High-pressure Phase Transformations Research Articles

Silicate perovskite from phase transformations of pyrope‐garnet at high pressure and temperature

Both natural and synthetic pyrope have been compressed to pressures above 300 kbar in a diamond‐anvil press and heated to temperatures above 800°C by a continuous YAG laser. After quenching and release of pressure, X‐ray diffraction shows that the natural pyrope transforms to a single phase with perovskite‐like structure whereas the synthetic pyrope disproportinates into a mixture of MgSiO3 (perovskite) plus Al2O3 (corundum). The orthorhombic perovskite lattice parameters for the natural pyrope are ao = 4.816 ± 0.004 Å, bo = 4.973 ± 0.004 Å, and co = 6.997 ± 0.006 Å with Z = 4. The orthorhombic cell dimensions for MgSiO3 (perovskite) are ao = 4.818 ± 0.005 Å, bo = 4.869 ± 0.005 Å, and co = 6.906 ± 0.007 Å with Z=4. The calculated density of MgSiO3 perovskite is thus 4.12 g/cm³, or 3.7% denser than the isochemical mixed oxides. These experiments are the first demonstration of static high pressure phase transformations in silicate garnets and the first examples of a pure silicate existing in the perovskite structure.

The system MgO-FeO-SiO 2 at high pressures and temperatures — phase equilibria and elastic properties

Advances, during the Upper Mantle Project, in investigations on phase equilibria and elastic properties of the mantle minerals in the MgO-FeO-SiO 2 system are reviewed. The experimental procedure for a phase equilibrium study is described. Some advantages and disadvantages of various high-pressure, high-temperature apparatus are discussed. Necessities of establishing a pressure calibration method at high temperatures are also pointed out. A brief description on a method, recently developed for the measurements of ultrasonic wave velocities of very small samples, is given. Olivine-spinel solid solution equilibria in the system Mg 2SiO 4-Fe 2SiO 4 have been studied at several laboratories over the pressure range 40–200 kbar at 800 and 1,000°C. Ringwood and Major first discovered a peculiar mode of the high-pressure transformation in compositions close to pure Mg 2SiO 4, resulting from the formation of β phase. Crystal structure of the β phase was clarified as modified spinel structure through a series of studies on the high-pressure transformation of Co 2SiO 4 and Mn 2GeO 4. The appearance of the modified spinel phase in the Mg 2SiO 4-Fe 2SiO 4 system was confirmed in further investigations at the author's laboratory. The isothermal section of the phase diagram for the Mg 2SiO 4-Fe 2SiO 4 system was constructed at 800 and 1,000°C on the basis of these recent experimental results. At 800°C a continuous series of spinel solid solutions was synthesizable from Fe 2SiO 4 to (Mg 0.9Fe 0.1) 2SiO 4. At 1,000°C, however, all the attempts to synthesize a true spinel phase of (Mg 0.9Fe 0.1) 2SiO 4 were unsuccessful up to about 140 kbar and coexistence of the modified spinel phase with true spinel was usually identified. This suggests that the stability field of the β(Mg, Fe) 2 SiO 4 is highly temperature dependent. A remarkable expansion of the β(Mg, Fe) 2SiO 4 region is expected at the higher temperatures. Based on the cell parameters of βMg 2SiO 4 and the extrapolated value for γMg 2SiO 4 (true spinel), the density increase associated with forsterite- βMg 2 SiO 4 transformation, and forsterite — γMg 2 SiO 4 transformation was calculated to be 7.9% and 10.8% respectively. Experimental data on the phase equilibria of the MgSiO 3-FeSiO 3 system are presented. A highpressure disproportionation of clino-pyroxene solid solutions into stishovite plus spinel solid solutions was found. It was established that the disproportionation curve for clino-ferrosilite (FeSiO 3) was well represented by the boundary curve for the coesite-stishovite transformation. A preliminary phase diagram for the MgSiO 3-FeSiO 3 system was constructed at 800 and 1,000°C from the data by Ringwood and Major and by the author's laboratory. Experimental data on the high-pressure phase transformations of SiO 2 are summarized. The boundary curve for the coesite-stishovite transformation was determined over the temperature range 550–1,200°C in the pressure range 83–101 kbar by means of a tetrahedral anvil press. The transition curve was fitted by the linear relation P( kbar) = 67 + 0.028 T(° C). This determination was found to be in reasonable agreement with the previous data. A stoichiometric compound, Fe 1,000O, was synthesized at high-pressures above 40 kbar at 775°C by a reaction between wüstite, Fe 0.950O, and metallic iron. The cell dimension of Fe 1,000O was determined to be 4.323 ± 0.001 A ̊ . Compressional- and shear-wave velocities of the synthetic (Mg, Fe) 2 SiO 4 olivine, Fe 2SiO 4 spinel, (Mg, Fe)SiO 3 orthopyroxene, eoesite, stishovite and Fe 0.98O were measured by means of the ultra-sonic pulse transmission method. Results are represented on Birch's diagram, where the wave velocities are plotted as a function of density. It was found that the wave velocities of these ferromagnesian silicates and oxide decrease linearly with the increase of the FeO/(FeO+MgO) ratio, and that the isomorphic lines of (Mg, Fe) 2 SiO 4 olivine, (Mg, Fe)SiO 3 orthopyroxene and (Mg, Fe)O magnesiowüstite are approximately parallel to each other. The Compressional- and shear-wave velocities of the true spinel phase of Mg 2 SiO 4 were estimated to be 10.0 km/sec and 5.7 km/sec respectively. Compressional- and shear-wave velocities of stishovite were determined to be 11.0 km/sec and 5.55 km/sec. The bulk modulus of stishovite was calculated from these values to be 3.43 Mbar. Compressional-wave velocity of the three polymorphs of silica, αquartz, coesite and stishovite, was found to increase regularly along Birch's mean atomic weight line of M = 21 .

