Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 unreacted calcium silicates in cement or soluble silica from clays or microcrystalline silts can also provide silicon needed for thaumasite formation in treated soils (15, 16). MECHANISMS OF REACTION Our current knowledge of the problem revolves around our understanding of the mechanisms of ettringite formation in cement chemistry, which is rapid during initial cement hydration (19). Of course, ettringite formation is expected early in the cement hydration process during hydration of tri-calcium aluminate and is not considered to be deleterious. This should not be confused with secondary ettringite formation that forms after the cement matrix has developed and is deleterious. The kinetics of ettringite formation during initial cement hydration is rapid because in cement the components in dry form are amorphous and uniformly blended. Also, due to the nature of their particle size distribution, they have a very large surface area. When mixed with water to form cement paste, this high surface area translates to a higher rate of reactivity and the reactants immediately become available in solution as soluble ions. Hence ettringite formation in Portland cement concrete is fast and is dependent solely on available sulfate content in the matrix. On the contrary, when soil systems are treated with calcium-based stabilizers, ion availability in solution is defined by mineralogy and dissolution properties of soil minerals. The flocculation/agglomeration during initial reaction periods contributes to reduction in surface area of soil particles. Hence the particles, or what may be better classified as agglomerates of particles, normally have a substantially smaller surface area when compared to Portland cement. Moreover, the soil minerals have a well-defined crystal structure and the effects of weathering and varying environmental conditions make the distribution of soil minerals more heterogeneous when compared to Portland cement. Hence the extent of ions available in solution to form ettringite is limited in stabilized soils when compared to cement pastes. Therefore, there is no reason to assume that because ettringite occurs rapidly in cement, say within the first day or so of cement hydration, it should occur as rapidly in stabilized soils. Furthermore, the behavior of treated soils and the extent of damage have to be considered to be soil specific and dependent on factors other than sulfate content alone (3, 5, 20). Research has proven that ettringite precipitation and the resulting volume changes in stabilized soils are higher in clays when compared to sandy soils under similar environmental conditions (5, 20). Mechanism of Formation There are two proposed mechanisms for ettringite formation in cement chemistry: by topochemical reaction and through solution reaction (21). In a topochemical reaction, ettringite crystal growth occurs at the solid solution interface (22, 23). In sulfate rich environments with a high concentration of lime, aluminum ions which dissolve cannot migrate far from the source due to the supersaturation of the liquid phase with respect to ettringite. Ettringite crystals therefore precipitate preferentially on the surface of the alumina-bearing phase in a topochemical reaction (24). Microscopy images of needle shaped crystals growing perpendicular to the surface of aluminum-bearing phases during cement hydration support the topochemical reaction mechanism (23). Other than a topochemical reaction, ettringite can form through a solution mechanism where the products precipitate randomly from the liquid phase after attaining a state of supersaturation. When the concentration of lime is low, the aluminum ions can migrate more freely in solution and ettringite can precipitate from solution under favorable conditions (24).
5 There are no published data suggesting that either of the mechanisms described above is the exact reason for ettringite formation in stabilized soils. Due to limited ion availability in solution, it is probable that the small amount of ettringite formed during the initial stabilization period or mellowing time may act as nucleation sites for future growth. Distribution of these nucleation sites may influence the extent of damage in stabilized soils. If a small number of random nucleation sites are created, then the limiting reagent, normally sulfate, may continue to migrate to these sites over time and form larger, concentrated crystals that ultimately cannot be accommodated by the soil matrix. The result would be a disruption of the matrix or heaving. On the other hand, if a larger number of well-dispersed nucleation sites are formed, the limiting reagent gets distributed over these sites and hence produces crystal growth that can be accommodated by the soil matrix (within the voids). This reaction can be promoted by using as much water as possible during construction and mellowing in order to place into solution as much sulfate as possible to form ettringite followed by thorough mixing to homogenize the system. Water moving through the system can solubilize unreacted lime and also act as a medium for ion migration. Water influx to the pavement layers and diffusion through the layers can therefore become a continuous source of reagents needed for future ettringite growth at these nucleation sites (1, 25). Mechanism of Expansion As one would expect, the discussion here is similar to that in the preceding section, âMechanism of Formation,â as formation and expansion are allied. Although the formation of ettringite is known to induce volume changes in stabilized soils, as with formation, the mechanism of expansion is not completely understood. Once again, two theories exist regarding the cause of expansion during ettringite formation. The first theory explains that expansion is due to topochemical formation of ettringite and the anisotropic growth of the crystals (22, 23). The second theory suggests that expansion is due to absorption of water by ettringite crystals (26). There is no intercrystalline chemical bonding among ettringite crystals, and hence water molecules can cause interparticle repulsion resulting in overall expansion of the matrix. Based on our current understanding it is difficult to select one of the above theories as the sole reason behind the observed expansion in soil systems. Hence it is reasonable to believe that either one of the above theories or a combination of both may result in expansive behavior when ettringite is present. Again, there is no simple proportionality between amount of ettringite and extent of expansion. Both theories described here suggest that water is the decisive factor in causing deleterious reactions. But the source of water that forms ettringite is critical in determining the extent of expansion in the matrix. Molar volume calculations based on ettringite stoichiometry show that the volume increases by 1.37 times the initial volume of reactants consumed from the matrix ( )432, SOandOAlCaO when the influx of external water triggers ettringite formation. In other words, the increase in molar volume of ettringite when fully hydrated by external water is 1.37 times that of the un-hydrated original reactants. On the other hand, ettringite formed from water available from within the matrix actually causes shrinkage, as the molar volume of ettringite is less than the components ( )OHandSOOAlCaO 2432 ,, when water is considered to be contained within the matrix and no external water is accounted. Moreover, it is noteworthy that a combination of negative charge and high surface area of ettringite crystals can attract water into the matrix (27). In fact, a recent study at Texas A&M University on controlled low
