PHILIPPINE RAW CEMENT MATERIALS. 217 



The limiting formula given above does not give entire satisfaction, 

 since there are two undetermined variables. Limits for the silica-alu- 

 mina (respectively iron oxide) ratio must also be set. . 



E. D. Campbell '- has shown in his experiments that the best cements were 

 produced from lean clays high in silica, with a ratio 3>1. He says "the 

 substitution of ALO3, or Fe^Os for SiO™, that is, the use of a rich clay, lowers 

 the overburning temperature" that with lean clays heavily limed there is a wide 

 margin between the proper clinkering temperature and the overburning tem- 

 perature, while with rich clays great care must be exercised in order to prevent 

 overburning. 



Bleininger," says, "The clay must have a percentage ratio of silica to alumina 

 of from 3 to 1 to 4 to 1" and in drawing conclusions from his own work, 

 continues " "The ratios of silica to alumina given on several occasions seems to 

 be correct, for the safest cements in the boiling test, though not the highest in 

 the tensile test are those with a silica-alumina ratio of from 3 to 1 to 4 to 1. 

 Aluminous cements are to be condemned." Experiments of this Bureau" lead 

 to the conclusion that the composition of Portland cement best adapted for use 

 in a tropical climate should have a high silica-alumina ratio; that is, at least 

 3 parts of silica to 1 part of alumina. R. K. Meade '° places the limits for a 

 freshly made American Portland cement as follows : 



Per cent. 

 Silica 20-24 



Alumina 5- 9 



Iron oxide 2- 4 



Lime 60-63.5 



Sulphur trioxide 1- 2 



Le Chatelier "' places the limits of the amount of individual constituents 

 usually present in good commercial Portland cement as follows : 



Per cent. 

 Silica 21-24 



Alumina 6- 8 



Iron oxide 2- 4 



Lime 60-65 



Magnesia 0.5- 2 



The above calculations show that cements agreeing nicely with the 

 theoretical analysis for Lehigh district cement could be produced from 

 materials such as those taken from the drill hole on Batan Island. On 

 the other hand^, the silica-alumina ratio is low and the composition of 

 the burned product shows that this would barely come within the highest 

 limits proposed by Meade and Le Chatelier. I am of the opinion that 

 it would be found necessary to add silica to the materials in some shape 

 to produce a wholly satisfactory cement. 



^Journ. Am. Chem. Soo. (1902) 24, 969. 



"The manufacture of Hydraulic Cements, Bull. Geol. Sur., Ohio 4th Ser. 

 (1904), 3, 223. 

 "lUd., 237. 



"Reibling, and Salinger, L. A. This Journal, Sec. A. (1908), 3, 185. 

 ''Chem. Eng. (1907), 5, 349. 

 "Trans. Am. Inst. Min. Eng. (1893), 32, 10. 



