electronic reprint .fr

channel-cut water-cooled monochromator was used to produce a bright ... given by dashed lines: liquid-to-cubic, cubic-to-tetragonal, tetragonal-to- monoclinic ... parameters and unit-cell volumes are listed in Table 1. The variations of ... mixture of ZrO2, KCl and Pt compressed at 98 GPa after laser annealing above 1200 K.
380KB taille 1 téléchargements 321 vues
research papers Journal of

Applied Crystallography

Phase relations and equation of state of ZrO2 to 100 GPa

ISSN 0021-8898

Received 27 January 2005 Accepted 8 June 2005

Osamu Ohtaka,a,b* Denis Andrault,b Pierre Bouvier,c Emmanuelle Schultzc and Mohamed Mezouard a

Department of Earth and Space Science, Osaka University, Osaka 560-0043, Japan, bLaboratoire des Geomateriaux, IPGP, Universite Paris 7, 75252 Paris, France, cLEPMI, INPG-CNRS, 38402 St Martin d’Heres CEDEX, France, and dESRF, BP220, 38043 Grenoble CEDEX, France. Correspondence e-mail: [email protected]

# 2005 International Union of Crystallography Printed in Great Britain – all rights reserved

The phase relations and equation of state of ZrO2 were investigated up to 100 GPa by means of in situ observation using laser heating in a diamond anvil cell and synchrotron radiation. A cotunnite (PbCl2)-type phase, which appears above 12.5 GPa, is stable to a pressure of 100 GPa and a temperature of 2500 K. No post-cotunnite phase was observed under the present experimental conditions. The unit-cell parameters and the volumes of the cotunnite-type ZrO2 were determined as a function of pressure at room temperature using a laser-annealing technique. The cotunnite-type ZrO2 shows rather isotropic compression. The bulk modulus calculated using the Birch–Murnaghan equations of state is 278 GPa, which suggests that high-density ZrO2 is a candidate for potentially very hard materials. In situ high-temperature experiments performed below 12.5 GPa revealed that a tetragonal fluorite (CaF2)-type phase is stable up to 3000 K, although a cubic fluorite-type phase has been assumed to exist in this high-temperature regime. The result suggests the possibility that stoichiometric ZrO2 does not show the cubic structure up to the melting temperature.

1. Introduction Zirconia (ZrO2), being one of the major components of modern ceramic materials, has been widely used in refractory, high-temperature solid-electrode and structural ceramics. Since it has proven to be the most important toughening agent for ceramics and the toughening mechanism is explained by the phase transitions induced by the stress field (Gupta et al., 1978; Kriven, 1988), the high-pressure behaviour of ZrO2 has attracted much interest from both the experimental and the theoretical point of view. The generalized pressure–temperature phase diagram of pure ZrO2 is shown in Fig. 1 by dashed lines (Ohtaka et al., 1991, 2001; Ondik & McMurdie, 1998). Pure ZrO2 crystallizes in the so-called baddeleyite structure (monoclinic, space group P21/c) under ambient conditions (Howard et al., 1988). At high temperatures, it transforms to a tetragonal (space group P42/nmc) (Teufer, 1962), and then to a cubic fluorite structure (space group Fm3m) (Wyckoff, 1963). On compression, monoclinic ZrO2 shows a sequential transition to two orthorhombic phases (denoted as orthoI and orthoII, respectively). OrthoI (space group Pbca) has a distorted fluorite structure similar to that of monoclinic ZrO2 with sevenfold polyhedral coordination (Ohtaka et al., 1990). OrthoII (space group Pnma) has a cotunnite (PbCl2)-type structure with ninefold polyhedral coordination (Haines et al., 1997). Both the orthoI and the tetragonal phase transform to J. Appl. Cryst. (2005). 38, 727–733

