686 EXPERIMENT STATION RECORD. 



Action of the salts in alkali water and sea water on cements, P. H. Bates, 

 A. J. Phillips, and R. J. Wig ( U. S. Dept. Com., Bur. Standards Technol. Paper 

 12, 1913, pp. 157, pU. 8, figs. ^5). — The results of extended field and laboratory 

 tests are given to determine tlie cause of the disintegration of concrete by the 

 salts present in sea water and the alkali salts. Both chemical and physical in- 

 vestigations were made in the laboratory, and in addition to sea water, solutions 

 of sodium chlorid, sodium sulphate, sodium carbonate, magnesium chlorid, 

 magnesium sulphate, and ferrous sulphate, and also solutions in which there 

 were present in equal parts by weight 2 of the salts, were used in the tests. 



In the physical laboratory investigation there were 3 series of tests which 

 consisted of making cement-mortar hollow cylinders closed at one end, 85 in. 

 outside diameter and 10 in. high, with walls and base i in. thick, and per- 

 mitting various solutions to percolate through. The chemical investigations 

 were made to determine the action of solutions on fresh and set cements. In 

 the field investigations concrete of varied composition was made with Portland, 

 natural, slag, and other special sea-water cements and exposed to sea water. 

 Concrete briquettes were also tested for the purpose of comparing various types 

 and brands of cement. 



From these investigations the following results and conclusions are stated : 



(1) Portland cement mortar or concrete, if porous, can be disintegrated by 

 the mechanical forces exerted by the crystallization of almost any salt in its 

 pores if a sufficient amount of it is permitted to accumulate and a rapid forma- 

 tion of crystals is brought about by drying ; and as larger crystals are formed 

 by slow crystallization there would be obtained the same results on a larger 

 scale but in greater time if slow drying were had. Porous stone, brick, and 

 other structural materials are disintegrated in the same manner. Therefore, 

 in alkali regions where a concentration of salts is possible a dense nonporous 

 surface is essential. 



(2) While in the laboratory an hydraulic cement is readily decomposed if 

 intimately exposed to the chemical action of various sulphate and chlorid 

 solutions, field inspection indicates that in service these reactions are much 

 retarded, if not entirely suspended, in most cases, due probably to the carboni- 

 zation of the lime of the cement near the surface or the formation of an im- 

 pervious skin or protective coating by saline deposits. 



(3) Properly made Portland cement concrete when totally immersed is ap- 

 parently not subject to decomposition by the chemical action of sea water. 



(4) Concrete sets and permanently hardens as satisfactorily in sea water as 

 in fresh water or in the atmosphere if it can be placed in the forms without 

 undue exposure to the sea water while being deposited. 



(5) Natural, slag, and other special cements tested in concrete mixtures 

 showed normal increase in strength with age both in sea water and in fresh 

 water. 



(6) In the form of neat briquettes most of the Portland cements of high 

 iron content, several of the cements of high or normal alumina content, and one 

 special slag cement did not show any marked difference in tensile strength, 

 whether exposed to fresh or sea water, for all periods up to 2 years. Other 

 cements of various compositions showed signs of disintegration after a few 

 weeks. 



(7) All cements resisted disintegration in sea water better in mortar mixtures 

 than in the form of neat briquettes. In most cases the mortar briquettes had 

 normal strength up to 2 years' exposure. 



(8) The physical qualities of the cement seem to determine its resistance to 

 decomposition when brought into intimate contact with the sulphate and chlorid 

 solutions. 



