“…Besides, that, the formalism allows incorporating microscopic properties as model parameters, such as vibrational frequencies, Grüneisen, and anharmonicity parameters, which are derived from spectroscopic measurements or ab initio techniques. This characteristic, absent in conventional methods based on independent parameterizations of 1-bar properties (e.g., Saxena 1996;Holland and Powell 2011), puts extra constraints on a thermodynamic analysis, and that appears to be especially useful for discriminating between the quality of different experimental data sets, such as shown for, for instance, the heat capacity of the wadsleyite and ringwoodite forms of Mg 2 SiO 4 , the akimotoite and perovskite forms of MgSiO 3 , and the coesite and stishovite forms of SiO 2 (Jacobs et al 2019(Jacobs et al , 2017 This characteristic is also advantageous when experimental data are missing, such as is the case for bulk modulus for both geikielite and ilmenite. Because thermodynamic properties are interdependent due to their relation to microscopic and static energy properties, missing experimental data on the bulk modulus can be compensated by available microscopic data, such as frequencies of vibrational modes in P-T space, in addition to thermodynamic data.…”