twin
domains is due to the cooperative mosaic tilt of monoclinic
unit cells, thereby producing þa and �a fine domains. On theother hand, the a-axis lattice parameter was slightly smallerthan that of the substrate, while the b-axis lattice parameterwas larger. A recurring appearance of a-axis and b-axisdomains, i.e., the macro-domain structure, is necessary to
effectively minimize interfacial misfit strain energy with the
substrate offering a nearly matched interface on average.
To investigate the cross sectional area of a WO3 film, a
transmission electron microscopy (TEM) study was per-
formed for a 70-nm-thick WO3 film (Fig. 3). In the weak-
beam-dark-field (WBDF) TEM image of the as-prepared
sample with a zone axis along [001]o (Fig. 3(a)), strained-
pillar structures were recognized through the contrast
depending on local strain. The widths of pillars were meas-
ured to be less than 10 nm. Although these dense twin walls
can be attributed to the relatively thin thickness of the film as
compared to the film used for the AFM image, one should be
cautious to conclude that the measured width in TEM repre-
sents the fine-domain width of the film accurately. The TEM
specimen was fabricated in a way that the foil thickness
along the zone axis was very thin for electron transmission,
which would lead to a different mechanical boundary condi-
tion from that of the as-grown film. In reality, the twin struc-
ture observed at the initial stage in the film was easily
relaxed by illumination with a weak electron beam with a
flux of 14 pA/cm2 (Fig. 3(b)). The illumination gradually
removed twin walls and associated individual fine domains
to form remaining larger domains. The susceptibility of the
twin structure to external perturbation leads us to conjecture
that the twin walls are primarily involved in large strains and
strain gradients being less relevant to interfacial reconstruc-
tion or/and defect accumulation in this thin film. The insta-
bility of twin walls in TEM specimen did not allow careful
high-resolution TEM studies on the twin walls, but it was
possible to examine the relaxed phase of WO3 regarding the
epitaxial coherency and interfacial structure between film
and substrate. Figure 3(c) shows a Z-contrast high-angle-an-
nular-dark-field scanning TEM (HAADF-STEM) image for
an interfacial region. It was observed that a WO2 sub-layer
was formed right on the YO layer of the substrate and W
atoms occupied only the B-sites of the perovskite keeping
the A-site empty. Figure 3(d) shows a selected-area-electron-
diffraction (SAED) pattern for the interfacial region. The red
spots can be indexed to the diffraction peaks of WO3 and the
green spots to the ones of YAlO3. The detection of exact
square reciprocal cell of WO3 indicates the monoclinic
ca-plane in this relaxed film was parallel to the zone axis.Misalignment between the reciprocal cells of YAlO3 and
WO3 by �1.5� along the horizontal reciprocal axis indicates
FIG. 3. TEM study for a 70-nm-thick WO3 film. (a) The WBDF TEM image
of the as-prepared sample showing a highly strained fine domain structure.
(b) The fine domain structure easily relaxed by illumination of a weak elec-
tron beam with a flux of 14 pA/cm2. (c) The HAADF-STEM image of the
interface of WO3 film and YAlO3 substrate. (d) A SAED pattern for an inter-
facial region. The dashed red (green) line box represents the pseudocubic
reciprocal unit cell of WO3 (the reciprocal unit cell of YAlO3). The scale
bars indicate 20 nm.
252904-3 Yun et al. Appl. Phys. Lett. 107, 252904 (2015)
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions. IP: 143.248.11.147 On: Wed, 17 Feb 2016 02:28:03
that the pseudocubic c-axis of substrate makes an angle of�1.5� with that of the film, which is expected in the ortho-rhombic substrate.
Next, the thickness dependence of the fine- and macro-
domain widths was examined using multiple samples with dif-
ferent thicknesses. As examples, in-plane piezoresponse force
microscopy (PFM) images of four representative samples with
thicknesses of 42 nm, 81 nm, 202 nm and �400 nm are shownin Figs. 4(a)–4(d). Remarkably, macro-domains were visible
in the PFM images enabling us to evaluate their macro-
domain widths by Fourier-transforming the PFM images
(insets of Figs. 4(a)–4(d)).37 Because the monoclinic phase of
WO3 with a space group of P21/n is centrosymmetric, the sig-
nificant enhancement of piezoresponse on either side of
macro-domain walls is unexpected. It is most likely due to the
fact that the twin wall areas are subject to large deformation
against the nearby macro-domain lowering local symmetry.
Moreover, films under the mosaic tilt are expected to create a
large strain gradient to reconcile with flat substrates under-
neath, leading to the involvement of flexoelectric polariza-
tions.8–10 It has been found that an oxygen deficient sheet
layer can be a non-centrosymmetric tetragonal phase.38 A
more rigorous mechanism of the unusual piezoresponse and
microscopic atomic arrangements at the twin walls and
interfaces still remain unexplored.
