A Treatise On The Principles And Practice Of Harbour Engineering

IOO HARBOUR ENGINEERING. tioiis where its possible failure would entail consequences of the most serions nature. The author has already, in another work, devoted some considerable space to a discussion of the subject,1 and he does not feel that it will be considered incumbent upon him to restate the particulars and details of the investigation which led him to the conclusion that any defects which have exhibited them- selves in the behaviour of Portland cement concrete in sea-water are due solely and entirely to inherent deficiencies in the quality of the cement used, and in the materials with which it was mixed, as also, in part, to indifferent and imperfect manipulation. Briefly stated, the facts in cases of known failure show that disintegration is (1) inaugurated by symptoms of expansion, and (2) subsequently accomplished by the solvent action of the sea. Now, the constituent elements of sea-water which have any effect upon cement are the sulphates and chlorides, especially the sulphate and chloride of magnesia. On coming in contact with lime, the magnesia in these salts is precipitated as a hydrate, with the simultaneous formation of chloride of calcium and sulphate of lime. Both these latter products are more or less soluble, especially the former, and the washing action of the sea soon occasions their removal. But, in order that this interchange of constituents may take place, it is essential that there should be present a certain amount of lime, either actually in a free, uncombined condition, or, at least, in the form of a very unstable compound. The silicates and aluminates of lime produced during calcination, at the proper temperature for the manufacture of Portland cement, are all more or less fixed; at any rate there is abundant evidence to show that neither of these combinations necessarily breaks down in a marine environment. It is usual, therefore, to attribute the change to some caustic lime which has failed to combine with the silica and alumina. The supposition is not out of consonance with observed phenomena. The slaking of lime causes expansion, with the formation of a hydrate which is readily soluble in water. One authority,2 however, suggests that a truer explanation of the expansion and disintegration is to be found in the formation of a “vitreous high-lime compound which slakes or hydrates so extremely slowly that it may be months before visible hydration even commences; but in such cases hydration is accompanied by enormous expansion, the increase in bulk amounting to many times the original size of the mass, sufficient to cause the disruption and total disintegration of the previously set particles in the cement.” The precise nature of this “ vitreous high-lime compound ” is difficult to identify, but from whatever source the lime be forthcoming, there seems no doubt that it exercises a deleterious effect, and that measures should be taken to ensure its absence wherever possible, and, in the second place, to limit the action of sea-water to the outer skin or surface. This can only be achieved 1 Dock Engineering, p. 123 et seq. 2 Butler, Portland Cement, London, Spon., 1905.