THE EFFECT OF CALCIUM SULPHATE ON CEMENT 
SECOND PAPEK 
By J. C. Witt 
{From the Chemical Laboratory, Bureau of Science, Manila) 
TWO TEXT FIGURES 
The first paper of this series was a report of an investiga- 
tion made at the Bureau of Science, Manila, to determine the 
effect of various amounts of calcium sulphate on a number of 
cements manufactured in the Orient. It was desired to learn the 
sensitiveness of these cements to calcium sulphate, the amount 
of this substance that could be present without harmful effects, 
and either to confirm or to refute some of the established prin- 
ciples of previous investigations. All the samples were prepared 
in the laboratory. Either finished cement or clinker was ob- 
tained, a calculated amount of gypsum or of some other form 
of calcium sulphate added, and then the two were placed in 
a small ball mill where they were ground and thoroughly mixed. 
The mixture was removed, analyzed for sulphuric anhydride 
(SO3), and subjected to the usual physical tests. 
In conformity with previous work it was found that in general 
the time required for the initial set was directly proportional to 
the amount of sulphuric anhydride present ^ until a certain 
point was reached, but further additions quickened the set.® 
This maximum point corresponded to from 1.5 to 2 per cent 
sulphuric anhydride. With more than 2 per cent sulphuric an- 
hydride the tensile strength was likely to decrease, and with 
more than 3 per cent considerable expansion in sea water was 
likely to result. The soundness was independent of the sul- 
phuric anhydride content. 
After the first paper was completed, some phases of the in- 
vestigation were continued. It developed that one of the ce- 
ments was very irregular in setting time. Several series of 
samples were prepared, and sulphuric anhydride-setting time 
^Philip. Joum. Sci., Sec. A (1917), 12, 133. 
” Many other electrolytes have this effect. See Philip. Journ. Sci., Sec. 
A (1916), 11, 273; (1918), 13, 29. 
^ Several references to the influence of substances on cement and con- 
crete are given in the papers referred to in footnotes 1 and 2 and are 
not repeated here. 
221 
