Browsing by Author "Sobhee-Beetul, Laxmee"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    Open Access
    A Comparative Study on Shear Strength Testing of Single and Multi-layer Interfaces using Large Direct Shear Apparatus
    (2021) Muluti, Shade; Kalumba, Denis; Sobhee-Beetul, Laxmee
    Geotechnical structures such as composite liner systems in landfills consist of multiple interfaces, which include a broad range of geosynthetics in conjunction with soil, rocks and any other related materials. This results in the introduction of many interface planes into the structure, which can potentially create instability especially along the slope and ultimately result in failure. To date, many laboratories use single interface testing instead of multi-layer interface testing to determine geosynthetic shear design characteristic values that are used in the design of structures such as landfill liners. A topic of discussion remains the preferred interface testing configuration and only a few studies have substantiated and quantified the significance of varying the different interface shear testing configurations. This study, therefore, aimed to evaluate and compare the effects of the use of the two interface test configurations on the shear strength of soil/geosynthetic and geosynthetic/geosynthetic interfaces. Furthermore, it was intended to identify the test configuration that provides the most critical shear strength results, while also understanding the fundamental mechanisms responsible for the shear strength observed. In this study, three geosynthetics were used: geotextile (GTX), geomembrane (GMB) and geosynthetic clay liner (GCL), which generally constitute the critical interface components of a lining system in a modern South African landfill liner. Two soils were utilised as a part of the materials required for the investigation and they were: river sand and red clay. The laboratory tests were conducted under saturated conditions in accordance with the ASTM D5321 and ASTM D6243 standards, using a 305 mm x 305 mm large direct shear box. The tests were carried out over a range of applied normal pressures of 50, 100, 200 and 400 kPa. A constant shear rate of 1.0 mm/min was used in the interface tests that did not involve GCLs or clay specimens and therefore no excess pore pressure was anticipated at the interface. On the other hand, for all other interface tests involving either clay or GCLs samples, a shearing rate of 0.1 mm/min was utilized. The results showed that nonlinear behaviour of the shear stress versus shear displacement responses was exhibited in both the single and multi-layer interface tests, regardless of the normal stress applied. However, it was noted that with an increase in normal stress applied, the deviation in mobilized shear stress between the two test configurations increased, with single interface tests yielding higher shear stress values compared to multi-layer interface tests. In single interface tests, the high shear stresses could be related to the clamping that confined each of the test specimens during shearing to one end of the shear block. On the other hand, only the top and bottom test specimens were clamped in multi-layer interface tests, thus allowing failure to have occurred at the weakest of the available interfaces. Moreover, for single interface tests, peak strengths were generally 9% lower for the range of normal stresses considered, whereas Large Displacement (LD) strengths were generally 24 % lower for the single interface tests, compared to the peak and LD strength values for multi-layer interface tests. This was particularly observed at low normal stresses between 50 and 200 kPa, and it could probably have been caused by the rigid clamping of the geosynthetics which results in some tensile strains in the geosynthetics. In addition, it was observed in multi-layer interface tests that a transfer of shear stresses within the system could have occurred, which could have led to higher overall shear resistance of the composite. As a result, single interface tests yielded a conservative estimate of the peak and LD shear strengths for the tested interfaces compared to multi-layer interface tests. This may be attributed to higher displacement along with the critical interface in single interface tests than in multi-layer interface tests. To allow the investigator to observe the displacement, as well as the possible transfer of shear stresses within the system during the shearing of the various geosynthetics, it was recommended that real-time monitoring of the displacement mobilization should be carried out in multi-layer interface tests during shearing.
