Structural and elastic properties of CaGeO3 perovskite at high pressures

Physics of the Earth and Planetary Interiors 189 (2011) 151–156

Contents lists available at SciVerse ScienceDirect

Physics of the Earth and Planetary Interiors journal homepage: www.elsevier.com/locate/pepi

Structural and elastic properties of CaGeO3 perovskite at high pressures Xiang Wu a,⇑, Shan Qin a, Ting-Ting Gu a, Jing Yang a, Geeth Manthilake b a b

Key Laboratory of Orogenic Belts and Crustal Evolution, Ministry of Education & School of Earth and Space Sciences, Peking University, Beijing 100871, China Bayerisches Geoinstitut, Universität Bayreuth, Bayreuth D-95440, Germany

a r t i c l e

i n f o

Article history: Received 7 March 2011 Received in revised form 17 August 2011 Accepted 21 August 2011 Available online 26 August 2011 Edited by Kei Hirose Keywords: CaGeO3 Perovskite Post-perovskite Phase transition Elasticity

a b s t r a c t A re-investigation on structural stability of CaGeO3 perovskite was conducted by synchrotron radiation X-ray diffraction with diamond anvil cell up to 48.8 GPa and first-principles calculations based on density functional theory. Both experimental and theoretical results reveal that CaGeO3 perovskite becomes more distorted deviated from the ideal perovskite with increasing pressure, in contradiction to previous experimental results. A phase transition from perovskite to post-perovskite of CaGeO3 is theoretically predicted to occur at 36–44 GPa, which is experimentally observed at 43 GPa quenched from high temperature. Elastic behavior of CaGeO3 at high pressures is similar to those of MgSiO3. A large positive jump (+4%) of shear velocity in CaGeO3 is predicted across the phase-transition pressure. Crown Copyright Ó 2011 Published by Elsevier B.V. All rights reserved.

1. Introduction The most abundant mineral component (Mg, Fe)SiO3 is perovskite (Pv) phase with space group Pbnm and Z = 4 in the Earth’s lower mantel. A phase transition of MgSiO3 from Pv to post-perovskite (PPv, Cmcm and Z = 4) observed around 125 GPa and 2500 K has particular implications for the Earth’s lowermost mantel with seismologically unusual properties (Murakami et al., 2004; Oganov and Ono, 2004; Tsuchiya et al., 2004a). Therefore, exploring the properties of this novel phase (PPv) becomes a hot topic in ABO3-type perovskite with a lower phase-transition pressure. So far, several perovskites have been observed experimentally to transform into PPv, such as MgGeO3, MnGeO3 and CaSnO3 (Hirose et al., 2005; Tateno et al., 2006, 2010; Ito et al., 2010). Theoretically, PPv phase is predicted to occur in a lot of perovskites, such as CaTiO3, CdTiO3 and CdGeO3 (Wu et al., 2005; Fang and Ahuja, 2006). Calcium germinate (CaGeO3) with larger size of Ca and Ge cations is believed to be one suitable low-pressure analog for MgSiO3, because it undergoes a sequence of pressure-induced phase transitions from triclinic wollastonite to tetragonal garnet, then to orthorhombic Pv (Ross et al., 1986; Ono et al., 2011), which is similar to that of MgSiO3. Several studies on high-pressure structural stability of Pv-CaGeO3 have been conducted by various methods: extended X-ray absorption fine structure (EXAFS) (Andrault and Poirier, 1991), X-ray diffraction (XRD) (Ross and Angel, 1999; Liu et al., 2008; Ono et al., 2011), far-infrared spectroscopy (Lu and Hofmeister, 1994), ultrasonic interferometry (Liu and Li, 2007) ⇑ Corresponding author. Tel.: +86 010 62757892; fax: +86 010 62752996. E-mail address: [email protected] (X. Wu).