orthoII at around 12.5 GPa (Ohtaka et al., 2001). The tetragonal-to-cubic phase boundary remains unknown, although the Clapeyron slope at room pressure is assumed to be negative (Ondik & McMurdie, 1998). Both experimental and theoretical studies have proposed that orthoII has a remarkably large bulk modulus (Cohen et al., 1988; Leger et al., 1993; Haines et al., 1995; Desgreniers & Lagarec, 1999; Mirgorodsky & Quintard 1999; Lowther et al., 1999; Ohtaka et al., 2001). The reported values based on experiments are in the range from 250 to 450 GPa, suggesting that there are large experimental uncertainties. Most of the previous data were collected by room-temperature compression, which is known to produce huge deviatoric stresses in the sample volume, and consequently the accuracy of both pressure determination and measurement of the unit-cell volumes seems to be deteriorated. Since orthoII is quenchable to ambient conditions, it is a potential candidate for a new superhard material. It is therefore of great importance to obtain reliable compression data for orthoII. Several dihalides, such as PbCl2, BaBr2, BaI2 and BaF2, show post-cotunnite transitions to Co2Si- or Ni2In-type structures (Leger et al., 1995a,b). Among dioxides, however, the cotunnite structure is the final step in the sequence of highpressure transitions to date. Diamond anvil cell (DAC) experiments performed at room temperature have confirmed the stability of orthoII up to 70 GPa (Desgreniers & Lagarec,

electronic reprint

doi:10.1107/S0021889805018145

727

research papers 2. Experimental

Figure 1 Pressure–temperature phase diagram of ZrO2. Reported boundaries are given by dashed lines: liquid-to-cubic, cubic-to-tetragonal, tetragonal-tomonoclinic, and tetragonal-to-orthoI are after Ondik & McMurdie (1998); monoclinic-to-orthoI is after Ohtaka et al. (1991); orthoI-toorthoII and tetragonal-to-orthoII are after Ohtaka et al. (2001). The present results are plotted as symbols. Open circles and closed circles represent orthoII and tetragonal, respectively. Data points at 14.3 GPa and 2200 K, and at 15.2 GPa and 2500 K are the two-phase mixture of orthoII and the tetragonal phase.

1999). It is well know, however, that ionic diffusion is usually required to initiate phase transitions in oxides, and these experiments are hence not relevant to establish the phase diagram at very high pressure. It is therefore of great interest to examine the phase relation and to search for the postcotunnite phase in an extended pressure regime under high temperature. Recent advances in high-pressure and high-temperature experiments using laser heating in a DAC and synchrotron radiation have opened the way for in situ X-ray diffraction measurements under extreme static pressure and temperature conditions (Andrault & Fiquet, 2001). It is now possible to obtain, from a laser-heated sample in a DAC, high-quality powder patterns that are needed for accurate determination of the phase boundaries and pressure–volume–temperature equations of state, as well as for structural refinements (Dewaele et al., 2000). In addition, a range of experiments can also be realised at room temperature after laser annealing of DAC samples, which has been proven to be very efficient in releasing stresses resulting from compression and in overcoming kinetic barriers of phase transitions (Fiquet et al., 2002; Andrault et al., 2003). In this work, we have attempted in situ X-ray diffraction experiments of ZrO2 under pressure using both the sample annealing and in situ laser-heating techniques. The main objectives were (i) to elucidate the phase relations of ZrO2 polymorphs at elevated temperature up to 100 GPa, focusing on the stability field of orthoII and its possible transition to a post-cotunnite phase; (ii) to obtain a reliable equation of state of orthoII in light of several results reported so far. The tetragonal-to-cubic phase boundary in a low-pressure region is also examined.

728

Osamu Ohtaka et al.



