This thesis reports the details and results of an
experimental study into the effect of porosity on the thermal conductivity of lightweight and normal-weight concretes. The concretes were made from either Pellite, limestone or quartzitic coarse aggregates, with or without air-entraining agent. Their density varied from 1550 to 2350 kg/m3 and their porosity from 10 to 39 per cent.
The thermal conductivity of all the concretes was
measured in accordance with B.S. 874: 1988, using a plain
hot-plate technique and the total porosity and pore size
distribution of each mix were determined using vacuum
saturation and Mercury Intrusion Porosimetry.
Micromorphological changes were additionally assesed
by:
a) using thermal analysis methods to monitor lime and
non-evaporable water content of mortars and also to monitor
the carbonation process by quantitative evaluation on mortar
samples.
b) assessment of the depth of carbonation on concrete cubes.
c) using Scanning Electron Microscopy on coarse aggregates
and mortar samples.
For characterization purposes, the experimental
programme also involved studying the properties of fresh
concrete, such as workability and air content and other
engineering properties of hardened concretes, such as
compressive strength.
The results obtained from the individual tests were
statistically analysed using a standard `Statistical Analysis System' (SAS) package. This enabled detailed examination of the relationship between the different properties to be assessed. Resulting from this, a model was developed which enabled the thermal cocductivity of a concrete to be estimated knowing its dry density, total porosity and median pore diameter.
The findings of this investigation confirm that in
general, the thermal conductivity of concretes are directly
related to density and inversely related to porosity and pore diameter and that density and total porosity have the highest coefficient of correlation respectively.