and theoretical simulation (Fang and Ahuja, 2006). EXAFS results show that structural distortion degree of Pv-CaGeO3 deviated from ideal cubic perovskite (Pm3m) presented a decrease with increasing pressure, resulting in a phase transition to a tetragonal phase at about 12 GPa (Andrault and Poirier, 1991). High-pressure single-crystal XRD results further supported Pv-CaGeO3 became less distorted with increasing pressure (Ross and Angel, 1999). But the phase transformation was not confirmed from orthorhombic to tetragonal phase in far-infrared experiment up to 24.4 GPa (Lu and Hofmeister, 1994). No any abnormal behavior of Pv-CaGeO3 was observed in recent multi-anvil experiments below 10 GPa and 1650 K (Liu et al., 2008; Ono et al., 2011). By contrast, theoretical calculation indicates that the lattice distortion increases with pressure for Pv-CaGeO3, and it will transform to PPv phase at 55 GPa and 0 K (Fang and Ahuja, 2006). In addition, compressional shear-wave velocities of Pv-CaGeO3 were measured up to 10 GPa at ambient temperature without any abnormal behavior (Liu and Li, 2007). The above available XRD experimental data are obtained below 10 GPa, which limits to identify the change of GeO6 octahedral tilt at high pressure (O’keeffe et al., 1979; Zhao et al., 2004). In addition compared with the information about isothermal equation of state (EoS) of Pv-CaGeO3, little is known of its elasticity at high pressure. A re-investigation at higher pressure, therefore, is required to clarify the high-pressure structural stability and reveal the elasticity of Pv-CaGeO3. In this manuscript we conducted an in situ highpressure XRD experiment up to about 48.8 GPa at room temperature or quenched form high temperature using synchrotron radiation X-ray with Diamond anvil cell (DAC), and underpin the data by density functional theory (DFT) based electronic structure calculations.

0031-9201/$ - see front matter Crown Copyright Ó 2011 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.pepi.2011.08.007

X. Wu et al. / Physics of the Earth and Planetary Interiors 189 (2011) 151–156

2. Materials and methods

Intensity (arb. unit)

a

48.8 GPa

32.5 GPa 13.7 GPa

6

8

10

12

14

16

18

20

22

10

Pv(004)/((220 20))

Au(002)

Pv(121) B2(011) P Pv(202 )

Au(111)

Pv(021)

Pv(020) Pv(112) Pv(200)

B1(002)

*

B1(111)

After releasing pressure to 0.0001 GPa Intensity / arb. unit

PPv022

B2(001)

Pv(111)

43.0 GPa quenched from high temperature

Pv(111)

b

Pv(020)/(112)/(200)

2 (°)

Intensity (arb. unit)

CaGeO3 perovskite samples were synthesized from CaGeO3 powder crystallized with wollastonite structure, which was prepared by mixing CaCO3 and GeO2 in a 1:1 mole ratio. The synthesis of CaGeO3 wollastonite involved two steps, first by slowly increasing the temperature in an atmospheric furnace over 12 h to 1000 °C in order to allow the decarbonation of CaCO3 and then keeping the mixture at 1400 °C for 48 h (Liu et al., 1991). Perovskite samples were then prepared from CaGeO3 wollastonite at 10 GPa and 1100 °C for 3 h. The sample assembly for synthesis consists of a pre-sintered MgO + 5% Cr2O3 octahedron with an edge length of 25 mm. Stepped LaCrO3 furnace was used to minimize the temperature gradient in the cell and the temperature was monitored using W97Re3–W75Re25 thermocouple wires. The assembly was compressed to desired pressure using tungsten-carbide anvils with truncated edge-length of 15 mm. Hot pressing experiments were performed using a KAWAI-type 1000-tonne multi-anvil apparatus at Bayerisches Geoinstitut, Germany. The first run of in situ high-pressure XRD experiment was conducted at AR-NE1 station of Photon Facility (PF), Japan. A 30 lm diameter collimator was applied, and Rigaku image plate was used to collect XRD data with k = 0.4113 Å, where exposure of time was 600 s for every spectrum. A modified four-pin Merrill-Bassett DAC was used to apply the pressure to the samples. The diamond culets have a diameter of 400 lm. A 200-lm diameter sample hole was drilled in Re gasket with an initial thickness of 260 lm, which were pre-indented to 55 lm in thickness. Ne was used as a pressuretransmitting medium. One small ruby sphere (3 lm diameter) was loaded in the center as pressure indicator. The second run was carried out at Beijing Synchrotron Radiation Facility (BSRF), China, with k = 0.6199 Å, where the FWHM of the beam is 26 lm  15 lm and exposure of time was 300 s for every spectrum. Double sides of the sample were covered by Au foil as a laser absorber, and then they were sandwiched by NaCl plates in the 120-lm diameter chamber of Re gasket, which was compressed by a pair of anvils with the 300-lm diameter culets. High temperature was obtained using a portable laser heating system for the DAC, where 1064 nm laser radiation was applied (Dubrovinsky et al., 2009). All collected images were integrated using the Fit2D program in order to obtain conventional one-dimensional diffraction spectra (Hammersley et al., 1996). Lattice parameters were refined by a full-profile model refinement (Le Bail method) implemented in the GSAS software (Toby, 2001). Refined parameters include background coefficients (8 variables), profile parameters (2 variables), and lattice parameters (3 variables). Theoretical simulations performed here are based on DFT with local density approximation (LDA) (Lundqvist and March, 1987) and generalized gradient approximation (PBE-GGA) (Perdew et al., 1996), implemented in the VASP packages (Kresse and Furthmüller, 1996). Three possible candidate phases of CaGeO3 were designed: Pv, PPv and T-Pv (tetragonal perovskite, I4/mcm and Z = 4) phases, respectively. We set the kinetic energy cut-off for plane wave expansion as 500 eV to obtain fully converged results. The Monkhorst– Pack scheme (Monkhorst and Pack, 1976) was used with 4  4  4k-points for Pv and T-Pv phases, and 8  4  4 for PPv phase, which yielded 8k-points for Pv, 6k-points for T-Pv phase, and 16k-points for PPv in the irreducible wedge of the Brillouin zone. The minimum total energy was obtained at a constant volume with the convergence criterion (the different energy/unit cell <0.0001 eV), where lattice parameters and internal coordinates of the atoms were fully relaxed to optimize. The energy-volume relationship was fitted by the third-order Birch–Murnaghan EoS (Birch, 1947). Single crystal elastic constants of Pv- and PPv-CaGeO3 were computed from the stress (r)–strain (e) relations (Karki et al.,

