Wordmark GFZ Potsdam

Hydrothermal Diamond-Anvil Cell

The hydrothermal diamond-anvil cell can be applied to study fluids and solids in situ at high pressures and temperatures (usually to about 2 GPa and 1100 K) using a wide range of techniques, e.g., optical microscopy and microthermometry, Raman spectroscopy, Brillouin spectroscopy, infrared absorption, X-ray diffraction, X-ray fluorescence, and X-ray absorption. These cells permit to obtain information on material properties that are altered upon quench, e.g., solubility, kinetics of mineral-fluid and fluid-fluid interaction, element partitioning, viscosity, density, complexation and speciation. Phase transitions and other changes in the sample are often easily observed optically, which is helpful or even essential for most of such studies (examples).

The currently used hydrothermal diamond-anvil cells are modifications of the original design by Bassett et al. (1993), which has the advantage of a small temperature gradient in the sample chamber. Accuracy and reproducibility of the temperature measurements are ± 0.1 to 0.2 °C for temperatures between -25 to 25 °C and better than ± 1 °C at temperatures between 100 and 600 °C. Pressure is determined from microthermometric measurements of phase equilibrium temperatures or from the frequency shift of Raman or fluorescence lines.

 

Hydrothermal Diamond-Anvil Cell
Hydrothermal Diamond-Anvil Cell


Examples for optical observation of phase transitions using a HDAC

View into the sample chamber of a HDAC at 380°C. It was loaded with a quartz crystal and a 5.4 molal NaHCO3 solution. The movie shows two heating runs for the same bulk composition but at 2 different densities. Each run starts at 380°C. In the case of the lower density, up to 3 fluids coexist upon heating, and homogenization into a single fluid phase occurred at 497°C. At the higher density, only two fluid phases are observed, with critical homogenization at 457°C.
View into the sample chamber of a HDAC at 380°C. It was loaded with a quartz crystal and a 5.4 molal NaHCO3 solution. The movie shows two heating runs for the same bulk composition but at 2 different densities. Each run starts at 380°C. In the case of the lower density, up to 3 fluids coexist upon heating, and homogenization into a single fluid phase occurred at 497°C. At the higher density, only two fluid phases are observed, with critical homogenization at 457°C.  

 

 

 

These pictures and the movie show the dehydration reaction of diaspore to corundum (2 AlO(OH) = Al2O3 + H2O). Initially, there were water and 2 diaspore crystals (a larger chip showing the [010] cleavage surface and a smaller fragment on top of it) in the sample chamber of the HDAC. The nicols were then crossed and the sample was heated to 610°C. At first, the reaction is relatively slow, and only slight changes are noticeable. After about 50 seconds, the reaction proceeds quickly from the rim towards the center of the larger crystal. In approximately 45 more seconds, all diaspore had reacted to fine-grained corundum. Interestingly, heating to a temperature about 200 K above the equilibrium temperature was required to observe the reaction in a short-time experiment. No reaction was recognizable in 2 hours after overheating by 100 K in similar experiments. The pressure at 610°C was determined from the liquid-vapor homogenization temperature of the pressure medium water measured after the experiment, and an equation of state of H2O (Wagner and Pruß, 2002).
These pictures and the movie show the dehydration reaction of diaspore to corundum (2 AlO(OH) = Al2O3 + H2O). Initially, there were water and 2 diaspore crystals (a larger chip showing the [010] cleavage surface and a smaller fragment on top of it) in the sample chamber of the HDAC. The nicols were then crossed and the sample was heated to 610°C. At first, the reaction is relatively slow, and only slight changes are noticeable. After about 50 seconds, the reaction proceeds quickly from the rim towards the center of the larger crystal. In approximately 45 more seconds, all diaspore had reacted to fine-grained corundum. Interestingly, heating to a temperature about 200 K above the equilibrium temperature was required to observe the reaction in a short-time experiment. No reaction was recognizable in 2 hours after overheating by 100 K in similar experiments. The pressure at 610°C was determined from the liquid-vapor homogenization temperature of the pressure medium water measured after the experiment, and an equation of state of H2O (Wagner and Pruß, 2002).


 

Pressure determination from the frequency shift of Raman or fluorescence lines

 

At -196 ≤ T (°C) ≤ 500 and pressures less than 2.5 GPa, the pressure can be determined from the frequency shift of the 464 cm-1 Raman line of quartz relative to its frequency at a reference pressure and temperature based on equations (2) and (3) given by Schmidt and Ziemann (2000). A better resolution can be achieved at temperatures close to 23 °C, if the frequency shift of the 206 cm-1 line of quartz relative its frequency at 0.1 MPa is used (e.g., eq. (5) in ref. Schmidt and Ziemann (2000)). The random error of the pressure determination is about 10 MPa if the frequency shift of the 206 cm-1 Raman line of quartz is used, or about 25 MPa in the case of the 464 cm-1 line. Prerequisites are a high-resolution Raman spectrometer, a good signal to noise ratio in the spectra, and calibration using plasma lines of the laser. Alternatively, the ruby fluorescence technique is applied at high P/low T conditions.
At -196 ≤ T (°C) ≤ 500 and pressures less than 2.5 GPa, the pressure can be determined from the frequency shift of the 464 cm-1 Raman line of quartz relative to its frequency at a reference pressure and temperature based on equations (2) and (3) given by Schmidt and Ziemann (2000). A better resolution can be achieved at temperatures close to 23 °C, if the frequency shift of the 206 cm-1 line of quartz relative its frequency at 0.1 MPa is used (e.g., eq. (5) in ref. Schmidt and Ziemann (2000)). The random error of the pressure determination is about 10 MPa if the frequency shift of the 206 cm-1 Raman line of quartz is used, or about 25 MPa in the case of the 464 cm-1 line. Prerequisites are a high-resolution Raman spectrometer, a good signal to noise ratio in the spectra, and calibration using plasma lines of the laser. Alternatively, the ruby fluorescence technique is applied at high P/low T conditions.

Pressure determination from measurement of phase equilibrium temperature

 

 

Pressure determination from measurement of phase equilibrium temperature
Pressure determination from measurement of phase equilibrium temperature



 

Contact person: Christian Schmidt

 




Last change: 15.06.2008  to top