pressure induced band gap narrowing and phase transitions...
TRANSCRIPT
1 © 2020 IOP Publishing Ltd Printed in the UK
Journal of Physics: Condensed Matter
Pressure induced band gap narrowing and phase transitions in Dy2Ti2O7
Xin Li1,2 , Zhipeng Yan2, Saqib Rahman2, Jinbo Zhang2,3 , Ke Yang4, Xiaodong Li5 , Jaeyong Kim6,10 , Xuefeng Sun7,8,10 and Lin Wang2,9,10
1 Department of Physics, Fudan University, Shanghai 200433, People’s Republic of China2 Center for High Pressure Science and Technology Advanced Research, Shanghai 201203, People’s Republic of China3 College of Physical Science and Technology, Yangzhou University, Yangzhou 225002, People’s Republic of China4 Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai 201203, People’s Republic of China5 Institute of High Energy Physics, Chinese Academy of Sciences, Beijing 100049, People’s Republic of China6 Department of Physics, HYU-HPSTAR-CIS High Pressure Research Center, Hanyang University, Seoul, 04763, Republic of Korea7 Department of Physics, Hefei National Laboratory for Physical Sciences at Microscale, and Key Laboratory of Strongly-Coupled Quantum Matter Physics (CAS), University of Science and Technology of China, Hefei, Anhui 230026, People’s Republic of China8 Institute of Physical Science and Information Technology, Anhui University, Hefei, Anhui 230601, People’s Republic of China9 Center for High Pressure Science (CHiPS), State Key Laboratory of Metastable Materials Science and Technology, Yanshan University, Qinhuangdao, Hebei 066004, People’s Republic of China
E-mail: [email protected] (L W), [email protected] (X F S) and [email protected] (J K)
Received 2 October 2019, revised 5 January 2020Accepted for publication 24 January 2020Published 25 February 2020
AbstractAn isostructural phase transition and a cubic to tetragonal phase transition in spin-frustrated pyrochlore Dy2Ti2O7 (DTO) at 35 GPa and 46 GPa, respectively, were detected by x-ray diffraction (XRD) and Raman spectroscopy studies at room temperature under high pressure. The band gap of DTO gradually increased with the increasing pressure up to 20 GPa, after which it narrowed significantly around 35 GPa, and further manifested with a slight discontinuity around 46 GPa. Below 35 GPa, the evolution of band gap with pressure maintained a close relationship with the Ti–O bond length and the distortion of TiO6 octahedron. The remarkable narrowing of the band gap around 35 GPa implied the changes of the band structure and confirmed the isostructural phase transition. These findings provide an in-depth understanding of the evolution of the structure and the band gap of DTO under high pressure.
Keywords: pyrochlore, high pressure, phase transition, band gap
S Supplementary material for this article is available online
(Some figures may appear in colour only in the online journal)
X Li et al
Printed in the UK
215401
JCOMEL
© 2020 IOP Publishing Ltd
32
J. Phys.: Condens. Matter
CM
10.1088/1361-648X/ab6f82
Paper
21
Journal of Physics: Condensed Matter
IOP
10 Author to whom any correspondence should be addressed.
2020
1361-648X
1361-648X/ 20 /215401+7$33.00
https://doi.org/10.1088/1361-648X/ab6f82J. Phys.: Condens. Matter 32 (2020) 215401 (7pp)
X Li et al
2
1. Introduction
Pyrochlores with the general formula of A2B2O7 (both A and B cations form two distinct interpenetrating sublattices of corner-sharing tetrahedra) are usually considered as geometri-cally-frustrated magnetic materials [1]. Due to their numerous compositions and structural flexibility, pyrochlores are widely used as electrolytes in solid oxide fuel cells and also as advanced nuclear materials for immobilization of actinides and inert matrix fuel [2, 3]. Some novel ground state proper-ties including spin ice state, spin liquid state, spin glass state, superconductivity, and metal–insulator transition have been discovered in these materials [1, 4–6].
