Conformity and precision of CO2 densimetry in CO2 inclusions: microthermometry versus Raman microspectroscopic densimetry

To assess the ability of densimetry for CO2 fluid in CO2 inclusions, we compare two methods, microthermometry and Raman microspectroscopic densimetry for CO2. The comparative experiment was performed for nine CO2 inclusions in three mantle xenoliths. The results are as follows: (1) microthermometry precisely determines CO2 density with the range of 0.65 to 1.18?g/cm3 compared with Raman microspectroscopic densimetry; (2) CO2 density obtained by Raman microspectroscopic densimetry is fairly consistent with that by microthermometry; (3) it is hard to determine CO2 density in CO2 inclusion with diameter of less than around 3 mu m using microthermometry; and (4) microthermometry can be applied only to the CO2 inclusion whose CO2 density ranges from around 0.65 to 1.18?g/cm3, whereas the Raman microspectroscopic densimetry is applicable to CO2 density ranging from 0.1 to 1.24?g/cm3. The above features carry the potential for estimation of depth origin of mantle-derived rocks. The depth where the rocks were trapped by host magma can be estimated using both geothermometric data and CO2 fluid density in CO2 inclusions in the rocks. Typical precisions of density of CO2 in CO2 inclusions obtained by the Raman microspectroscopic densimetry (similar to 0.01?g/cm3) and by the microthermometry (< 0.001?g/cm3) correspond to uncertainties in the depth origin of 2.4?km and?