Measurement of moisture in concrete and masonry with special reference to neutron scattering techniques

Abstract

Failures of surfacing materials on concrete floors (likewise on concrete or masonry walls) are usually due to dampness. A rapid and reliable method of determining the dryness of such structures could prevent this type of failure.

Present methods depend on conditions in the atmosphere adjacent to the floor etc. (indirect methods) or on changes in the properties of the concrete (direct methods). Indirect methods can be satisfactory if their criteria of dryness are soundly based. A well based procedure is the quasi-equilibrium method of British Standard C.P. 203, which has a long history of successful field use. Recently methods using the rate of emission of moisture have appeared. These may also be satisfactory but the bases of their criteria are not known.

Gravimetric methods are unsatisfactory because of the difficulty of obtaining a representative sample. Electrical conductivity has been used but is sensitive to factors other than moisture content. Dielectric properties vary with moisture content and have given reasonable results under laboratory conditions. Further work is needed to apply this method to field conditions. Micro-wave absorption has been used to determine moisture in walls but at present cannot be used for floor slabs on grade. The results vary with the nature of the material.

Neutron scattering has been widely used to determine moisture in soils and has been advocated for other materials including concrete. The method is direct and non-destructive and requires access to only one side of the speciment. Results have been claimed to depend solely on the water content, i.e., to be independent of the nature and density of the material. Experiments with concrete and timber have shown that this is not correct. Instrument response rises with density to such an extent that the same instrument response was obtained on a foamed concrete and a dense concrete with water contents in the ratio of 2:1. The composition of the sample also affects the response. This method therefore cannot safely be used with a common calibration curve for all substances.