On the Question of the Contact Melting of Substances Forming Eutectics

P. A. Savintsev

The formation of binary eutectics is explained by a number of authors [1, 2, 3] by the character of intermolecular interaction. Thus, at a temperature somewhat below the eutectic temperature, the interaction between like atoms (A—A or B—B) is greater than the interaction between unlike atoms (A—B).

At a temperature somewhat above the eutectic temperature, the interaction between unlike atoms is greater than between like atoms. Such a change in interaction takes place within a certain temperature interval [1, 3]. Apparently, this hypothesis may be accepted for explaining the formation of ternary and quaternary eutectics and the contact melting of components that form eutectics in experiments [4, 5]. In these works it is shown that a eutectic structure is by no means obligatory for eutectic melting. The latter does not determine the low-melting character of eutectics.

It should be recalled that the possibility of such experiments for binary eutectics was pointed out earlier by the Kazan professor F. M. Flavitskii [6]. He considered that a eutectic represents a chemical compound, and that the formation of a liquid phase at the contact of two crystals represents a manifestation of chemism. Flavitskii’s works were forgotten by many because the theory of the chemical nature of eutectics had been rejected.

N. S. Kurnakov and N. N. Efremov [7], who observed the formation of binary liquid eutectics at the contact of two crystals, returned again to this question. In their experiments, pieces of camphene, having a melting temperature of 49.3°, and methyl mustard oil (melting temperature 34°) were brought into contact in a test tube at room temperature, 15°, as a result of which, after some time, liquid appeared in the test tube. The state diagram of this system, obtained by the authors, has no solid solutions at all; the eutectic contains 84% camphene and melts at a temperature of minus 8°. In the experiments the ratio of the components was not eutectic. The authors explain the appearance of liquid under these conditions by the fact that the temperature of the experiment considerably exceeds the eutectic temperature.

Because of this, “the region of existence of the liquid phase at ordinary room temperature lies in a very large concentration interval—between 14–76.5 mol. % of mustard oil.” For this reason, in the experiment under consideration, “liquefaction is not connected with narrow limits of weight ratios, but readily occurs over broad limits of change in the composition of the mixture” [7].

The authors then investigated the systems camphene–naphthalene, camphene–phenanthrene, and methyl mustard oil–naphthalene. The eutectic concentrations and temperatures of these systems are given in Table 1.

In these three systems the extreme limits of existence of the liquid phase at 15° are much closer to one another, in comparison with the system

camphor–methyl mustard oil. Thus, for example, camphor with phenanthrene gives a liquid at 15° only within the concentration range between 9–10 mol.% phenanthrene (emphasis ours) [7].

Table 1

Thus Kurnakov and Efremov assert that, when the concentration of the components differs considerably from the eutectic concentration and the overheating is small, melting in contact will not be observed.

Saratovkin and Savintsev [4] observed contact melting at a concentration of the components differing considerably from the eutectic one, and with a small excess of the experimental temperature over the eutectic temperature. The experiments were carried out with substances forming binary, ternary, and quaternary eutectics. Later, similar experiments were performed by Avakyan and Lashko [5].

However, in these experiments [4, 5] the phenomenon of contact melting for ternary and quaternary eutectics was not observed in pure form. Thus, in order to observe contact melting of components giving ternary eutectics, a piece of a binary eutectic was taken and brought into contact with a third component, and the entire system was heated several degrees above the melting temperature of the ternary eutectic. The authors proceeded analogously when observing contact melting for components giving a quaternary eutectic. Here melting was observed in contact between a piece of a ternary eutectic and a fourth component. The same result could be reached by bringing into contact two pieces of binary eutectics.

Direct observations of melting in contact of three or four components have not been described by anyone. At the same time, obtaining the effect of contact melting under direct contact of three or four components appears interesting from the standpoint of the hypothesis concerning the interaction of like and unlike atoms.

The experiments described by us were designed to prove the possibility of contact melting for three and four components.

In contrast to the previous authors [4, 5], we used not massive pieces of metal but powders.

An increase in the dispersity of the components leads to an increase in the total contact surface between them, thereby facilitating contact melting. Low dispersity does not lead to these results.

We carried out experiments with powders ranging in size from 0.1 mm to 1.0 mm. The powders were prepared from Pb, Bi, Sn. The required temperatures were maintained by an ultrathermostat with an accuracy up to ±0.05°C. The comminuted metal was sifted through sieves of the corresponding sizes. The selected samples were placed in test tubes and poured over with a solution of boric acid and glycerin to remove oxides and ensure the cleanliness of the contacting surfaces. In all other respects the experimental procedure did not differ from that described in [4]. The results of the experiments are given in Table 2.

Table 2

Dependence of the contact-melting temperature of lead–tin–bismuth powders on particle diameter

The increase in the melting temperature with increasing particle size is apparently explained by the fact that, in this case, the probability of the simultaneous meeting of three unlike particles decreases.

