Diffusion-limited reactive wetting: effect of interfacial reaction behind the advancing triple line F. Hodaj Æ O. Dezellus Æ J. N. Barbier Æ A. Mortensen Æ N. Eustathopoulos Received: 28 March 2007 / Accepted: 4 June 2007 / Published online: 3 July 2007 Ó Springer Science+Business Media, LLC 2007 Abstract Using the ‘‘dispensed drop’’ variant of the sessile drop technique, spreading kinetics of dilute Cu–Cr alloys on smooth vitreous carbon substrates are measured under helium microleak conditions. In this system, it is known that the drop spreading rate is controlled by diffusion of the reactive atom species (Cr) from the bulk liquid to the triple line, where wetting is induced by formation of an interfacial layer of chromium carbide. Microstructural characterization of rapidly cooled drops shows that growth of the interfacial reaction product layer continues behind the moving solid– liquid–vapor triple line. The spreading velocity is modeled by finite-difference numerical analysis of diffusion near the triple line in the presence of continued interfacial reaction, simplifying the growth rate as being constant and using realistic parameter values. We show that continued interfa- cial reaction explains the dependence of the triple line spreading rate on the instantaneous wetting angle that is observed in this system. Introduction Non-metallic solids, such as oxides, carbides, or carbon, are generally poorly wetted by liquid engineering metals such as tin or copper: contact angles typically exceed 90° [1, 2]. The complications this produces in materials processes such as infiltration or brazing have motivated the search for various means to lower the contact angle in such systems. One approach to this end is to alloy the metal with elements that, by reaction with the solid, form a better- wetted compound. On substrates of carbon for example, chromium or titanium alloyed into molten copper form, by reaction with the carbon, a chromium or titanium carbide. These carbides, of partly metallic character, are relatively well wetted by the metal [3, 4]. When a droplet of such reactive alloys comes in contact with the solid, it first spreads very rapidly, at a rate determined by the (low) viscosity of the metal. It thus quickly reaches a first metastable equilibrium contact an- gle, characteristic of the metal on the unreacted solid substrate. After a short transient stage (a few seconds) a thin continuous reaction layer is then formed along the solid/liquid interface [5]. Thereafter, since the reaction layer is better wetted than the original substrate, the drop starts spreading again as the layer expands outward of the triple line, parallel to the interface. In this second spreading process, the drop spreading rate is controlled, not by the metal viscosity, but rather by the rate at which the reaction product layer can grow at the triple line, parallel to the solid surface. Spreading continues in this manner until the contact angle becomes equal to the equilibrium contact angle of the alloy on the reaction product, h e . Following this, the drop remains stable even though the reaction layer may continue to grow, both in thickness and further out- wards of the triple line, along the substrate free surface. F. Hodaj (&)  J. N. Barbier  N. Eustathopoulos SIMAP – UMR CNRS 5266, INP Grenoble-UJF, Domaine Universitaire, BP 75 – 1130, rue de la Piscine, 38402 Saint Martin d’Heres, Cedex, France e-mail: fhodaj@ltpcm.inpg.fr O. Dezellus LMI – UMR CNRS No. 5615, Universite ´ Claude Bernard Lyon 1, 43 Bd du 11 novembre 1918, 69622 Villeurbanne Cedex, France A. Mortensen Laboratory for Mechanical Metallurgy, Ecole Polytechnique Fe ´de ´rale de Lausanne, Lausanne 1015, Switzerland 123 J Mater Sci (2007) 42:8071–8082 DOI 10.1007/s10853-007-1915-0