High-Pressure Phase Transformations in Hexagonal and Amorphous Selenium

Hexagonal selenium transforms to a new phase at 140 ± 10 kbar. The c/a ratio of 1.42 at the transformation corresponds to the same c/a ratio for tellurium at its transformation at 40 kbar. It has not been possible to decide unambiguously a unit cell from the diffraction pattern of the new phase. The high-pressure phase is irreversible being retained upon releasing the pressure. Amorphous Se slowly transforms to the crystalline state under pressure and eventually develops the same unknown high-pressure structure as the hexagonal form. The pressure-cycled amorphous material retains its crystalline structure upon releasing the pressure. This structure appears to be identical to the irreversible hexagonal phase. It is thought that the high-pressure structure is superconducting.

High-pressure phase transformations in MnAl 2S 4 and MnGa 2S 4

Phase transformations in MnAl 2S 4 and MnGa 2S 4 have been investigated at high pressure up to 95 kb and temperature up to 1000°C. The decomposition of MnAl 2S 4 to MnS with rock salt structure and the spinel phase at moderately high pressure has been observed. The lattice parameter for the spinel phase is shown to decrease with increasing pressure. It is interpreted due to the formation of nonstoichiometric spinel expressed by Mn 1− x Al 2S 4− x . Above about 90 kb, an Al-rich unknown phase appears instead of spinel, which can not be identified on account of its extreme unstableness in air. Six polymorphs have been observed for MnGa 2S 4, designated as I ∼ VI in the order of increasing pressure. MnGa 2S 4 IV is identified to be identical with an atmospheric pressure phase of MnAl 2S 4 with the ZnIn 2S 4 structure. The crystal structures of MnGa 2S 4 I, II and III are thought to be due to the polytypism in the ZnIn 2S 4 structure. The identification of the other two polymorphs can not be made.

High mobility of heavy rare gases in potassium halides at low temperatures (20–100°C)

