PACROFI VI - Electronic Program

Precautions on the use of the Hydrothermal Diamond-Anvil Cell for the Acquisition of Volumetric and Phase Relation Data of Geologic Fluids

I-Ming Chou

959 National Center, U.S. Geological Survey, Reston, VA 22092
Phone: (703) 648-6169; e-mail:

As the hydrothermal diamond-anvil cell (HDAC, Bassett et al., 1993) gradually gains popularity in fluid-inclusion related research (Chou et al. 1994), the following precautions and procedures maybe helpful for new users:

(1) the use of pressure calibrants:
Currently available calibrants include quartz (Shen et al., 1993), PbTiO3 (Chou and Haselton, 1994), Pb3(PO4)2 (Chou and Nord, 1994), and BaTiO3 (Chou et al., 1993), and their applicable P-T ranges are shown in Fig. 1. The exact P-T relations for the phase transitions used in these calibrants depend on the purity (and therefore, the source) of the material. It is necessary to calibrate these material before use, and the use of the equation of state of H2O (Haar et al., 1984; Saul and Wagner, 1989) for this purpose is convenient and reliable for pressures up to 10 kbar and temperatures up to 850oC (see Fig. 1). Also, the question of chemical compatibility of the calibrants with the sample system (e.g., solubility and possible chemical reactions) needs to be considered.

(2) the isochoric assumption:
When Re is used as a gasket material to house the sample between diamond anvil faces, it is very difficult, if not impossible, to keep the sample chamber at constant volume during the first heating. However, during cooling and the subsequent heating, the volume change can be small. The interferometric method described by Chou et al. (1994) is recommended for monitoring the volume change; a change of about 1% is to be expected.

(3) the loading of samples of known compositions:
It is straight forward and routine for loading stable compounds (e.g., H2O, CO2, NH3, etc.) in the HDAC. However, loading fluid mixtures of known compositions is a challenging task. This is because, in some cases, it is not easy to prepare the mixtures (e.g., H2O-CO2, H2O-CH4), and in other cases, when such a mixture is available, the composition of the sample may change before sealing (e.g., H2O-NaCl). Therefore, it is important to develop methods for determining sample composition after sealing. For example, in dealing with samples in the system H2O- NaCl-CO2, where a certain amount of Ag2C2O4 is sealed together with an NaCl solution in the sample chamber. The NaCl/H2O ratio can be determined by the freezing point depression before the generation of CO2, whereas the CO2 content after the decomposition of Ag2C2O4 can be determined by the melting point of either ice or clathrate, if liquid CO2 is not present (see Fig. 1 of Barton and Chou, 1993). However, for samples containing liquid CO2, a more involved procedure is required (Diamond, personal communication, 1996).

Fig. 1. A P-T plot showing the phase relations in pure H2O and four pressure calibrants (quartz, PbTiO3, Pb3(PO4)2, and BaTiO3). I, III, V, and VI are ice polymorphs, and their melting data are from Wagner et al. (1994); the open rectangles are triple points along the melting curves; LV is the liquid-vapor curve of H2O; CP is critical point of H2O; the dashed and dotted curves are isochores of H2O derived from equations of state formulated by Haar et al. (1984) and Saul and Wagner (1989), respectively.