007s.pdf

Welding Journal | January 2014

and semiconducting — with molten metals that will allow doing further steps in understanding metal-perovskite ceramics interaction and elaboration of some brazing alloys and technological processes for joining (brazing) of BaTiO3 materials. Semiconducting BaTiO3-x Nonstoichiometric semiconducting BaTiO3-x can be obtained by means of annealing in high vacuum, as is mentioned above. It is believed that BaTiO3-x nonstoichiometry is insignificant and will not exceed such parameters for pure titanium oxide TiO2-x. For this oxide, the x value is between 0.04 and 0.07 (Ref. 12). Such deviation only has a minor effect on mechanical and thermodynamic properties of the compound, except for the electrophysical characteristics. Such nonstoichiometry variation can be attributed to oxygen vacancies compensated mostly by background and/or intrinsic acceptors within higher oxygen partial pressure (p(O2)) regions and by electrons within lower p(O2) regions (Ref. 13). Semiconducting BTO has specific resistivity value near 300 Ω⋅cm (compared to about 106–10 Ω⋅cm for ferroelectric BaTiO3). The technology of vacuum metalization and brazing by melts containing titanium as a chemically active element was tested for preliminary annealed semiconducting barium titanate ceramics. Experimental and Discussion The main experiments consist of the wettability measurements of BTO by liquid metals. Wettability studies were carried out by a sessile drop method in vacuum (~ 10–4 Pa) at temperature 870–1870 K. This method allows determining the values of the wetting contact angle and interphasic surface energy at the liquid-gas interface. The sessile drop method essence was discussed in detail earlier (Refs. 14–16). The main requirement for measuring wetting contact angle by the sessile drop method is in the placement of a symmetrical drop of the melt on the solid surface. The sample should be in controlled gaseous atmosphere or in at the temperature specified. Standard equipment for the wettability of solid ceramic specimens by a liquid metals study using the sessile drop method is shown in Fig. 1. A wide variety of metals and alloys having a broad application range in electroceramic devices was used. Fourteen pure metals (Cu, Ag, Au, Ge, Sn, Pb, Ga, In, Al, Si, Ni, Co, Fe, Pd) and several titaniumcontaining alloys (Cu-Sn-Ti, Ag-Cu-Ti, Cu-Ga-Ti, In-Ti) were tested. Metal samples for wetting experiments typically have approximately 0.5–0.9 g. Metal alloys were formed in-situ by alloying. Polycrystalline barium titanate has been specially fabricated by the method of solid-phase synthesis. In this study, we used BaTiO3 ceramic discs 20 mm in diameter and ~ 3 mm thick. The sample’s porosity was 3.5±0.03%. BTO substrates were ground and polished with sandpaper and abrasive powder. The average surface roughness value (Ra) was equal to 0.02 μm. Before experiments, BTO samples were annealed in vacuum at ~ 1740 K during 60 min. The wetting of BTO by molten alloys Cu 8.6% (at.) Sn, Ag 39.9% (at.) Cu, and Cu 17.6% (at.) Ga (which was used to create the many braze alloys) with active titanium additive (from 3 up to 25% (at.)) was studied as well. Results of the wetting studies of BTO by pure metals melts and some alloys are presented in Table 1. Most of the investigated pure metals did not wet the barium titanate ceramics surface (contact angles exceeded 90 deg). Silicon and aluminum wet BTO (alu- WELDING JOURNAL 7-s WELDING RESEARCH Fig. 1 — Scheme of the apparatus for determining surface tension and wetting angles of metallic liquids. The labeled numbers represent the following: vacuum chamber (1); stream-oil pump (2); vacuum valve (3); furnace (4); metal sample on the ceramic substrate studied (5); next samples (6); horizontal rod (7); quartz prism (8); vertical rod (9); and digital camera (10). Fig. 2 — Dependence of contact angle of BaTiO3 for the pure metals melts on free energy of their oxides formation. Fig. 3 — Contact angle/titanium concentration dependence for melted Ti-containing systems on BaTiO3 at 1270 K.


Welding Journal | January 2014
To see the actual publication please follow the link above