Dissertation (PhD)--University of Stellenbosch, 2000. / ENGLISH ABSTRACT: The use of foamed bitumen and bitumen emulsion as binders for use in road rehabilitation is gaining favour globally. High-level road facilities through to unpaved roads requiring attention are being treated with these binders due to environmental, economic and practical benefits in the use of cold bituminous mixes. In addition, static and mobile plant with the capability of performing stabilisation using bitumenemulsion and foamed bitumen has become commercially available and widely utilised, as a result of development in recycling technology and lapse in patent rights on foam nozzles. An understanding of the behaviour and failure mechanisms of these cold mixes, as well as sound guidelines for the mix design procedures of cold mixes, especially foamed bitumen, and design guidelines for pavements structures incorporating these materials, are lacking however. Mix designs are carried out primarily on the basis of experience and pavement designs are empirically based. The main objective of this dissertation is to address the need for a fundamental understanding of foamed bitumen and foamed bitumen mixes, and in so doing to develop techniques for adjudicating mixes, optimising their composition and rationalising their design both as mixes and as layers in road pavements. At the same time the exploration of new applications for foamed bitumen and the possibilities for progressive related technology, is a priority. To commence, this study includes an appraisal of most of the literature available on foamed bitumen. This is followed by a fundamental investigation of the colloidal mass of foam that is produced when small quantities of cold water are added to hot bitumen. Factors influencing the quality of the foam are identified and a Foam Index is developed for characterisation and optimisation of the foam. The spatial composition of a cold foamed bitumen mix, including Interaction of the foam with moist mineral aggregate, is also addressed in this dissertation. In particular, the stiffening of the filler mastic using foamed bitumen as binder is analysed. Techniques of optimising the sand type and content in the mix are also developed and guidelines for desired aggregate structures for foam treatment are established. The temperature of the mineral aggregate has been shown to have a profound influence on the behaviour and performance of a foamed bitumen mix. This has been selected as a focal area of further investigation and the research has lead to the development of a new process called "The half-warm foamed bitumen treatment process" that can produce mixes with almost the quality of hot mix asphalt with up to 40% less energy consumption. Other processes developed in this research include the use of cold mix asphaltic blocks for construction of road pavements in developing areas. This technology enhances the use of a high labour component in road construction in an economically competitive manner. The dissertation provides details for mix design and construction of the cold mix blocks. Finally, the study includes models for the performance prediction of foamed bitumen mixes. In particular, foamed mixes that exhibit stress-dependent behaviour have been investigated and models established on the basis of triaxial testing and accelerated pavement testing for the prediction of permanent deformation of such layers under repeated loading. Practical applications of the research findings are summarised in Appendix F. This includes: • methods for optimisation of the foamed bitumen properties, • guidelines for the selection of the ideal aggregate structure for cold foamed mixes, • procedures for carrying out cold mix design in the laboratory (including mixing, compaction and curing), • procedures for manufacturing half-warm foamed mixes in the laboratory, • methods for manufacturing cold mix blocks, and • pavement design methods for road structures incorporating foamed mix layers. Appendix G outlines statistical techniques that are relevant to the design of experiments in pavement engineering including examples of applications of these procedures. The techniques are applied selectively in the relevant chapters of the dissertation. / AFRIKAANSE OPSOMMING: Die gebruik van skuim bitumen en bitumen emulsie as bindmidel in pad rehabilitasie begin groter voorkeur wêreldwyd geniet. Van hoë vlak padfasiliteite tot ongeplaveide paaie wat aandag benodig, word met hierdie binders behandel vanweë die omgewings-, ekonomiese en praktiese voordele wat hierdie koue bitumen mengsels inhou. Voorts is statiese en mobiele masjinerie wat die vermoë het om stabilisasie in die gebruik van bitumenemulsie en skuimbitumen te bewerkstellig, in die handel verkrygbaar. Dit word algemeen gebruik as 'n uitvloeisel van ontwikkelings in herwinningstegnologie en die verslapping van patenteregte op skuim sproeikoppe. 