An experimental study of the deformational and performance characteristics of foamed bitumen stabilised pavements

The research presented in this thesis studies the effects of foamed bitumen on the deformational behaviour and performance of pavement materials. The research was conducted in the laboratory and the field, using specific New Zealand materials. The aggregate used is a blend of a coarse aggregate imported from the Auckland region with a crushed dust from the Canterbury region. The bitumen selected for the study is an 80/100 bitumen grade, and the active filler was a Portland Cement, both commonly used for foamed bitumen stabilization in New Zealand. In the laboratory, samples of mixes with different foamed bitumen content were tested under various loading and stress conditions to investigate the effects of foamed bitumen on the deformational behaviour of the mix. The tests performed were: Indirect Tensile Strength (ITS), Indirect Tensile Resilient Modulus (ITM), Repeat Load Triaxial compression (RLT) and Monotonic Load Triaxial compression (MLT). Preliminary ITS and RLT tests conducted on mixes with 1% and 0% cement, at different foamed bitumen contents, indicated that mixes without cement performed poorly compared to the mixes with 1% cement. Therefore, the rest of the laboratory study was on mixes with 1% cement. ITS tests were conducted on 150 mm specimens prepared with 0% 1%, 2%, 3% and 4% bitumen content, with a common 1% cement. Results indicated that foamed bitumen increases the ITS values of the mix, up to an estimated optimum of 2.8% bitumen content. Similar trends were obtained with ITM tests, in which a diametrical load pulse was applied on 150 mm specimens, showing an estimated resilient modulus peak near to 2.8% bitumen content. RLT specimens were prepared at 0%, 2% and 4% bitumen content, at two compaction efforts, creating specimens at low and high bulk density. Permanent deformation RLT tests involved the application of seven stages of 50,000 load cycles each (4 Hz), with increasing deviator stress (from 75 kPa in the first stage, up to 525 kPa in the seventh stage) and at constant confining pressure of 50 kPa. Results of RLT permanent deformation tests indicated that the increase in the foamed bitumen content resulted in an increase in the permanent deformation of the material. MLT tests were conducted on specimens at 0%, 2% and 4% bitumen contents, at two compaction efforts, creating specimens of low and high bulk density, at confining pressures ranging from 50 kPa to 300 kPa, with a deformation rate of 2.1% per minute. Results indicated that the effect of foamed bitumen was a reduction of the peak vertical stress, or a reduction in the peak strength. The peak stresses obtained in MLT tests were plotted in stress diagrams, and the failure was approximated as linear function of the confining stress. The fundamental shear parameters (angle of internal friction and apparent cohesion) were estimated, and results indicated that foamed bitumen has no apparent effect in cohesion but does reduce the angle of internal friction. The reduction of the angle of internal friction explains the general trends observed in the laboratory, that on one hand the compressive strength decreases with increasing bitumen content, but on the other hand, the tensile strength increases up to an optimum. A full-scale experiment was carried out using an accelerated testing of foamed bitumen pavements at the Canterbury Accelerated Pavement Testing Indoor Facility (CAPTIF). In the full-scale experiments, the same materials that were tested in the laboratory (aggregates, bitumen, cement) were used to construct six different pavement sections, each with different contents of bitumen and cement. Three were constructed using foamed bitumen contents of 1.2%, 1.4% and 2.8% respectively, plus a common active filler content of 1.0% cement. Two more pavements were constructed adding cement only (1.0%), and foamed bitumen only (2.2%). In addition, one control section with the untreated unbound material was tested. Strains were collected using a 3D Emu soil strain system installed in each pavement section. The curing time between construction and pavement loading was approximately three months. The pavement response, such as surface deformation (rutting), surface deflections and strains were periodically recorded during the execution of the test. The strains were collected at different depths by using an array of Emu strain gauges. Deflections were recorded using both a Falling Weight Deflectometer (FWD) and CAPTIF Beam deflectometer, which is a modified Benkelmann beam. A total number of approximately 5.6 million equivalent standard axles were applied on the pavement sections. The rutting measured in the sections stabilised with foamed bitumen and cement was the lowest, showing that the addition of foamed bitumen significantly improved the performance of materials with 1% cement. The sections stabilised with cement only, foamed bitumen only, and the control untreated section showed large amounts of rutting and heaving by the end of the test. Deflection measurements showed that the effect of foamed bitumen content is a reduction of pavement deflections, with the lowest deflection measured in the section stabilised with 2.8% bitumen and 1% cement. The elastic pavement strains showed that foamed bitumen reduced the tensile strains in the basecourse but did not have a significant effect on vertical compressive strains. During the construction of pavements, material samples were taken for ITS and RLT testing. Results indicated that the highest ITS was measured in the section with 2.8% foamed bitumen content and 1% cement, and the ITS in the section without cement and foamed bitumen only was about 4-5 times lower than the ITS measured in specimens with cement. RLT specimens without cement performed poorly in comparison with the specimens with 1% cement. The specimens with 1% cement showed higher permanent deformation with increase in the foamed bitumen content, supporting the results from the previous laboratory study. To interpret and relate the results observed in the laboratory and the field, stress path analysis was used, in which the stress ratio of the foamed bitumen layers was calculated at different depths. The analysis showed that foamed bitumen content decreases the maximum stress ratio, hence reducing the proximity to failure and relative damage of the layer. Three-dimensional and two-dimensional finite element modelling of the CAPTIF pavements, were used to further investigate the stress and strain fields induced by the loading and to explain the pavement performance observed in the full-scale experiment