Mechanism of Internal Frost Damage (George W. Scherer, Princeton University)
Frost does billions of dollars worth of damage to the infrastructure of the US every year, and has been the subject of intensive study for a century, but the cause of the damage is still in dispute. The first goal of this work is to identify the mechanisms by which the crystallization of water within the pores of cement and concrete produces damage. Theoretical arguments suggest that the primary problem is not the increase in volume as water transforms into ice; rather, it is the pressure exerted directly on the walls of pores by ice crystals. Air voids are conventionally introduced into concrete to protect against frost damage, and it is widely thought that their role is to provide sinks for water displaced by the expansion as ice is formed. However, there is reason to believe that the more important role of the voids is to provide sites for nucleation; ice formed in the voids then sucks water out of the surrounding pores and suction in the water compresses the body. In this research, we will study the sites where crystals nucleate, and the kinetics and thermodynamics of the process by which the crystals propagate through the pores. The second goal of the work is to develop improved ways to protect concrete from frost damage. To that end, we will introduce nucleating agents into the voids to promote nucleation, and will manipulate the pore structure surrounding the voids to control the suction exerted by the ice; both of these factors will help to reduce the risk that ice crystals will exert stress on concrete. To model the freezing process and the development of stress, it is necessary to know the permeability of the pore network as the body freezes, and the pores become progressively blocked by ice. We will use a novel beam-bending method to measure the changes in permeability during freezing. The principle is that bending a thin plate causes the top surface to contract and the bottom surface to stretch; if the plate is a saturated porous body, then the liquid in the pores is compressed near the top of the plate and put into tension near the bottom. The pressure gradient caused by bending leads to flow of the liquid through the pores until the pressure equilibrates. Since the force needed to sustain a constant deflection of the plate decreases as the pore pressure equilibrates, measurements of the change in force with time can be analyzed to determine the permeability of the body, as we have demonstrated using porous glass, cement, and stone. In the present work, the sample will be progressively frozen, and the effect of ice on the stiffness and permeability of the body will be determined.
