Development of Fracture Based Geotechnical Design of Clay Barriers
| dc.contributor.author | Zaid, Mohammad | |
| dc.date.accessioned | 2026-08-26T16:58:01Z | |
| dc.date.issued | 2026-08-26 | |
| dc.date.submitted | 2026-08-24 | |
| dc.description.abstract | Compacted clay barriers are central components of landfill liners and covers, clay cores in embankment dams and tailings facilities, and other geoenvironmental containment systems. Their design is traditionally governed by classification, plasticity, compaction, hydraulic conductivity, shear strength, compressibility, and construction-quality requirements. These criteria are essential, but they do not directly measure the resistance of a compacted clay barrier to the initiation and extension of open-mode discontinuities and the subsequent formation of connected preferential flow paths. This omission is critical because an intact clay layer may meet the required low hydraulic conductivity, whereas the same layer can lose containment capacity when tensile cracks develop due to desiccation, differential settlement, shrinkage, hydraulic loading, or construction-induced fabric defects. The problem is particularly consequential for persistent contaminants such as per-and polyfluoroalkyl substances (PFAS), where landfill leachate can migrate through a liner with an open discontinuity, creating long-term exposure pathways to groundwater. The present study develops a fracture based geotechnical design framework for compacted clay barriers by combining laboratory fracture testing, effective single edge notched beam (SENB) calibration, state-dependent geotechnical characterization, numerical comparison of crack-propagation formulations with selected experimental measures, and sequential one-way fracture-to-flow modelling. The work is organized around three connected technical questions. First, can Mode I fracture toughness of compacted clay be measured in a repeatable manner using a specimen geometry and loading system appropriate for weak, moisture-sensitive soils? Second, can the single edge notched beam (SENB) geometry correction function, 𝑓(𝑎/𝑊), be derived for compacted soils rather than adopted directly from metallic fracture standards? Third, can the measured fracture parameters be incorporated into numerical crack propagation and hydraulic conductivity models to evaluate clay barrier performance as a sequential one-way mechanical-fracture-to-flow coupling problem rather than as an intact permeability problem alone? A displacement-controlled SENB testing system is developed for compacted clayey soils. The system includes a low-load resolution load cell, vertical displacement measurement, dual crack mouth opening displacement measurement, roller supports, and machined moulds that produce repeatable notch depths during specimen preparation. The experimental matrix used laboratory-prepared kaolin and sand mixtures, denoted LS0, LS15, and LS25, together with a field clay barrier soil, denoted FS. The tested SENB geometries were selected to match practical soil sample dimensions while preserving fracture-mechanics similarity through 𝑆/𝑊 = 4, 𝐿/𝑊 = 5, and notch ratios 𝑎/𝑊 = 0.45, 0.50, and 0.55. A balanced geometry and replication programme showed that the 𝑊 = 𝐵 = 50 mm SENB specimen provides the most repeatable corrected fracture toughness response for the tested compacted soils, because it provides sufficient ligament size for a representative fracture process zone while remaining practical for soil preparation and handling. The present study develops and cross-checks an effective SENB load-to-𝐾I correction within the tested geometry and material-state domain using three complementary routes: three-dimensional Abaqus® contour integral extraction, analytical derivation of the weight function, and experimental compliance validation. The finite element route first reproduced the ASTM metallic benchmark to verify mesh convergence and stress intensity extraction, and then used the measured soil elastic properties to evaluate finite body amplification in compacted soils. The analytical route preserved the LEFM crack-tip singularity and introduced soil-specific finite body scaling using measured the ranges of Poisson’s ratio. The experimental route used the measured load-CMOD response as an independent compliance check. The three routes provide a cross-consistency assessment of the adopted load-to-𝐾I correction within the tested geometry and