FLOORING-PLASTERS AND HARD-FINISH PLASTERS. 69 



tradictory as to temperatures reached, composition of product, etc. 

 Fortunately, however, a detailed account * of the chemical changes 

 involved was published during 1903 by Van't Hoff in the Transactions 

 of the Berlin Academy of Sciences. As this paper is practically inac- 

 cessible to the American engineer or manufacturer, a translation f is 

 here appended: 



"As a complement to the investigations on gypsum and anhydrite 

 we have turned our attention to a kindred product, usually designated 

 by the term hydraulic gypsum or floor-gypsum (Estrichgips) . It is 

 obtained by burning natural gypsum, CaS0 4 .2H 2 0. When natural 

 gypsum is worked up into stucco gypsum (CaSO 4 )2.H 2 0, in the process 

 called cooking, the temperature of 120-130 C. is not exceeded; but 

 in the preparation of floor-gypsum higher temperatures are applied. 

 Accordingly the product is free from water, but we are not in this case 

 dealing with dead-burned gypsum, since the capacity to bind water has 

 not yet been lost. However, the time required for setting is much longer 

 than in the case of stucco gypsum; the latter, as every one knows, hardens 

 in about a quarter of an hour, while the hardening of the floor-gypsum 

 takes place only after some days, and the complete absorption of the 

 amount of water theoretically required may take weeks. 



"Composition and structure of floor-gypsum. Some indications 

 in literature suggested that in this floor-gypsum we are dealing with 

 a basic sulphate. Accordingly a commercial floor-gypsum was first 

 analyzed; it showed 38.6% CaO, 54.3% S0 3 , and foreign ingredients. 



" From this the ratio of CaOiSOs in molecules is found to be 1.01 :1. 

 The product therefore is calcium sulphate anhydrite without any notable 

 surplus of lime. The fact that but slight quantities of lime are present 

 was also demonstrated in another way: 1.134 gr. of the substance was 

 boiled for 3^ hours with 75 ccm. of 0.425 normal potassium hydroxide, 

 whereupon everything went into solution except slight impurities. 

 To titrate back, 47.8 ccm. of potassium hydroxide were then required, 

 which indicates a content in free lime corresponding to 125 (75 + 47.8) 

 = 2.2 ccm. in weight 0.026 gr. or about 2 per cent of the whole. 



" Having ascertained this composition, it only remained to answer 

 the question as to the relation which the floor-gypsum bears to the 

 two known modifications of anhydrous calcium sulphate, namely, the 

 natural and the soluble anhydrite. The difference between these two 

 lies in this, that the former is practically incapable of binding water, 

 that is to say, it does so with extreme slowness, and thus acts like dead- 

 burned gypsum, while the soluble anhydrite takes up water even more 

 rapidly than stucco gypsum does. 



"A first hint was obtained from the microscopic examination. As 

 shown by this, the floor-gypsum consists for the most part of distinctly 



* Van't Hoff and Just, G. Der hydraulicshe oder sogennante Estrichgips. 

 Sitzungsberichte der Kgl. Preuss. Akad. der Wissenschaften, 1903, B. I, pp. 249-258. 



f For this translation the writer is indebted to Dr. Robert Stein, formerly of 

 the U. S. Geological Survey. 



