Cohesion, Adhesion, and the Durability of Stabilized Materials

Stabilization projects have been implemented in various parts of the world to improve the strength and bearing capacity of foundations and to control the permeability, shrink–swell potential, and related characteristics of soil materials. A primary stabilization method is using amendments that have textural, mineralogical, and chemical characteristics that are capable of generating the required changes in physicochemical characteristics of soils when they are mixed. Stabilization agents that are currently available for soil improvement include portland cement, bituminous cement, fly ash, slags, lime, lignins, calcium salts, and polymers. The composite materials formed through soil stabilization vary in degree of cementation from free particles to monoliths in which soil and other introduced particles are bound. Whatever the degree of cementation attained, pores within particles (intragranular pores) and pores among particles (intergranular pores) cannot be completely eliminated from stabilized materials. Invariably, solid–solid, solid–liquid, solid–air, liquid–air, and triple interfaces exist in stabilized materials. These interfaces are the primary locations of physicochemical and biological processes that have significant implications on the initial strength and durability of stabilized materials.

Across solid–solid interfaces, particles may be of the same or different materials, the statistics of which depend on the mix proportions of the host material (often, a multicomponent soil) and the stabilization agent, if it is initially in solid form. Usually, the agent is applied in the liquid form, but regardless, subsequent precipitation of new solid phases derived from substances in the agent as well as in the host material, can still create new solid–solid interfaces. Obviously, the prominence of interfaces involving liquids and air depends on the porosity and permeability of the stabilized material and pore fluid chemistry. Interfaces are weak links in materials, because they provide greater opportunities than intact portions of materials for stress concentrations through a variety of mechanochemical processes. These mechanisms and processes are briefly discussed herein, with respect to the durability of chemically stabilized geomaterials.

The surficial environment (typically, $0–10\mathrm{m}$ deep) in which field stabilization projects are usually performed is subjected to cyclical stresses and reversals of environmental conditions over hours, days, and months. Within the bounds of the microclimate of the region of concern, daily and seasonal reversals in ground temperature, as well as moisture conditions, occur. As a result of contaminant emissions from industrial and civil facilities and contact of precipitation runoff with soluble materials on the ground, moisture that comes in contact with in situ stabilized materials is never neutral in chemistry. The long-term durability and environmental performance of a stabilized material depends on the response of its interfaces to loading and physicochemical attack in the surficial environment. The stabilization of unstable clayey soils with such chemical agents as cement and lime usually generates new solid phases, although the intergranular and intragranular porosities mentioned in a preceding paragraph still exist. At the fundamental level, the parameters of interest are cohesion (in terms of the soundness of individual particles) and adhesion, in regard to bonding of adjacent articles across solid–solid interfaces. In this context, cohesion is not defined as in soil mechanics, where it is considered to be the binding strength of particles, which is most significant in clays.

With respect to the load-deformation response, interface flaws that are distributed in a stabilized material as a result of incomplete cementation during stabilization and damage by the previously discussed environmental stresses, can grow. For soils, flaw sizes are likely to be somewhat directly proportional to soil lump sizes in plastic clays and to particle sizes in cohesionless soils (silt, sand, and gravel). On this account, flaws in boundary areas among large particles or internally sound lumps are likely to degrade more because of greater opportunity for their extension. However, a Poisson distribution of flaw sizes should be expected in stabilized, poorly sorted, cohesionless soils, since the larger pores will be fewer than the smaller pores, as a reflection of the typical distribution of particle sizes in soils. During the mixing of stabilization agents with a lumpy clay soil, disintegrating the lumpy clay soil is necessary so that particles or aggregates of particles that are internally weaker than the major particle interfaces because of their nonpermeation by the stabilization agent are not significantly present. In clay, nonpermeation can allow the swelling and weakening of lumps on contact with liquids.

The allowance of pores by incomplete cementation has many material durability implications. First, new materials can precipitate within the pores and can close both pores and debonded particle contacts so that the strength of the material is increased. This rationale is often used for stabilizing soils with cement and lime. However, if the crystallizing minerals exert expansion pressures that are beyond the tensile strength of the stabilized material, fracturing may intensify even without external loads. This phenomenon is exemplified in aggregate soundness tests in which the extent of aggregate damage by pressures exerted by anhydrous sodium sulfate is assessed. Pore fluids that have aggressive chemistry can attack solid–solid interface bonds and weaken stabilized materials through a suite of processes that are recognized in mechanochemistry of materials as the “Rehbinder effect.” As an example, dispersion of ineffectively stabilized clay of appropriate mineralogy can lower the magnitudes of shear modulus and other strength-related parameters of clay. This phenomenon occurs with Na-montmorillonitic soils that are exposed to a high pH environment. The hydroxyl ions (OH-) present in pore fluid can attach to positive charges on broken edges of clay particles (those that have not been neutralized by stabilization agents), causing the expansion of double layers around affected clay particles, and leading to dispersion and lowering of the shear strength of some portions of the clay soil.

Another significant aspect of pore fluid action in stabilized materials relates to contaminant leachability. Ashes produced by municipal waste incinerators and coal combustion in electric power plants are used in soil stabilization in many countries. Some papers in this special edition focus on ash use for this purpose. A concern with ash use in exposed structural systems is the excessive leachability of chemical substances that are known to be distributed internally and on the surfaces of ash particles. When ash is bound with soils in cemented monoliths, the diffusion coefficient of a potential contaminant through soil lumps and ash particles into the percolating leachant (rainwater or groundwater) in the intergranular pore fluid is likely to be rather small. Of course, the contaminant diffusion coefficient is directly proportional to material porosity when all other factors are held constant. This porosity is controlled operationally by the design of the stabilized material mix in terms of its component mix proportions. Consequently, for an ash-stabilized soil, the mix proportions of the ash, soil, water, and any additionally applied binder affect the porosity and hence the leachability of chemical substances (e.g., the contaminant) from the stabilized soil. The concentration gradient that drives the flow of the diffusing contaminant from the host particles (ash) to the intergranular pores where flushing occurs is partly dependent on the volumetric fraction of ash in the stabilized soil mix. The greater the ash content, the greater the gradient and the contaminant leaching time required to dissipate it. Furthermore, greater particle or lump sizes mean greater distances for diffusion, with a consequent high probability of transport constraint to the introduction of the contaminant into the flushing pore fluid at large flow rates.

Semi-plastic materials such as bitumen have also been used in soil stabilization to improve the bearing capacity of soils. Sandy or silty soils can be effectively stabilized through this method. Progressive aging of the bituminous binder can cause the stabilized soil to be brittle and fail in the long run. The effects of aging are most prevalent at solid–solid interfaces.

The significance of interfaces with respect to the durability of stabilized materials should be given greater recognition in stabilization projects. On the basis of scenarios of the environment in which the stabilized material is expected to perform, the most significant stress-inducing mechanisms and relevant parameter magnitudes should be identified for use in designing simulative tests. Subjecting the stabilized material (with varied mix composition) to field and laboratory tests is likely to enable the optimization of material mix composition for cost-effective project implementation. Mix design can then be combined with monitoring and maintenance planning to improve stabilization programs. This special edition of the ASCE Journal of Materials in Civil Engineering on “Stabilization of Geomedia Using Cementitious Materials” exhibits research findings and practical projects that represent advances in this direction.