ZrO2

The starting material was a fine powder of >99.9% pure ZrO2 provided by Tosoh Co. It was mixed with about 10 mol% of powdered Pt (platinum black), used as an internal pressure calibrant as well as infrared absorber. We used a membranetype DAC with a large optical aperture (Chervin et al., 1995), equipped with beveled diamond anvils with 80–200 mm inner diameter and 300 mm diameter culets. Re gaskets were preindented to a thickness of 40 mm and drilled to a diameter of 30–100 mm depending on the pressure range of interest. Two types of pressure-transmitting medium were used: KCl for laser annealing experiments at very high pressures up to 100 GPa and Ar for in situ high-temperature experiments at medium pressures below 40 GPa, respectively. In the laser-annealing experiments, the sample was thoroughly annealed at each pressure with a multimode infrared YAG laser. An optical microscope was used to scan the laser over the entire sample, and the temperature was estimated from the colour of the sample. The annealing temperature was mostly around 1200 K, which is high enough to release deviatoric stresses accumulated during the room-temperature compression (Weidner et al., 1992), whereas the sample was annealed up to 2500 K at several pressure points to promote any possible phase transitions. X-ray powder diffraction was carried out in an angledispersive mode at the ID30 beamline of the ESRF. A channel-cut water-cooled monochromator was used to ˚ produce a bright monochromatic X-ray beam at 0.3738 A wavelength. Two-dimensional images were recorded on an imaging plate over a 2 interval from 4 to 25 , and angularly integrated with the Fit2D program (Hammersley, 1996). The obtained one-dimensional spectra were analysed in Lebail and Rietveld modes using the general structure analysis program package GSAS to refine the lattice parameters of each phase. The generated pressure was determined from the unit-cell volume of Pt using the equation of state (Jamieson et al., 1992; Holmes et al., 1989). A typical diffraction image and its 2 integration recorded after laser annealing are shown in Fig. 2. Diffraction rings appear slightly spotty as a result of the high-temperature annealing, which promotes grain growth. Nevertheless, the rings remain very clear, indicating that the number of grains is sufficiently large to allow accurate unit-cell refinements. The integrated one-dimensional patterns show sharp diffraction profiles that are almost free from deviatoric stresses. Whether or not deviatoric stresses remain in a sample can be confirmed in the course of Rietveld refinements; if annealing is insufficient and deviatoric stresses exist, several peaks show relatively large standard deviations compared with other peaks in the whole-profile fitting, and accordingly the resulting cell parameters are eliminated in the calculation of the equation of state. The in situ high-temperature and high-pressure experiments were performed by focusing both the monochromatic X-ray beam and an infrared laser beam on the sample in a DAC. The temperature was evaluated by the analysis of thermal emission

electronic reprint

J. Appl. Cryst. (2005). 38, 727–733

research papers spectra of the sample, and the absolute error was estimated to be within 100 K below 2000 K, and within 200 K up to 3000 K. The generated pressure was determined from the unit-cell volume of Pt using the equation of state. When reliable dspacings of Pt were not recorded at high temperature due to the crystal growth or melting, the generated pressure was estimated from the nominal pressure measured at room temperature and the thermal pressure in the laser-heated DAC (Andrault et al., 1998). The pressure change induced by laser heating of the present sample in an Ar pressure-transmitting medium is demonstrated in Fig. 3. On laser heating, diffraction lines from Pt and ZrO2 shift to low 2 angle, whereas those from Ar shift to high 2 angle. The observed peak shift and broad peak profile of Ar, which is clearly shown by the peak around 13.3 , evidences the thermal pressure as well as the large temperature gradient in the pressure-transmitting medium. The nominal pressure measured at room temperature was 6 GPa, and the pressure value was increased to 12 GPa at 2300 K due to the thermal pressure. We used a

value of 3  103 GPa K1 for the corrections of the thermal pressure. More experimental details of the laser-heated DAC at the ESRF are given elsewhere (Andrault et al., 1998; Andrault & Fiquet, 2001).

3. Results and discussion 3.1. Stability of orthoII

Three ranges of the sample annealing experiments were performed up to 100 GPa. No post-cotunnite phase is observed and orthoII is found to be stable to a pressure of 100 GPa and a temperature of 2500 K. The obtained lattice parameters and unit-cell volumes are listed in Table 1. The variations of a/a0, b/b0 and c/c0 as a function of pressure are plotted in Fig. 4, along with our previous determination (data below 25 GPa). The present result is in accord with the previous data collected from well annealed samples in a multianvil-type high-pressure device (Ohtaka et al., 2001). OrthoII has very small linear compressibilities: 6% for the c axis, 7% for the a axis, and 8.5% for the b axis at 100 GPa. This result also indicates that orthoII shows rather isotropic compression. Cotunnite-structured compounds can be divided into several groups based on the lattice parameter ratios a/c and (a + c)/b (setting Pnma) (Jeitschko, 1968). One group, with by far the largest number of representatives [a/c from 0.81 to 0.89; (a + c)/b from 3.3 to 4.0], contains dihalides such as SnCl2, Pb(F,Cl,Br)2, Ba(Cl,Br,I)2 and BaF2 (high-pressure form), as well as hydrides and phosphides. Leger et al. (1995a,b) performed a series of high-pressure experiments of these dihalides and found that there are two types of postcotunnitephase: Co2Si and Ni2In types. In these dihalides, the

Figure 2

Figure 3

X-ray diffraction image (a) and its 2 integration (b) recorded from a mixture of ZrO2, KCl and Pt compressed at 98 GPa after laser annealing above 1200 K.