In an ordered pyrochlore structure, all atoms occupy spe-cial positions except the oxygen site at 48f . The structural stability of pyrochlore mainly depends on the ratio of cation radii (rA/rB) [1]. An ordered pyrochlore is generally formed when rA/rB is greater than 1.46, whereas a disordered fluo-rite structure can be obtained when rA/rB is smaller than 1.46. Furthermore, ordered pyrochlore structures are sensitive to some extreme conditions, such as low temperature, high temperature, high pressure, and high radiation. An ordered pyrochlore structure is often transformed into a disordered fluorite structure (amorphous state) under a high-radiation field [7, 8]. Moreover, the anion disorder transition is gener-ally observed in cubic pyrochlore structures under high pres-sure [9, 10]. Saha et al [11] studied the effects of pressure on Gd2Ti2O7 through Raman scattering and x-ray diffraction (XRD) and observed that two Raman modes became negli-gibly small beyond 9 GPa, indicating the pressure-induced subtle distortion of the lattice. Zhao et al [12] observed a pres-sure-enhanced insulating state with the trigonal distortion and relaxation in Eu2Sn2O7. These rare-earth pyrochlores usually experience lattice deformation in their cubic pyrochlore phase due to the variation in fractional parameter (x) of O48f under high pressure.
DTO has attracted tremendous scientific interests due to its ‘spin-ice’ magnetic ground state [13] and flexible struc-tural stability at extreme conditions. Low-temperature Raman spectr oscopy and XRD studies reveal a ‘subtle’ structural trans-formation in DTO near 110 K [14]. High-pressure (HP) XRD studies reveal that DTO undergoes an ordered to dis ordered phase transition (from cubic pyrochlore to orthorhombic cotunnite (Pnma)), and transition pressure maintains a close relationship with the existing defects in the starting material [15, 16]. However, it is very difficult to determine whether the final material possesses a cotunnite structure as its all peaks are almost identical to those of a cubic pyrochlore at high pressure [17]. Therefore, the HP structure of DTO has not yet been explored. In the present work, the structural evolution of DTO at room temperature under high pressure was revealed by XRD and Raman spectroscopy. In addition, the variation in the band gap of DTO under high pressure was investigated by UV–vis absorption spectroscopy.
2. Experimental details
High-quality DTO samples were grown in an optical floating-zone furnace equipped with four 1000 W halogen lamps (Crystal System Incorporation, Japan) [18]. The samples were loaded into a diamond anvil cell (DAC) inside a tung-sten gasket and ruby chips were used for pressure calibra-tion. Neon was used as the pressure medium for XRD and Raman spectroscopy studies, whereas silicon oil was used for UV–vis measurements. XRD patterns were collected on the 4W2 beamline of the Beijing Synchrotron Radiation Facility (BSRF) at a monochromatic wavelength of 0.6199 Å. Raman measurements with a resolution of 1 cm−1 were performed in a Renishaw Raman spectrometer with a laser excitation wave-length of 532 nm. UV–vis absorption spectra were measured by the UV–vis spectrometer with a UV–VIS-NIR light source, absorption optical system and a CCD detector which can receive the signal with a wavelength from 180 nm to 850 nm.