  • Thumbnail Image
    Item
    Open Access
    A study on ground improvement using a combination of stone and concrete columns
    (2018) Pudaruth, Yogendra; Kalumba, Denis; Sobhee-Beetul, Laxmee
    Stone column is a cost-effective ground improvement technique that is typically employed for low-rise buildings and road embankments. This technique mainly uses naturally occurring materials as its load transferring medium. However, stone columns have some constraints because of the loose interactions between their aggregates which can lead to uncontrolled settlements, especially in soft soils. As a result, their performance is usually improved by the inclusion of geosynthetics either in layers or as a confinement. However, there was a lack of studies that used a binder within the stone column aggregates with a view to limit the bulging/lateral spreading of its aggregates in such soils. In this study, the upper portion of the stone columns was replaced by different grades of unreinforced concrete. The length of the concrete, as well as the depth of the soil beneath the columns, were varied. The effects of these different variables, when the resulting column was subjected to an applied load, were investigated. The optimum configuration of the above was identified and its resulting change in performance when it was combined with a reinforced bedding layer was studied. Considering application/installation procedures on site, it was best deemed to install and test a geosynthetic-reinforced bedding layer on top of, rather than within, the stone column. It was observed that increasing the grades of concrete did not have any consistent influence on the performance of the resulting columns when there was a considerable layer of soil beneath them. The hybrid stone columns (combination of stone and concrete) performed better than the normal stone column and even to a full concrete column of the same length in several cases. Physical modelling revealed that the bulging length ranges from 2.0-2.4D (D is the diameter of the column). Test results for the optimum hybrid stone column yielded a maximum load improvement factor of 3 to 6 folds (200% to 500% increase in bearing capacity) depending on their respective configuration compared to the unreinforced soil. The improvement factor was further increased to 9.9-fold (nearly 900% increase in bearing capacity) when the optimum hybrid stone column was tested in combination with a reinforced bedding layer. The findings from this research can be used to enhance and promote the stone column ground improvement technique while still providing an economical advantage as well.
  • Thumbnail Image
    Item
    Open Access
    An investigation into the volume change characteristics of loess like soil in Mount Moorosi Village in Lesotho
    (2019) Damane, Monica; Kalumba, Denis; Sobhee-Beetul, Laxmee
    The Mount Moorosi village is situated in the Senqu River Valley of southern Lesotho, within the Stormberg landform. The integrity and aesthetic appearance of nearly all the structures in this area are undermined by recurrent cracks. At present, no apparent institutionalised effort had been conducted to investigate the source of this problem. The crack patterns were associated with the possible volume change of the underlaying loess like soil. This soil has caused a disastrous failure to brittle civil engineering structures in various parts of the world. Its behaviour is attributed to sand and silt particles bonded by minerals, which become active upon saturation and induce hydrocollapse settlement. This study characterised the volume change properties of the underlaid deposits in Mount Moorosi. The research utilised representative samples from trial pits in the study region to perform laboratory experiments such as the Atterberg limits, wet sieving, sedimentation, free swell, x-ray diffraction, scanning electron microscope and slaking. The consolidated undrained tests and hydrocollapse potential were also determined from the GEOCOMP triaxial and Global Digital System oedometer, respectively. Results revealed that Mount Moorosi is generally underlaid by a more than 3 m thickness of low plasticity (9 to 17 %) silty-sandy loess. The material had significant warping (up to 27 mm) in linear shrinkage that illustrated potential inducement of detrimental stresses to the superimposed structures during drying. The identification and quantification of the mineralogy composition clearly evidenced the passive minerals (quartz, feldspar and mica) to be predominant (86 %), while the active phases (kaolinite, carbonates, sulfates, halides, the oxides and hydroxides of aluminum and iron) were subordinate (14 %), which substantiated potential soil settlement upon wetting. Furthermore, the micrographs depicted structures that synergistically enhanced the collapse properties of the tested deposits. These included the porous clays, silts bonded by clay and silts coated with clay, which all rendered a metastable fabric. A comparison of the stressstrain graphical plots from the consolidated undrained tests at the field and saturated moisture contents indicated a drastic reduction (up to 73 %) in deviator stress at saturated water content. This was attributed to the augmentation of the interparticle spaces, caused by a rise of up to 337 kPa in pore water pressure. Shear strength parameters obtained from Mohr’s failure envelopes were also decreased by up to 80 %. The hydrocollapse index measured from the oedometer tests ranged from 10 to 15 % at a vertical stress of 200 kPa. It indicated severe settlement problems for structures constructed on this soil. This was caused by the loss in shear strength of the soil under the saturated conditions and a high slaking mechanism that reached a maximum rating of 4. Generally, the mineralogy composition, morphology, saturated shear strength, slaking and hydrocollapse index collectively indicated the possibility of soil volume decrease. In fact, the check for serviceability limit state demonstrated a settlement that exceeded the tolerable value of 50 mm. The cracks observed on structures were, therefore, related to soil settlement. This study recommends further research on suitable ground techniques to minimise settlement, thereby improving the durability of structures. Moreover, investigations should be conducted to understand the pressure induced by warping during shrinkage.