The double logarithmic plot of domain widths versus
film thickness was carefully obtained (Fig. 4(e)). The linear
slope in the log-log plot determines the scaling exponent to
be 0.60 (60.01) for the macro-domain. Similarly, the valueof the fine-domain is determined to be 0.42 (60.05). Thesescaling exponents significantly deviate from the 0.5 expected
in the typical Landau-Lifschitz-Kittel law.39 Similar devia-
tion has been reported in the ferroelectric domains of multi-
ferroic BiFeO3; irregular domain walls characterized by a
roughness exponent and a fractal Hausdorff dimension ex-
hibit the domain size scaling with an exponent �0.6.40,41 Itwill be an interesting topic to investigate the fractal-like
behavior in this hierarchical twin structure in future studies.
Finally, we address the rearrangement of the macro-
domains by the AFM tip-based mechanical force. For an as-
grown 53-nm-thick WO3 film, in-plane PFM image was
captured to visualize (at nominally low loading force of
approximately 50–100 nN) the initial ferroelastic domain
microstructure as shown by Figure 5(a). Thereafter, within
the initially visualized region (shown by dashed-line frame
in Figs. 5(a) and 5(b)), a contact mode scan with a grounded
AFM-tip exerting a comparatively high-loading force of
approximately 2.5 lN was performed. All fast scan axes arealong [1�10]o including the compressive scan. In Fig. 5(a),super-macro-domain boundaries are indicated by white
solid lines in the selected area (the dashed box). After scan-
ning the dashed-line region with a compressive force of
2.5 lN, the in-plane PFM image (Fig. 5(b)) was capturedunder the same conditions as before, and allows a direct
FIG. 4. Thickness-dependent domain widths. (a)–(d) In-plane PFM images
at the selected WO3 thicknesses. The interval between neighboring bright
(or dark) contrasts corresponds to the macro-domain width. It has a tendency
to be wider with increasing film thickness. Bright (dark) contrast represents
in-plane piezoresponse pointing to the positive (negative) vertical direction.
Tip orientations during measurements are depicted at the right corner. All
the scale bars indicate 500 nm. (Inset) The Fourier transform of the in-plane
PFM image. All the Fourier transform images have the same size and the
unit of tick labels in (d) is lm�1. (e) The double logarithmic plot of themacro-domain (m) and fine-domain (f) widths versus the film thickness.Because of the AFM resolution limit, fine-domain widths of four samples
were obtained through AFM images (solid blue circles), and the others from
the horizontal satellite peaks observed in the H-scans of x-ray diffraction
(open blue circles).
FIG. 5. Domain reorientation induced by the AFM tip-based nano-mechani-
cal force. (a) In-plane PFM image in the as-grown state of a 53-nm-thick
WO3 film. The orientation of AFM tip during the experiment is expressed
by the tip cartoon. Fast scan axis of the tip was along [1�10]o and the slowscan direction was [001]o. Dashed-line box represents the area where the
scan with a loading force of 2.5 lN was performed. White solid lines aresuper-macro-domain boundaries. As shown in the (b) in-plane PFM images,
macro-domains were rearranged by the mechanical force. All the white scale
bars indicate 500 nm.
252904-4 Yun et al. Appl. Phys. Lett. 107, 252904 (2015)
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions. IP: 143.248.11.147 On: Wed, 17 Feb 2016 02:28:03
comparison with initially observed domain-microstructure.
Within the area where a high mechanical pressure was
applied, the super-macro-domains and their boundaries
were displaced, which means mechanical-force induced
rearrangements of macro-domains (Fig. 5(b)). Because the
irregular-shaped bundles do not fit with each other exactly,
some local domains and their boundaries near the edges of
the super-macro-domains could be highly strained. The
more unstable boundaries are likely to be rearranged toward
the stable state.
In summary, the comprehensive structural study of epi-
taxial WO3 thin films on YAlO3 substrates provided useful
insights into their inter-relevance among the monoclinic cell,
four-fold mosaic tilt, misfit strain, and the emergence of a
ferroelastic hierarchical twin structure (including fine-,
macro-, super-macro-domains).
This work was supported by the National Research
Foundation of Korea Grant funded by the Korean
Government (Contract Nos. NRF-2014R1A2A2A01005979
and NRF-2013S1A2A2035418) and the Global Frontier
Hybrid Interface Materials of the NRF of Korea funded by
the Korea Government (2013M3A6B1078872).
1A. Saxena and A. Planes, Mesoscopic Phenomena in MultifunctionalMaterials (Springer, Berlin, 2014), Chap. 8.
