Aerial
photograph of patterns on Howe Island in the Arctic Ocean. Larger
circular patterns are about 2 meters apart, the smaller polygons are
about 30 cm across.
Frost heave refers to the uplifting of the ground surface owing to the freezing of water within the soil. Its typical magnitude exceeds the mere expansion of water upon freezing (~9%) due to freezing of additional water drawn upward from the unfrozen soil below to the freezing point. The driving force for drawing water through the soil is called crysostatic suction, and can be due to several different phenomena including soil texture, composition, and intermolecular forces.
Differential frost heave (DFH) is laterally non-uniform frost heave that results in part of the ground surface heaving more relative to an adjacent area. DFH can be caused by: (1) variations in soil texture, saturation, thermal conductivity or consolidation, (2) differences in surface thermal conditions, and (3) variations in overburden pressure. DFH can be particularly destructive to engineering infrastructure such as roads, pipelines, and building foundations by causing fracture due to high frost heave pressure. DFH has also been implicated in forming some types of patterned ground.
The existing mathematical models that describe frost heave are fairly well developed but still not perfect. The system is complicated due to the wide range of length scales from micron-scale unfrozen water films, to the millimeter-scale frozen fringe, to meter-scale soil deformation. The discrete nature of individual ice lenses forming, growing, and then stopping when the subsequent one begins elsewhere makes for a non-continuous system. Most all recent models still focus on one-dimensional heave.
Multi-dimensional models of frost heave are less prevalent, perhaps due to the complicating factor of describing the unfrozen and frozen soil deformation. Frozen soil exhibit elastic, plastic and creep behavior on different time scales, and is also a strong function of temperature near the freezing point, which is most relevant when heave is occurring.
We have been working on modeling DFH using a few different techniques, including stability theory, finite element technique, and scaling analysis. Goals include predicting the rate and degree of DFH observed in both the field and laboratory, and predicting the highly-regular patterns observed in many arctic landscapes.
Aerial
photograph of patterns on Howe Island in the Arctic Ocean. Larger
circular patterns are about 2 meters apart, the smaller polygons are
about 30 cm across.
Although patterned ground is found in many diverse arctic locations and appears to be robust in many of them, reproducing the pattern formation in the laboratory is still elusive. Some pilot experiments have shown promise, but consistent and reproducible results do not currently exist. Although the physical mechanisms that cause patterned ground to form are not entirely agreed upon in the scientific literature, DFH is often implicated as a contributing, if not the major factor.
We have laboratory facilities for simulating the arctic freeze/thaw thermal conditions that occur where patterned ground is observed. We are using different size systems for simulating different types of patterning due to DFH. Because natural patterned ground is formed only over long time spans of 10's to 100's of years, we take advantage of some scaling properties of the system to speed up the pattern formation process.
Soil
is recurrently frozen and thawed in an 8-ft3
environmental chamber.
Larger
freezing simulations occur in 40-ft3 tank where the
atmosphere is controlled by an external environmental chamber.
Concrete continues to be one of the most widely used engineering infrastructure materials today, used in foundations, bridges, well casing, sidewalks and some roads and highways. The use of concrete in cold regions is common, although some extra precautions are taken, such as air entrainment to encourage small (~100 microns) air bubbles that can accommodate the 9% volume increase of water upon freezing. Concrete has a low porosity of 10% or less, and some of this involved in an interconnected pore structure that gives concrete a permeability similar to clay. Furthermore, concrete can be subject to thermally-induced water migration (TIWM) in much the same way that soils are.
Whereas discrete ice lenses are formed in soils when particles separate, because concrete is a rigid solid, stress is induced in thin pre-existing cracks where water and ice accumulate. When this stress is great enough, fracture can occur which may result in failure of the concrete structure. Similar behavior has been observed in rock materials and caused much of the brecciation seen in arctic regions.
A
30-cm block of chalk that fractured due to recurrent uni-directional
freezing in a wet environment. No visible cracks existed before the
experiment.
There are different modeling approaches for rigid, porous materials undergoing freeze/thaw in a wet environment. We are investigating using the finite element technique and the discrete element technique to calculate the temperature, saturation, and pressure regime in concrete. Using fracture mechanics models, we can then determine where crack formation and ultimate failure might occur.
We are simulating the freeze-thaw behavior in the laboratory using cement paste, mortar, and concrete. Fracture can occur at discrete locations that is a function of temperature regime, rate of freezing, and material properties. Our goal is the determine the conditions that lead to fracture due to TIWM, and the techniques that help suppress it, such as air entrainment.
A
3-inch cylinder of mortar is tested in an environmental chamber.
Instrumentation and insulation has been removed.