Mechanisms of Freezing Damage to Concrete

Princeton University, Princeton NJ

Investigators

Abstract

Freeze/thaw cycles can damage concrete by causing internal cracking or superficial scaling. The physical mechanisms responsible for these two types of damage are apparently different, but the details are controversial. The purpose of this project is to examine the fundamental mechanisms responsible for the damage, and to test novel methods for preventing it. Concrete is conventionally protected from frost damage by introducing surfactants that create air voids with diameters on the order of 50-100 um and spacing of about 200um The voids serve as sinks into which water can flow when ice forms in the pores of the cement paste; this prevents development of destructive hydraulic pressures created by the expansion as water transforms to ice. However, the voids do not protect the paste against crystallization pressure when ice forms in the surrounding mesopores. We intend to examine the relative importance of crystallization pressure and hydraulic pressure with respect to internal cracking. To control the occurrence of crystallization, we will introduce nucleating agents for ice into the air voids, so that ice forms preferentially in those open spaces, where it can grow without creating stress. If effective, such nucleating agents could be incorporated into concrete along with the surfactants that create the voids. To examine surface scaling of concrete, we will employ a novel experimental approach. A pool of liquid is established on the surface of a plate of cement paste, and the system is cooled until freezing occurs. As the ice forms, internal stresses develop that are revealed by warping of the plate; by measuring the deflection with an optical probe, we can study the kinetics of stress development. Preliminary experiments have revealed several stress-generating mechanisms: thermal contraction of the pore liquid relative to the solid paste, transfer of pore water toward the ice at the exterior surface of the body, and thermal expansion mismatch between the superficial ice and the paste. The relative importance of these mechanisms will be quantified, and we will explore the effectiveness of air voids, nucleation control, and other methods for minimizing the damage. This research will be the subject of a doctoral thesis, and will be integrated into the undergraduate curriculum by having parts of the study done as senior theses and independent research projects. Freeze/thaw testing will use special facilities and expertise in the labs of the W.R. Grace Corporation. We will also collaborate with Prof. Leo Pel (Eindhoven) to obtain magnetic resonance imaging of the freezing process within the concrete samples.