Jean-Claude Rifflet

Amorphous materials: Properties, structure, and durability† Structure of Mg- and Mg/Ca aluminosilicate glasses: 27Al NMR and Raman spectroscopy investigations

Abstract The structure and properties of glasses and melts in the MgO-Al2O3-SiO2 (MAS) and CaO-MgOAl2O3- SiO2 (CMAS) systems play an important role in Earth and material sciences. Aluminum has a crucial influence in these systems, and its environment is still questioned. In this paper, we present new results using Raman spectroscopy and 27Al nuclear magnetic resonance on MAS and CMAS glasses. We propose an Al/Si tetrahedral distribution in the glass network in different Qn species for silicon and essentially in Q4 and VAl for aluminum. For the CMAS glasses, an increase of VAl and VIAl is clearly visible as a function of the increase of Mg/Ca ratio in the (Ca,Mg)3Al2Si3O12 (garnet) and (Ca,Mg)AlSi2O8 (anorthite) glass compositions. In the MAS system, the proportion of VAl and VIAl increases with decreasing SiO2 and, similarly with calcium aluminosilicate glasses, the maximum of VAl is located in the center of the ternary system.

Density of superheated and undercooled liquid alumina by a contactless method

The density of liquid alumina drops maintained in levitation with an aerodynamic device and heated with CO2 lasers is determined by analysis of high-speed video digital images between 2000 and 3100 K in various gases. It is shown that consistent results can be achieved for the lighter drops (m<100 mg) which do not depend on the nature of the gas. Experiments performed with lasers impinging the drop surface or during free cooling of the preheated drop gave similar results. The density is represented by the following expression: d=(2.79±0.01)(l−α(T−2500)) g·cm−3, where α=(4.22 ±0.14) × 10−5K−1.

Surface Tension of Liquid Alumina from Contactless Techniques

New data for the surface tension of liquid alumina from 2300 to 3200 K are reported. Aerodynamic levitation of CO2 laser-heated liquid drops allowed contactless measurement of vibration frequencies directly related to surface tension. Consistent data were obtained on drops of different mass ranging from 20 to 160 mg. It was also shown that the oxydo-reducing character of the atmosphere does not modify the results within experimental uncertainty.

Thermophysical Properties of Containerless Liquid Iron up to 2500 K

AbstractThermophysical properties of high temperature liquid iron heated with a CO2 laser have been determined in an aerodynamic levitation device equipped with a high-speed camera and a three-wavelength pyrometer. Characteristic curves of the free cooling and heating of the drop can be used to determine the same apparent emissivity of solid and liquid iron and to calibrate pyrometers based on the known value of the melting point of iron, i.e., 1808 K. Examination of the recalescence of undercooled liquid iron and further solidification are used to obtain the ratio of the melting enthalpy versus the heat capacity of liquid iron as

On the role of carbon dioxide in the combustion of aluminum droplets

\frac{{\Delta H_m }}{{c_P^l }} = 306 \pm 2.5{\text{ K}}

Probe for magnetic resonance spectrometric measures at very high temperatures

. The surface tension σ was determined from an analysis of the vibrations of liquid drops. Results are accurately described by σ (mJ⋅m−2)=(1888±31)−(0.285±0.015) (T−Tm) between 1750 K (undercooled liquid) and 2500 K. The density of liquid iron has been deduced from the image size and the mass of the liquid iron drops.

E.P.R. of Gd3+ in vitreous yag and CaOGa2O3SiO2 glasses

The density, surface tension, and spectral and total hemispherical emissivities of liquid boron obtained with contactless diagnostics are reported for temperatures between 2360 and 3100 K. It is shown that, contrary to previous expectations, liquid boron is denser than the solid at its melting point. It is also shown that the high total emissivity of 0.36 is not consistent with that of a liquid metal as recently claimed. Finally, good agreement is found with previously reported surface tensions and spectral emissivities of liquid boron.

Time Resolved Very High Temperature NMR Study of the Cooling Process of CaO-Al2O3 Liquids

ABSTRACT In this paper, the influence of carbon dioxide on the combustion of aluminum droplets is investigated. Millimeter-sized droplets were heated and ignited by a laser in an aerodynamic levitation system in several CO2 containing atmospheres (H2O/CO2, H2O/CO2/N2) with a large range of compositions (wet – xH2O < 3%, 80/20, 50/50, 12.5/87.5, 50/25/25, 20/40/40). The combustion processes were observed with a high-speed CCD camera, and the droplet radiation was recorded by two optical pyrometers. The ignition occurs with the oxide coating breakdown, which liquefies to form an initial cap moving on the droplet surface. Aluminum vaporizes and burns with the oxidizers as a detached diffusion flame. The droplet regression rate, i.e., the burning rate, strongly depends on the oxidizing atmosphere, from β = 0.58 mm2/s in wet 50% CO2/50% N2 to β = 2.45 mm2/s in 80% H2O/20% CO2. It is shown that CO2 is the worst oxidizer with a smaller “oxidizer efficiency” compared to H2O and O2 . The burning droplet temperature in wet CO2 and in the H2O/CO2 mixtures is around T ≈ 2600 K, and is smaller in H2O/CO2/N2 (T ≈ 2450 K). The non-correlation between the burning rates and the droplet temperatures confirms that the combustion processes are limited by molecular diffusion, and highlights the influence of H2 in the gas-phase transport. An estimation of the exponent n of the “d n law” shows that n decreases with the increase of CO2 from n = 1.7 to n = 0.6. The stagnant burning rates are evaluated as being 7 to 9 times smaller than the measured ones with convective effects. Furthermore, during the droplet regression, the oxide cap dimensions also regress, and it is generally completely removed. Oxide cap regression rates are estimated and are slower than their respective burning rates. Nevertheless, there is a good correlation between the oxide cap regression rate and the droplet temperature that shows that the oxide cap regression results from the chemical decomposition of Al2O3 by the liquid Al droplet producing gaseous AlxOy. For CO2 concentration higher than 40%, a solid phase suddenly appears on the liquid Al surface which entirely covers the droplet leading to the end of the gas-phase combustion. This phenomenon is the consequence of the massive dissolution of carbon in the droplet during burning. Analyses of the unburnt residues showed amounts of dissolved carbon up to 20–23% molar, which is near the saturation concentration limit. Thus, the solid coating corresponds to the ejection of carbon from the droplet because of its continuing regression and is expected to be also present for smaller particles. Therefore, carbon dioxide plays a double role. First, it participates in the gas-phase combustion, but it is the worse oxidizer with smaller burning rates. Second, CO2 causes the carbon dissolution in the Al droplet and finally stops the gas-phase burning. A further implication could be that carbon dioxide may promote the appearance of a combustion regime with surface reactions only.