grained fill, overlying topsoil and peat, overlying granular marine deposits, overlying very soft to soft fine-grained marine deposits consisting of clayey silt to silty clay interbedded with silt layers, overlying sandy silt to silty sand and gravelly sand to sand and gravel. Bedrock was encountered beneath the granular deposits and generally comprised granodiorite with occasional andesite dykes. Within the DM area, the subsurface conditions were highly variable, ranging from deep (up to 21 m [69 ft] thick) deposits of very soft to soft clayey silt to silty clay, overlying layers of dense to very dense gravelly sand to sand and gravel, to compact deposits of silty sand (up to 15 m [49 ft] thick), and overlying dense sand and gravel. The key geotechnical parameters used for static and seismic analyses to iterate equivalent strength and elastic modulus properties for the composite DM zone are summarized in Table 2, with the design values for the main soil units presented. For static analyses, the key soil parameter is the undrained shear strength (S ) of the u clayey silt to silty clay soils. Design S u values considered rapid loading of the MSE walls and temporary fill surcharges and ignored any beneficial gains in shear strength due to consolidation effects. Pseudo-static and dynamic analyses were carried out considering long-term post-consolidation strength values, assuming full consolidation under the MSE walls and the bulk earthworks. Material parameters representative of the post-OBE and post-SSE conditions, considering the effect of cyclic loading, strain-softening of the clayey silt to silty clay soils, and pore pressures developed during shaking, were analyzed. The design considered the composite DM zone extending approximately 5 m (16 ft) beyond the toe and heel of the MSE walls (reinforced section of the wall). The design depths of the CSM barrettes were initially based on global and external stabil- ity analyses, followed by internal stability and static FE stress-deformation analyses. Further analyses on global stability, BCM10 cutter head with mixing wheels fitted with TungStuds Deep Mixing Design The DM design consisted of rectangular CSM panels typically 2.8 m (9.2 ft) long and 1.0 m (3.3 ft) wide ranging from about 5 m (16 ft) to about 30.7 m (101 ft) in depth, penetrating typically 2 m (6.6 ft) into competent, dense coarse-grained soils, with a nominal overlap of about 200 mm (8 in) to form continuous walls, termed “barrettes.” These barrettes were constructed in parallel at 3.4 m (11 ft) centre-to-centre spacings to form a “composite DM zone,” with equivalent strength and elastic modulus properties estimated based on an area replacement ratio of about 0.3. The composite DM zone was connected to a continuous longitudinal wall of CSM panels with material properties estimated from the laboratory test results completed on representative wet grab and trial panel cored soil-cement samples for the project design mix. Design Parameters Strata Structural Fill (MSE wall and bulk fill behind MSE wall) Compact Fill / Marine Granular Deposits Fine-Grained Marine Deposits (clayey silt to silty clay) Sandy Silt to Silty Sand Gravelly Sand to Sand and Gravel 22.0 20.0 18.5 19.0 20.0 42 33 30 to 34 33 to 35 35 to 37 N/A N/A 15 to 55 N/A N/A Table 2: Summary of design soil parameters for static DM analysis DEEP FOUNDATIONS • JAN/FEB 2015 • 57 0.32 to 0.35 38 to 170 0.28 0.48 0.30 0.33 20 to 29 6 to 24 30 to 56 93 to 133 Unit Weight (KN/m3) Shear strength Drained (˚) Undrained (kPa) Poisson’s Ratio Elastic Modulus (MPa) external and internal stability, static finite element, liquefaction assessment and seismic site response are detailed in the paper, “Deep Mixing Ground Improve- ment Design to Support Large MSE Walls for Kitimat LNG, BC” presented at the 2014 DFI Annual Conference on Deep Foundations in Atlanta. The CSM Process In order to consistently achieve the design strength in the very weak marine deposits at the site, it was anticipated that a relatively high cement content would be required for the deep mixing elements. In addition to bench scale laboratory testing, a comprehensive trial phase was undertaken, which included collection and testing of both wet-grab and cored samples of panels mixed using a range of cement contents. The results of the review of the test data indicated that an average of 375 kg of ordinary Portland cement per cubic metre (632 lb per cubic yard) of mixed soil would be required to con- sistently meet the design specifications. Consideration was briefly given to the use of blended cements, but given the volume required and the logistics of delivery and handling, this was not pursued. For this project, a total of 1,645 CSM 73,900m (96,660 yd ) were constructed over a period of 15 months using two CSM rigs, with one rig typically working 24 hours per day, and the other, 12 hours per day, 7 days per week. During this period, some 32,500 tonnes (35,830 US tons) of cement were supplied and processed. Due to the lack of road access to the site, elements with a total volume of around 3 3 bulk cement bag delivery comprising 18,000 double lined 1.8 tonne (4,000 lb) bags was required. The adoption of bulk delivery using “big-bags” introduced additional issues to the project. It was noted