A similar picture is observed for powders of the components forming a quaternary eutectic (Pb, Sn, Bi, Cd).

Thus, our experiments have shown the possibility of eutectic contact melting directly at the place of contact of three (four) components forming eutectics. Since the phenomenon of contact melting for two, three, and four components is observed under the same experimental arrangement, this argues in favor of the possibility of extending the hypothesis concerning the interaction of like and unlike particles in binary eutectics to ternary and quaternary eutectics.

It is of interest to study the effect, on the interaction of atoms giving a eutectic, of certain impurities that form solid solutions with one of the components. A system satisfying these conditions is the lead–tin–bismuth–antimony system.

Here bismuth and antimony form unlimited solid solutions, while tin, bismuth, and lead form a eutectic. If pieces of the solid solution antimony–bismuth and of the lead–tin eutectic are brought into contact, then at the boundary three kinds of atoms (lead, tin, bismuth) will meet, the interaction of which will be complicated by the fact that bismuth enters the lattice of antimony or, conversely, when there is more bismuth than antimony.

We set ourselves the task of determining the melting temperature at the above-mentioned boundary. The antimony–bismuth solid solution was taken in various concentrations from 10 to 90% Bi. To obtain it, technically pure bismuth and antimony were fused in a porcelain crucible under a layer of flux. The samples, in the form of plates of bismuth and antimony, were then brought into contact with plates of the lead–tin eutectic, and the contact-melting temperature was determined with the aid of an ultrathermostat. The results of the experiments are given in Table 3, from which an increase in the melting temperature with an increase in the percentage of antimony in the antimony–bismuth solid solution follows.

This can apparently be explained by the fact that an increase in the percentage of antimony means a decrease, at the contact surface, in the number of bismuth atoms forming a ternary eutectic with tin and lead.

Table 3

According to G. Tammann [8], the lowest-melting alloy in the lead—antimony—tin system melts at a temperature of 189°, whereas the lead—tin eutectic melts at a temperature of 183°. It seems to us that the explanation for this should be sought in the experiments just described.

Experiments were then carried out to study contact melting at the boundary between a solid lead—bismuth solution and pure bismuth. According to the phase diagram, the solubility of bismuth in lead in the solid state is 37%. Consequently, by alloying lead and bismuth in quantities not exceeding 37% bismuth, we obtain a single-phase system. If, however, the amount of bismuth is greater than 37%, then a eutectic will be present in the solidified alloy.

By melting lead and bismuth under a layer of flux, we obtained a solid solution containing 5, 10, and 30% bismuth. These alloys were brought into contact with bismuth specimens, and the temperature of contact melting was determined with the aid of an ultrathermostat. The results are given in Table 4.

Table 4

These experiments showed that, with an increase in the percentage of bismuth in the solid bismuth—lead solution, the temperature of contact melting at the boundary of this solid solution and bismuth decreases.

The results obtained can apparently be explained by a change in the interaction between atoms in the solid solution as compared with the interaction of atoms in the pure metal.

This is consistent with the assertion that the melting temperature is determined, in a first approximation, by the bonding forces between atoms. In our case, however, the solid solution has a lower melting temperature than pure lead.

The experiments we have performed lead to the following conclusions:

1. The hypothesis of molecular interactions, which explains the formation of binary eutectics, can be extended to ternary and quaternary eutectics.

2. The temperature of contact melting depends on the degree of dispersion of the components. The smaller the dimensions of the contacting crystallites, the lower the melting temperature.

3. Atoms of foreign substances that do not form eutectic mixtures with the given pair can significantly raise the temperature of contact melting by hindering the encounter of the corresponding atoms. Thus, the lead + tin—bismuth + antimony system has a higher melting temperature than the lead + tin—bismuth system.

I express my deep gratitude to the supervisor of this work, Associate Professor A. F. Gorodetskii, for valuable guidance and critical comments during its execution.

LITERATURE

1. V. I. Danilov. Izv. AN SSSR, Physical Series, 5, 1, 1941.
2. Ya. I. Frenkel. Introduction to the Theory of Metals, 1948.
3. K. Bunn. Izv. AN SSSR, Division of Technical Sciences, 2, 1946.
4. D. D. Saratovkin and P. A. Savintsev. DAN SSSR, 33, 4, 1941; 58, 9, 1947.

1. S. V. Avakyan and N. F. Lashko. DAN SSSR, 65, 1, 1949.
2. F. M. Fawitskii. Investigation of a eutectic alloy of nitrogen-oxygen salts of silver and ammonium by the method of melting. ZhRKhO, 41, 739, 1909.
3. N. S. Kurnakov and N. N. Efremov. Proceedings of the Petrograd Polytechnic Institute 18, 105, 1912.
4. G. Tammann. Metallography, p. 419, 1931.