3. Results
3.1. XRD of DTO under high pressure
The structural stability of DTO was investigated at room temperature under high pressure. Figure 1(a) displays the detected XRD patterns of DTO. At lower pressures, XRD patterns of a stable cubic pyrochlore structure were observed. All diffraction peaks shifted toward higher angles with the increasing pressure up to 45.9 GPa (no peaks disappeared or no additional peaks appeared). However, at 45.9 GPa, a small peak appeared near the (2 2 2) peak at 2θ = 13.22°, thus implying the formation of a HP phase. With further compres-sion, three more peaks appeared between the (1 1 1) and the (3 1 1) peaks, and it indicates the degeneration of the cubic phase. The diffraction patterns near the (1 1 1) and the (2 2 2) peaks are presented in figure 1(b). A clear splitting of these two peaks was noticed, and the pyrochlore structure almost disappeared and transformed into a HP phase at 58.3 GPa. An orthorhombic structure (a = 13.1332 Å, b = 5.5985 Å, c = 5.4454 Å) and a tetragonal structure (a = b = 6.5712 Å, c = 9.7314 Å) could be indexed by the new peaks at 58.3 GPa. We compared these two phases with XRD pattern at 58.3 GPa in figure S1 (stacks.iop.org/JPhysCM/32/215401/mmedia) and it can be seen that the tetragonal structure matches well with the XRD pattern. In our studies, we used the x-ray with a longer wavelength of 0.6199 Å than 0.4133 Å in previous studies [17] which increases the diffraction resolution so the diffraction rings could be separated and some possible over-lapped peaks could be distinguished clearly. Furthermore, the maximum pressure in our XRD study is higher than in previous studies [17] so the phase transition is almost com-plete. Therefore, a different phase transition from previous studies in DTO was observed and the high pressure phase was suggested to be with the tetragonal structure. This cubic
J. Phys.: Condens. Matter 32 (2020) 215401
X Li et al
3
to tetragonal phase transition was reversible when the pres-sure was released (figure 1(a)).
The XRD patterns of DTO below 45.9 GPa were well refined with the cubic structure in the GSAS program based on the Reitveld method (figure 2(a)). Although the cubic pyrochlore structure remained stable up to 45.9 GPa, the pressure-induced variation in d-spacing of diffraction pat-terns displayed a change in the slope at about 35 GPa (marked by red and blue lines) (figure 2(b)). The third-order Birch–Murnaghan equation of state was used to fit the pressure versus volume data obtained below 35 GPa, and the values of
B0 and B′0 were calculated as 144 GPa and 14.9, respectively.
The large B′0 may be due to the different distortions of TiO6
octahedron at different pressures and this abnormal com-pressibility of pyrochlore under high pressure has also been reported in Eu2Sn2O7 with a B′
0 of 8 [12] and Gd2Ti2O7 with a B′
0 of 6.9 [11]. But when the fitting curve was extended to 50 GPa, the experimental data deviated from the curve obvi-ously from 35 GPa to 45.9 GPa (figure 3(b)). We also use the third-order Birch–Murnaghan equation of state to fit the pres-sure versus lattice parameter before 35 GPa and extend the fitted curve to 50 GPa, and similar change was observed above
Figure 1. (a) Selected XRD patterns of Dy2Ti2O7 at high pressures. The cubic pyrochlore structure can be stable below 45.9 GPa. Above 45.9 GPa, a tetragonal high pressure structure starts to form and this phase transition is almost complete after 58.3 GPa (peaks marked with the star are from neon). The structure of quenched sample agrees with the cubic pyrochlore structure. (b) Detail changes of peaks (1 1 1) and (2 2 2) above 45.9 GPa.
Figure 2. (a) Plots of refined XRD patterns of Dy2Ti2O7 measured at 2 and 44 GPa, respectively. The difference between the observed (black-open cirle) and the calculated (red-solid line) data is shown below the plots in a green line; (b) Pressure dependence of d-spacings of (1 1 1), (3 1 1), (2 2 2) and (4 0 0).
J. Phys.: Condens. Matter 32 (2020) 215401
X Li et al
4
35 GPa (figure 3(a)). As these XRD patterns were obtained using neon as the pressure medium which provided a satisfac-tory hydrostatic pressure condition in the proposed pressure range, it can be speculated that these observed changes imply an isostructural phase transition at about 35 GPa.
