Engineered Water Repellency for Infiltration Control in Coal Fly As

In 2014 approximately 47% of the 190 million short tons of coal combustion product (CCP) was recycled. As a construction material, coal fly ash (CFA) finds considerable use as a partial replacement for Portland cement, as structural fill, and as an additive for waste stabilization and environmental remediation (ACAA 2014). But still the larger portion is sent to disposal in ash impoundments or landfills. Recent regulations are accelerating the industry trend of dry ash handling and the closure of impoundments. In the case of ash that remains unencapsulated (e.g., in a structural fill, landfill, or dewatered/capped in place impoundment), concerns have been raised regarding the leaching of naturally occurring trace elements into groundwater. One mechanism to eliminate this concern is to treat ash so that it is water repellent, thereby preventing infiltration and leachate generation. Recent research has demonstrated promise, however little has been done regarding the relationship of governing parameters for water repellency, such as contact angle, infiltration pressure, grain size, and the mineralogical and chemical composition of fly ash. This study provides data and analysis to serve as a design basis for mitigating leachate generation through engineered water repellency. Engineering water repellency involves modifying the surface of CFA particles with organo-silane. The degree of water repellency is defined through contact angle (CA) and breakthrough pressure (BP) measurements. This study provides the first results of dynamic contact angle measurements for a modified coal fly ash (CFA) and compares the results with static contact angle measurements using three organo-silane (OS) chemicals. Five types of CFA were tested using three different OS chemicals at different mix ratios (weight based). This dissertation reports on the pattern that each of the modified CFA exhibits as a function of treatment type and level as well as the unique relationship between steady state motion of the three-phase contact line and drop volume. The proposed dynamic approach reduces the standard error of apparent contact angle measurements from +/-20° or greater with conventional approaches to less than +/-5°. The reduction in error is attributed to improvements in experimental design that addresses substrate heterogeneity and roughness as well as drop size and motion. For each sample a minimum drop size at which the three-phase contact line motion reaches a stable motion has been identified. It was identified that for a certain combination of CFA and OS, there exists a threshold drop size that overcomes factors that otherwise render apparent CA measurements unreliable. The extent of water repellency varied considerably as a function of OS treated dosage, which ranged from 1:500 to 1:140 (OS: CFA, by weight). The CAs of all five unmodified CFAs were below 90° whereas modification results in values greater than 135° on average. This change in CA from hydrophilic (CA 90°) is indicative of water repellency, (no infiltration), by which leachate control is possible. Multiple breakthrough pressure (BP) experiments were conducted for different samples at different mix ratios to identify the performance of the modified surface to resist a positive water entry pressure. Results indicated that CFA can be modified and made sufficiently water resistant against infiltration using OS. Regression analysis between measured and calculated (from the Washburn equation) BP revealed that the calculated BP underestimates the measured one by a specific scale factor. The magnitude and pattern of the scale factor varies based on the type of CFA and OS used. It was found that when the CA of the modified CFA becomes slightly greater than a threshold value, the relationship between CA and BP abruptly changes from linear to exponential. Based on these results; the Washburn equation should be adjusted for both a scale factor and the effect of pore CA. While the scale factor is related to changes in the surface energy of the material, the pore CA is related to the prevailing pressure of pore water, i.e., higher water entry pressure results in greater disparity. The proposed exponential relationship between measured apparent CA and BP is more robust and is consistent with the significance of pore (vs. apparent) contact angle as a function of surface energy. For example, at lower surface energy levels, the radius of water drop curvature in the pore space does not change much, and so pore CA approximates apparent CA as measured on flat surface. The relative abundance of chemicals and minerals in CFA and their corresponding sensitivity to OS modification has been investigated. In addition, surrogate indicators, such as the degree of alkalinity, amorphous content and reactivity are identified for each CFA. In terms of composition, a group of four variables were found to be most significant when using OS to modify CFA. Results showed that a semi log-transformed regressions model can effectively describe the relationship between water repellency and mineral composition (e.g., oxides, other minerals and their surrogates) in CFA. It has been found that the degree of water repellency of an OS treated CFA can be predicted using a limited number of minerals (only two) in CFA instead of the primary substrate (silica) that forms siloxane bonds with OS. Further, this study suggests that OS engenders water repellency with a variety of oxides. Class F fly ash is more easily rendered water repellent than Class C. When mixed with water, lime raises the pH of the solution, which in turn may adversely affect the bonding of OS molecules to Class C CFA. Results also indicate that the presence of magnesium oxide can partially improve OS bonding. In summary; these data can be used by engineers to evaluate induced water repellency as an approach to increase ash use in applications where leachability is a concern.