X-ray diffraction patterns showing the thermal pressure induced by laser heating. Diffraction lines of Ar shift to high 2 angle, indicating that the thermal pressure is introduced.

J. Appl. Cryst. (2005). 38, 727–733

electronic reprint

Osamu Ohtaka et al.



ZrO2

729

research papers Table 1 Obtained lattice parameters and unit-cell volumes. Data at ambient pressure are from Ohtaka et al. (2001). P (GPa)

˚) a (A

˚) b (A

˚) c (A

˚ 3) V (A

a/a0

b/b0

c/c0

V/V0

a/c

(a + c)/b

27.18 31.00 35.52 38.84 40.22 42.13 44.00 46.24 49.02 51.89 55.66 57.84 52.61 55.78 58.18 61.62 64.86 70.32 71.83 83.47 81.54 84.30 87.84 90.95 94.99 97.66 100.59 101.83 Ambient

5.4274 (10) 5.4105 (11) 5.3935 (13) 5.3811 (12) 5.3725 (12) 5.3655 (13) 5.3595 (13) 5.3492 (13) 5.3385 (12) 5.3244 (12) 5.3105 (13) 5.3015 (13) 5.3076 (11) 5.3005 (11) 5.2932 (10) 5.2769 (10) 5.2682 (10) 5.2454 (12) 5.2356 (11) 5.1924 (19) 5.2023 (45) 5.1886 (16) 5.1792 (17) 5.1715 (20) 5.1637 (20) 5.1540 (19) 5.1476 (20) 5.1541 (17) 5.5795

3.2143 (5) 3.2021 (5) 3.1912 (6) 3.1854 (6) 3.1810 (6) 3.1767 (6) 3.1720 (6) 3.1653 (6) 3.1570 (6) 3.1501 (6) 3.1410 (7) 3.1357 (6) 3.1437 (5) 3.1387 (5) 3.1330 (5) 3.1238 (5) 3.1213 (5) 3.1079 (6) 3.1032 (6) 3.0759 (10) 3.0799 (23) 3.0728 (9) 3.0699 (9) 3.0648 (11) 3.0616 (10) 3.0573 (10) 3.0514 (10) 3.0508 (9) 3.3259

6.3116 (11) 6.2963 (12) 6.2839 (13) 6.2712 (13) 6.2632 (13) 6.2575 (13) 6.2523 (14) 6.2444 (13) 6.2333 (13) 6.2239 (13) 6.2116 (13) 6.2012 (13) 6.2132 (12) 6.2071 (11) 6.1975 (11) 6.1841 (11) 6.1782 (11) 6.1555 (13) 6.1472 (12) 6.1044 (20) 6.1106 (47) 6.1026 (18) 6.0937 (18) 6.0915 (22) 6.0822 (21) 6.0760 (19) 6.0651 (21) 6.0629 (18) 6.4655

110.11 (11) 109.08 (12) 108.16 (14) 107.49 (14) 107.04 (14) 106.66 (14) 106.29 (14) 105.73 (14) 105.05 (14) 104.39 (14) 103.61 (14) 103.09 (14) 103.67 (12) 103.26 (12) 102.78 (12) 101.94 (11) 101.59 (11) 100.35 (14) 99.88 (12) 97.50 (20) 97.91 (47) 97.30 (17) 96.89 (18) 96.55 (21) 96.15 (21) 95.74 (19) 95.27 (21) 95.33 (17) 119.98

0.97274 0.96971 0.96667 0.96445 0.96290 0.96165 0.96056 0.95872 0.95680 0.95428 0.95179 0.95018 0.95127 0.95000 0.94868 0.94576 0.94421 0.94012 0.93836 0.93062 0.93240 0.92994 0.92825 0.92687 0.92547 0.92374 0.92259 0.92375 1.0