3.2. Raman spectra of DTO under high pressure
In situ Raman measurements were executed to understand local structural changes in DTO. Figure 4 exhibits the Raman spectra of DTO under high pressure. Six Raman modes at 213 cm−1 (F2g), 310 cm−1 (F2g), 325 cm−1 (Eg), 518 cm−1 (A1g), 547 cm−1 (F2g), and 701 cm−1 (F2g) were observed at 0.9 GPa [19, 20]. As the pressure increased, the modes at 213 cm−1, 547 cm−1, and 701 cm−1 (corresponding to the interactions between Ti and O48f ) gradually became weaker and, finally, disappeared at high pressures. When the pressure reached 27.1 GPa, a broad and weak peak (marked by stars) appeared at about 700 cm−1 was attributed to distortions to the oxygen positions of the TiO6 octahedron [13]. Moreover, three more peaks (marked by triangles) appeared at around 200 cm−1 above 35 GPa and shifted slowly with the further increase in pressure. From the Raman studies of DTO at low temperatures [19], they revealed that the mode near 200 cm−1 was commonly assigned as F2g phonon and a disorder-induced Raman active mode involving Ti4+ vibrations [19]. Therefore, these changes are related with TiO6 octahedron and indicate the local structure change of TiO6 octahedron (without any change in lattice periodicity), thus confirming the isostructural phase transition at 35 GPa. With further compression above 42.4 GPa, more Raman modes started to appear. Furthermore, above 45.7 GPa, two bands at 400 cm−1 and 600 cm−1 corresponding to the O′–Dy–O′ bending mode and the Dy–O′ stretching mode, respectively, lost their inten-sities significantly and became comparable with other Raman modes at high pressures; it indicates that DTO started to lose its cubic pyrochlore structure. Hence, the weakening of the
O′–Dy–O′ bending mode and the Dy–O′ stretching mode and the appearance of new Raman modes provide evidence of phase transition in DTO. When the pressure was released to 1 atm, DTO went back to its cubic structure, thus suggesting a reversible phase transition.
Figure 3. Variation of the (a) lattice parameter and (b) the unit cell volume with pressure. The red lines are the fitting results of pressure versus lattice parameter and volume by 3rd Birch–Murnaghan equation of state.
Figure 4. Raman spectra of Dy2Ti2O7 at different pressures during compression. The top spectrum is collected at the ambient pressure after decompressing from 54.4 GPa. Raman peaks marked with triangles and stars represent the phonon modes involving the vibrations of Ti4+ ions and the distortion of TiO6 octahedron, respectively.
J. Phys.: Condens. Matter 32 (2020) 215401
X Li et al
5
3.3. The evolution of band gap of DTO under high pressure
The band gap of DTO under high pressure was measured by UV–vis absorption spectroscopy. Theoretical studies suggest that the valence and the conduction bands of DTO are gen-erally separated by a direct band gap with the band extrema located at the Γ point [21]. The Tauc plots of the absorption spectra were used to determine the band gap at high pressures. Theoretical studies reveal that the optical absorption strength depends on the difference between the photon energy and the band gap can be described as [22]
αhν = B(hν − Eg)n,
where B is the slope of the Tauc plot in the linear region, α is the absorption coefficient, hʋ is the photon energy, and Eg is the band gap. n possesses different values corresponding to different electron transitions, and n = 1/2 and 2 are usually used for the direct and indirect electron transitions, respec-tively. Figure 5(a) displays the UV–vis absorption spectrum of DTO during compression. Absorption edges manifested with a blue shift with the increasing pressure up to 20.8 GPa, signifying an increase in band gap. After 20.8 GPa, absorp-tion edges followed a red shift up to 54.2 GPa; hence, it suggests a decrease in band gap with the increasing pres-sure. Interestingly, a sudden shift of the absorption edge was observed from 36.5 GPa to 39.4 GPa due to the iso-structural transition, thus implying a close relationship between the iso-structural phase transition and the band structure. The variation in the band gap under high pressure is plotted in figure 5(b). Below 36.5 GPa, the absorption edge was very steep, thus suggesting the presence of a direct band gap in DTO. The evolution of band gap below and above 20.8 GPa occurred due to the change in the Ti–O48f bond length and the distortion of TiO6 octahedron. After 36.5 GPa, the absorption edge became sluggish; it implies a direct to indirect transition of the band gap due to the distortion of the local structure [23].
