POROSITY: III. WATER AS AN ABSORPTION LIQUID

Classes of absorption method .—The ordinary immersion method has been shown by previous investigators to be unreliable. Simple immersion in low vacuo without boiling is also shown theoretically to be unreliable. An analysis of the results of previous investigators indicates the presence of unsuspected sources of error. Effect of adsorbed gases upon dry weights of test pieces .—Dry air is shown to be without appreciable influence. Depending upon the humidity of the atmosphere, adsorbed water vapor may cause errors up to 2 per cent in the porosity value. Perfectly dry fired clay will remove water from conc. H₂SO₄, and from fused CaCl₂. Saturation by boiling at atmospheric pressure .—A one hour's (and in one instance a 5 hours') boiling failed to saturate completely. On continued boiling the saturated weight increases linearly with the time and this in spite of the fact that appreciable quantities of dissolved materials are removed from the test piece by the hot water. This result is shown to be due to a gradual and continuous rehydration of the clay by the hot water. The error from this factor may amount to as much as 3% and the error from dissolved materials to as much as 2%, during a 3 hours' boiling. The nature of the dissolved material was determined. Method of cooling the test piece .—It is recommended that the test piece be kept in a closed vessel over 95% sulphuric acid for several hours before its dry weight is taken. Saturation procedure .—A vacuum method is described in which water may be employed as the saturation liquid under conditions where the above sources of error are reduced to a minimum. The method is not however recommended as a primary standard. Necessity of a soaking period .—It is shown from theoretical considerations that a soaking period is necessary and methods are given for calculating and for measuring the minimum soaking period required for a given test piece and a given liquid of known penetrativity.