CONSTITUTION, ETC., OF PORTLAND CEMENT. 571 



cement clinker are the materials identified under the microscope by 

 Tornebohm and named alit and celit; that alit is a solid solution 

 of tricalcic aluminate (SCaO.A^Os) in tricalcic silicate (CaO.SiC^), 

 while celit is a solid solution of dicalcic aluminate (2CaO.Al 2 03) in 

 dicalcic silicate (2CaO.Si02). 



" Having determined that alit and celit are solid solutions of alu- 

 minates in silicates, the aluminates being present in less than an amount 

 sufficient to make a saturated solution of aluminate in the silicate, 

 it becomes of interest to consider how these solutions are formed during 

 the conversion of a raw mixture or of a mixture of pure chemicals into 

 a clinker. It would be simple to understand this if fusion took place in 

 its formation, but this does not happen, the material is only sintered. 

 If two gases are brought together they diffuse into each other with very 

 great rapidity. If two liquids are poured one upon the other in layers 

 without mixing, they diffuse more slowly. If solids are brought into 

 contact it would be naturally assumed that diffusion would cease. Experi- 

 ments of Robert-Austen have shown that molecular mobility in solids 

 exists, since when carefully polished surfaces of gold and lead are brought 

 into contact and left under pressure for some months, at the ordinary 

 temperatures, gold is diffused into the lead and the lead into the gold 

 for an appreciable distance. Mixtures of the components which would 

 produce a fusible wood metal when subjected to pressure at ordinary 

 temperature become converted into this alloy. Anhydrous sulphate 

 of soda and carbonate of barium also diffuse when brought into close 

 contact with the formation of barium sulphate and carbonate of soda. 

 It is not difficult to understand, therefore, how at a temperature of 

 1650 C. the particles of silica, alumina, and lime may diffuse below 

 the melting-point of the resulting clinker to form a Portland cement, 

 and the fact that such a clinker is stable depends not only on its com- 

 position, but upon the fact that the diffusion has been complete, even 

 in material which is only sintered. Sintering, therefore, may be denned 

 as diffusion at a temperature below the melting-point of the compo- 

 nents or of the resulting solid solution. That diffusion under such con- 

 ditions is surprisingly rapid is seen by placing a particle of ferric oxide 

 on the surface of white Portland-cement clinker, and then submitting 

 it to a moderately high temperature. The rapid diffusion of iron through 

 the white clinker can readily be noticed by the color which spreads 

 through the mass. It is evident that the higher the temperature the 

 more rapid the diffusion until it becomes very rapid on fusion. From 

 this it may be concluded that the length of time during which it is 

 necessary to expose any mixture of silica alumina and lime to a tern- 



