Janz, M.
1997
Doctoral Thesis, Unviersity of Lund, Lund, Sweden
http://www.byggnadsmaterial.lth.se/pdf/TVBM-3076.pdf
| The purpose of this licentiate study is to analyze capillarity transport in building materials, and in order to do that, develop a new measurement method by which the moisture diffusivity at high moisture levels can be measured in an easy way.
The material used in this development procedure was a sedimentary calcareous sandstone named Uddvide emanating from the island of Gotland. The porosity and density of the sandstone were measured to 23% and 2059 k g/M3 , respectively.
Two new evaluation methods to determine the moisture diffusivity at high moisture contents have been developed. From the first method the moisture diffusivities are calculated exactly. This method is based on the Boltzmann-transformation and was first presented by Claesson (1994). The second method is based on an analytical solution of step response with two moisture capacities given in (Arfvidsson 1994). This method gives an approximate solution of the Kirchhoff flow potential. The advantage of this method is that the calculations are easy and quick to perform. A presumption for both methods is that the moisture flow can be mathematically described by Fick's law, i.e. the moisture flow is linearly proportional to the gradient of the water content.
Both methods give reasonable and rather similar results at high water contents, but it seems as if neither of the methods is useful on the actual sandstone at low moisture contents. Both methods must therefore be supplemented by the measurement of the moisture diffusivity at low moisture levels, e.g. by measurements performed by the cup-method. However, the purpose of this work was to find a measurement method to determine the moisture diff-usivity at high moisture contents which cannot be obtained by traditional methods.
The input used in both methods described is a series of sorption coefficients, A], A2,..., AN [kg/(M2_S~)j, corresponding to different initial water contents, Win = WI, W2,---, wN [k g/M3] and to the water content at capillary saturation, w,,p. Thus, the only experiment that must be carried out is a series of capillary water uptake tests performed on pre-conditioned specimens with no initial moisture gradients. This entails a limited technical effort, principally the only equipment needed is a balance. The laboratory set-up used in this study recorded automatically, on-line, the amount of absorbed water.
If the two methods of evaluating the moisture diffusivity from the relation A(wi,) are to be of an extensive use, a rapid method of conditioning the specimens must be used. A method for this was developed. The specimens were conditioned by letting them suck water to the desired initial moisture content whereupon they were sealed and stored in at least 14 days before the capillary water uptake test was performed. To what extent this rapid conditioning method results in moisture gradients that influence the capillary water test was however uncertain. Therefore, the results of capillary water uptake tests performed on specimens conditioned with this rapid method were compared with two other conditioning methods; one in the hygroscopic range and one in the superhygroscopic range. Moisture gradients in the specimen body were prevented with these two conditioning methods since the specimens were conditioned through absorption or desorption from considerably lower and higher moisture contents than the initial moisture content aimed at. Conditioning was made by letting the material reach equilibrium with a certain pore water pressure.
In the hygroscopic range the specimens were conditioned both by absorption and desorption in eight different climate chambers. Different saturated salt solutions were used to obtain the desired relative humidity in the climate chambers. In the superhygroscopic range the specimens were conditioned in an "extractor". The water -was forced out of the capillary saturated specimen by an applied air pressure. The conditioning also resulted in information about the sorption isotherm (moisture content versus relative humidity, RH) and the water retention curve (moisture content versus capillary pressure) of the sandstone. These two measurements of the moisture storage capacity were related to each other by the Kelvin equation.
No significant difference were noticed between the different methods of conditioning. This indicates that there was not any moisture gradient in the specimen body after the rapid conditioning, thus, this seems to be suitable on the sandstone used in this study. |
