PLASTIC MAGNESIA CEMENTS1

Abstract

Preparation of the Cement.—Crystalline magnesite from Washington was crushed to three sizes—to pass a No. 6 sieve but retained on a No. 10; between a No. 10 and a No. 30 and to pass a No. 60 sieve. These sizes were burned separately in an electric rotary kiln at 600°, 650°, 700°, 800°, and 900°C. Tests of the Cement.—The cement was tested in three flooring mixes in which the MgO was 35.0, 42.5 or 50.0% by weight, and in three stucco mixes in which the MgO was 11, 22 or 33% by weight. Tensile and compressive strength specimens were broken at 24 hours, 7 and 28 days. The coefficient of expansion was determined at 48 hours, 4, 7, 28 and 90 days. Other properties determined were time of set, consistency, soundness, fineness and effect of density of the chloride solution used. Results.—The property of the calcined cement is materially affected by the size of ore, temperature and duration of burning. The rate of reaction of MgO with chloride materially decreases with increased temperature of burning. Decreasing the concentration of the magnesium chloride solution (down to 22° Bé) accelerates the set of freshly calcined magnesite and retards the set of magnesia which has become hydrated to a considerable extent through exposure. Exposure to moisture before using as well as increasing the consistency of the mixture lengthens the setting lime. A composition which expands excessively and warps or buckles is not necessarily “unsound,” but one which disintegrates within a comparatively short time may be considered as such. Used in this sense, “unsoundness” is believed to result from the presence of free magnesia which hydrates after the mixture has hardened, and not from the presence of lime. Therefore, soundness is not a property of the magnesia alone but depends upon the extent to which it reacts with the magnesium chloride, water, or carbon dioxide before hardening takes place and upon the amount of hydration which subsequently occurs. The steam or cold water tests are unsatisfactory as accelerated tests for soundness of magnesia mixtures. Under the conditions of these tests, the best material with regard to setting time and strength was produced at a temperature of 800°C. However, the magnesia burned at 650°C and which gave comparatively very low strength when gaged with a 22° Bé solution of magnesium chloride, gave excellent strengths when gagrd with more concentrated solutions. Materials tested with 22° Bé solution and giving satisfactory results would not necessarily be satisfactory with a higher or lower concentration. It seems that general specifications should also include limits corresponding with tests in which higher and lower concentrations are used. No relation was found between the volume change and any other property of the magnesia. Tests of any particular mixture are no indication of the behavior of the same magnesia in other mixtures. However, the laboratory tests indicate that the leaner mixtures undergo less change in volume than the richer ones. Furthermore, a number of strict comparisons between laboratory and field tests indicate that the laboratory tests of volume changes as made are no index to the behavior of the material under actual service. The lean mortar mixture proved to be the most suitable of any used for testing magnesia, and in this case the tensile strength furnished as much, if not more, information than the compressive strength.