Thesis (MScEng)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: This research investigates the flexibility and the performance properties of bitumen stabilised materials under the influence of mix variables. The laboratory testing consisted of two main phases. During the first phase (mix design), the strength and the flexibility of the mixes were assessed through ITS (Indirect Tensile Strength), UCS (Unconfined Compressive Strength), displacement at break, strain at break and fracture energy. The second phase consisted of a series of triaxial tests done to assess the performance properties (shear strength: cohesion and angle of internal friction; and stiffness: resilient modulus) of the mixes. The mineral aggregates used in this study were milled from different locations of the R35, near Bethal. This was a blend of granular material (dolerite, from various locations of the existing base and subbase layer of the R35) and Reclaimed Asphalt (RA) milled from the existing surfacing. During the mix design phase, two types of bituminous binders were used (bitumen emulsion and foamed bitumen) at bitumen content ranges of 2%, 2.4% and 2.8% each. Two types of active filler were used separately and in combination at a proportion of 1% and 2%. Finally, specimens were tested in wet and dry conditions for each mix combination. During the triaxial testing phase, only the optimum bitumen content of 2.4% was used, both for bitumen emulsion and foamed bitumen, with only cement as active filler in a proportion 1% and 2%. The specimens were tested at different ranges of densities and saturation levels. The flexibility of the mix was assessed through the fracture energy, the strain and the displacement at break parameters. An analysis of variance (ANOVA) was conducted on the data to assess the significance of experimental variables on this property. This property was found to be very sensitive to bitumen and cement content added to the mix. When assessing the combined effect and the significance of the variables on the flexibility of the mixes, it was found that fracture energy is mostly influenced by the cement content, followed by the bitumen content, then the type of treatment and finally the testing condition. However, the level of significance was not in the same order for the other two parameters (displacement and strain at break). It was also found that the combined effect of some independent variables (cement content + testing condition, type of treatment + cement content + bitumen content) had a significant effect on the fracture energy and the strain at break respectively. From the ITS and UCS tests, an increase in strength was noticed with the increase of cement content. On the other hand, the increase in bitumen content led to a decrease in strength of the material. The statistical analysis on the ITS and UCS values show that the independent variable with the most significant effect on the ITS is the cement content, followed by the testing conditions, then the bitumen content and finally the type of treatment. The combined effect of cement content + bitumen content was found to be significant both for ITS and UCS. In the second phase triaxial tests were performed in order to evaluate the performance properties of the mixes. It was found that the increase of the active filler content significantly improves the shear strength of the material. It was also found that at a fixed cement content, specimens tested at low density and/or high level of saturation show low shear strength. The Mr-θ model was used to model the resilient modulus of the mixes and the model coefficients used to evaluate the effect of experimental variables on the resilient modulus. It was found that the resilient modulus of the mixes increases as the bulk stress increases. This confirms the stress dependent behaviour of bitumen stabilised materials. The analysis show that increasing the percentage of active fillers content results in a significant increase in the resilient modulus values. An increase in relative density also resulted in an increase in the resilient modulus of the mixes, while the opposite effect was observed with the increased of the saturation level. Besides the engineering properties and the mechanical test parameters, other parameters such as the Tensile Strength Ratio (TSR) was calculated in order to evaluate the moisture sensitivity of the mixes. Weakening due to moisture was found to be more predominant in the mixes with less active filler. In addition, bitumen emulsion mixes were found to have a better resistance to moisture weakening effects compared to foamed bitumen. In addition, a comparison between the rapid curing and the accelerated curing was done. Higher ITS and UCS results were obtained for specimens cured using long term curing compared to specimens cured using the accelerated curing method. In conclusion, flexibility is an important property of bitumen road construction material (bitumen stabilised material include) however, it is not an easy property to measure. Although, displacement/strain at break and fracture energy from ITS and UCS were