Pv(002)/(110 0)

152

11

12

13

14

2 /

11

12

13

14

15

16

17

18

19

20

21

2 (°) Fig. 1. X-ray diffraction patterns of CaGeO3. (a) Collected at PF with k = 0.4113 Å up to 48.8 GPa and room temperature. The bottom vertical bars represent the positions of the diffraction peaks from Pv-CaGeO3 at 13.7 GPa. (b) Collected at BSRF with k = 0.6199 Å at 43 GPa quenched from high temperature (at least above 2000 K). The inset is the pattern after relaxing pressure to 0.0001 GPa. Pv: perovskite, PPv: post-perovskite; Au: gold; B1 and B2: NaCl.

Table 1 Lattice constants of Pv-CaGeO3 at various pressures. Pressure/GPa

a (Å)

b (Å)

c (Å)

V (Å3)

0.0001 6.3 13.7 16.3 19.4 23.3 27.6 32.5 34.8 36.3 39.5 47.1 48.8

5.2630(2) 5.2161(2) 5.1485(1) 5.1367(5) 5.1282(3) 5.1112(3) 5.0835(4) 5.0542(3) 5.0370(4) 5.0244(6) 5.0124(7) 4.9750(6) 4.9648(10)

5.2841(2) 5.2586(10) 5.2067(14) 5.1788(13) 5.1579(16) 5.1392(12) 5.1311(12) 5.1151(21) 5.1016(6) 5.0914(9) 5.0793(7) 5.0491(8) 5.0346(8)

7.4395(2) 7.3588(2) 7.2746(4) 7.2595(5) 7.2384(4) 7.2069(7) 7.1765(10) 7.1442(7) 7.1251(8) 7.1192(9) 7.1031(9) 7.0494(6) 7.0231(10)

206.89(1) 201.84(2) 194.99(5) 193.12(4) 191.46(5) 189.31(3) 187.19(3) 184.70(10) 183.09(4) 182.12(2) 180.84(4) 177.08(3) 175.55(8)

2001). In orthorhombic system, nine independent elastic constants (c11, c12, c13, c22, c23, c33, c44, c55, and c66) are requested to apply six strain tensors (e11, e22, e33, e44, e55, and e66). We applied positive and negative strains of magnitude 1% in order to accurately determine the stresses in the appropriate limit of zero strain.