material-state domain; they do not establish a universal material-dependent geometry function. The experimental results show that fracture toughness is governed by the coupled effects of water content, dry density, fabric, specimen geometry, and soil composition. Higher dry density was generally associated with higher apparent Mode-I initiation resistance, whereas higher water content was associated with lower resistance within the tested soil series; matric suction and contact-scale resistance were not measured independently. Near the representative design state, corrected Mode I fracture toughness values were approximately 0.86-0.87 kPam^0.5 for LS0, 0.99-1.01 kPam^0.5 for LS15, 0.51-0.62 kPam^0.5 for LS25, and 1.70-1.83 kPam^0.5 for FS. These values demonstrate that the selection of clay barrier material cannot be based solely on hydraulic conductivity or fracture toughness. Clay-rich mixtures may provide excellent sealing but only moderate crack resistance, whereas field soils or modified mixtures may exhibit higher fracture resistance but different hydraulic and compressibility behaviour. Nanomaterial improvement is therefore assessed using fracture toughness, hydraulic conductivity, strength, and consolidation metrics together rather than through a single optimized property. A proposed 𝐾Ic-𝑘-𝑎v interrelationship is developed to connect crack initiation resistance, saturated hydraulic conductivity, and coefficient of compressibility for clay barrier screening. For the investigated materials, a preliminary design screening space defined by 𝑘 ≤ 1.0×10−9 m/s and 𝐾Ic ≥ 0.8 kPam^0.5 identifies LS0 and FS as the most favourable barrier candidates, classifies LS15 as fracture toughness favourable but hydraulically marginal, and identifies LS25 as unsuitable under the coupled criterion. This interpretation demonstrates that crack resistance, intact permeability, and compressibility are mutually constrained design variables. The numerical part of the present study compares four fracture formulations: the extended finite element method, the space-time finite element method, the phase-field fracture method, and the lattice element method. All methods were evaluated using the same SENB cases and selected experimental response measures. For the examined cases, the lattice element method provided the closest combined agreement with the corrected apparent initiation-toughness trends and visible post-test paths and supplied the failed-link graph and displacement jumps required for hydraulic mapping. The subsequent pore-network calculations predict changes in equivalent hydraulic conductivity from the mapped opening, throat conductance, and connectivity; these predictions were not independently validated against flow measurements on mechanically fractured specimens. The final outcome is a design methodology in which conventional geotechnical tests remain the first screening level, but fracture toughness, the soil-specific SENB correction function, the 𝐾Ic-𝑘-𝑎v relationship, and sequential fracture-to-flow predictions provide the additional checks required for crack sensitive clay barriers. The proposed framework advances clay barrier design from intact-hydraulic conductivity acceptability to sequential one-way mechanical-fracture-to-flow coupling performance. | |
| dc.identifier.uri | https://hdl.handle.net/10012/24062 | |
| dc.language.iso | en | |
| dc.pending | false | |
| dc.publisher | University of Waterloo | en |
| dc.subject | PFAS | |
| dc.subject | SENB | |
| dc.subject | Fracture | |
| dc.subject | Clay Barrier | |
| dc.subject | Lattice Element Method | |
| dc.subject | Geometry Correction Factor | |
| dc.subject | Hydraulic Conductivity | |
| dc.title | Development of Fracture Based Geotechnical Design of Clay Barriers | |
| dc.type | Doctoral Thesis | |
| uws-etd.degree | Doctor of Philosophy | |
| uws-etd.degree.department | Civil and Environmental Engineering | |
| uws-etd.degree.discipline | Civil Engineering | |
| uws-etd.degree.grantor | University of Waterloo | en |
| uws-etd.embargo.terms | 0 | |
| uws.contributor.advisor | Basu, Dipanjan | |
| uws.contributor.advisor | Wuttke, Frank | |
| uws.contributor.affiliation1 | Faculty of Engineering | |
| uws.peerReviewStatus | Unreviewed | en |
| uws.published.city | Waterloo | en |
| uws.published.country | Canada | en |
| uws.published.province | Ontario | en |
| uws.scholarLevel | Graduate | en |
| uws.typeOfResource | Text | en |