The vast majority of U.S. river cities, now growing at increasing rates, are protected from flooding by earthen levees that are at risk from many sources of failure including seepage (both underseepage and through seepage), erosion and instability due to seismic loading. This grant provides funding to develop a novel type of slurry cut-off wall for levees that possesses multiple risk mitigation, including seepage resistance, high seismic resistance, constructability, and sustainable fabrication. The project aims at establishing the feasibility of such a cut-off wall through a combination of advanced materials design and non-linear dynamic numerical simulation of failure resistance against seismic loading. The cement and bentonite based material reinforced with nano/micro fibers will be designed using a micromechanics composite design methodology combined with numerical analyses of cut-off walls under seismic loading. To confirm the expected benefits, the project will include finite-element numerical analyses of levees and embankments with cut-off walls to understand the mechanical behavior of these slurry cut-off walls under seismic loading, as well as the soil-cut-off wall interaction.
The results of this research will directly impact the design of vertical cut-off walls for levees in seismic regions, providing a safer, more efficient and sustainable alternative. The coupled material tailoring, mechanical and transport property testing, and soil-structure analyses represent a novel departure from past practice and provide an intellectual platform to attain significantly enhanced safety for U.S. river cities. This interdisciplinary project is expected to bring about new approaches to addressing the performance of broad geotechnical infrastructure systems beyond levees, including e.g. piles, below grade pipes and underground construction that have common features of soil/structure/materials interactions and are often subjected to large imposed deformation and water seepage.
In an effort to address the need for more advanced cutoff wall materials for levees in seismic regions, a feasibility study was conducted to investigate the use of an engineering cementitious composite (ECC) as a levee cutoff wall material. Two sets of numerical modeling parametric analyses were conducted to investigate how the presence of a cutoff wall affects the dynamic response of a levee and what type of demands are placed on the cutoff wall during the seismic event. One set of analyses was for levees founded on non-liquefiable soils and the other was for levees founded on liquefiable soils. In collaboration with materials science researchers, an ECC material was then tailored for use in the construction of levee cutoff walls. The results of materials testing on the resulting candidate mix design were then used as input in further numerical modeling, in which the dynamic response of levees with ECC cutoff walls was investigated. This numerical model was used to assess the feasibility of the candidate mix design. The main conclusion, addressing the primary focus of this study, is that ECC is a feasible material option for high performance cutoff walls for levees in seismic regions. Table 1 presents a summary of the key aspects of the final candidate mix design. The results of the numerical modeling indicate that the structural demands placed on an ECC levee cutoff wall by an expected range of seismic events are within the capacity of the proposed material. The proposed mix design was shown to be feasible in terms of economics and constructability, with a material cost well within the target cost and fresh properties consistent with other materials that are implemented using the slurry construction method. In addition, the tailoring process also showed that the environmental impact of the material can be decreased with a variety of mix alterations. In addition to demonstrating the feasibility of ECC as a cutoff wall material, this study has also resulted in findings related to the dynamic response of levees with cutoff walls. Levees founded on non-liquefiable soils - The ratio of vertical to horizontal slope displacement was observed to deviate significantly from the commonly assumed value of 0.7 and was observed to by systematically different for levees with and without cutoff walls. For analyses of clay profiles with Ricker wavelet inputs, the ratio was observed to vary between 0.25 and 1.2. The ratio was consistently lower (by approximately 0.1) for levees with sheet pile cutoff walls, as opposed to levees with no cutoff wall. The average value of the ratio for levees with no cutoff wall ranged between 0.7 and 0.9 and the average value of the ratio for levees with a steel sheet pile cutoff wall ranged between 0.6 and 0.8. (Fig.1and 2) - Comparison of horizontal slope displacements for ground motion recording inputs with simplified relationships for Newmark-type displacements indicate that the simplified relationship tends to overestimate displacements for levees with cutoff walls. The overestimation was observed to be as great as 80%. - Earth pressures on deep portions of the cutoff wall behave similarly to what has been observed for braced retaining structures. However, at shallower locations, particularly within the levee itself, the wall pressures were strongly influenced by adjacent levee slope movements. A simple monotonically increasing lateral earth pressure profile is not adequate to characterize the dynamic wall pressures. Levees founded on liquefiable soils - For dense sandy levees (Levee A), no systematic difference was observed in horizontal displacements between levees with and without cutoff walls. The difference in horizontal displacements (normalized by displacements for a levee with no wall) between the two cases ranged from -10% to 10% and averaged approximately 0%. For medium dense to loose sandy levees (Levee B), the presence of a wall was seen to result in a reduction in horizontal displacements by 15% on average, but as high as 32%. - Similar behavior was observed for vertical crest displacements. For Levee A, there was little difference between levees with and with cutoff walls. The difference ranged from ?8% to 11%, with an average of approximately 2%. For Levee B, however, the presence of a cutoff wall resulted in a reduction of vertical displacements of 15%, on average. - No consistent difference was observed for the ratio of vertical to horizontal displacements for levees with and without cutoff walls. However, the ratio ranged between 0.45 and 1.0 and varied for different numbers of cycles of loading, although it converged to 0.7 for input motions with greater than 9 cycles of loading (Fig. 3). - With regard to the structural demands placed on the cutoff wall during and after shaking, the greatest cutoff wall demands were observed when only portions of the soil beneath the levee liquefied, as opposed to cases with very widespread liquefaction beneath the levee.