153

X. Wu et al. / Physics of the Earth and Planetary Interiors 189 (2011) 151–156

3. Results In the first run, the sample was compressed step by step to 48.8 GPa at room temperature, where no phase transition was observed (Fig. 1a). In the second run, the sample was scanned twice by the laser heating at 45 GPa. The XRD pattern measured at 43 GPa quenched from high temperature presents a weak but unambiguous new peak (2h = 12.88°, d = 2.76 Å), denoted by the star symbol (Fig. 1b). This value is close to the d022 of PPv phase obtained from theoretical simulations (2.78 Å at 41 GPa in GGA or 2.72 Å at 45 GPa in LDA). Thus, this new peak is indexed to the (0 2 2) typical diffraction peak of the PPv phase (Fig. 1b). The peak disappears after decompression to ambient pressure (the inset of Fig. 1b), meaning a reversible transition. In this run experiment, CaGeO3 samples were not fully covered by Au. Thus some CaGeO3 samples were not heated during the experimental progress. In addition, the PPv-CaGeO3 was enclosed by gold with strong absorption coefficient. The size of X-ray beam spot on the sample is about 50 lm  30 lm. The above factors cause the pattern with a large Pv-CaGeO3 component and a small PPv-CaGeO3 component (Fig. 1b). Here the density of Pv phase is 5.95 mg/ cm3, while the density of PPv phase is estimated to be 6.00 mg/ cm3 assuming that the axial ratios are the same to those from

Pseudocubic sub-cell parameters (Å)

3.75

ap bp cp

3.70

3.65

3.60

3.55

3.50

a 0

10

20

30

40

50

Pressure (GPa) 220

Exp. (this study) Ross and Angel, 1999 Liu and Li, 2007 LDA (this study) GGA (this study)

3

Unit-cell volume (Å )

210

theoretical calculations (a/b = 0.3056 and c/b = 0.7519 in GGA at 41 GPa, or a/b = 0.3024 and c/b = 0.7438 at 45 GPa in LDA). Thus the density of CaGeO3 shows a jump of 0.8% from Pv to PPv at 43 GPa. Lattice constants of CaGeO3 at various pressures are listed in Table 1, obtained from the first run. Fig. 2a shows the lattice parameters reduced to those of a pseudocubic sub-cell as a function of pressure. The values of ap and cp are very close in the pressure range. It is very hard to obtain the accurate results from the broadened diffraction peaks (Fig. 1a), which is attributed to the deviatoric stress increase. This is why ap and cp show a abnormal trend at high pressure. However, the difference between ap/cp and bp is divergent with pressure increasing, indicating that Pv-CaGeO3 becomes more distortion in contrast to previous experimental results (Andrault and Poirier, 1991; Ross and Angel, 1999). About 20 GPa, P–V data shows a slight discontinuity also caused by the pressure gradient. The isothermal EoS was fitted to the third-order Birch–Murnaghan EoS (Fig. 2b). If the bulk modulus pressure derivative (K00 ) is fixed to 6.1 according to the single crystal experimental value (Ross and Angel, 1999), the bulk modulus (K0) is 178 GPa and the volume per unit cell (V0) is 208.3 Å3, which deviate from experimental values 206.36 Å3 (Sasaki et al., 1983). If K00 is fixed to 4, V0 (206.9 Å3) is consistent with previous experimental values and K0 is slightly larger than pervious experiment results (Ross and Angel, 1999; Liu and Li, 2007). Therefore, the later results compare more favorably with the experimental data and theoretical data (seen below), Table 2. EoS parameters of CaGeO3 from theoretical simulations were obtained and also listed in Table 2. Fig. 3 presents the different enthalpy relative to Pv phase of both LDA and GGA results. It is very clear that T-Pv is metastable with higher energy at high pressure and zero K. A phase transformation from Pv to PPv occurs at 36 GPa in LDA computation and at 44 GPa in GGA computation, which is smaller than 55 GPa also based on GGA reported by Fang and Ahuja (2006), but in consistent with our experimental result. In Pv-CaGeO3 system, the V0 in LDA based computations is smaller by 1.8% than our experimental result, while in GGA calculation overestimated by 5.5% (Table 2). These values are satisfactory and the errors are typical of ab initio DFT methods. The experimental P–V curve lies in the region between P–V curves from LDA and from GGA (Fig. 2b), which supports each other that these data are reliable. But on the whole, the data from LDA calculation are closer to our experimental values. Elastic constants were thus computed only based on LDA. At ambient pressure, the diagonal elastic constants (c11, c22 and c33) of Pv-CaGeO3 are very closer, which means a slight distortion deviated from the ideal perovskite (Fig. 4a). With increasing Table 2 Equation of state parameters for Pv- and PPv-CaGeO3. V0, K0, and K00 are the volume per unit cell at zero pressure, the bulk modulus at V0, and its pressure derivative, respectively.