able to give us some indications on the main factors governing the flexibility of bitumen stabilised materials (the bitumen and active filler content), more accurate and adequate tests are required to evaluate the parameter. / AFRIKAANSE OPSOMMING: Die buigsaamheid en gedragseienskappe van bitumen gestabiliseerde materiale was getoets om sodoende die invloed van verskeie mengselveranderlikes te evalueer. Die ondersoek het uit twee fases bestaan. Tydens die eerste fase (mengfase) is die sterkte en buigsaamheid deur middel van indirekte treksterkte toetse (ITS), onbegrensde druksterkte toetse (UCS), verplasing – en vervorming by breekpunt sowel as breek-energie toetse gedoen en ondersoek. Die tweede fase het bestaan uit ʼn reeks drie-assige triaksiaal toetse. Triaksiaaltoetse is uitgevoer om die gedragseienskappe soos die skuifsterkte, kohesie, hoek van interne wrywing, styfheid en weerstand modulus te ondersoek. Die gemaalde mineraal-aggregaat wat in hierdie ondersoek gebruik is, was verkry op verskeie areas van die R35, geleë naby Bethal. Die materiaal is ʼn mengsel van granulêre materiaal (van die bestaande kroonlaag en stutlaag van die pad) en herwonne asfalt (RA). Tydens die mengontwerp fase is twee tipes bitumen gebruik naamlik bitumenemulsie en skuimbitumen in hoeveelhede van 2%, 2.4% en 2.8%. Twee tipes aktiewe vulstof (hoeveelhede van onderskeidelik 1% en 2%) was saam met elk van die verskeie bitumen-hoeveelhede gebruik. Proefstukke van elk van hierdie mengsel kombinasies is onder beide nat en droë kondisies getoets. Tydens die tweede fase, is slegs die optimum binder inhoud (2.4%) gebruik vir beide emulsie- en skuimbitumen, gekombineer met 1% en 2% aktiewe vulstof. Proefstukke was getoets by ʼn reeks van verskillende digthede en versadigingvlakke. Die buigsaamheid was ondersoek deur middel van breek-energie, vervorming en die verplasing by breekpunt. ʼn Analise van variasie (ANOVA) is uitgevoer op die toetsdata om sodoende die te evalueer of die veranderlikes beduidend is ten opsigte van buigsaamheid. Daar is gevind dat die buigsaamheideienskap sensitief is vir beide bitumen en sement inhoud. Met assessering van die gekombineerde effek en betekenis van die veranderlikes op die buigsaamheid van die mengsels, is daar gevind dat die hoogste beduidende veranderlike t.o.v breek-energie die sement inhoud is, gevolg deur die bitumeninhoud, tipe behandeling en laastens die toetskondisie. Die orde van belangrikheid verskil vir die ander twee parameters (verplasing en vervorming by breekpunt). Daar is ook gevind dat die gekombineerde effek van sommige veranderlikes (sement inhoud en toets kondisie, tipe behandeling en sement inhoud tesame met bitumen inhoud) ook beduidend was t.o.v breek-energie en vervorming by breekpunt. Vanuit die ITS en UCS toetse was daar ʼn toename in sterkte waargeneem soos die sementinhoud toeneem. Aan die anderkant, het ʼn toename in bitumeninhoud ‘n afname in sterkte veroorsaak. Die statistiese analise van ITS en UCS resultate, toon dat die grootste beduidende onafhanklike t.o.v ITS waardes ook die sement inhoud was, gevolg deur toets kondisies die grootste effek, bitumen inhoud en die tipe behandeling. Die gekombineerde effek van sementinhoud en bitumeninhoud, was betekenisvol vir beide ITS en UCS. Drie-assige triaksiaaltoetse was uitgevoer om die gedragseienskappe van die mengsels te evalueer. Daar is gevind dat die toename in sement inhoud, die skuif sterkte van die materiaal grootliks verbeter. By ʼn konstante sementinhoud, wys toetsresultate van proefstukke wat getoets is by lae digthede en hoë vlakke van versadiging, lae skuif sterkte. Die Mr – θ model was gebruik om die veerkragsmodulus van die mengsels te moduleer en die modelkoëffisiënte is gebruik om die effek van eksperimentele veranderlikes op die weerstand modulus te evalueer. Met toename in die omhullende spanning is ‘n toename in die veerkragsmodulus waargeneem, wat bevestig dat die gedrag van bitumen gestabiliseerde materiale spannings afhanklik is. ʼn Toename in die sement en relatiewe digtheid het ʼn merkwaardige toename in die veerkragsmodulus tot gevolg gehad, terwyl die teenoorgestelde waargeneem is met toename in versadigingsvlakke. Buiten die ingenieurseienskap en meganiese toetsfaktore, is ander faktore (soos die trekspanning verhouding) bereken om die vogsensitiwiteit van die mengsels te evalueer. Mengsels met laer sement inhoud het groter verswakking ervaar met blootstelling aan water. Bitumenemulsie proefstukke toon beter weerstand teen water as skuimbitumen. Vergelyking tussen versnelde en korttermyn nabehandelingsprosedure van proefstukke, toon hoër ITS en UCS waardes vir die versnelde nabehandelingsprosedure prosedure. Buigsaamheid is ‘n belangrike eienskap van bitumen in padkonstruksie materiale (insluitend bitumen gestabiliseerde materiale), maar word moeilik gemeet. Alhoewel verplasing/vervorming by breekpunt en breek energie, bepaal vanaf ITS en UCS, ‘n indikasie toon van die hooffaktore (binder en sement) wat buigsaamheid van bitumen gestabiliseerde materiaal beïnvloed, word meer akkurate toetse benodig om die eienskap te ondersoek.