0.96646 0.96279 0.95949 0.95774 0.95643 0.95515 0.95372 0.95170 0.94923 0.94714 0.94440 0.94281 0.94523 0.94371 0.94199 0.93924 0.93849 0.93444 0.93306 0.92485 0.92604 0.92391 0.92302 0.92148 0.92052 0.91924 0.91746 0.91728 1.0

0.97620 0.97383 0.97192 0.96995 0.96870 0.96783 0.96702 0.96581 0.96408 0.96263 0.96073 0.95912 0.96097 0.96003 0.95855 0.95647 0.95557 0.95205 0.95077 0.92485 0.94511 0.94388 0.94249 0.94216 0.94072 0.93976 0.93807 0.93772 1.0

0.91773 0.90919 0.90145 0.89593 0.89213 0.88898 0.88590 0.88122 0.87560 0.87005 0.86357 0.85922 0.86407 0.86069 0.85660 0.84962 0.84675 0.83636 0.83244 0.81261 0.81604 0.81096 0.80752 0.80469 0.80141 0.93976 0.79401 0.79456 1.0

0.85990 0.85931 0.85831 0.85807 0.85780 0.85745 0.85720 0.85664 0.85645 0.85548 0.85493 0.85492 0.85425 0.85394 0.85408 0.85330 0.85271 0.85215 0.85170 0.85059 0.85137 0.85022 0.84993 0.84897 0.84897 0.84825 0.79401 0.85010 0.86296

3.6521 3.6559 3.6593 3.6581 3.6579 3.6588 3.6607 3.6628 3.6654 3.6660 3.6683 3.6683 3.6647 3.6664 3.6677 3.6689 3.6672 3.6684 3.6680 3.6726 3.6731 3.6745 3.6721 3.6750 3.6733 3.6732 3.6746 3.6767 3.6216

from the values that these ratios move towards upon compression. In contrast to the cotunnite-structured dihalides, the linear compressibility of orthoII follows the order b > a > c, and a/c decreases very slightly (from 0.863 to 0.848) and (a + c)/b increases slightly (from 3.62 to 3.68) by compression up to 100 GPa, as listed in Table 1. Under pressure, these parameter ratios still remain in the regime of the dihalides group, and there are no indications for post-cotunnite transitions. If a/c decreases and (a + c)/b increases by further compression, they move towards a region [a/c is 0.75; (a + c)/b is 4.25] where another group of cotunnite-structured compounds appears; it is interesting to note that this group is composed of (Ti,Zr,Hf)P2 and (Ti,Zr,Hf)As2 (Jeitschko, 1968). OrthoII may be stable up to extremely high pressure. 3.2. EOS of orthoII

Figure 4 Variation of a/a0, b/b0 and c/c0 as a function of pressure plotted along with previously determined data (below 25 GPa) obtained by Ohtaka et al. (2001).

a parameter decreases more rapidly than the b and c parameters under pressure, and thereby the ratio a/c decreases faster than the ratio (a + c)/b. The Co2Si structure has a/c of around 0.7 and (a + c)/b of around 3.3, whereas the Ni2In structure has a/c of around 0.75 and (a + c)/b of 3.1. Leger et al. (1995b) proposed that whether a compound adopts the Co2Si or Ni2In type as the post-cotunnite phase can be inferred

730

Osamu Ohtaka et al.



ZrO2

The volume of orthoII as a function of pressure at ambient temperature is plotted in Fig. 5 along with our previous data. These data are fitted to the third-order Birch–Murnaghan equations of state (Birch, 1947): PðVÞ ¼ 1:5 B0 ½ðV=V0 Þ7=3  ðV=V0 Þ5=3   f1 þ 0:75ðB00  4Þ½ðV=V0 Þ2=3  1g; where B0 and B00 are the room-temperature bulk modulus and its first pressure derivative, respectively. The obtained results are listed in Table 2 together with those previously reported.

electronic reprint

J. Appl. Cryst. (2005). 38, 727–733

research papers Table 2 Parameters of the Birch–Murnaghan equations of state of ZrO2. DAC: diamond anvil cell. MA: multi anvil. EDX: energy-dispersive X-ray diffraction. ADX: angular-dispersive X-ray diffraction. Phase