'n Leemte bestaan in die begrip van die gedrags- en swigtingsmeganismes van hierdie koue mengsels, asook goeie riglyne vir die mengselontwerp van koue mengsels en in besonder skuimbitumen, en ontwerpriglyne vir plaveisel strukture waar hierdie materiaal geinkorporeer is. Mengselontwerpe word hoofsaaklik uitgevoer op grond van ondervinding, terwyl plaveiselontwerpe empiries gebaseer is. Die hoofdoel van die verhandeling is om die behoefte vir 'n fundamentele begrip van skuim bitumen en skuimbitumen mengsels aan te spreek, en daardeur tegnieke te ontwikkel vir die be-oordeling van mengsels, optimsering van hul samestelling en rasionalisering van hulontwerp vir beide mengsels en plaveisellae. Terselfdertyd is die ondersoek na nuwe toepassings van skuim bitumen en die moontlikhede van nuwe tegnologie 'n prioriteit. As beginpunt sluit hierdie studie 'n waardeskatting van die meeste literatuur beskikbaar op skuim bitumen in. Dit word gevolg deur 'n basiese ondersoek na die kolloidale massa van skuim wat geproduseer word wanneer klein hoeveelhede koue water by warm bitumen gevoeg word. Faktore wat die gehalte van die skuim beinvloed word uitgewys en 'n Skuim Indeks is ontwikkel vir die karakterisering en optimisering van die skuim. Die ruimtelike samestelling van 'n koue bitumen mengsel, wat die interaksie van die skuim met vogtige minerale samevoegings (aggregate) insluit, word ook aangespreek. Besondere aandag word gewy aan die verharding van die vuiler mastiekgom wat gebruik word as bitumen binder. Tegnieke om die sandtipe en inhoud van die mengsel te optimiseer is ontwikkel en riglyne vir die verlangde samevoegingstrukture (aggregate strukture) vir skuimbehandeling is opgestel. Daar is bevind dat die temperatuur van die minerale aggregaat 'n duidelike invloed op die gedrag en verrigting van 'n skuimbitumen mengsel het. Dit is gekies as In fokuspunt vir verdere studie en die navorsingswerk daarop het gelei tot die ontwikkeling van In nuwe proses wat "Die half-warm skuim bitumen behandelings proses" genoem word. Hierdie proses produseer mengsels wat byna gelykstaande is aan die gehalte van warm gemengde asfalt, maar met tot 40% minder energie verbruik. Ander prosesse wat met dié navorsing ontwikkel is, sluit koue gemengde asfaltiese blokke in, wat gebruik word in die konstruksie van padplaveisel in ontwikkelende gebiede. Hierdie tegnologie bevorder die ekonomiese gebruik van 'n hoë arbeidskomponent in padkonstruksie. Die studie stel besonderhede vir die mengselontwerp en konstruksie vir koue gemengde blokke voor. Laastens sluit die studie modelle in vir die werkverrigtingsvoorspelling van skuimbitumen mengsels. Op basis van drie-assige proewe in die laboratorium en versnelde belasting van paaie is modelle ontwikkel vir skuim bitumen mengsels wat spanningsafhanklike gedrag vertoon (die sogenaamde "granulêre" groep) om die permanente deformasie in die lae as gevolg van herhaalde verkeersbelasting te voorspel. 'n Opsomming van al die praktiese toepassings van die ontwikkelings van die navorsing word in Appendix F verskaf. Dit sluit in: • optimesering van die eienskappe van skuimbitumen, • riglyne vir gewenste samestelling van aggregaat in skuimbitumen mengsels, • ontwerp metodes vir koue skuimbitumen mengsels in die laboratorium (meng, verdigting en curing), • produksie metodes vir half-warme mensels in die laboratorium, • produksie metodes vir blokke wat met koue mengsels gemaak word, en • plaveiselontwerp metodes van strukture wat skuimbitumen lae inkorporeer. Appendix G gee 'n oorsig van statistiese tegnieke wat relevant is vir die ontwerp van eksperimente, insluitend voorbeelde van toepassings van die prosedures in plaveiselingenieurswese. Die tegnieke word selektief toegepas in die relevante hoofstukke van die verhandeling.

