Wetting and Reactivity between Nitride Ceramics and Molten Copper and Silver Alloys

Wetting and reactivity were studied for molten Cu-Cr, Cu-Zr and Ag-Zr alloys on two nitride ceramics AlN and Si₃N₄. The aim of the study was to understand the interfacial behaviour of these metal alloys in molten state on each ceramic. The potential for the successful application of these material combinations can be evaluated by measuring the contact angle formed in the wetting test. Wetting test is also a suitable method for producing interfacial reactions and analysing them after the experiment. If good bonding is the most important requirement, the wetting should be as good as possible and the reactivity should not produce cracking in the materials. On the other hand, if a high temperature material is needed for metal alloy handling applications, the wetting should not be too good and the reactivity should be limited to avoid the cracking of the materials. High vacuum sessile drop experiments were carried out with polished substrate samples by in-situ alloy manufacturing. During the experiment the development of contact angles and other dimensions were determined as a function of time and the spreading rate was used to evaluate the kinetics of the reactive wetting. After the experiment the adhesion of the drop to the surface was evaluated through possible crack formation and chemical analyses for the interfaces were carried out with SEM and EDS. Thermodynamic calculations for predicted reactions were carried out to determine the driving forces for these reactions. The results were used to identify the phases revealed at the interface after the wetting experiments. Among the studied Cu- and Ag- base metals with Cr- and Zr-additions only the Ag-Zr alloy on Si₃N₄ showed perfect wetting. Same alloy on AlN produced the contact angle of 15°±2°. The best wetting with copper alloys was obtained with the CuCr1,5 alloy on Si₃N₄ showing the contact angle of 75°±2°. Other studied combinations did not produce wetting. Interfacial reactivity was revealed with all Zr-and Cr-additions. The formation of ZrN and Cr₂N was confirmed by SEM analyses. Significant amounts of oxygen were also found in most cases and it was concluded that oxygen interfered wetting, even if no separate oxygen-containing phases were found in interfacial layers. The AlN substrate cracked after wetting experiments due to the strong bonding of interfacial layers. The Si₃N₄ substrate did not crack, even if the observed reaction layers were several micrometers thick. Since oxygen was found in the interfacial layers even in these high vacuum experiments it will probably have a strong influence on the behaviour in real applications. It may even prevent some of the observed interactions by interrupting the process of wetting