200

190

Phase

180

b

Pv Pbnm

170 0

10

20

30

40

V0 (Å3)

K0 (GPa)

K00

Method and reference

206.490 (17) 208.3 (7)

194 (2)

6.1

XRD Ross and Angel (1999)

178 (7)

XRD (this study)

206.9 (4) 206.41 (3) –

220(5) 194.6 (11) 220 (10)

6.1 (fixed) 4 (fixed) 6.4 (2) 4 (fixed)

203.09 218.25 218.20 201.97 217.47 216.91

215 173 182 191 160 179

4.2 4.4 – 4.3 4.2 –

50

Pressure (GPa) Fig. 2. Lattice constants of Pv-CaGeO3 vs. pressure. (a) Pseudocubic sub-cell p p parameters of CaGeO3 as a function of pressure. ap = a/ 2, bp = b/ 2, and cp = c/ p 2 where a, b and c are the lattice constants in Pbnm (Table 1). (b) Unit cell volumes at various pressures. Curves are the fit of experimental data (solid squares) and computed data (LDA: open triangles; GGA: open circles) using the third-order Birch–Murnaghan EoS.

PPv Cmcm

XRD (this study) Ultrasonic Liu and Li (2007) EXAFS Andrault and Poirier (1991) LDA (this study) GGA (this study) GGA Fang and Ahuja (2006) LDA (this study) GGA (this study) GGA Fang and Ahuja (2006)

154

X. Wu et al. / Physics of the Earth and Planetary Interiors 189 (2011) 151–156

900

a

Pv C11

Elastic Contants, Cij (GPa)

800

C22 C33

700

PPv C11

600

C22 C33

500

400

300 0

20

40

60

80

100

80

100

80

100

Pressure (GPa) Fig. 3. The calculated different enthalpies relative to Pv phase a function of pressure. The phase-transition pressure from Pv to PPv is 36 GPa for LDA model and 44 GPa for GGA model.

4. Discussion Our XRD experiments and theoretical simulations both show that Pv-CaGeO3 becomes more distorted at high pressure. This result is in contradiction with pervious experimental reports of in situ single-crystal XRD up to 8.6 GPa and of EXAFS up to 24.4 GPa (Ross and Angel, 1999; Andrault and Poirier, 1991), but in consistent with that of recent theoretical computations (Fang and Ahuja, 2006) and our present simulations. The discrepancy reason is mostly attributed to Pv-CaGeO3 structure itself with smaller distortion and stiffer compressibility. Tilting angle U, representing rotation of GeO6 octahedron about pseudo-cubic axes of pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi [1 1 1] obtained from the formula cos1 ð 2a2 =bcÞ, is about 3.86° at

b

Pv C12

Elastic Contants, Cij (GPa)

450

C13

400

C23 PPv

350

C12 300

C13 C23

250 200 150 100 0

20

40

60

Pressure (GPa)

280

c

Pv C44

Elastic Contants, Cij (GPa)

pressure they present a divergent tendency, indicating that Pv-CaGeO3 becomes more distorted. A crossover between c55 and c66 occurs above 35 GPa (Fig. 4c), suggesting a structural instability. In PPv system, the elastic constants exhibit obvious anisotropy, such as c11 and c33 is 1.5 times greater than c22 reflecting significant compressibility in the stacking direction of GeO3 layers parallel to [0 1 0]. The shear c66 (related to the lateral shift of GeO6 layers) is larger by 100 GPa than the c55 (related to the deformation of the GeO6 layers themselves), consistent with those of the layered structures (Tsuchiya and Tsuchiya, 2006). These behaviors are also observed in MgSiO3, MgGeO3, CaIrO3 PPv phases (Tsuchiya and Tsuchiya, 2006, 2007; Usui et al., 2010). The bulk (K) and shear (G) moduli of the aggregate were calculated from the single crystal elastic constants by the Voigt– Reuss–Hill approximation (Hill, 1952), plotted in Fig. 5a. The bulk modulus (202 GPa) of Pv-CaGeO3 at ambient pressure obtained here is in the average of those from the E–V EoS fitting, the P–V EoS fitting, and ultrasonic interferometry (Table 2), providing a cross-check with the compression results. We obtained compressional wave pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi velocity (Vp) and shear wave velocity (Vs) by Vp ¼ ðK þ 4=3GÞ= q pffiffiffiffiffiffiffiffiffi and Vs ¼ G=q (Fig. 5b), where q is the density. Below 10 GPa, the pressure dependence of K, G, and VP of Pv-CaGeO3 shows an excellent agreement with experimental results obtained from ultrasonic interferometry (Liu and Li, 2007), only VS shows a little difference. There is a discontinuous change for Vs of Pv-CaGeO3 above 40 GPa (Fig. 5b), reflecting that Pv phase is unstable at higher pressure (ie. the Pv-to-PPv transition). At the static phase-transition pressure (PT = 36 GPa), the Vp values are almost the same between Pv and PPv, but Vs shows a positive jump of 4%.

