64 CEMENTS, LIMES, AND PLASTERS. 



Rate of Set and Hardening. 



A pure plaster of Paris will normally harden or set in from five to 

 fifteen minutes after having been mixed with water. Plasters made 

 from impure gypsum will be considerably slower setting than this, 

 setting usually in from one to two hours. 



When plasters are to be used for structural work, they must be 

 either naturally slow-setting, like some of the cement plasters, or be 

 made slow-setting by proper treatment. Retarders are therefore used 

 at plaster-plants in preparing their product for the market. Occasion- 

 ally, though rarely, a plaster will be naturally too slow setting for the 

 particular use to which it is to be applied. In this case an accelerator 

 must be used. 



Theory of the action of retarders and accelerators. As will be noted 

 later, the materials most commonly used as retarders are glue, tankage, 

 and other organic and uncrystallized materials, while accelerators are 

 usually inorganic and crystallized. This fact suggested a theory as 

 to the cause of the action of accelerators and retarders. The theory, 

 as set forth * by its originator, Dr. Grimsley, is as follows: 



Dr. Grimsley assumes that the set of plaster is due to the presence 

 of a few small crystals which have escaped dehydration during burning 

 and which set the example, so to speak, to the other crystals to form; 

 and, further, that the strength of the set material is due to the formation 

 of a mass of interlacing crystals. The action of retarders and accelerators 

 is therefore explained by assuming that "any substance [added to the 

 water with which the calcined plaster is mixed, or to the dry plaster] 

 which will keep the molecules apart or from too close contact will 

 retard the setting. Such substances are dirt or organic matter that 

 is not of a crystalline character". On the other hand, the action of 

 accelerators is ascribed to the fact that, being of crystalline character, 

 they induce crystallization in the plaster to which they are added. 



It is probable that the researches of Rohland on the effect of various 

 substances on the speed of hydration of plasters, Portland cement, etc., 

 are directly applicable to the question on hand. In summing up his 

 conclusions, Rohland decided that substances which increase the solu- 

 bility of the cementing material accelerate its speed of hydration, while 

 substances which decrease the solubility of the cementing material 

 retard its hydration. 



Materials used as retarders. The materials used as retarders are 

 usually of animal or vegetable origin. Glue, sawdust, blood, and packing- 



* Vol. 5, Reports Geological Survey Kansas, pp. 167, 168. 



