A new concept, the "energetic parameter of solution", is suggested for molten salt solutions. For the A_n B_m—A_n' C_m' system (valency of A ion is Z):f(r)=(Z~2nn')/(m(r_A+r_B))+(Z~2n'~2)/(m'(r_A+r_C))-(2Z~2nn')/(2r_A+r_B+r_C)(n'/nm'+1/m)n>n'. For A_nB_m—C_n'D_m' system (valency of C ion is Z'): f(r)=(n'Z'Z((n+nZ)/(n'Z')m'))/(m'(r_A+r_D))+nZZ'(m+nZ/Z')/(m(r_B+r_C))-(nZ~3(n+m))/(m(r_A+r_B))-(n'Z'Z(n'+m'))/(m'(r_C+r_D)). The "energetic parameter of solution" is an approximate function of electrostatic energy of mixing of molten salts. When f(r) is positive, the solution exhibits negative deviation; when f(r) is negative, the solution shows positive deviation. The function f(r) may be used as an approximate method to estimate the thermodynamic properties of molten salt solutions, as well as phase diagrams of binary salt systems.

The systems In-Sb and Sn-Te are similar to each other in that both contain a compound of definite composition and no range of solid solution. These two systems have been investigated by various authors by emf measurements on cells represented by In(1)

The equilibria between calcium carbide and liquid lead, tin and silver at 850℃, 950℃ and 1050℃ under an argon atmosphere have been investigated. From the equilibrium data obtained, the activity coefficients of calcium in the aforementioned liquid metals have been evaluated. It has been found that the calcium content of the liquid metals in equilibrium with calcium carbide and graphite increases as the temperature is lowered. Of the three metals studied, tin has been found to contain the highest amount of calcium under the same equilibrium conditions. These findings may prove useful in the commercial production of Pb-Ca and Sn-Ca alloys by the calcium carbide process.

The two-stage sulphating roasting process is a modified process in which sulphuric acid is used as the sulphatizing agent. This process has been proposed for treating certain highly siliceous, low grade cobalt and copper oxide ores. As a result of such a treatment, over 90％ of cobalt and copper can be converted into water-soluble products while over 90％ of iron will be left insoluble during subsequent leaching. Taking advantage of rate tests, differential weight analysis as well as differential thermal analysis, the authors examined the necessity of carrying out the roasting process in two stages. In the first stage (230—300℃) the sulphatization of oxides on the one hand and the prevention of acid evaporation loss on the other should be both satisfied. The main reactions are (1) dehydration, (2) sulphatization of different oxides with acid vapor, and (3) partial decomposition of ferric sulphate. In the second stage (550—600℃) reaction between cobalt and copper oxides and SO_3 dissociated from iron sulphate is initiated. The temperature is controlled so as to reduce the decomposition of cobalt and copper sulphate to a minimum. Certain oxides, such as Fe_2O_3, Cu_2O, act catalytically in the reactions and have been found to lower appreciably the starting temperature of cobalt sulphate decomposition, while SiO_2, being neutral to SO_3, shows little effect.

Reaction between carbon and oxygen in melts of iron was studied, using crucibles made of electrically prefused magnesia, in a medium-frequency induction furnace of 10kg. capacity. It was found that the melts continually absorbed oxygen from the crucible, so that during the heat its oxygen content dropped to a minimum and then rose again. Under a furnace pressure of 10-4 (and in a few cases 10~(-1) mm Hg, the [％C].[0％] product fell between 6×10~(-6) and 3.5×10~(-5). By limiting the carbon content of the charged iron to 0.03—0.05％, with a C:O ratio of 2, a high-purity product containing less than 0.003％ oxygen and less than 0.006％ carbon was satisfactorily obtained. The melt-crucible reaction became less pronounced with each heat after a new crucible had been put into use. The oxygen content of the melt had a marked effect on its final nitrogen content. A high oxygen concentration retarded de-nitrogenation.

As regards the effect of the iron oxide content of molten slags upon desulphurization in iron- and steelmaking, different conclusions have been given in the literature: (1) pure iron oxide slag has a desulphurizing effect and gives a sulphur-distribution ratio of 3.6 at 1600℃; (2) variation of the iron oxide content of openhearth slags from 3 to 70％ has practically no effect upon the sulphur-distribution ratio; (3) with electric furnace (or blast furnace) slags, the less the iron oxide content, the higher the sulphur-distribution ratio; and (4) with less basic slags within a certain range of basicity, increase in the iron oxide content brings about an increase in the sulphur-distribution ratio. In the present paper, through calculations with the help of the sulphur-distribution ratio formula: (％S)/[％S]=2.56fs/γ_s--·(Σn+Σn_-)/n_(FeO) derived from the complete-ionization theory of slag constitution and with the introduction of activity coefficients, it has been shown that the above-mentioned four conclusions are not contradictory against one another, and between them there exists an intimate relation. They simply signify that under different conditions iron oxide exerts different actions, and all these different actions could be amalgamated and generalized under the same formula. On the assumption that SiO_2 is present as SiO_4~(4-)in the slag, the formula is not suitable for application to acid slags, while at the same time the effect of temperature upon the sulphur-distribution ratio needs further investigation. The formula has been used to evaluate the sulphur-distribution ratio of basic cupola and basic side-blown converter slags from data in actual practice, and the following resuits have been obtained: (1) desulphurization in the basic cupola practice has practically reached equilibrium; and (2) in basic side-blown converter practice, owing to the presence of gaseous desulphurization, the sulphur-distribution ratio found in actual practice is less than that calculated from the formula.

The magnetic properties and the degree of crystal orientation of a 3％ silicon-iron have been investigated. The effects of rolling methods, the degree of cold deformation, and the intermediate and final annealing temperatures on the inhomogeneities in magnetic properties and the spread of preferred orientation have been considered.

By means of the Bitter pattern, the domain structure of the ferromagnetic ζ-phase in Mn-Al system was investigated by using polycrystalline specimens. The patterns show either fairly straight lines or wavy and even more complicated forms. No lace-like pattern has been observed, however. Based on the magnetic properties of the phase and on the effects of orientations and sizes of the crystallites, it has been inferred that the ζ-phase has a domain structure typical of uniaxial ferromagnetics with large magnetic anisotropy but owing to the minute size of the crystallite the domain structure is more or less simplified as no spike-in domains are formed. In the pulverized specimen, regions deposited with magnetic powders can be easily observed and they have been tentatively identified as single domain particles.

~~