T (K)

B0 (GPa)

B00

Technique (pressure range)

Reference

OrthoII OrthoII OrthoII OrthoII OrthoII OrthoII OrthoII OrthoII OrthoII OrthoII

298 298 298 1273 298 298 298

278 (11) 267 (3) 265 (10) 265 (15) 332 (8) 306 (10) 444 (15) 305 314 254

3.70 (22) 4 (fixed) 4 (fixed) 4 (fixed) 2.3 3.66 (fixed) 1 (fixed) 4.68 3.66

DAC + ADX (0–100 GPa) DAC + ADX (0–100 GPa) MA + EDX (0–24 GPa) MA + EDX (0–24 GPa) DAC + ADX (0–50GPa) DAC + ADX (0–50 GPa) DAC + EDX (0–70 GPa) Ab initio calculations Ab initio calculations LD calculations

This study This study Ohtaka et al. (2001) Ohtaka et al. (2001) Haines et al. (1995) Haines et al. (1995) Desgreniers & Lagarec (1999) Lowther et al. (1999) Cohen et al. (1988) Mirgorodsky et al. (1999)

Figure 5 Volume of orthoII as a function of pressure at ambient temperature.

First, B0 is refined with fixed B00 at 4 in order to compare it with our previous study (Ohtaka et al., 2001) where B00 could not be refined reasonably because of the limited pressure range of data collection. The calculated B0 of 267 GPa agrees very well with our previous B0 of 265 GPa, indicating that both data sets are consistent and can be treated together. Both B0 and B00 are then refined simultaneously, which yields 278 GPa and ˚ 3 with 3.70 (22). In this study, V0 is also refined: 120.12 (30) A 3 0 0 ˚ with B0 = 4 (fixed). These two B0 = 3.70, and 120.37 (19) A values are the same within the experimental error and are ˚ 3 deterconsistent with those previously reported: 120.65 A mined by neutron diffraction for a quenched sample (Haines ˚ 3 by X-ray diffraction for a quenched et al., 1997); 120.13 A ˚ 3 by X-ray diffraction in a sample (Devi et al., 1987); 119.98 A multianvil apparatus at 0.1 MPa (Ohtaka et al., 2001). The present value of B0 is smaller than those determined by DAC experiments at room temperature without sample annealing. Since uniaxial compression by DAC introduces deviatoric stress, special attention must be given to it because apparently higher pressures could yield higher values of the B0 (Meng et al., 1993). The obtained B0, 278 GPa, is discernibly higher than the value of 254 GPa for corundum, Al2O3 (Ohno et al., 1986). If the correlation between hardness and high bulk J. Appl. Cryst. (2005). 38, 727–733

Figure 6 Typical diffraction patterns taken at very high temperature. Diffraction angles for the orthoII and tetragonal phases are represented by bars.

modulus in material is accepted (Yang et al., 1987), orthoII can be considered as a candidate for potentially very hard materials. 3.3. Cubic fluorite-type phase

In situ high-temperature experiments under medium pressure below 40 GPa were attempted to examine the phase relations, focusing on the tetragonal-to-cubic phase boundary, where the transition is reversible with temperature. Typical observed diffraction patterns are shown in Fig. 6. The profile taken at 12 GPa and 2300 K shows pronounced doublets indexed according to tetragonal symmetry. These tetragonal doublets are clearly observed up to 3000 K. The observed phases are plotted in Fig. 1. The two-phase mixture of orthoII and tetragonal phase was observed at 14.3 GPa and 2200 K, and 15.2 GPa and 2500 K. The cubic phase, which has been assumed to exist in the high-temperature regime, is not observed in this study. Instead, the tetragonal phase is confirmed to be stable up to 3000 K.

electronic reprint

Osamu Ohtaka et al.



