Indirect measurement of the ice content curve of partially frozen cement based materials

- Measured ice contents are compared to estimated ones from MIP and BJH tests.
- Results from the desorption curves are close to the direct measurement.
- Use of the MIP data overestimates the ice content near 0 °C.
- A new way to use the MIP data is proposed in order to reduce this difference.

The goal of this paper is to investigate the validity of indirectly built ice content curves from porous network characterization tests to predict the behavior of a freezing cement based material. To do so, the results obtained by the two classical methods, namely the mercury intrusion porosimetry (MIP) and the gravimetric water desorption (WD) tests, are compared with the direct measurement of the ice content with a dielectric capacitive apparatus. A corrective approach, based on an interface energy analysis between the in-pore phases, is proposed in order to increase the accuracy of MIP based prediction of the ice saturation curve. The impact of the method used to estimate the ice saturation curve on the behavior of freezing porous medium is evaluated and discussed in the framework of poromechanics.