Improved 1.5-sided model for the weakly nonlinear study of Bénard-Marangoni instabilities in an evaporating liquid layer.

Bénard-Marangoni instability, with coupled gravity and surface tension effects, in an evaporating liquid layer surmounted by its vapor and an inert gas is investigated theoretically. We show that this system can be described by a model that consists in the liquid layer equations plus the diffusion equation for the vapor in the gas (the so-called 1.5-sided model) and that this model is equivalent to a one-sided model when the vapor mass fraction field can be considered as quasi-stationary, provided that the equivalent Biot number is a nonlocal function of the interface temperature. A comparison of weakly nonlinear results for the 1.5-sided model with a previous one-sided model [M. Dondlinger, J. Margerit, P.C. Dauby, J. Colloid Interface Sci. 283 (2) (2005) 522-532] that considered a Biot number depending on the wavenumber evaluated at the threshold is performed. Very good agreement is found between both models. For this reason, the present analysis can also be considered as a detailed theoretical justification for the use of a one-sided model in the study of evaporative thermoconvection.

Weakly nonlinear study of Marangoni instabilities in an evaporating liquid layer.

We propose a theoretical study of Marangoni driven convection in an evaporating liquid layer surmounted by an inert gas-vapor mixture. After reduction of the full two-layer problem to a one-sided model we use a Galerkin-Eckhaus method leading to a finite set of amplitude equations for the weakly nonlinear analysis of the problem. We analyze the stability of the roll, square, and hexagonal patterns emerging above the linear stability threshold for a water-air and for an ethanol-air system.

Interfacial nonequilibrium and Bénard-Marangoni instability of a liquid-vapor system.

We study Bénard-Marangoni instability in a system formed by a horizontal liquid layer and its overlying vapor. The liquid is lying on a hot rigid plate and the vapor is bounded by a cold parallel plate. A pump maintains a reduced pressure in the vapor layer and evacuates the vapor. This investigation is undertaken within the classical quasisteady approximation for both the vapor and the liquid phases. The two layers are separated by a deformable interface. Temporarily frozen temperature and velocity distributions are employed at each instant for the stability analysis, limited to infinitesimal disturbances (linear regime). We use irreversible thermodynamics to model the phase change under interfacial nonequilibrium. Within this description, the interface appears as a barrier for transport of both heat and mass. Hence, in contrast with previous studies, we consider the possibility of a temperature jump across the interface, as recently measured experimentally. The stability analysis shows that the interfacial resistances to heat and mass transfer have a destabilizing influence compared to an interface that is in thermodynamic equilibrium. The role of the fluctuations in the vapor phase on the onset of instability is discussed. The conditions to reduce the system to a one phase model are also established. Finally, the influence of the evaporation parameters and of the presence of an inert gas on the marginal stability curves is discussed.

Physical properties and self-setting mechanism of calcium phosphate cements from calcium bis-dihydrogenophosphate monohydrate and calcium oxide.

An apatitic calcium phosphate cement was developed from calcium bis-dihydro-genophosphate monohydrate (or monocalcium phosphate monohydrate, MCPM) and calcium oxide (CaO). The powder had a Ca/P molar ratio of 1.67, and the liquid was either pure water or 0.25 M-1 M sodium phosphate buffer, pH 7.4. The influence of the powder-to-liquid (P/L) ratio on the setting time and the mechanical strength were studied. The best results were obtained for the 1 M phosphate buffer with a P/L ratio of 1.53; the setting time was 7 min and the compressive strength was 25 MPa after 24 h and 33 MPa after 11 d. The mechanism and kinetics of the setting reaction were investigated by X-ray diffraction, differential scanning calorimetry, 31P magic angle spinning-nuclear magnetic resonance and infrared spectrometry. The setting reaction was found to be biphasic: in the first step, during the mixing time, MCPM reacted with CaO immediately to give calcium hydrogenophosphate dihydrate (or dicalcium phosphate dihydrate, DCPD) which, in the second step, reacted more slowly with the remaining CaO to give hydroxyapatite. The conversion of the starting materials to hydroxyapatite was complete within 24 h when the liquid was water, but was slower and incomplete with the phosphate buffers. Of the starting materials, 30% remained after 3 d.

Chemical characterization of in vivo aged zinc polycarboxylate dental cements.

The chemical composition of zinc polycarboxylate dental cements aged in vivo was studied. Thirty samples aged from one to 17 years were investigated using X-ray diffraction, infrared spectroscopy, thermogravimetric analysis and differential scanning calorimetry. Evidence for the presence of zinc oxide, amorphous zinc polycarboxylate and water of hydration was found. No correlation with age concerning either the chemical structure of the components or their relative amounts was found. Zinc polycarboxylate dental cements show very good chemical stability on long-term use.

Chemical and structural changes in zinc polycarboxylate cements after immersion in dilute organic acid solutions.

The behaviour of zinc polycarboxylate cements in contact with dilute aqueous solutions of organic acids at concentrations close to those existing in buccal medium, was studied. The organic acids were acetic, citric, tartaric and lactic acids, at 0.01 M and 0.001 M. The elution of zinc and magnesium was 10-1000 times greater in acid than in pure water, and correlated with the concentrations and the dissociation constants, pK1, of the acids tested. In all cases, important water losses were observed. In the 0.01 M acids, the cement structure collapsed to form a viscous, compact and homogeneous layer on the cement surface. In this layer, the polymeric carboxylic chains were regenerated from the zinc and magnesium polycarboxylate cement. Comparison with pure water showed that even the smallest concentration of the weak acids greatly modified the cement behaviour. This could explain the well-known differences in erosion processes between theoretical erosion predicted by standard specification tests and the in vivo situation.