A Treatise On The Principles And Practice Of Dock Engineering

125 ACTION OF SEA-WATER UPON CEMENT. Analyses of Samples of Cement. Original Cement of Test Briquette. Decomposed Cement. L II. III. IV. V. Alumina and oxide of iron, 13 TO 26-76 28-42 1-05 1-53 5-60 Silica 20-92 18-04 19-55 1-33 1-31 10-87 Carbonate of lime, . 8-18 6-61 15-78 45-72 35-42 38-37 Hydrate of lime, 11-26 30-54 16-94 27-85 17-17 19-21 Caustic lime, . 45-39 Magnesia, Hydrate of magnesia, 0-33 13-57 15-08 21 03 39-96 22-30 Sulphuric acid, Soluble in water, . 0-82 2-98 1-50 4-23 1-31 1-71 0-90 3-71 0'85 2-80 Mr. Pattinson’s report, based on a separate series of samples, contained the following conclusions :— “ On comparing the analyses of the concrete used in the work with those of the original briquettes, it is evident that very considérable changes have occurred in the composition of the cement used in the concrete. Ist. Much of the lime has disappeared from six samples. 2nd. A great increase of the magnesia has taken place in the same samples. 3rd. An increase in the amount of sulphuric acid has taken place in the same samples. This sulphuric acid exists as hydrated sulphate of lime. “ There can be no doubt, I think, that this deterioration is caused by the action of the sea water with which the cement has corne in contact. According to Thorpe and Morton’s analysis,* 1,000 grains of sea water contains 3T51 grains of chloride of magnesium and 2-066 grains of sulphate of magnesia. The magnesia of both these salts is precipitated as hydrate of magnesia on coming into contact with lime, with the simultaneous formation of soluble chloride of calcium and partially soluble sulphate of lime. This chemical action of sea water has evi- dently taken place in the cement used in the six samples, and notably in one of theni, from which about two-thirds of the lime has been removed, and in which about twenty times the original quantity of magnesia, and more than three times the original quantity of sulphate of lime, have been deposited, thereby causing the friable and disintegrated condition which marked this sample. The same result, in a lesser degree, is observable in the other samples.” Mr. Messent, commenting on these analyses, observed :—“ In their exam. ination of the deteriorated concrete, both agree that the presence of too mucli magnesia in the cement is the cause of the deterioration, and that, as tlie same proportion or quantity was not found in the briquettes made of the neat cement used, the additional quantity found in the spoiled concrete must have been supplied by the sea-water, in contact with the * Chem. Soc. Joum., vol. xxiv., p. 506.