Akimoto, S., 1972. The system MgO–FeO–SiO2 at high pressures and temperatures – phase equilibria and elastic properties. In: A.R. Ritsema (Editor), The Upper Mantle. Tectonophysics, 13 (1–4): 161–187.Advances, during the Upper Mantle Project, in investigations on phase equilibria and elastic properties of the mantle minerals in the MgO–FeO–SiO2 system are reviewed. The experimental procedure for a phase equilibrium study is described. Some advantages and disadvantages of various high-pressure, high-temperature apparatus are discussed. Necessities of establishing a pressure calibration method at high temperatures are also pointed out. A brief description on a method, recently developed for the measurements of ultrasonic wave velocities of very small samples, is given.Olivine–spinel solid solution equilibria in the system Mg2SiO4–Fe2SiO4 have been studied at several laboratories over the pressure range 40–200 kbar at 800 and 1,000°C. Ringwood and Major first discovered a peculiar mode of the high-pressure transformation in compositions close to pure Mg2SiO4, resulting from the formation of β phase. Crystal structure of the β phase was clarified as modified spinel structure through a series of studies on the high-pressure transformation of Co2SiO4, and Mn2GeO4. The appearance of the modified spinel phase in the Mg2SiO4–Fe2SiO4 system was confirmed in further investigations at the author's laboratory. The isothermal section of the phase diagram for the Mg2SiO4–Fe2SiO4 system was constructed at 800 and 1,000°C on the basis of these recent experimental results. At 800°C a continuous series of spinel solid solutions was synthesizable from Fe2SiO4 to (Mg0.9Fe0.1)2SiO4. At 1,000°C, however, all the attempts to synthesize a true spinel phase of (Mg0.9Fe0.1)2SiO4 were unsuccessful up to about 140 kbar and coexistence of the modified spinel phase with true spinel was usually identified. This suggests that the stability field of the β(Mg, Fe)2 SiO4 is highly temperature dependent. A remarkable expansion of the β(Mg, Fe)2SiO4 region is expected at the higher temperatures. Based on the cell parameters of βMg2SiO4 and the extrapolated value for γMg2SiO4 (true spinel), the density increase associated with forsterite–βMg2SiO4 transformation, and forsterite– γMg2SiO4 transformation was calculated to be 7.9% and 10.8% respectively.Experimental data on the phase equilibria of the MgSiO3–FeSiO3 system are presented. A high-pressure disproportionation of clino-pyroxene solid solutions into stishovite plus spinel solid solutions was found. It was established that the disproportionation curve for clino-ferrosilite (FeSiO3) was well represented by the boundary curve for the coesite–stishovite transformation. A preliminary phase diagram for the MgSiO3–FeSiO3 system was constructed at 800 and 1,000°C from the data by Ringwood and Major and by the author's laboratory.Experimental data on the high-pressure phase transformations of SiO2 are summarized. The boundary curve for the coesite–stishovite transformation was determined over the temperature range 550–1,200°C in the pressure range 83–101 kbar by means of a tetrahedral anvil press. The transition curve was fitted by the linear relation P(kbar) = 67 + 0.028 T(°C). This determination was found to be in reasonable agreement with the previous data.A stoichiometric compound, Fe1.000O, was synthesized at high-pressures above 40 kbar at 775°C by a reaction between wüstite, Fe0.950O, and metallic iron. The cell dimension of Fe1.000O was determined to be 4.323 ± 0.001 å.Compressional- and shear-wave velocities of the synthetic (Mg, Fe)2SiO4 olivine, Fe2SiO4 spinel, (Mg, Fe)SiO3 orthopyroxene, coesite, stishovite and Fe0.98O were measured by means of the ultra-sonic pulse transmission method. Results are represented on Birch's diagram, where the wave velocities are plotted as a function of density. It was found that the wave velocities of these ferromagnesian silicates and oxide decrease linearly with the increase of the FeO+(FeO+MgO) ratio, and that the isomorphic lines of (Mg, Fe)2 SiO4 olivine, (Mg, Fe)SiO3 orthopyroxene and (Mg, Fe)O magnesiowüstite are approximately parallel to each other. The compressional- and shear-wave velocities of the true spinel phase of Mg2 SiO4 were estimated to be 10.0 km/sec and 5.7 km/sec respectively.Compressional- and shear-wave velocities of stishovite were determined to be 11.0 km/sec and 5.55 km/sec. The bulk modulus of stishovite was calculated from these values to be 3.43 Mbar. Compressional-wave velocity of the three polymorphs of silica, αquartz, coesite and stishovite, was found to increase regularly along Birch's mean atomic weight line of M¯=21.

Defects in crystalline quartz: Electron paramagnetic resonance of E' vacancy centers associated with germanium impurities