Thesis (MScEng (Civil Engineering))--Stellenbosch University, 2008. / There are currently well established compaction methods being used in laboratories globally to prepare specimens for material testing. None of these methods provides the repeatability and reproducibility, ease of execution or simulation and correlation to field compaction desired by engineers. The research presented in this report was aimed at the development of a new or adapted compaction method for bituminous stabilized materials (BSM) that would address the aforementioned factors, by making use of a vibratory hammer. Along with this, a new protocol was to be established. The initial vibratory hammer that was tested was the Kango 637®. This specific vibratory hammer suffered irreparable damage to the gearbox during the research. A replacement Kango hammer could not be purchased, therefore a substitute hammer was purchased i.e. a Bosch GSH 11E®, for which back-up service and replacement parts are readily available throughout South Africa. Significant progress had been made with the development of a laboratory compaction protocol for BSM using the Kango Hammer. The specifications of the Bosch® hammer showed it was superior in terms of power, weight and other technical features. Comparative testing was therefore carried out. This allowed for the adaptation of the results achieved to that point. Extensive experimentation was then carried out using two types of BSM i.e. foamed bitumen (80/100 bitumen) and bitumen emulsion (60/40 Anionic Stable Grade) stabilized material. The initial material used for the experimentation was a G2 quality graded crushed stone. Additional material was also obtained from a recycling project taking place along the N7 near Cape Town. The N7 material was used to perform correlation experiments so as to determine how representative the laboratory compacted specimens were to field compacted material. Results showed that the vibratory hammer is capable of producing specimens for testing in the laboratory as well as providing a possible benchmark method for accurately controlling the quality of work on site i.e. field density control. This was done by identifying the time to and level of refusal density compaction. The level of refusal density compaction was expressed as a percentage of Mod AASHTO compaction and using current specifications, a potentially new site compaction level specification was determined. In order to asses the material applicability of the vibratory hammer compaction method, tests regarding moisture sensitivity analysis were carried out on a G5 material. The vibratory compaction protocol includes a specification for the type of hammer, guide-frame, surcharge weight, compaction moisture and number of layers. Vibratory compaction can be used to prepare two types of specimens: • Specimens for triaxial testing with a diameter of 150mm and a height of 300mm • Specimens for laboratory testing with a diameter of 150mm and a height of 125mm. Tests showed that the material properties prove to have an influence on the compactability of the material. Material from the N7 recycling project had been milled out thus altering the grading and including some RAP. This in turn influenced compaction. The vibratory hammer moisture curve was found to shift slightly to the left when compared to the Mod AASHTO moisture curve. The variability of the vibratory hammer was found to be well below the specified variability of 15%. Repeatability experiments on G5 material indicate that vibratory hammer compaction may be used on lesser quality granular materials. A recommended procedure for the compaction of BSM was developed following the experimentation results.