ZrO2

731

research papers ZrO2 under high pressure and high temperature were investigated. OrthoII is found to be stable to a pressure of 100 GPa and a temperature of 2500 K. The unit-cell parameters and the volumes of orthoII were determined as a function of pressure. OrthoII does not show the anisotropic compression followed by post-cotunnite transitions as is observed for the cotunnitestructured dihalides, but contracts rather isotropically. OrthoII may be stable up to extremely high pressure. The bulk modulus of orthoII calculated using the Birch–Murnaghan equation of state is 278 GPa, which indicates that orthoII is highly incompressible and thus a candidate for ultra hard materials. The cubic phase is not observed but the tetragonal phase is confirmed to be stable up to 3000 K. The present result suggests the possibility that stoichiometric ZrO2 does not show the cubic structure up to the melting temperature. Figure 7 Hypothetical P–T phase diagram of stoichiometric ZrO2 assuming that there is no cubic form. The present results are plotted as symbols. Open circles and closed circles represent the orthoII and the tetragonal phase, respectively.

On survey of the literature, it seems unclear whether or not the cubic form occurs in pure ZrO2. Before and during the 1960s, whether or not pure ZrO2 shows a cubic structure at high temperature was a controversial subject (Wolten, 1958; Weber, 1962; Boganov et al., 1965; Collongues et al., 1971; Ondik & McMurdie, 1998). Several high-temperature X-ray diffraction studies confirmed the cubic fluorite-type structure (Boganov et al., 1965; Collongues et al., 1971). However, because these experiments were performed in vacuum or in reducing atmosphere due to the very high temperature, and because contamination from furnace and thermal insulator materials was inextricable, it was suggested that the cubic form exists only with an O deficiency or some impurity (Weber, 1962; Collongues et al., 1971). This concept led to so-called stabilized ZrO2, which has become a major ceramic material, whereas the study of pure ZrO2, addressing the hightemperature phase relation, has been sidelined. The tetragonal-to-cubic transition in ZrO2 is affected greatly by the nature of the atmosphere and by the O2 partial pressure (Ondik & McMurdie, 1998). The present sample, being tightly sealed in a DAC, was surrounded by an Ar pressure medium and a small amount of Pt, both of which show low reactivity with ZrO2 and O. Furthermore, high pressure suppresses the evaporation of O from ZrO2. Consequently, it is most likely that the present ZrO2 is close to the stoichiometric composition, even at very high temperature. The present result, therefore, suggests that pure ZrO2 does not show the cubic structure up to the melting temperature. Fig. 7 illustrates a schematic pressure–temperature phase diagram of stoichiometric ZrO2 assuming that there is no cubic form.

4. Conclusion Using both the sample annealing and in situ laser-heating techniques, the phase relations and compression behaviour of

732

Osamu Ohtaka et al.



ZrO2

This study was done while OO was an invited researcher in the Laboratoire des Geomateriaux of IPGP, and he thanks the staff for their support during the stay. Financial support from the IPGP is gratefully acknowledged.

References Andrault, D., Angel, R. J., Mosenfelder, J. L. & Le Bihan, T. (2003). Am. Mineral. 88, 301–307. Andrault, D. & Fiquet, G. (2001). Rev. Sci. Instrum. 72, 1283–1288. Andrault, D., Fiquet, G., Itie, J.P., Richet, P., Gillet, P., Hausermann, D. & Hanfland, M. (1998). Eur. J. Mineral. 10, 931–940. Birch, F. (1947). Phys. Rev. 71, 809–824. Boganov, A. G., Rudenko, V. S. & Makarov, L. P. (1965). Dokl. Acad. Nauk SSSR, 160, 1065–1068; Dokl. Chem. (Engl. Transl.) 160, 146– 148. Chervin, J. C., Canny, B., Besson, J. M. & Pruzan, J. M. (1995). Rev. Sci. Instrum. 66, 2595–2598. Cohen, R. E., Mehl, M. J. & Boyer, L. L. (1988). Physica B, 150, 1–9. Collongues, R., Revcolevschi, A., Foex, M., Traverse, J. P. & Rouanet, A. (1971). Colloq. Int. CNRS, 205, 241–246. Desgreniers, S. & Lagarec, K. (1999). Phys. Rev. B, 59, 8467–8472. Devi, S. R. U., Ming, L. C. & Manghnani, M. H. (1987). J. Am. Ceram. Soc. 70, C218–C219. Dewaele, A., Fiquet, G., Andrault, D. & Hausermann, D. (2000). J. Geophys. Res. 105, 2869–2877. Fiquet, G., Guyot, F., Kunz, M., Matas, J., Andrault, D. & Hanfland, M. (2002). Am. Mineral. 87, 1261–1265. Gupta, T. K., Lange, F. F. & Bechtold, J. H. (1978). J. Mater. Sci. 13, 1464–1470. Haines, J., Leger, J. M. & Atouf, A. (1995). J. Am. Ceram. Soc. 78, 445–448. Haines, J., Leger, J. M., Hull, S., Petitet, J. P., Pereira, A. S., Perottoni, C. A. & Da Jornada, J. A. H. (1997). J. Am. Ceram. Soc. 80, 1910– 1914. Hammersley, J. (1996). Fit2D. ESRF, Grenoble, France. Holmes, N. C., Moriarty, J. A., Gathers, G. R. & Nellis, W. J. (1989). J. Appl. Phys. 66, 2962–2967. Howard, C. J., Hill, R. J. & Reichert, B. E. (1988). Acta Cryst. B44, 116–120. Jamieson, J. C., Fritz, J. N. & Manghnani, M. H. (1992). High Pressure Research in Geophysics, pp. 27–48. Tokyo: Center for Academic Publications. Jeitschko, W. (1968). Acta Cryst. B24, 930–934.

