Fiqiri Hodaj

Chemical reaction-limited spreading: the triple line velocity versus contact angle relation

The mechanisms of reaction-limited spreading of molten copper and nickel silicides on carbon substrates are studied by the dispensed drop variant of the sessile drop technique under high vacuum. Both standard wetting experiments and experiments which separate the effect of time and instantaneous contact angle on spreading rate are carried out. Equations are derived for the time-dependent variation of contact angle and spreading rate.

Progress in modelling of chemical-reaction limited wetting

In reactive metal/ceramic systems the wetting rate of small, millimeter sized droplets on smooth ceramic surfaces is controlled by the slower of two successive phenomena that intervene in the reaction process: diffusive transport of reacting species to or from the triple line, and local reaction kinetics at the triple line. The first case, of diffusive wetting, was modelled by Mortensen et al. [Scripta mater. 36 (1997) 645] a few years ago. The purpose of this paper is to present progress accomplished during the last few years in modelling the second type of wetting controlled by the reaction process at or close to the triple line. The predictions of equations derived from a model describing the change in contact angle and spreading rate with time are compared with experimental results obtained for different silicon alloys on carbon substrates.

Mechanism of solid-state dissolution of WC in Co-based solutions

Abstract The dissolution mechanism of WC in solid solutions (Co, W, C) with different C/W ratios and Cr or V additions is studied at 1200°C using the diffusion couple technique. Depending on the composition of the Co-based phase (β), three sequences are generated: with no intermediate phase layer (β/WC), or with one (β/η/WC) or two (β/η/η′/WC) layers of η (Co3W3C) and η′ (Co6W6C). The experimental results — diffusion profiles, interface compositions, layer widths — show that WC dissolution is limited by diffusion and lead to values of the diffusion coefficient of 3×10 −15 , 2×10 −15 and 6.5×10 −15 m 2 s −1 for β, η and η′, respectively. The Cr or V addition does not significantly change the WC solubility in Co but it reduces the limit W content corresponding to the formation of η and η′.

Spreading of Cu-Si alloys on oxidized SiC in vacuum: experimental results and modelling

Abstract The wetting behaviour of a Cu–40 at.% Si alloy, non-reactive with SiC, on oxidized (0001) faces of α-SiC single crystals is studied at temperatures close to the melting point of copper by the “dispensed drop” technique under high vacuum. The experiments focused on wetting kinetics to determine the mechanisms controlling the rate of spreading under conditions allowing for “in situ” deoxidation of silicon carbide. It is shown that spreading occurs in three stages: (i) an initial stage consisting of a rapid spreading of the alloy on the oxidized SiC up to a contact angle which is close to that on vitreous silica, (ii) a second stage with a zero spreading rate during which the alloy goes through the oxide layer by dissolution near the triple line allowing an intimate contact between the alloy and the SiC surface to be established, and (iii) a final stage during which the alloy spreads with a constant rate up to a contact angle equal to that of clean SiC. These results are interpreted on the basis of a dissolution–diffusion–evaporation process occurring in the vicinity of the solid–liquid–vapour triple line.

Wetting and interfacial reactivity in Ag–Zr/sintered AlN system

Abstract Wetting (both final contact angles and spreading kinetics) of pure Ag and Ag–Zr alloys diluted in Zr on sintered AlN is studied by sessile drop experiments under high vacuum at 970 °C. The results are related to the interfacial reactions between Zr and AlN. The role of thermodynamic activity of Zr on wetting is discussed.

DIFFUSION-LIMITED REACTIVE WETTING: SPREADING OF Cu-Sn-Ti ALLOYS ON VITREOUS CARBON

When wetting is promoted by interfaces, the rate at which the liquid spreads over the solid substrate is controlled by the rate at which the well-wetted interfacial product can be formed along the liquid/solid/atmospheric triple line. If this reaction involves a reactive alloy addition to the liquid, the rate of spreading can be limited by two phenomena. A simplified analysis of diffusion- limited reactive spreading of Cu-Sn-Ti alloys on vitreous carbon was conducted based on the assumption that the interfacial reaction is strictly localized at the triple line. The addition of 3at.%Ti to Cu-15at.%Sn led to very good wetting on vitreous carbon. This was due to the formation, at the interface, of a continuous layer of TiC.

