Wetting at High Temperature - Drops and Bubbles in Contact with Solid Surfaces - page 324

minimum oxygen diffusion in the experimental 'cell', (4) the specimen is back-lit
to obtain a constant contrast in the recorded image, whereas the self-emitted light
is filtered out, (5) vacuum turbomolecular pumps are used to produce the required
high-vacuum conditions, (6) the specimen support can be accurately leveled at any
moment.
To perform reliable wetting tests a certain number of requirements should be fol-
lowed strictly. First of all, it should be taken into account that an extremely impor-
tant role is played by the atmosphere surrounding the specimens as it can strongly
affect the wetting results (the final or equilibrium contact angle and the spreading
kinetics [211]) as discussed in the previous paragraphs. Ultra high vacuum condi-
tions can assure that contamination from residual gases (oxygen in particular) is
kept to a minimum, but, at the same time, evaporation from the liquid and/or the
solid surfaces can alter the residual atmosphere and thus the interfacial chemistry.
For this reason, a getter (e.g., Zr), is often placed near the specimen to assure the
right oxygen pressure to obtain stable oxide-free surfaces.
In particular, under low pressures, also the materials forming the experimental
cell can enter into play, for example giving rise to volatile oxides which impose
a specific oxygen partial pressure, as in the presence of silica-based materials. If
gas mixtures are used, their oxygen content should be monitored; it is suggested
to measure both the inlet and the outlet oxygen partial pressures, when working
under flux, if the 'local' values are not accessible (solid-state oxygen gauges do no
work above about 900 ◦ C). It should be reminded that, when the PO 2 is dictated
by an equilibrium reaction in gas phase (buffer systems) the value measured by the
pressure gauges must be re-calculated at the experimental temperature.
Another critical point is linked to the sequences leading to the melting of the
metallic phases and its coming into contact with the solid surface under study. In-
deed, if the metal and the solid support are heated up together, many interaction
phenomena can occur during this, usually long, phase. In particular, surface oxi-
dation can easily occur, even if vacuum or protecting gases are used, because the
oxygen equilibrium pressure for oxide formation is very low at low temperatures,
at levels (often below 10 − 15 -10 − 20 Pa) which cannot be 'corrected' by protective
gases. At the same time, chemical interactions can occur between the two solid
phases, leading to mutual diffusion phenomena which can arrive at causing local
contact melting processes. To avoid this, two possibilities exist: the first one is to
introduce the metal/ceramic couple, charged in a pre-evacuated chamber, into the
preheated furnace (e.g., by a magnetically operated push-rod) only when all pa-
rameters ( T ,PO 2 ) have reached equilibrium; the time necessary to complete the
melting process is usually of the order of tens of seconds, depending on the set
temperature, the atmosphere, the mass of the test pieces and their specific heat.
At the end of the pre-set holding period, the couple is moved into the colder part
of the experimental chamber without opening it, where it is cooled down to room
temperature (this choice is shown in Fig. 19).