500

C55

240

C66 PPv C44

200

C55 C66 160

120

0

20

40

60

Pressure (GPa) Fig. 4. Elastic constants of Pv- and PPv-CaGeO3 as a function of pressure. (a) Longitudinal elastic constants; (b) off-diagonal elastic constants; and (c) shear elastic constants.

ambient condition (Sasaki et al., 1983). It became 3.43° at the highest pressure of 8.6 GPa calculated from high-accurate experimental data by Ross and Angel (1999). The slighter change of U is difficultly distinguished by XRD method with some uncertainty in a narrow pressure range. However the experimental data in a larger pressure range up to 24.4 GPa are with larger measurement errors by EXAFS method. Our present synchrotron radiation XRD experiment extends the pressure range to 48.8 GPa, and we obtained

X. Wu et al. / Physics of the Earth and Planetary Interiors 189 (2011) 151–156

from Brillouin scattering measurement (Murakami et al., 2007)) and that of MgGeO3 (1.02%) (Usui et al., 2010). The D00 layer shows a 2.5–4.0% increase of shear velocity in seismic observations (Wysession et al., 1998), which can not be simply explained using the isotropic Pv-to-PPv transition of MgSiO3 (Murakami et al., 2007). Our results suggest that a large positive jump of Vs across the Pv-to-PPv phase transition can be achieved by varying chemical composition in ABO3-type perovskite. Therefore, the chemical composition effect is important to clarify the unusual properties of the D00 layer, as well as the lattice preferred orientation of PPv phases (Miyagi et al., 2010).

550

Pv K G K (Liu et al., 2007) G (Liu et al., 2007)

500

Aggregate moduli (GPa)

450 400

K

PPv K G

350 300 250

G

200 150

Acknowledgements

a

100 0

20

40

60

80

100

Pressure (GPa) 11.0

VP 10.5

Isotropic wave velocity (km/s)

155

10.0

PPv Pv Pv

VS

5.6

References Liu et al. 2007

5.4

9.5 9.0

VS

VP

4.0%

8.5

5.2

5.0

8.0 4.8 7.5

b

7.0 0

X. Wu and S. Qin acknowledge the financial support of the National Natural Science Foundation of China (Grant No. 41072027). In-situ high-pressure XRD measurements were carried out at PFKEK (proposal No. 2010G041) and BSRF.

20

40

60

80

4.6 100

Pressure (GPa) Fig. 5. The aggregate properties of Pv- and PPv-CaGeO3 vs. pressure. (a) Bulk and shear moduli; (b) compressional and shear isotropic wave velocities.

tilting angle U of 9.69° at this highest pressure (U is equal to 12° at 52 GPa in theoretical simulation). Such large change of tilting angle can be undoubtedly determined by XRD method. In addition, ab initio calculation based on DFT has proved to be a powerful tool for studying structural stability at high pressure, especially for similar structures with small different enthalpy. Here crystal chemistry and elastic properties of Pv-CaGeO3 from theoretical simulations clearly illustrate an increase in structural distortion at high pressure, agreement with present experimental results. Anisotropic PPv-MgSiO3 phase provides particular insight into the seismic anisotropy observed in some regions of the Earth’s lowermost mantel (Tsuchiya et al., 2004b). One thermodynamically stable compound even at ambient condition, PPv-CaIrO3, is firstly suggested to be a good analog of PPv-MgSiO3 (Hirose and Fujita, 2005). But previous studies about CaIrO3 present many different properties, such as axial ratio, compressibility and elastic behavior (Tsuchiya and Tsuchiya, 2007; Lindsay-Scott et al., 2007; Kubo et al., 2008). Quenchable PPv-CaSnO3 is proposed to be a better analog with similar behaviors (DV of about 1.5% at phase transition and the Clapeyron slope of +17 ± 2.0 MPa/K in PPv-CaSnO3; 1.2% and +13.3 ± 1.0 MPa/K in those of PPv-MgSiO3 respectively) (Tateno et al., 2010). For CaGeO3 case, the volume reduction is about 1.7% at the static PT (36 GPa) from theoretical simulations or 0.8% at 43 GPa from experimental results, which is also similar to that of MgSiO3. Surprisingly, a larger positive jump (+4%) of Vs in CaGeO3 occurs across the PT, which is higher than that of MgSiO3 (+1.5% from theoretical calculation (Tsuchiya et al., 2004b), 0.5%