Standard enthalpies of formation of some Mn–Y and Mn–Sc intermetallic compounds

Abstract The standard enthalpies of formation of Mn2Y, Mn23Y6 and Mn2Sc have been determined by aluminium drop solution calorimetry in an isothermal double cell Tian–Calvet high temperature microcalorimeter at 1123 K. The heats of dissolution of the elements and the intermetallic compounds have been measured at infinite dilution in liquid aluminium. Mn, Y and the three intermetallic compounds dissolved rapidly. It was not possible to determine the value for Sc directly due to the low solubility of this element in Al, which led to a precipitation of Al3Sc without further dissolution. The value has therefore been determined indirectly by the dissolution of two intermetallic compounds — Sc5Ge3 and Al2Sc — for which the enthalpies of formation are known from the literature.

Flux-driven nucleation at interfaces during reactive diffusion

Nucleation of intermetallic compounds and of voids at interfaces during reactive diffusion is treated with account of influence of the flux divergence in the nucleation regions of the real space as an additional term for drift in the size space (in Fokker–Planck equation for nucleation). Such approach enables the construction of effective Gibbs nucleation barrier which may (in the broad region of parameters) increase to infinity meaning the full suppression of nucleation, or, by the contrast, decrease assisting the nucleation. The introduced effective nucleation barriers depend on kinetic factors – on the ratio of diffusivities in nucleating and in neighboring phases. Thus, the competition of stable and metastable phases is reconsidered, as well as nucleation of Kirkendall/Frenkel voids at the interfaces.

On thermal effects in reactive wetting

LTPCM-INP Grenoble, UJF, URA 29, BP 75, 38402, Saint Martin d’He`res, Cedex, France(Received December 29, 1997)(Accepted January 14, 1998)IntroductionHigh-temperature capillary phenomena intervene in many material processing operations, includingbonding operations such as soldering and brazing, and infiltration processes used to produce metal orceramic matrix composites. Of particular importance in this class of processes is the issue of wettingof solids by molten inorganic materials: good wetting, as manifest by a low contact angle of the liquidon the solid in the processing environment, drives flow of the liquid over the solid and eases processingsignificantly. This issue has therefore motivated a large body of research aimed at developing methodsfor improving wetting in high-temperature systems of practical interest. Results from contact anglemeasurements, and also from process development work, indicate that chemical interaction between thesolid and the liquid tends to promote good wetting (1–5).Wetting is usually characterized by measurement of contact angles in sessile drop experiments,wherein a drop of the liquid is placed at fixed temperature on a smooth and flat substrate of the solid.When interfacial reactions drive wetting, these cause transitions with time in the contact angle while thetriple line advances at a rate determined by the rate of reaction product formation along the substratesurface (e.g., (6–9)): reactive wetting is thus generally a dynamic phenomenon. It is customarilyassumed in the interpretation of sessile drop data that the wetting process is isothermal, even in dynamicsituations typical of reactive wetting; however, interfacial reactions encountered in high-temperaturematerials processing can be highly exothermic. These could, therefore, cause a significant increase inthe local temperature at the triple line and, hence, influence both reaction kinetics and capillary forcesin reactive wetting.We examine here this influence using a simplified description of the problem, in which we take therate of heat evolution behind the triple line of solid/liquid/vapor contact to be constant and limited intime, and the wetting process to be fully steady-state. More specifically, we address two very basicsituations, encountered both in materials processes and in experimental measurements of wetting. Thefirst is that of a liquid spreading over a bulk solid substrate, typical in particular of sessile dropexperiments; the second is that of a liquid infiltrating a porous solid such as a preform of fibers orparticles, or a thin capillary. We seek, in the most simple and general terms possible, to assess whetherheat evolution at reacting interfaces could cause significant departures in the local temperature at thetriple line.

Cu3Sn suppression criterion for solid copper/molten tin reaction

The formation of Cu3Sn phase in the soldering reaction is believed to be harmful for the reliability of solder contacts on account of Kirkendall voiding in the compound. In this study, a criterion for the suppression of the growth of this phase by the fast growing scallop-like Cu6Sn5 compound is presented. The average thickness of the η-Cu6Sn5 phase above which the ε-Cu3Sn phase starts to grow as a continuous layer at the Cu/Cu6Sn5 interface during liquid Sn/solid Cu interaction has been evaluated from thermodynamic and kinetic considerations.