2G. Catalan, J. Seidel, R. Ramesh, and J. F. Scott, Rev. Mod. Phys. 84, 119(2012).
3S. Y. Yang, J. Seidel, S. J. Byrness, P. Shafer, C.-H. Yang, M. D. Rossell,
P. Yu, Y.-H. Chu, J. F. Scott, J. W. Ager III, L. W. Martin, and R.
Ramesh, Nat. Nanotechnol. 5, 143 (2010).4S. Conti, S. Muller, A. Poliakovsky, and E. K. H. Salje, J. Phys.: Condens.
Matter 23, 142203 (2011).5T. Lottermoser and M. Fiebig, Phys. Rev. B 70, 220407 (2004).6K.-E. Kim, B.-K. Jang, Y. Heo, J. H. Lee, M. Jeong, J. Y. Lee, J. Seidel,
and C.-H. Yang, NPG Asia Mater. 6, e81 (2014).7Y. Gu, M. Li, A. N. Morozovska, Y. Wang, E. A. Eliseev, V. Gopalan,
and L.-Q. Chen, Phys. Rev. B 89, 174111 (2014).8D. Lee and T. W. Noh, Philos. Trans. R. Soc. A 370, 4944 (2012).9K. Chu, B.-K. Jang, J. H. Sung, Y. A. Shin, E.-S. Lee, K. Song, J. H. Lee,
C.-S. Woo, S. J. Kim, S.-Y. Choi, T. Y. Koo, Y.-H. Kim, S.-H. Oh, M.-H.
Jo, and C.-H. Yang, Nat. Nanotechnol. 10, 972 (2015).10G. Catalan, A. Lubk, A. H. G. Vlooswijk, E. Snoeck, C. Magen, A.
Janssens, G. Rispens, G. Rijnders, D. H. A. Blank, and B. Noheda, Nat.
Mater. 10, 963 (2011).11J. Seidel, L. W. Martin, Q. He, Q. Zhan, Y.-H. Chu, A. Rother, M. E.
Hawkridge, P. Maksymovych, P. Yu, M. Gajek, N. Balke, S. V. Kalinin,
S. Gemming, F. Wang, G. Catalan, J. F. Scott, N. A. Spaldin, J. Orenstein,
and R. Ramesh, Nat. Mater. 8, 229 (2009).12W. Wu, Y. Horibe, N. Lee, S.-W. Cheong, and J. R. Guest, Phys. Rev.
Lett. 108, 077203 (2012).13S. Farokhipoor and B. Noheda, Phys. Rev. Lett. 107, 127601 (2011).14Y. Kim, M. Alexe, and E. K. H. Salje, Appl. Phys. Lett. 96, 032904
(2010).15A. Aird and E. K. H. Salje, J. Phys.: Condens. Matter 10, L377 (1998).16A. Aird and E. K. H. Salje, Eur. Phys. J. B 15, 205 (2000).17M. Calleja, M. T. Dove, and E. K. H. Salje, J. Phys.: Condens. Matter 13,
9445 (2001).18R. Brusetti, P. Haen, and J. Marcus, Phys. Rev. B 65, 144528 (2002).19S. Reich, G. Leitus, R. Popovitz-Biro, A. Boldbourt, and S. Vega,
J. Supercond. Nov. Magn. 22, 343 (2009).20P. M. Wu, C. Hart, K. Luna, K. Munakata, A. Tsukada, S. H. Risbud, T.
H. Geballe, and M. R. Beasley, Phys. Rev. B 89, 184501 (2014).21Y. Levi, O. Millo, A. Sharoni, Y. Tsabba, G. Leitus, and S. Reich,
Europhys. Lett. 51, 564 (2000).22Z. Barkay, E. Grunbaum, G. Leitus, and S. Reich, J. Supercond. Nov.
Magn. 21, 145 (2008).23R. S. Crandall and B. W. Faughnan, Appl. Phys. Lett. 28, 95 (1976).24B. W. Faughnan and R. S. Crandall, Appl. Phys. Lett. 31, 834 (1977).25C. G. Granqvist, Sol. Energy Mater. Sol. Cells 60, 201 (2000).26M. Penza, M. A. Tagliente, L. Mirenghi, C. Gerardi, C. Martucci, and G.
Cassano, Sens. Actuators, B 50, 9 (1998).27X. L. Li, T. J. Lou, X. M. Sun, and Y. D. Li, Inorg. Chem. 43, 5442
(2004).28X. Leng, J. Pereiro, J. Strle, A. T. Bollinger, and I. Bozovic, APL Mater.
