Hiroshi Kojitani

Abstract High‐temperature calorimetry (HTC) originated in the 20th century as a niche method to enable measurements not easily accomplished with acid solution calorimetry, combustion calorimetry, vapor pressure, or EMF methods. Over time, HTC has evolved into a versatile approach to accurately quantify formation, phase transition, surface and interfacial enthalpies of a wide range of materials including minerals and refractory inorganic compounds. This evolution has been the result of numerous adjustments to experimental setups and procedures, followed by rigorous testing. The commercial availability and the scientific success of this technique have led to an increase in the number of laboratories applying HTC. However, the knowledge acquired by researchers over the past 70 years is scattered throughout the literature or only available as laboratory internal documentation and personal experience. This publication is a collaborative effort among several leading HTC laboratories to summarize and unify current state‐of‐the‐art HTC techniques and procedures. The text starts by summarizing various HT techniques that are commonly used for readers with an interest in HTC in general. It is then directed toward HTC users and includes a brief section on data evaluation procedures as well as a comprehensive compilation of reference data utilizing molten sodium molybdate and lead borate solvents. Finally, for experienced HTC users, an in‐depth discussion of some common difficulties and a discussion of uncertainties are presented.

Abstract Three single crystals of CaTi2O4 (CT)-type, CaFe2O4 (CF)-type, and new low-density CaFe2O4 (LD-CF) related MgAl2O4 were synthesized at 27 GPa and 2500 °C and also CT-type MgAl2O4 at 45 GPa and 1727 °C using conventional and advanced multi-anvil technologies, respectively. The structures of CT-type and LD-CF related MgAl2O4 were analyzed by single-crystal X-ray diffraction. The lattice parameters of the CT-type phases synthesized at 27 and 45 GPa were a = 2.7903(4), b = 9.2132(10), and c = 9.3968(12) Å, and a = 2.7982(6), b = 9.2532(15), and c = 9.4461(16) Å, respectively, (Z = 4, space group: Cmcm) at ambient conditions. This phase has an AlO6 octahedral site and an MgO8 bicapped trigonal prism with two longer cation-oxygen bonds. The LD-CF related phase has a novel structure with orthorhombic symmetry (space group: Pnma), and lattice parameters of a = 9.207(2), b = 3.0118(6), and c = 9.739(2) Å (Z = 4). The structural framework comprises tunnel-shaped spaces constructed by edge- and corner-sharing of AlO6 and a 4+1 AlO5 trigonal bipyramid, in which MgO5 trigonal bipyramids are accommodated. The CF-type MgAl2O4 also has the same space group of Pnma but a slightly different atomic arrangement, with Mg and Al coordination numbers of 8 and 6, respectively. The LD-CF related phase has the lowest density of 3.50 g/cm3 among MgAl2O4 polymorphs, despite its high-pressure synthesis from the spinel-type phase (3.58 g/cm3), indicating that the LD-CF related phase formed via back-transformation from a high-pressure phase during the recovery. Combined with the previously determined phase relations, the phase transition between CF-and CT-type MgAl2O4 is expected to have a steep Clapeyron slope. Therefore, CT-type phase may be stable in basaltic- and continental-crust compositions at higher temperatures than the average mantle geotherm in the wide pressure range of the lower mantle. The LD-CF related phase could be found in shocked meteorites and used for estimating shock conditions.

XRD and TEM observations of recovered high-pressure phase Mg₂Cr₂O₅ revealed split spots around the position of h = n + 1/2 (n: integer) in the diffraction patterns of the basic structure of this phase. It was proved that these split spots were formed by the periodic arrangement of the anti-phase boundaries in the structure. Then, the relation between the formation of these periodic anti-phase boundaries and the high-pressure phase transformation of Mg₂Cr₂O₅ is discussed.

A new high-pressure phase with the composition of Mg2Al2O5 was found. A crystal structure model of the Mg2Al2O5 phase was constructed based on that of ludwigite because the X-ray diffraction pattern of the former is very similar to that of the latter. In the model, (Mg, Al)O6 octahedra connected by edge-sharing and corner-sharing form triangular tunnels in which Mg ions are accommodated. Mg ions in the tunnels have a coordinate environment with trigonal prism-type MgO6. Systematic absences of reflections indicate that the space group of the structure is Pba2 or Pbam. The Rietveld analysis showed that R factor in the case of Pbam was smaller than that in the case of Pba2. This result suggests that the space group of the Ma2Al2O5 phase is Pbam.

High pressure experiments and calorimetric measurements were performed to clarify the stability field of CaAl4Si2O11 CAS-phase. It was shown that CAS-phase is stable above about 13 GPa and 1100°C. Combining the measured enthalpy data with published ones, dissociation boundary of CAS-phase into a mixture of Ca-perovskite, corundum and stishovite was calculated. Assuming reasonable geotherms, it was concluded that CAS-phase is stable in the transition zone and upper part of the lower mantle.

Using a multianvil apparatus, high pressure phase relations in NaAlSi3O8 and in the system NaAlSi3O8-KAlSi3O8 were examined up to 24 GPa. Synthesized samples were examined using microfocused and powder X-ray diffraction, and were analyzed with SEM-EDS. At 21-22 GPa and 800-2000 C, NaAlSi2O6 jadeite and stishovite changed to NaAlSiO4 calcium ferrite and stishovite, but NaAlSi3O8 holladite was not observed. Maximum solubility of NaAlSi3O8 in KAlSi3O8 holladite was 40-50 mol% at 1400 C, which was almost the same as 40 mol% at 1000 C. Above results suggest that NaAlSi3O8 hollandite would be stable at higher temperature than 2000 C. This interpretation is consistent with natural observation of NaAlSi3O8-rich hollandite in shocked meteorites.