In the present work, the indirect gap model was used to fit the spectrum above 36.5 GPa, the fitting results of Tauc plots were shown in figure S2. It is discernible from figure 5(b) that the band gap had an obvious narrowing of 1.26 eV from 36.5 GPa to 38.2 GPa. Moreover, the direct model was also used to fit the spectrum (figure S3), and the band gap manifested a narrowing of 0.7 eV from 36.5 GPa to 39.4 GPa. Hence, the narrowing of the band gap expresses the electronic structural transition in DTO at about 35 GPa. With further compression, a slight change in the band gap was observed at about 46 GPa.
4. Discussions
Structural changes in DTO around 35 GPa (the changes in lattice parameters, the appearance of the Raman modes, and the narrowing of the band gap) could be a sign of the Lifshitz transition (an isostructural phase transition corresponding with the electronic structural transition) [24]. In order to explore the mechanism of this isostructural phase transition, the relationship between the structural parameters and the band gap of DTO under high pressure was investigated. In the cubic pyrochlore structure, Ti atoms occupied the 16c (0, 0, 0) positions and formed TiO6 octahedron with six equivalent O48f atoms, whereas Dy atoms occupy the 16d (1/2, 1/2, 1/2) positions, which were surrounded by eight O (6O48f + 2O′) atoms. The bond length of Dy–O48f was found to be much longer than that of Dy–O′. Therefore, the cubic pyrochlore structure was composed of two alter-nating 3D networks—corner-shared TiO6 octahedron and Dy–O′–Dy zig-zags (figure 6(a)). Theoretical calculations revealed that the direct band gap of DTO was located at the Γ point. The Ti 3d states hybridized with the O48f 2p states formed conduction bands near the Fermi level, and above these bands, the Dy 5d states hybridized with the O′ 2p states also formed other conduction bands. At the same
Figure 5. (a) UV–vis absorption spectra of Dy2Ti2O7 during compression. (b) Optical band gap as a function of pressure derived from the absorption spectra. Direct and indirect gap model are used to fit the absorbtion spectrum before and after 36.5 GPa, respectively.
J. Phys.: Condens. Matter 32 (2020) 215401
X Li et al
6
time, the O48f 2p states are contributed to valence bands near the Fermi level (figure 6(b)) [21]. Hence, the inter-actions between Ti and O48f atoms manifested great influ-ences on the band gap of DTO. Moreover, the distortion of TiO6 octahedron and the Ti–O bond length maintained a very close relationship with the band gap of DTO. A strong distortion of TiO6 octahedron and a large Ti–O bond length resulted in a small band gap. As all atoms of the cubic pyro-chlore structure occupied their specific positions (except for the x parameter of O48f atoms), the distortion of the TiO6 octahedron was reflected by the variation in the x parameter. Based on the results of GSAS refinement, the x parameter of O48f atoms and the Ti–O and Dy–O bond lengths of the cubic pyrochlore structure below 45 GPa were analyzed.