electronic reprint

J. Appl. Cryst. (2005). 38, 727–733

research papers Kriven, W. M. (1988). J. Am. Ceram. Soc. 71, 1021–1030. Leger, J. M., Haines, J. & Atouf, A. (1995a). Phys. Rev. B, 51, 3902– 3905. Leger, J. M., Haines, J., Atouf, A., Schulte, O. & Hull, S. (1995b). Phys. Rev. B, 52, 13247–13256. Leger, J. M., Tomaszewski, P. E., Atouf, A. & Pereira, A. S. (1993). Phys. Rev. B, 47, 14075–14083. Lowther, J. E., Dewhurst, J. K., Leger, J. M. & Haines, J. (1999). Phys. Rev. B, 60, 14485–14488. Meng, Y., Weidner, D. J. & Fei, Y. (1993). Geophys. Res. Lett. 20, 1147–1150. Mirgorodsky A. P. & Quintard, P. E. (1999). J. Am. Ceram. Soc. 82, 3121–3124. Ohno, I., Yamamoto, S. & Anderson, O. L. (1986). J. Phys. Chem. Solids, 47, 1103–1108. Ohtaka, O., Fukui, H., Kunisada, T., Fujisawa, T., Funakoshi, K., Utsumi, W., Irifune, T., Kuroda, K. & Kikegawa, T. (2001). Phys. Rev. B, 63, 174108.

J. Appl. Cryst. (2005). 38, 727–733

Ohtaka, O., Yamanaka, T., Kume, S., Hara, N., Asano, H. & Izumi, F. (1990). Proc. Jpn. Acad. Ser. B, 66, 193–196. Ohtaka, O., Yamanaka, T., Kume, S., Ito, E. & Navrostky, A. (1991). J. Am. Ceram. Soc. 74, 505–509. Ondik, H. M. & McMurdie, H. F. (1998). Editors. Phase Diagrams for Zirconium + Zirconia Systems. pp. 35–38. Ohio: The American Ceramic Society. Teufer, G. (1962). Acta Cryst. 15, 1187. Weber, B. C. (1962). J. Am. Ceram. Soc. 45, 614–615. Weidner, D. J., Vaughan, M. T., Ko, J., Wang, Y., Liu, X., YeganehHaeri, A., Pacalo, R. E. & Zhao, Y. (1992). High Pressure Research: Application to Earth and Planetary Science, pp. 13–17. Washington, DC: American Geophysical Union. Wolten, G. M. (1958). J. Am. Chem. Soc. 80, 4772–4775. Wyckoff, R. W. G. (1963). Crystal Structures, 2nd ed. Vol. 1. New York: Wiley. Yang, W., Parr, R. G. & Uytterhoeven, L. (1987). Phys. Chem. Miner. 15, 191–195.

electronic reprint

Osamu Ohtaka et al.



ZrO2

733