Andrault, D., Poirier, J.P., 1991. Evolution of the distortion of perovskites under pressure – an exafs study of BaZrO3, SrZrO3 and CaGeO3. Phys. Chem. Mineral. 18, 91–105. Birch, F., 1947. Finite elastic strain of cubic crystals. Phys. Rev. 71, 809–824. Dubrovinsky, L., Glazyrin, K., McCammon, C., Narygina, O., Greenberg, E., Ubelhack, S., Chumakov, A.I., Pascarelli, S., Prakapenka, V., Bock, J., Dubrovinskaia, N., 2009. Portable laser-heating system for diamond anvil cells. J. Synchrotron Radiat. 16, 737–741. Fang, C.M., Ahuja, R., 2006. Structures and stability of ABO3 orthorhombic perovskites at the Earth’s mantle conditions from first-principles theory. Phys. Earth Planet. Inter. 157, 1–7. Hammersley, A.P., Svensson, S.O., Hanfland, M., Fitch, A.N., Hausermann, D., 1996. Two-dimensional detector software: from real detector to idealized image or two-theta scan. High Press. Res. 14, 235–248. Hill, R., 1952. The elastic behaviour of a crystalline aggregate. Proc. Phys. Soc. London, Sect. A 65, 349–355. Hirose, K., Fujita, Y., 2005. Clapeyron slope of the post-perovskite phase transition in CaIrO3. Geophys. Res. Lett. 32, L13313. Hirose, K., Kawamura, K., Ohishi, Y., Tateno, S., Sata, N., 2005. Stability and equation of state of MgGeO3 post-perovsktie phase. Am. Miner. 90, 262–265. Ito, E., Yamazaki, D., Yoshino, T., Fukui, H., Zhai, S.M., Shatzkiy, A., Katsura, T., Tange, Y., Funakoshi, K., 2010. Pressure generation and investigation of the postperovskite transformation in MgGeO3 by squeezing the Kawai-cell equipped with sintered diamond anvils. Earth Planet. Sci. Lett. 293, 84–89. Karki, B.B., Stixrude, L., Wentzcovitch, R.M., 2001. High-pressure elastic properties of major materials of Earth’s mantle from first principles. Rev. Geophys. 39, 507–534. Kresse, G., Furthmüller, J., 1996. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 54, 11169–11186. Kubo, A., Kiefer, B., Shim, S.H., Shen, G.Y., Prakapenka, V.B., Duffy, T.S., 2008. Rietveld structure refinement of MgGeO3 post-perovskite phase to 1 Mbar. Am. Miner. 93, 965–976. Lindsay-Scott, A., Wood, I.G., Dobson, D.P., 2007. Thermal expansion of CaIrO3 determined by X-ray powder diffraction. Phys. Earth Planet. Inter. 162, 140– 148. Liu, W., Li, B.S., 2007. Compressional and shear-wave velocities of the polycrystalline CaGeO3 perovskite to 10 GPa. Phys. Rev. B 75 (024107), 1–4. Liu, X., Wang, Y., Liebermann, R.C., Maniar, P.D., Navrotsky, A., 1991. Phase transition in CaGeO3 perovskite: evidence from X-ray powder diffraction, thermal expansion and heat capacity. Phys. Chem. Mineral. 18, 224–230. Liu, W., Kung, J., Wang, L.P., Li, B.S., 2008. Thermal equation of state of CaGeO3 perovskite. Am. Miner. 93, 745–750. Lu, R., Hofmeister, A.M., 1994. Infrared-spectroscopy of CaGeO3 perovskite to 24 GPa and thermodynamic implications. Phys. Chem. Mineral. 21, 78–84. Lundqvist, S., March, N.H., 1987. Theory of the Inhomogeneous Electron Gas. Plenum, New York. Miyagi, L., Kanitpanyacharoen, W., Kaercher, P., Lee, K.K.M., Wenk, H.R., 2010. Slip systems in MgSiO3 post-perovskite: implications for D’’ anisotropy. Science 329, 1639–1641. Monkhorst, H.J., Pack, J.D., 1976. Special Points for Brillouin-zone integrations. Phys. Rev. B 12, 5188–5192. Murakami, M., Hirose, K., Kawamura, K., Sata, N., Ohishi, Y., 2004. Post-perovskite phase transition in MgSiO3. Science 304, 855–858. Murakami, M., Sinogeikin, S.V., Bass, J.D., Sata, N., Ohishi, Y., Hirose, K., 2007. Sound velocity of MgSiO3 post-perovskite phase: a constraint on the D’’ discontinuity. Earth Planet. Sci. Lett. 259, 18–23. Oganov, A.R., Ono, S., 2004. Theoretical and experimental evidence for a postperovksite phase of MgSiO3 in Earth’s D00 layer. Nature 430, 445–448.