Thesis (MScEng (Civil Engineering))--University of Stellenbosch, 2009. / The need for a more reliable testing procedure for the characterisation and Quality Assurance/ Control of Bitumen Stabilised Materials (BSMs), besides UCS and ITS testing, has long been recognised by the roads industry. In fact, at CAPSA 2004 and CAPSA 2007, discussions of improved test methods for granular materials, i.e. possible replacement tests for CBR procedures, were conducted in workshops. Triaxial testing for the evaluation of shear parameters is widely recognised as a reliable method of measuring these critical performance properties of granular and Bitumen Stabilised Materials (BSMs). However, the triaxial test in its current state as a research test has little chance of extensive use by practitioners and commercial laboratories, because of complexity, cost and time issues. Major adaptations to the research triaxial test are necessary, before this useful test can have a chance of being accepted by road practitioners. The main aim of this study is to investigate possibilities of developing a simple, affordable, reliable and robust test for characterizing granular and bitumen stabilized materials thus linking test outcome with in-situ performance. This is achieved through the innovative design and manufacture of a prototype triaxial cell capable of accommodating 150 mm diameter by 300 mm deep specimens. The cell is simpler than the research (geotechnical) triaxial cell and the operational protocols have been streamlined, thereby reducing the time and steps required in assembling specimens and testing them. In order to ensure the development of an appropriate triaxial cell for industry, a survey was conducted aimed at investigating currently available facilities, testing capacity and resources within civil engineering laboratories in South Africa. Findings of the survey (Appendix 4) have provided guidance with regard to the nature and sophistication of any new tests to be developed. The survey highlighted some of the limitations and lack of sophistication of the current loading frames used for CBR and UCS testing such as lack of electronic LVDTs, limited overhead space, limited loading capacity and others. Most laboratories would need to invest in new loading facilities to carry out triaxial tests. A review of the test procedure for monotonic triaxial test showed that two main factors contribute to the complexity of the research (geotechnical) triaxial cell namely, time taken to assemble the specimen accurately in the cell and secondly the inherent design of the cell which makes it water and/or air tight at relatively high pressures. The design of the Simple Triaxial Test, therefore, was aimed at overcoming the drawbacks of research triaxial test e.g. fitting a membrane to each specimen to be tested, through considerable simplification by means of a new structure design and procedure of assembly of specimen into the cell. The advantage of addressing these issues would be reduction in the number of steps required in the test procedure and therefore reduction in testing time. The design of the cell particularly was preceded by a conceptualization process that involved investigation of numerous options. Concepts such as the bottle, encapsulated-tube, bottle and sandwich concepts were considered and given reality checks. In addition, available triaxial procedures of a similar nature e.g. Texas Triaxial, were evaluated and analyzed. Ultimately, with some trials and innovation, a design was developed for a simple triaxial cell comprising a steel casing with a latex tube which is then introduced around the specimen sitting on a base plate. It is based on the ‘tube concept’ in which the specimen acts like a ‘rim’ and the cell acts like a ‘tyre’ providing confinement to the triaxial specimens for testing, within the tube. This approach eliminates the use of O-rings and membranes for the specimen and tie-rods for the triaxial cell, thus reducing testing time considerably. The overall dimensions of the cell are 244 mm diameter by 372 mm height (Appendix 5). The cell was manufactured at Stellenbosch University Civil Engineering workshop and preliminary tests were conducted under this study. Parallel tests were also conducted with the Research Triaxial Test setup at Stellenbosch University in order to determine if preliminary results obtained with the Simple Triaxial Test setup were comparable therefore providing a means of validating the data. Results of analysis of variance (ANOVA) show that variability between Simple Triaxial Test (STT) and Research Triaxial Test (RTT) results is less significant whilst that within samples of STT and RTT results is quite significant. Comparisons also show that good correlation were obtained from Reclaimed Asphalt Pavement (RAP) Hornfels + 3.3 % Emulsion + 0 % Cement mix and mixes with the G2 base course aggregate whilst completely different correlation was obtained from RAP + 3.3 % Emulsion + 1% Cement. It is evident however that the differences observed stem from material variability i.e. random variability to one degree or the other and not to the STT apparatus. It is recommended for future research that more STT versus RTT testing be done especially on a mix with known mechanical properties when compacted to a specified dry density, e.g. graded crushed stone (G1) compacted to 100% mod. AASHTO. In summary, a locally made, low cost, relatively durable triaxial cell with relatively easy and quick specimen assembly procedures has been developed. It is now possible to perform triaxial tests on 150 mm diameter by 300 mm high specimen relatively easily and quickly. However, the challenge of validating results obtained, as well as improving the manufacture process of its main component, the tube, still remains.