Akimoto, S., 1972. The system MgO–FeO–SiO2 at high pressures and temperatures – phase equilibria and elastic properties. In: A.R. Ritsema (Editor), The Upper Mantle. Tectonophysics, 13 (1–4): 161–187.Advances, during the Upper Mantle Project, in investigations on phase equilibria and elastic properties of the mantle minerals in the MgO–FeO–SiO2 system are reviewed. The experimental procedure for a phase equilibrium study is described. Some advantages and disadvantages of various high-pressure, high-temperature apparatus are discussed. Necessities of establishing a pressure calibration method at high temperatures are also pointed out. A brief description on a method, recently developed for the measurements of ultrasonic wave velocities of very small samples, is given.Olivine–spinel solid solution equilibria in the system Mg2SiO4–Fe2SiO4 have been studied at several laboratories over the pressure range 40–200 kbar at 800 and 1,000°C. Ringwood and Major first discovered a peculiar mode of the high-pressure transformation in compositions close to pure Mg2SiO4, resulting from the formation of β phase. Crystal structure of the β phase was clarified as modified spinel structure through a series of studies on the high-pressure transformation of Co2SiO4, and Mn2GeO4. The appearance of the modified spinel phase in the Mg2SiO4–Fe2SiO4 system was confirmed in further investigations at the author's laboratory. The isothermal section of the phase diagram for the Mg2SiO4–Fe2SiO4 system was constructed at 800 and 1,000°C on the basis of these recent experimental results. At 800°C a continuous series of spinel solid solutions was synthesizable from Fe2SiO4 to (Mg0.9Fe0.1)2SiO4. At 1,000°C, however, all the attempts to synthesize a true spinel phase of (Mg0.9Fe0.1)2SiO4 were unsuccessful up to about 140 kbar and coexistence of the modified spinel phase with true spinel was usually identified. This suggests that the stability field of the β(Mg, Fe)2 SiO4 is highly temperature dependent. A remarkable expansion of the β(Mg, Fe)2SiO4 region is expected at the higher temperatures. Based on the cell parameters of βMg2SiO4 and the extrapolated value for γMg2SiO4 (true spinel), the density increase associated with forsterite–βMg2SiO4 transformation, and forsterite– γMg2SiO4 transformation was calculated to be 7.9% and 10.8% respectively.Experimental data on the phase equilibria of the MgSiO3–FeSiO3 system are presented. A high-pressure disproportionation of clino-pyroxene solid solutions into stishovite plus spinel solid solutions was found. It was established that the disproportionation curve for clino-ferrosilite (FeSiO3) was well represented by the boundary curve for the coesite–stishovite transformation. A preliminary phase diagram for the MgSiO3–FeSiO3 system was constructed at 800 and 1,000°C from the data by Ringwood and Major and by the author's laboratory.Experimental data on the high-pressure phase transformations of SiO2 are summarized. The boundary curve for the coesite–stishovite transformation was determined over the temperature range 550–1,200°C in the pressure range 83–101 kbar by means of a tetrahedral anvil press. The transition curve was fitted by the linear relation P(kbar) = 67 + 0.028 T(°C). This determination was found to be in reasonable agreement with the previous data.A stoichiometric compound, Fe1.000O, was synthesized at high-pressures above 40 kbar at 775°C by a reaction between wüstite, Fe0.950O, and metallic iron. The cell dimension of Fe1.000O was determined to be 4.323 ± 0.001 å.Compressional- and shear-wave velocities of the synthetic (Mg, Fe)2SiO4 olivine, Fe2SiO4 spinel, (Mg, Fe)SiO3 orthopyroxene, coesite, stishovite and Fe0.98O were measured by means of the ultra-sonic pulse transmission method. Results are represented on Birch's diagram, where the wave velocities are plotted as a function of density. It was found that the wave velocities of these ferromagnesian silicates and oxide decrease linearly with the increase of the FeO+(FeO+MgO) ratio, and that the isomorphic lines of (Mg, Fe)2 SiO4 olivine, (Mg, Fe)SiO3 orthopyroxene and (Mg, Fe)O magnesiowüstite are approximately parallel to each other. The compressional- and shear-wave velocities of the true spinel phase of Mg2 SiO4 were estimated to be 10.0 km/sec and 5.7 km/sec respectively.Compressional- and shear-wave velocities of stishovite were determined to be 11.0 km/sec and 5.55 km/sec. The bulk modulus of stishovite was calculated from these values to be 3.43 Mbar. Compressional-wave velocity of the three polymorphs of silica, αquartz, coesite and stishovite, was found to increase regularly along Birch's mean atomic weight line of M¯=21.

Coordination and volume changes accompanying high-pressure phase transformations of oxides

Although attention has always been given to changes of cation coordination in high-pressure phase transformations, changes of anion coordination have generally been ignored even though they are as characteristic and probably as important as those of the cations. The importance of the anion coordination change is illustrated for transformations of In203 and the transition metal molybdates. The relative change of molar volume or density can frequently be used to distinguish high-pressure phase transformations involving a coordination change from those not involving a coordination change.

