Yu. P. NIKITIN, V. G. KORPACHEV, A. N. SAFRONNIKOV
Unknown
Submitted 1963-01-01 | RussiaRxiv: ru-196301.28473 | Translated from Russian

Full Text

PHYSICAL CHEMISTRY

Yu. P. NIKITIN, V. G. KORPACHEV, A. N. SAFRONNIKOV

SURFACE PROPERTIES OF MELTS BASED ON CaF₂

(Presented by Academician A. N. Frumkin, 7 IX 1962)

The study of the surface properties of oxide melts makes it possible to establish a connection between them and the nature of the cations of the melt. The introduction of CaF₂ into the latter leads to systems in which it is possible to vary the relative amounts not only of different cations, but also of anions.

Investigation of the dependence of the surface tension \((\sigma)\), measured by the sessile-drop method \((^{1,2})\), on the composition of the melt at 1470–1550° C showed that, to a first approximation, there is a definite parallelism between \(\sigma\) and the ionic fraction of oxygen \((N_{\mathrm{O}^{2-}})\) in the melt \((^3)\) (see Table 1). Indeed—

Table 1

Surface tension of oxide–fluoride melts at 1470–1550°

No. CaF₂ CaO Al₂O₃ SiO₂ MgO Ionic fraction of oxygen \(N_{\mathrm{O}^{2-}}\) \(\sigma\), erg/cm²
1 100 0 280
2 95 5 0.036 290
3 85 15 0.110 300
4 74 26 0.200 315
5 86 14 0.180 285
6 80 20 0.225 295
7 71 26 0.325 330
8 65 7 10 12 6 0.380 340
9 52 21 27 0.440 375

the addition of calcium oxide to CaF₂ up to 26 wt.% raises \(\sigma\) from 280 to 315 erg/cm². This increase in \(\sigma\) may be associated with the appearance in the surface layer of the melt of oxide anions with a charge twice as large as that of the fluoride anion. Evidence in favor of this assumption is provided by the results of a comparison, by the method \((^4)\), of the surface potentials of liquid CaF₂ and of a melt with 26 wt.% CaO. Their difference \((\Delta \varepsilon_\omega)\) was determined from the relation

\[ \Delta \varepsilon_\omega = \Delta \varepsilon'' - \Delta \varepsilon' - \Delta \varepsilon_x . \tag{a} \]

To determine \(\Delta \varepsilon_x\), a galvanic cell was assembled in corundum crucibles:

\[ \mathrm{Pt}\,(\mathrm{I}) \;|\; \underset{(\text{traces of CaO})}{\mathrm{CaF}_2} \;|\; \mathrm{CaF}_2-\mathrm{CaO} \;|\; \mathrm{Pt}\,(\mathrm{II}), \]

for which

\[ \Delta \varepsilon_x = \varepsilon_{\mathrm{II}} - \varepsilon_{\mathrm{I}} = 32\ \text{mV}. \tag{b} \]

The values of \(\Delta \varepsilon''\) and \(\Delta \varepsilon'\) were determined as the difference in potentials, respectively, between electrodes I or II and platinum electrode III, located above the surface of the melts under study, i.e.,

\[ \Delta \varepsilon'' = \varepsilon_{\mathrm{III}} - \varepsilon_{\mathrm{I}} = 150\ \text{mV}, \tag{c} \]

and

\[ \Delta \varepsilon' = \varepsilon_{\mathrm{III}} - \varepsilon_{\mathrm{II}} = 90\ \text{mV}. \tag{d} \]

All electrodes were in a stream of unpurified nitrogen (2–4% \(O_2\)). As a result of the measurements it was found that the surface potential of the melt with 26% CaO is 28 mV more negative than that for \(\mathrm{CaF}_2\).* The introduction of aluminum oxide into \(\mathrm{CaF}_2\) up to 15 wt.% reduced \(\Delta \varepsilon_\omega\) almost to zero. Apparently, in the \(\mathrm{CaF}_2-\mathrm{Al}_2\mathrm{O}_3\) system the surface layer also contains an appreciable amount of oxygen anions, which increases \(\sigma\) of the melt when \(\mathrm{Al}_2\mathrm{O}_3\) is introduced into it. The addition of \(\mathrm{SiO}_2\) to \(\mathrm{CaF}_2\) has almost no effect on \(\sigma\), which is promoted by the form in which \(\mathrm{SiO}_2\) exists in the melt \((^5)\). The simultaneous introduction of various oxides into \(\mathrm{CaF}_2\) appreciably increases \(\sigma\).