Thesis (MEng)--University of Stellenbosch, 2001. / ENGLISH ABSTRACT: The use of polymer modified binders have increased drastically over the past twenty years. It is generally accepted that current empirical specification tests do not adequately characterise the improvement in binder properties when bitumen is modified with polymers. At the University of Stellenbosch, a programme was initiated to relate binder properties to asphalt mix performance. This study forms part of that programme. This study focuses on three binders and is an attempt to characterise binder properties fundamentally. The binders that were tested are: • 60/70 penetration grade bitumen • SBS modified 60/70 bitumen • Bitumen-Rubber Two rheometers were used to characterise the binder properties: • Sliding Plate Rheometer (SPR) • Dynamic Shear Rheometer (DSR) The Sliding Plate Rheometer was modified to allow testing at elevated temperatures. The modification is described in this thesis. Three types of testing were used to test the binders: • Creep tests (SPR and DSR) • Dynamic tests (Strain and Frequency Sweeps - DSR) • Flow tests (high temperature viscosity tests - DSR) The results of the tests were used to characterise differences between the binders. Shear susceptibility lines were plotted and viscosities calculated from the creep test results; Master Curves and Black Diagrams were plotted from frequency sweep results; flow test results were used to characterise high temperature behaviour. A further development in the asphalt industry is the move towards the development of performance based specifications. A programme has been initiated in this regard and this study forms part of the programme. As a step towards the development of performance based specifications filler was included in this study. BinderlFiller systems were subjected to the same test programme the binders were, to characterise the influence on binder properties of filler addition. Some of the conclusions that were drawn at the end of the study are: • Polymer modification improves binder properties related to rutting. • Filler addition does not affect binder properties negatively. • Shear rate is very important during production of asphalt mixes and construction of asphalt layers containing polymer modified binders. / AFRIKAANSE OPSOMMING: Die gebruik van polimeer gemodifiseerde bitumen het drasties toegeneem oor die afgelope twintig jaar. Dit word algemeen aanvaar dat die huidige empiriese spesifikasie toetse nie die verbetering in bindmiddeleienskappe van polimeer modifisering aanvaarbaar karakteriseer nie. 'n Program is geïnisieer by die Universiteit van Stellenbosch om 'n verband te vind tussen bindmiddeleienskappe en asfaltmengseleienskappe. Hierdie studie vorm deel van die program. Hierdie werk fokus op drie bindmiddels en is 'n poging om die bindmiddeleienskappe fundamenteel te karakteriseer. Die bindmiddels wat gebruik is in hierdie werk, is: • 60/70 penetrasiegraad bitumen • SBS gemodifiseerde 60/70 bittimen • Bitumen-Rubber Twee reometers is gebruik om bindmiddeleienskappe te karakteriseer: • Glyplaat Reometer (SPR) • Dinamiese SkuifReometer (DSR) Die glyplaat reometer is aangepas om toetse by hoër temperature te doen en die aanpassing wat gemaak is word in die tesis beskryf. Drie tipes toetse is uitgevoer om die bindmiddels mee te toets: • Kruiptoetse (SPR en DSR) • Dinamiese toetse (DSR) • Vloei toetse (DSR) Die resultate van die toetse is gebruik om die verksille tussen die bindmiddels te kwalifiseer. Verskillende grafieke en figure is getrek van die resultate om die onderskeie bindmiddels te karakteriseer. Die beweging weg van empmese spesifikasies na werkverrigting gebaseerde spesifikasies in die asfalt industrie is 'n verder nuwe ontwikkeling. 'n Program is geïnisïeer ten opsigte hiervan en hierdie werk vorm deel van die program. As 'n stap in die ontwikkeling van die werkverrigting gebaseerde spesifikasies is 'n vuIler ingesluit in die toetsprogram. Bindmiddel/vuIler stelsels is onderwerp aan dieselfde toetse waaraan die bindmiddels onderwerp is om die invloed van die vuIler op die bindmiddels te karakteriseer. Sommige van die gevolgtrekkings waartoe gekom is aan die einde van die tesis, is: • Polimeer modifisering verbeter bindmiddeleienskappe met betrekking tot wielsporing. • Die toevoeging van vuIler affekteer nie die bindmiddeleienskappe negatiefnie. • Die vervormingstempo is van groot belang wanneer asfaltmengsels gemeng en lae gebou word van mengsels wat polimeer gemodifiseerde bitumen bevat.