156

X. Wu et al. / Physics of the Earth and Planetary Interiors 189 (2011) 151–156

O’Keeffe, M., Hyde, B., Bovin, J.O., 1979. Contribution to the crystal chemistry of orthorhombic perovskites: MgSiO3 and NaMgF3. Phys. Chem. Mineral. 4, 299– 305. Ono, S., Kikegawa, T., Higo, Y., 2011. In situ observation of a garnet/perovskite transition in CaGeO3, Phys. Chem. Miner. doi:10.1007/s00269-011-0446-z. Perdew, J.P., Bruke, K., Ernzerhof, M., 1996. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868. Ross, N.L., Angel, R.J., 1999. Compression of CaTiO3 and CaGeO3 perovskites. Am. Mineral. 84, 277–281. Ross, N.L., Akaogi, M., Navrotsky, A., Susaki, J.I., Mcmillan, P., 1986. Phase-transitions among the CaGeO3 polymorphs (wollastonite, garnet, and perovskite structures) – studies by high-pressure synthesis, high-temperature calorimetry, and vibrational spectroscopy and calculation. J. Geophys. Res. 91, 4685–4696. Sasaki, S., Prewitt, C.T., Liebermann, R.C., 1983. The crystal structure of CaGeO3 perovskite and the crystal chemistry of the GdFeO3-type perovskites. Am. Mineral. 68, 1189–1198. Tateno, S., Hirose, K., Sata, N., Ohishi, Y., 2006. High-pressure behavior of MnGeO3 and CdGeO3 perovskites and the post-perovskite phase transition. Phys. Chem. Mineral. 32, 721–725. Tateno, S., Hirose, K., Sata, N., Ohishi, Y., 2010. Structural distortion of CaSnO3 perovskite under pressure and the quenchable post-perovskite phase as a lowpressure analogue to MgSiO3. Phys. Earth Planet. Inter. 181, 54–59.

Toby, B.H., 2001. EXPGUI, a graphical user interface for GSAS. J. Appl. Cryst. 34, 210– 213. Tsuchiya, T., Tsuchiya, J., 2006. New high-pressure phase relations in CaSnO3. Am. Miner. 91, 1879–1887. Tsuchiya, T., Tsuchiya, J., 2007. Structure and elasticity of Cmcm CaIrO3 and their pressure dependences: ab initio calculations. Phys. Rev. B 76 (144119), 1–4. Tsuchiya, T., Tsuchiya, J., Umemoto, K., Wentzcovitch, R.M., 2004a. Phase transition in MgSiO3 perovskite in the Earth’s lower mantle. Earth Planet. Sci. Lett. 224, 241–248. Tsuchiya, T., Tsuchiya, J., Umemoto, K., Wentzcovitch, R.M., 2004b. Elasticity of post-perovskite MgSiO3. Geophys. Res. Lett. 31 (L14603), 1–4. Usui, Y., Tsuchiya, J., Tsuchiya, T., 2010. Elastic, vibrational, and thermodynamic properties of MgGeO3 postperovskite investigated by first principles simulation. J. Geophys. Res. - Solid Earth 115 (B03201), 1–13. Wu, X., Dong, Y.H., Qin, S., Abbas, M.I., Wu, Z.Y., 2005. First-principles study of pressure-induced phase transition in CaTiO3. Solid State Commun. 136, 416– 420. Wysession, M.E., Lay, T., Revenaugh, J., Williams, Q., Garnero, E., Jeanloz, R., Kellog, L., 1998. The D’’ discontinuity and its implications. In: Gurnis, M., Wysession, M.E., Knittle, E., Buffet, B.A. (Eds.), The Core-Mantle Boundary Region, vol. 28. AGU, Washington, DC, pp. 273–297. Zhao, J., Ross, N.L., Angel, R.J., 2004. New view of the high-pressure behavior of GdFeO3-type perovskites. Acta Crystallogr., Sect. B 60, 263–271.