It is observable from figure S4(a) that the x parameter of O48f atoms decreased from 0.338 to 0.328 below 6 GPa and then started to vibrate around a value of 0.334 below 20 GPa. When the value of x was equal to 0.3125, TiO6 attained a perfect octahedronl structure. Under high pres sure, the x parameters of DTO were larger than 0.3125, thus indicating the distortion of TiO6 octahedron. The changes in x param-eter below 20 GPa signify that the distortion of TiO6 gradu-ally became weaker during compression. It is also noticeable from figure S4(b) that the Ti–O bond length decreased first below 6 GPa and then reached a relatively stable value below 20 GPa, thereby resulting in an increase in the band gap of DTO. In the proposed pressure range, Ti atoms were sur-rounded by six equivalent O48f atoms, and the lengths of six
Ti–O bonds were equal; therefore, the states of the Ti 3d orbital were highly degenerative. Above 20 GPa, the band gap of DTO started to decrease with the increasing pressure. The x parameter and the bond length of Ti–O48f increased in the pressure range of 28 to 35 GPa, whereas the band gap of DTO decreased gradually, and this change of band gap under high pressure still agrees with the theoretical studies. Above 35 GPa, although the x parameter and the bond length of Ti–O48f increased gradually, the band gap of DTO expe-rienced a sudden drop. This abnormal change of band gap implied the electronic structural transition in DTO under high pressure. Besides, the changes of Raman spectra around 35 GPa also indicate the local structural change in the cubic structure corresponding to the isostructural phase trans-ition. From figure 4, it can be noted that the Raman modes associated with the vibration of Dy and O′ atoms gradually weaken above 46 GPa and additional Raman modes appear, revealing the occurrence of the cubic to tetragonal phase transition. This mechanism is completely different from pre-viously reported phase transitions, in which pyrochlore com-pounds underwent an ordered to disordered phase trans ition (from cubic to an amorphous, defected fluorite, or cotun-nite structure), and no site preferences for different species of cations occurred by the distortion of cations [3, 9, 25]. The as-described phase transition in DTO has been also observed in some pyrochlore materials at low temperatures. Ruff et al [26] studied the structural behavior of Tb2Ti2O7 at low temperatures and noticed a continuous broadening of the
Figure 6. (a) Crytal structure of Dy2Ti2O7. The structure can be viewed as two alternate 3D networks, consisting of corner shared TiO6 octahedron and Dy–O′–Dy zig-zags. (b) The sketch band gap of Dy2Ti2O7 at ambient pressure.
J. Phys.: Condens. Matter 32 (2020) 215401
X Li et al
7
Bragg peaks below 20 K due to a cubic to tetragonal phase transition.
5. Conclusions
In summary, in situ XRD and Raman spectroscopy studies were performed in the present research to analyze the phase transitions of DTO under high pressure. An isostructural phase transition and a cubic to tetragonal phase transition were observed around 35 GPa and 46 GPa, respectively. Both these phase transitions were reversible, and DTO regained its cubic phase structure when the pressure was released. The band gap evolution of DTO under high pressure was inves-tigated by UV–vis absorption spectroscopy. The band gap of DTO increased with the increasing pressure below 20.8 GPa and then started to decrease on compression up to 54.2 GPa. The band gap of DTO had a clear narrowing due to the iso-structural phase transition. With further compression, the interaction between Dy and O48f was enhanced that led to the cubic to tetragonal phase transition. The analysis of the bond lengths of Ti–O48f and Dy–O48f revealed that the distortion of TiO6 octahedron and also the hybridization between the 3d orbital of Ti and the 2p orbital of O48f had profound influ-ences on the band gap of DTO below 35 GPa. As the pressure exceeded 35 GPa, the band structure was changed, which led to the remarkable narrowing of band gap. These results pro-vide an in-depth understanding of the evolution of the struc-ture and the band gap of DTO under high pressure.
Acknowledgments
This work was mainly supported by Natural Science Foun-dation of China (Grant No. 11874076), National Science Associated Funding (NSAF, Grant No. U1530402), and Sci-ence Challenging Program (Grant No. TZ2016001), National Research Foundation of Korea (2016K1A4A3914691). X.F.S. acknowledges support from the National Natural Science Foundation of China (Grant Nos. U1832209 and 11874336), the National Basic Research Program of China (Grant No. 2016YFA0300103), the Innovative Program of Hefei Science Center CAS (Grant No. 2019HSC-CIP001), and Users with Excellence Project of Hefei Science Center CAS (Grant No. 2018HSC-UE012). The authors thank the Beijing Synchro-tron Radiation Facility (BSRF) (4W2 beamline) and Shanghai Synchrotron Radiation Facility (SSRF) (BL15U1 beamline) for the use of the synchrotron radiation facilities.
