The relative Raman scattering cross sections of liquid H 2 O and D 2 O were measured using Na 2 SO 4 as an internal standard. Measurements were carried out both at ambient conditions and at elevated temperatures and pressures inside a diamond anvil cell. The Raman spectra of a 1:1 mixture of H 2 O and D 2 O were also studied. Under all conditions, the Raman scattering of H 2 O was found to be significantly larger than for D 2 O, with the precise ratio depending on measurement conditions. On the other hand, in a series of haplogranitic glasses containing up to 5 wt % of H 2 O or D 2 O, the Raman scattering cross section of D 2 O was found to be equal to that of H 2 O within experimental uncertainty. Our data demonstrate that the frequently made assumption of equal Raman scattering cross section for different isotopic species may not always be valid. In situ measurements of isotope fractionation using Raman spectroscopy would require an extremely careful calibration of scattering cross sections. Apparent fractionation factors obtained without such a calibration are probably unreliable. The significant influence of subtle changes in measurement conditions on Raman intensity ratios observed in this study also suggest that Raman calibrations (e.g., for determining water contents in glasses) may not always be transferable between different laboratories.

No abstract

No abstract

Subsurface magmatic–hydrothermal systems are often associated with elevated electrical conductivities in the Earthʼs crust. To facilitate the interpretation of these data and to allow distinguishing between the effects of silicate melts and fluids, the electrical conductivity of aqueous fluids in the system H2O–HCl was measured in an externally heated diamond anvil cell. Data were collected to 700 °C and 1 GPa, for HCl concentrations equivalent to 0.01, 0.1, and 1 mol/l at ambient conditions. The data, therefore, more than double the pressure range of previous measurements and extend them to geologically realistic HCl concentrations. The conductivities $$\sigma$$ σ (in S/m) are well reproduced by a numerical model log $$\sigma$$ σ  = −2.032 + 205.8 T−1 + 0.895 log c + 3.888 log $$\rho$$ ρ  + log$$\Lambda_{0}$$ Λ 0 (T,$$\rho$$ ρ ), where T is the temperature in K, c is the HCl concentration in wt. %, and $$\rho$$ ρ is the density of pure water at the corresponding pressure and temperature conditions. $$\Lambda_{0}$$ Λ 0 (T,$$\rho$$ ρ ) is the limiting molar conductivity (in S cm2 mol−1) at infinite dilution, $$\Lambda_{0}$$ Λ 0 (T,$$\rho$$ ρ ) = 2550.14 − 505.10$$\rho$$ ρ  − 429,437 T−1. A regression fit of more than 800 data points to this model yielded R2 = 0.95. Conductivities increase with pressure and fluid densities due to an enhanced dissociation of HCl. However, at constant pressures, conductivities decrease with temperature because of reduced dissociation. This effect is particularly strong at shallow crustal pressures of 100–200 MPa and can reduce conductivities by two orders of magnitude. We, therefore, suggest that the low conductivities sometimes observed at shallow depths below the volcanic centers in magmatic–hydrothermal systems may simply reflect elevated temperatures. The strong negative temperature effect on fluid conductivities may offer a possibility for the remote sensing of temperature variations in such systems and may allow distinguishing the effects of magma intrusions from changes in hydrothermal circulation. The generally very high conductivities of HCl–NaCl–H2O fluids at deep crustal pressures (500 MPa–1 GPa) imply that electrical conductors in the deep crust, as in the Altiplano magmatic province and elsewhere, may at least partially be due to hydrothermal activity.