January io, 191 8] 



NATURE 



373 



of stars of increasing, as well as stars of decreasing, 

 temperature, as has so long been advocated by Sir 

 Norman Lockyer. Chiefly through the work of Hertz- 

 sprung and Russell, it has, in fact, been found that 

 many well-known stars are in a highly diffuse state, 



ind when such stars contract under the influence of 

 their own gravitation, they must rise in temperature 

 until they cease to approximate to perfectly gaseous con- 

 ditions (density 02 to 04 compared with water). Prof. 

 Eddington has made an important modification of 

 Lane's theory- by taking account of radiation-pressure, 

 md he shows that as a star contracts the diminishing 

 -urface is compensated by increasing radiation, so that 

 ;he total radiation remains nearly constant, until the 



iensity becomes too great for the theory to apply. 

 Afterwards the star passes to the descending branch 

 of the tempyerature curve, and the total radiation falls 

 iiff very rapidly. Radiation-pressure has further been 

 fdund to provide a reasonable explanation of the ap- 

 proximate uniformity of stellar masses. 



THE VISCOSITY OF SLAG AT HIGH 

 TEMPERATURES. 



ALTHOUGH the data which have in recent years 

 been accumulated on the behaviour of silicates at 

 high temperatures possess a great interest and value 

 from the points of view of the mineralogist and the 

 geophysicist, a knowledge of the melting points and 

 fields of stability of the silicates is not the most impor- 

 tant factor for consideration in so far as applica- 

 tion to the metallurgy of iron is concerned. In a study 

 of the blast-furnace process what is of particular in- 

 terest is the behaviour of the slag from the time it 

 enters the zone of fusion until it is flushed from the 

 slag-notch. In passing through this region where the 

 smelting process occurs the most important physical 

 property of the slag is its viscosity, while its most 

 important chermcal property is its desulphurising 

 power. 



It has long been known to furnacemen that molten 

 blast-furnace slag is much more viscous than molten 

 iron and most fused salts, and that the slag undergoes 

 a gradual softening on heating rather than a sudden 

 change to a mobile liquid. This particular character- 

 istic was from the first rightly attributed to the silica 

 content of the slag, and it appears to be due to the 

 nature of the molecule, SiOj, rather than to the 

 element silicon itself. X-ray analysis in the hands of 

 Prof. W. H. and Mr. W. L. Bragg has furnished an im- 

 portant confirmation of this hypothesis. Instead of find- 

 ing, as is the case with the crystals of most chemical 

 compounds, that 'he atoms are arranged separately at 

 definite points of a space-lattice, they concluded that 

 three silica molecules are associated with each point 

 of the space-lattice. It is a matter of common know- 

 ledge that highly associated or polymerised liquids 

 possess unusually high viscosity, and hence it seems 

 plausible to argue that, since silica appears to be un- 

 usuallv complex in the solid state, this association or 

 polymerisation tendency must be the fundamental cause 

 of the extreme viscosity of silica itself and the high 

 viscosity of the silicates in the liquid state. When a 

 silicate gradually softens with rising temperature and 

 passes entirely over into the liquid state it is probable 

 that the increased fluidity is due to a weakening of the 

 "residual-valence" attraction between the group-mole- 

 cules, while the relatively high viscosity of the melt, as 

 compared with that of molten metals and ordinary 

 salts, is due to the preponderance of the group-mole- 

 rules silica, alumina, and lime, and possibly in a par- 

 ticularly large degree to a highly polymerised condition 

 of the silica group-molecule. 



Considerations of this kind are set out in a valuable 

 paper by Mr. A. L. Feild, assistant metallurgist at the 

 United States Bureau of Mines, in a recent paper pre- 

 sented to the Faraday Society. ^ Mr. Feild points out 

 that while it is theoretically possible to render any sili- 

 cate mixture whatever sufficiently fluid to flow from the 

 slag-notch of a blast-furnace it is necessary in practice 

 that the slag should attain this necessary fluidity at a 

 temperature which i-s not beyond the working limit of 

 the blast-furnace lining, and does not demand 

 an unusually high fuel consumption. It is obvious 

 that if, for instance, a slag requires a minimum tem- 

 perature of 1400° C. in order to attain a working 

 fluidity, no iron will be produced in a furnace using 

 this slag, regardless of the number of B.Th. units 

 developed within the furnace, unless the temperature 

 distribution is such that the slag acquires the neces- 

 sary temperature at 1400° C. Thus the fuel economy 

 of the blast-furnace is to a great extent dependent upon 

 the temperature-viscosity relations of the slag. Apart 

 from the question of mining cost and freightage, the 

 value of an iron ore suflficiently rich in iron to be 

 considered marketable largely depends on whether it 

 can be made to yield econorriically a slag of desirable 

 viscosity and desulphurising power. 



Mr. Feild has worked out a method of determining 

 the viscosities of slags up to 1600° C, this limit being 

 imposed by the furnace refractories and not by inherent 

 limitations of the apparatus. He has used a modifica- 

 tion of the method originated by Margules in 188 1, in 

 which the liquid is confined between two concentric 

 cylinders. The outer cylinder is rotated at a constant 

 speed, and the torque exerted upon the inner cylinder 

 measured. The method is applicable to liquids of a 

 wide range of viscosity, and has been applied in this 

 case to measurements on slags over a range of viscosity 

 from 200 to 3000 (water at 20° C. = i). Acheson 

 graphite was used in the construction of all parts sub- 

 jected to high temperatures, and the suspended system 

 was damped so as to ^ive it the stability and aperiodicity 

 of the familiar damped D'Arsonval galvanometer. The 

 outer cylinder was rotated about a vertical axis at a 

 constant speed. The inner cylinder was suspended 

 coaxially within the outer one by means of a steel or 

 phosphor-bronze ribbon. Experimental data derived by 

 this method are capable of an easy direct mathematical 

 interpretation. A cylinder of radius h, rotating with a 

 constant angular velocity w, will exert upon an inner 

 fixed concentric cylinder of radius a — the space between 

 them being filled with the liquid — a couple r given by 

 the relation 



♦vhere t; is the coefficient of viscosity and L the common 

 fength of the two cylinders. With cylinders of fixed 

 dimensions the viscosity is proportional to the torsion 

 couple and inversely proportional to the speed of rota- 

 tion In revolutions per second. Conversely the torsion 

 couple is proportional to the speed of rotation and to 

 the viscosity. 



Viscosity values are given for eight commercial slags, 

 two synthetic slags, and an artificial diopside. The 

 temperature-viscosity curve approximates in form to 

 that of the rectangular hyperbola, while the tempera- 

 ture-fluidity curve approaches a straight line in form. 

 The average viscositv at 1500° C. of eight commercial 

 slags was found to be 30T (water at 20° C. = i). Re- 

 fractory slags have been found to be not necessarily 

 more viscous at high temperature than more fusible 

 ones. H. C. H. C. 



1 " The Viscosity of Blast-fumace Sl.ie and its Relation to Iron Metal- 

 lurgy, including a Description of a New Method of Measuring Slag Viscosity 

 at High Temperatures." Proceedings, December, 1917. 