ORCID iDs
Xin Li https://orcid.org/0000-0001-6190-1716Jinbo Zhang https://orcid.org/0000-0002-1079-249XXiaodong Li https://orcid.org/0000-0002-2290-1198Jaeyong Kim https://orcid.org/0000-0003-1787-3775
Lin Wang https://orcid.org/0000-0003-2931-7629
References
[1] Gardner J S, Gingras M J P and Greedan J E 2010 Rev. Mod. Phys. 82 53–107
[2] Kumar R S, Cornelius A L, Nicol M F, Kam K C, Cheetham A K and Gardner J S 2006 Appl. Phys. Lett. 88 031903
[3] Zhang F X, Wang J W, Lian J, Lang M K, Becker U and Ewing R C 2008 Phys. Rev. Lett. 100 045503
[4] Hanawa M, Muraoka Y, Tayama T, Sakakibara T, Yamaura J and Hiroi Z 2001 Phys. Rev. Lett. 87 187001
[5] Melko R G, den Hertog B C and Gingras M J 2001 Phys. Rev. Lett. 87 067203
[6] Mirebeau I, Goncharenko I N, Cadavez-Peres P, Bramwell S T, Gingras M J P and Gardner J S 2002 Nature 420 54–7
[7] Sickafus K, Minervini L, Grimes R, Valdez J, Ishimaru M, Li F, McClellan K and Hartmann T 2000 Science 289 748–51
[8] Weber W J 2000 Science 289 2051 [9] Zhang F X, Manoun B, Saxena S K and Zha C S 2005 Appl.
Phys. Lett. 86 181906[10] Zhang F X and Saxena S K 2005 Chem. Phys. Lett. 413 248–51[11] Saha S, Muthu D V S, Pascanut C, Dragoe N,
Suryanarayanan R, Dhalenne G, Revcolevschi A, Karmakar S, Sharma S M and Sood A K 2006 Phys. Rev. B 74 064109
[12] Zhao Y, Yang W, Li N, Li Y, Tang R, Li H, Zhu H, Zhu P and Wang X 2016 J. Phys. Chem. C 120 9436–42
[13] Ramirez A P, Hayashi A, Cava R A, Siddharthan R and Shastry B 1999 Nature 399 333–5
[14] Saha S, Singh S, Dkhil B, Dhar S, Suryanarayanan R, Dhalenne G, Revcolevschi A and Sood A K 2008 Phys. Rev. B 78 214102
[15] Xiao H Y, Gao F and Weber W J 2009 Phys. Rev. B 80 212102
[16] Rittman D R, Turner K M, Park S, Fuentes A F, Park C, Ewing R C and Mao W L 2017 Sci. Rep. 7 2236
[17] Rittman D R, Turner K M, Park S, Fuentes A F, Yan J, Ewing R C and Mao W L 2017 J. Appl. Phys. 121 045902
[18] Li Q J, Xu L M, Fan C, Zhang F B, Lv Y Y, Ni B, Zhao Z Y and Sun X F 2013 J. Cryst. Growth 377 96–100
[19] Saha S, Ghalsasi P, Muthu D V S, Singh S, Suryanarayanan R, Revcolevschi A and Sood A K 2012 J. Raman Spectrosc. 43 1157–65
[20] Gupta H C, Neelima Rani S B and Gohel V B 2001 J. Raman Spectrosc. 32 41–4
[21] Nemoshkalenko V V, Borisenko S V, Uvarov V N, Yaresko A N, Vakhney A G, Senkevich A I, Bondarenko T N and Borisenko V D 2001 Phys. Rev. B 63 075106
[22] Coulter J B and Birnie D P 2018 Phys. Status Solidi b 255 1700393
[23] Wang T, Daiber B, Frost J M, Mann S A, Garnett E C, Walsh A and Ehrler B 2017 Energy Environ. Sci. 10 509–15
[24] Occelli F, Farber D L, Badro J, Aracne C M, Teter D M, Hanfland M, Canny B and Couzinet B 2004 Phys. Rev. Lett. 93 095502
[25] Zhang F X, Manoun B and Saxena S K 2006 Mater. Lett. 60 2773–6
[26] Ruff J P, Gaulin B D, Castellan J P, Rule K C, Clancy J P, Rodriguez J and Dabkowska H A 2007 Phys. Rev. Lett. 99 237202
J. Phys.: Condens. Matter 32 (2020) 215401