Phase transition induced deformation in porous media

Abstract

Capillary condensation-evaporation and freeze-thaw processes are the most familiar examples of first-order phase transitions in equilibrium thermodynamics. Porous, amorphous materials are widely used in everyday life, yet their complex spatial structures lead to unresolved questions for statistical mechanics, both in- and out-of equilibrium. In this thesis I examine the mechanical consequences of capillary force and freezing transition in porous media, using cement, a construction material of pivotal importance, as an example of practical concern. During changes in relative humidity, the amount of water absorbed in a porous material varies and typically displays hysteresis, i.e differences between adsorption and desorption at the same relative humidity. This process is accompanied by mechanical deformations, such as drying shrinkage in cement. A parallel computing library is developed to simulate the adsorption/desorption processes using a lattice gas model.

Based on my derivation of the generalized Maxwell-Korteweg stress tensor from Landau-Ginzburg theory, capillary forces are obtained and coupled into nano-particle movements using Molecular Dynamics simulation technique. I then investigated the poromechanics of wet cement using this framework. There I tested the continuum postulate at different length scales, and show local irreversible deformations despite linear elastic response to capillary forces on a macroscopic scale. Freezing poses threats to both living systems and infrastructures. Conventional thinking attributes the damage to water volume expansion upon freezing. However, multiple field observations/experimental evidence conflict with this thinking. To resolve the paradoxes, a thermodynamically consistent theory that highlights the role of charged pore surfaces as well as multiscale porosity is presented, predicting freezing point depression and pressures in different limits of salt behaviors.

Explanation of damage based on nano-fluidic salt trapping mechanism is qualitatively consistent with experimental observations. Further implications on freezing tolerance of biological materials are discussed.