By an approximate estimate, based on the shape of solidified drops \((^6)\), of the interfacial tension \((\sigma_m)\) of certain metallic alloys based on nickel and iron, a parallelism was found between \(\sigma_m\), \(\sigma\), and \(N_{\mathrm{O}^{2-}}\) in the melt (see Table 2). From the electrocapillary motion \((^7)\) of metallic drops in the bulk of oxide-fluoride melts \((^8)\) toward the anode, it was established that the surface of the metal at the interphase boundary with the melt has a negative electric charge.

Table 2

Interfacial tension of metallic alloys at the boundary with an oxide-fluoride melt

Composition of melt \(N_{\mathrm{O}^{2-}}\) 20% Cr, 77% Ni, 2% Ti, 1% Al 20% Cr, 79.5% Ni, 0.5% (Ti+Al) 75% Fe, 15% Cr, 9% Ni, 1% Al 73.5% Fe, 18% Cr, 8.8% Ni, 0.5% Ti
100% \(\mathrm{CaF}_2\) 0.000 1230 1300 1315 1150
74% \(\mathrm{CaF}_2\), 26% CaO 0.200 1250 1300 1350 1200
71% \(\mathrm{CaF}_2\), 29% \(\mathrm{Al}_2\mathrm{O}_3\) 0.325 1300 1370 1520 1300
65% \(\mathrm{CaF}_2\), 12% \(\mathrm{SiO}_2\), 6% MgO, 10% \(\mathrm{Al}_2\mathrm{O}_3\), 7% CaO 0.380 1350 1430 1500 1380
52% \(\mathrm{CaF}_2\), 21% CaO, 27% \(\mathrm{Al}_2\mathrm{O}_3\) 0.440 1300 1380 1530 1310

Apparently, replacement in the melt of \(\mathrm{F}^-\) ions by \(\mathrm{O}^{2-}\) decreases the negative charge of the metal surface \((^9)\) and thereby increases \(\sigma_m\) \((^{10})\).

The parallelism between \(\sigma_m\) and \(\sigma\) indicates, in particular, the constancy of the work of adhesion of metals to oxide-fluoride melts

\[ W=\sigma+\sigma' - \sigma_m, \]

where \(\sigma'\) is the surface tension of the metal at the boundary with the gas. Indeed, a simultaneous increase in \(\sigma\) and \(\sigma_m\) does not change \(W\).

Wetting of these same solid metals by oxide-fluoride melts in a furnace atmosphere with a carbon heater was very good (\(\theta = 3 \div 5^\circ\)), which can be explained by some oxidation of the metal. This is confirmed under the same conditions by the direction of the electrocapillary motion of metal drops over the surface of the melt toward the cathode.

Ural Polytechnic Institute
named after S. M. Kirov

Received
25 VII 1962

CITED LITERATURE

  1. H. K. Adam, Physics and Chemistry of Surfaces, Moscow, 1947.
  2. S. I. Popel, Yu. P. Nikitin, S. M. Ivanov, Graphs for Calculating Surface Tension from Drop Dimensions, Sverdlovsk, 1961.
  3. M. I. Temkin, ZhFKh, 20, 105 (1946).
  4. A. I. Manakov, O. A. Esin, B. M. Lepinskikh, DAN, 136, No. 3, 644 (1961).
  5. O. A. Esin, P. V. Geld, Physicochemical Theory of Metallurgical Processes, 2, Moscow, 1954.
  6. A. A. Leont’eva, ZhFKh, 19, 388 (1945).
  7. A. N. Frumkin, V. G. Levich, ZhFKh, 19, 573 (1945).
  8. V. V. Khlynov, O. A. Esin, DAN, 120, 134 (1958).
  9. Yu. P. Nikitin, O. A. Esin et al., Proceedings of Higher Educational Institutions, Ferrous Metallurgy, No. 5 (1962).
  10. A. N. Frumkin, V. S. Bagotskii et al., Kinetics of Electrode Processes, Moscow, 1952.

* In determining \(\Delta \varepsilon_\omega\), the minimum value \(\Delta \varepsilon''\) and the maximum values \(\Delta \varepsilon'\) and \(\Delta \varepsilon_x\) observed over the course of 10 min were used.

Submission history

Yu. P. NIKITIN, V. G. KORPACHEV, A. N. SAFRONNIKOV