158 The Philippine Journal of Science wis 



that more of the calcium compounds in the cement would be 

 hydrated. This is on the basis of the results presented in this 

 paper, which show that the presence of calcium hydroxide in 

 solution tends to inhibit further hydration. Further, it is known 

 that if an imperfect crystal is suspended in a saturated solution 

 of the same substance, it does not change in weight, though it 

 may change in form sufficiently to become again regular. 



The principal components of cement are compounds of calcium 

 with aluminium and of calcium with silicon. Indeed, according 

 to Rankin :^^ 



Microscopical examination of commercial Portland cement clinker shows 

 it to be made up largely (over 90 per cent) of the three compounds, 

 2CaO.Si02, SCaO.SiO. and SCaO.AUOs. It would therefore appear that 

 the value of Portland cement as a cementing material when mixed with 

 water is largely due to one or more of these compounds. 



Now since about 90 per cent of the total calcium in a cement 

 is found in solution after treating with water under given con- 

 ditions, with indications that still more could be dissolved, it 

 follows that (1) all the important compounds may be rapidly 

 hydrated under favorable conditions and that (2) one product 

 of the hydration is always calcium hydroxide. 



The colloid theory for the setting of cement was advanced by 

 Michaelis." His idea is that the most important step is the 

 formation of a gelatinous mass containing calcium oxide, silica, 

 and water. Later this colloid dries and hardens, and to it is 

 due the principal strength of the cement. Considerable work 

 has been done by others on the basis of this theory, using cement 

 itself or one of the calcium aluminates. 



Schott ^' and Keiserman ^^ found that, when certain calcium 

 aluminates are hydrated, aluminium hydroxide is split off. 

 Stem ^^ found that aluminates were decomposed by water form- 

 ing the hydroxide of calcium and aluminium. Later he dialyzed 

 the filtrate and found that calcium passed the membrane, but 

 with only a trace of aluminium. Klein and Phillips -" repeated 

 the work of Stern, taking great care to exclude carbon dioxide 

 during the operation. They used tricalcium aluminate and found 



"Rankin, George A., Journ. Franklin Inst. (1916), 181, 770. 

 "Michaelis, W., Cement & Eng. News (1909), 21, 298, 338. 

 "Schott, 0., ibid. (1910), 22, 515. 

 "Keiserman, ibid. (1911), 23, 10. 

 " Stern, E., loc. cit. 



-° Klein, A. A., and Phillips, A. J., Tech. Paper, U. S. Bitr. Standards, 

 (1914), No. 43, 18. 