3, 096102 (2015).29P. Tagtstrom and U. Jansson, Thin Solid Films 352, 107 (1999).30S. C. Moulzolf, L. J. LeGore, and R. J. Lad, Thin Solid Films 400, 56
(2001).31A. Garg, J. A. Leake, and Z. H. Barber, J. Phys. D: Appl. Phys. 33, 1048
(2000).32See supplementary material at http://dx.doi.org/10.1063/1.4938396 for the
materials synthesis and experimental methods including the process of
finding macro-domain and fine-domain widths.33P. M. Woodward, A. W. Sleight, and T. Vogt, J. Solid State Chem. 131, 9
(1997).34K. R. Locherer, I. P. Swainson, and E. K. H. Salje, J. Phys.: Condens.
Matter 11, 6737 (1999).35S. I. Hamazaki, N. Tashiro, Y. Fukurai, F. Shimizu, M. Takashige, and S.
Kojima, Ferroelectrics 219, 183 (1998).36R. Diehl, G. Brandt, and E. Salje, Acta Cryst. B 34, 1105 (1978).37C.-S. Woo, J. H. Lee, K. Chu, B.-K. Jang, Y. B. Kim, T. Y. Koo, P. Yang,
Y. Qi, Z. Chen, L. Chen, H. C. Choi, J. H. Shim, and C.-H. Yang, Phys.
Rev. B 86, 054417 (2012).38A. Aird, M. C. Domeneghetti, F. Mazzi, V. Tazzoli, and E. K. H. Salje,
J. Phys.: Condens. Matter 10, L569 (1998).39C. Kittel, Rev. Mod. Phys. 21, 541 (1949).40G. Catalan, H. B�ea, S. Fusil, M. Bibes, P. Paruch, A. Barth�el�emy, and J. F.
Scott, Phys. Rev. Lett. 100, 027602 (2008).41P. Paruch, T. Giamarchi, and J.-M. Triscone, Phys. Rev. Lett. 94, 197601
(2005).
252904-5 Yun et al. Appl. Phys. Lett. 107, 252904 (2015)
Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions. IP: 143.248.11.147 On: Wed, 17 Feb 2016 02:28:03
http://dx.doi.org/10.1103/RevModPhys.84.119http://dx.doi.org/10.1038/nnano.2009.451http://dx.doi.org/10.1088/0953-8984/23/14/142203http://dx.doi.org/10.1088/0953-8984/23/14/142203http://dx.doi.org/10.1103/PhysRevB.70.220407http://dx.doi.org/10.1038/am.2013.72http://dx.doi.org/10.1103/PhysRevB.89.174111http://dx.doi.org/10.1098/rsta.2012.0200http://dx.doi.org/10.1038/nnano.2015.191http://dx.doi.org/10.1038/nmat3141http://dx.doi.org/10.1038/nmat3141http://dx.doi.org/10.1038/nmat2373http://dx.doi.org/10.1103/PhysRevLett.108.077203http://dx.doi.org/10.1103/PhysRevLett.108.077203http://dx.doi.org/10.1103/PhysRevLett.107.127601http://dx.doi.org/10.1063/1.3292587http://dx.doi.org/10.1088/0953-8984/10/22/003http://dx.doi.org/10.1007/PL00011037http://dx.doi.org/10.1088/0953-8984/13/42/305http://dx.doi.org/10.1103/PhysRevB.65.144528http://dx.doi.org/10.1007/s10948-009-0443-3http://dx.doi.org/10.1103/PhysRevB.89.184501http://dx.doi.org/10.1209/epl/i2000-00375-2http://dx.doi.org/10.1007/s10948-008-0309-0http://dx.doi.org/10.1007/s10948-008-0309-0http://dx.doi.org/10.1063/1.88653http://dx.doi.org/10.1063/1.89566http://dx.doi.org/10.1016/S0927-0248(99)00088-4http://dx.doi.org/10.1016/S0925-4005(98)00149-Xhttp://dx.doi.org/10.1021/ic049522whttp://dx.doi.org/10.1063/1.4930214http://dx.doi.org/10.1016/S0040-6090(99)00379-Xhttp://dx.doi.org/10.1016/S0040-6090(01)01447-Xhttp://dx.doi.org/10.1088/0022-3727/33/9/303http://dx.doi.org/10.1063/1.4938396http://dx.doi.org/10.1006/jssc.1997.7268http://dx.doi.org/10.1088/0953-8984/11/35/312http://dx.doi.org/10.1088/0953-8984/11/35/312http://dx.doi.org/10.1080/00150199808213515http://dx.doi.org/10.1107/S0567740878005014http://dx.doi.org/10.1103/PhysRevB.86.054417http://dx.doi.org/10.1103/PhysRevB.86.054417http://dx.doi.org/10.1088/0953-8984/10/33/002http://dx.doi.org/10.1103/RevModPhys.21.541http://dx.doi.org/10.1103/PhysRevLett.100.027602http://dx.doi.org/10.1103/PhysRevLett.94.197601