Phase transformations in the mantle

This paper presents a review of the constitution of the mantle between depths of 200 and 1200 km in the light of recent experimental investigations on high pressure phase transformations in silicate minerals and on their germanate analogues. Phase transformations occurring at depths to 600 km can be studied directly using apparatus which develops >200 kb at elevated temperatures. These studies indicate that the rapid increase of seismic velocity around 350–450 km is caused mainly by the transformation of pyroxenes into a new type of garnet structure and the transformation of olivine to the spinel (or related) structure. Current understanding of the constitution of the mantle below 600 km rests heavily upon interpretation of phase transformations in germanate analogue systems and on shock wave investigations. These suggest that around 600–700 km, garnets and spinels transform to new phases possessing ilmenite, perovskite and strontium plurabate structures with densities and clastic properties resembling those of isochemical mixed oxides (MgO + SiO2 (as stishovite) + Al2O3 + CaO, etc.). At greater depths, a further set of transformations to new phases appears likely, causing the “zero-pressure density” of the mantle to attain values about 5% higher than an isochemical mixture of oxides. The phase transformations which have been directly observed or inferred provide a quantitative explanation of the seismic velocity distribution between 300 and 800 km. From a study of the elastic properties of the mantle it is concluded that the average FeO/(FeO + MgO) ratio of mantle silicates is probably approximately uniform at 0.11 (molecular) throughout the entire mantle.

High-pressure phase transformations in germanate pyroxenes and related compounds

Crystal chemical relationships between germanates and silicates are reviewed. It is observed that germanates tend to behave as high-pressure models for the corresponding silicates. Accordingly, the investigation of pressure transformations in germanates provides valuable information on the probable modes of transformation which would be displayed by the corresponding silicates under higher pressures. We have applied this technique to germanate pyroxenes in order to infer the behavior of silicate analogs—particularly enstatite. The following germanate pyroxenes were synthesized: MgGeO3, MnGeO3, CoGeO3, FeGeO3 (Mg.75 Ni.25)GeO3, CaMgGe2O6. The related compounds CaGeO3 and CdGeO3 were also prepared. At 700°C all these compounds were found to transform to much denser phases at pressures which were mostly less than 40 kb. MgGeO3 and MnGeO3 pyroxenes transformed to ilmenite structures; CoGeO3, FeGeO3, and (Mg.75Ni.25)GeO3 broke down into orthogermanate spinels+GeO2 (rutile structure). The compounds CaGeO3 and CdGeO3 transformed to garnet structures, and CaMgGe2Oe broke down into a mixture of MgGeO3 ilmenite+CaGeO3 garnet. In the system MgGeO3-MgSiO3, pyroxene solid solutions containing between 10 and 50 per cent of MgSiO3 broke down into Mg2(GeSi)O2 spinels+(GeSi)O2 rutile solid solutions. These results are discussed in detail and the probable behavior of MgSiO3 at high pressures is inferred with the aid of thermodynamic data. It is concluded that enstatite will probably break down into forsterite+stishovite at a pressure of approximately 120 kb in the temperature range 500° to 2000°C.

High-pressure Phase Transformations in Laboratory Mechanical Mixers and Mortars

THE mechanical action of simple laboratory grinders, mortars and similar devices has occasionally been used to assist in chemical reactions in addition to performing their primary physical functions. Among these chemical reactions are phase transformations. That the effect is due to some kind of a pressure component in the mechanical action may appear obvious, but the magnitude of the pressures is not readily appreciated, nor can it be calculated or measured. However, in the course of work in this laboratory on the high-pressure polymorphism of lead dioxide (PbO2), the information that a phase identical with the new high-pressure polymorph, which had been shown to be stable only in the region above 10,000 bars, had been formed by simple grinding1 was noted with no little surprise. The observation was promptly confirmed by grinding a small amount (2 gm.) of the common rutile form (I) of lead dioxide in a mechanical mortar and pestle combination of laboratory pattern. Grinding in air for a few hours converted an estimated one-third to the denser orthorhombic form (II). After preliminary work, it transpired that nothing new in principle had been added to some similar results which had been reported by Burns and Bredig2 on the transformation of calcite to aragonite by grinding in a mortar. The high-pressure phase was formed from the low-pressure one, the amount of change was dependent on time, and subsequent heating would form the low-pressure phase. A significant difference is that phase equilibrium and thermochemical studies3 place the calcite–aragonite transformation at about 3,000 bars at room temperature, which is considerably lower than the 10,000 bars necessary for the lead oxide I ⇌ II transformation.