  • Thumbnail Image
    Item
    Open Access
    Ground improvement using PET bottle waste as a potential reinforcement material for granular columns: an experimental approach
    (2019) Sobhee-Beetul, Laxmee; Kalumba, Denis
    Out of the many ground improvement methods aimed at ameliorating the weak engineering properties of certain soils, granular columns are often preferred due to their cost effectiveness and environment friendliness. Despite their high usage in other countries, this technology remains rather unpopular in South Africa. Therefore, this study was undertaken to extend the associated knowledge of the granular column within the local context. In line with the need to develop environmentally sustainable construction technologies, the study incorporated Polyethylene Terephthalate (PET) bottle waste as a reinforcement material for these columns. Several laboratory experiments were conducted to improve the load carrying capacity and settlement characteristics of a local fine silt. A wet silt bed (prepared at optimum moisture content or liquid limit) was created within a steel cylindrical tank. An ordinary granular column (OGC) or reinforced granular column (RGC) was then installed within the tank and a compressive vertical load was applied to the prepared sample up to a settlement of 50 mm. Reinforcement for the columns was used in different forms and arrangements. The stresssettlement characteristics were electronically captured and subsequently analysed. Posttesting, the deformation of the column was physically modelled by vacuuming out the column material to create an empty opening. A prepared wet mix of plaster of Paris and sand was then poured into the empty hole until it was filled to the top. Once set, the casted column was removed from the tank and its circumference was measured at different intervals up along the length of the column. This process was repeated after each test and these measurements were later used for determining the respective maximum bulging. The study confirmed that the inclusion of granular columns generally improved load carrying capacity, as well as reduced settlement in weak soils such as fine silts. Also, certain conditions of reinforcing of the columns further improved their performances. From the results, it appeared that reinforcing a column with a concentration of 0.1 % of randomly mixed fibres, and installed in a base soil at liquid limit, produced the largest percentage improvement of 244 % in load carrying capacity. Furthermore, the diameter of maximum bulging was reasonably low and was measured as 144 mm, compared to 150 mm for an OGC which was tested under similar conditions. The outcome of this study considerably extended the understanding of the reinforcement of granular columns using PET bottle waste. Since the concept of reinforcing granular columns with waste is new, several areas were identified for future research to further increase knowledge pertaining to this ground improvement method.
  • Thumbnail Image
    Item
    Open Access
    An investigation into using rammed stone columns for the improvement of a South African silty clay
    (2012) Sobhee-Beetul, Laxmee; Kalumba, Denis
    Ground improvement is the term used to describe the act of modifying the soil properties in geotechnical engineering. It is often required when the existing ground conditions do not meet the requirements for a construction project. The technique of improving the ground are many and they usually aim at reducing settlement, increasing bearing capacity, mitigating liquefaction, improving drainage, retaining unstable soils or remediating contaminated soils. Among these techniques, stone column technology which was pioneered in the 1950s' aimed at improving both cohesive soils and silty sands. Although the technique has been used successfully in many advanced countries, its application in South Africa is minimal. This limited use is associated with a lack of research, instrumented case studies and design specifications pertaining to local ground conditions. In this investigation, the behaviour of rammed stone columns installed in a South African clay (sourced from District Six in Cape Town) was studied through extensive laboratory tests conducted in a specifically designed rectangular wooden tank. A testing programme was established whereby the majority of the tests were conducted on the local clay, with a few performed on Kaolin for comparison purposes. The effect of moisture content (OMC, LL and 1.2LL) of the base soil specimen, the column diameter (50 mm, 70 mm and 100 mm) and the column material (Klipheuwel sand, Cape Flats sand and crushed aggregate) on the normal compressive stress applied up to a settlement of 50 mm were studied. The vertical stresses and settlements were recorded electronically and analysed. Results indicated an increase in vertical applied stress with a simultaneous reduction in settlement when improving Cape Town clay with rammed stone columns. The in-depth analysis showed that the vertical bearing stresses were generally higher with larger columns, irrespective of the column material and the moisture content of the base soil. Furthermore, 100 mm diameter crushed aggregate columns repeatedly exhibited higher improvement compared to identical columns made out of sand. Nevertheless, for smaller columns the coarser sand (Klipheuwel) performed better than all three column materials. In general, columns failed in bulging with the maximum bulge occurring within the top third of the column. This study deepened the understanding of the performance of rammed stone columns in fine grained soils. The results obtained recommended full scale investigations with a view of applying the technology in slope stability, strip footing foundations and supporting lightweight structures.