Teve como objetivo verificar e comparar a alteração das características do carvão antracitoso e do carvão betuminoso utilizados em meio filtrante, devido ao efeito abrasivo causado entre os grãos decorrentes de lavagens exclusivamente com água e lavagens independentes com ar e água. Foram estudados dois tipos de carvão antracitoso e um tipo de carvão betuminoso, para os quais foram determinadas suas características químicas e físicas. Os ensaios foram realizados em uma instalação ao piloto montada na escola de engenharia de São Carlos-EESC, USP, na qual foram feitas simulações de lavagens admitindo-se uma lavagem diária durante um ano para cada método, de modo intermitente, em que o meio granular era compactado ao final de cada lavagem, e continuo, sem compactação do meio granular. A velocidade ascensional utilizada para cada simulação de lavagem foi de 0,9 m/min, promovendo expansão no meio granular de 35 a 40%. A taxa de ar utilizada para as simulações de lavagens com aplicação de ar foi de 15 l/sm2. Após o término das 365 lavagens, foram determinados os valores médios do coeficiente de esfericidade e da porosidade do meio granular estratificado e construída a curva de distribuição granulométrica das amostras utilizadas nos ensaios. Conclui-se que o carvão betuminoso apresentou comportamento semelhante ao dos carvões antracitosos e que a metodologia de lavagem continua produz praticamente o mesmo efeito abrasivo comparada a de lavagem intermitente. / This work had the objetive to verify and to compare the alterations on the properties of Anthracite coal and Betuminous coal, due to the abrasive effect caused between the grains resulted from washing exclusively with water and independent washing with air and water. There were studied two types of Anthracite coal and one type of Betuminous coal, their chemical and physical properties were determined. The tests were performed at a pilot instalation setted at the School of Engineering of São Carlos - EESC,USP; where, were done simulations of washing, it was admitted one washing a day during one year (365 washing) for each method, in an intermitent way, in which the granular medium was compacted at the end of each washing, and continous without compactation of the granular medium. The ascending speed used for each simulation of washing was 0,9 m/min, promoting an expansion in the granular medium of 35 to 40%. The rate of air used for the simulations of washing with the aplication of air was 15 I/sm2. After the last of the 365 washing, were determined the average values of the coefficient of sphericity and porosity of the granular estratified medium and was constructed a curve of the granulometric distribution of the samples used during the tests. It was conclued that Betuminous coal shows similar behaviour as Anthracite coaI and the continous method produce the same effect abrasive compared intermitent method.

As the most important materials in civil engineering, bituminous and cementitious materials have been used widely in pavements and constructions for many years. Accurate determination of adhesion is important to the bonding properties of bituminous and cementitious materials. In this work, we presented a novel approach to measure the adhesion between binders and aggregate mineral particles at microscopic scale. Particle probe scanning force microscopes (SFM) were used to study the adhesion between mineral microspheres representing the primary aggregate constituents (Al2O3, SiO2 and CaCO3) and various control (PG 64-22 and PG 58-22) and modified binders. Results showed that these modified SFM probes could effectively measure the adhesion between binders and aggregate minerals. Consistent adhesion measurements were obtained between different asphalt binders and aggregate mineral particles. Statistical analyses were performed to evaluate effects of different factors on the aggregate-modified binder adhesion, including aggregate constituents, binder types, modifier types and cantilever properties. Due to the stronger polarity of alumina particles, stronger interactions occur within alumina-binder pairs than within silica- and calcium carbonate-binder pairs. Meanwhile, morphologies of different modified binders clearly demonstrated microstructural variations in these binders. The adhesion between steel and different cement hydrated products was measured using particle probe SFM. Adhesive forces are collected between steel microspheres and new (four-week old) and old (six-month old) cement in air and saturated lime water. Mixed Gaussian models were applied to predict phase distributions in the cement paste, i.e., low density C-S-H, high density C-S-H, CH, other hydrated products and the unreacted components. For new cement in saturated lime water, adhesive forces between steel and low density C-S-H, high density C-S-H and other hydrated products are intermediate among all groups selected. The adhesive forces between steel and calcium hydroxide are smallest, whereas the adhesive forces between steel and the unreacted phases are largest. For the six-month old cement, the interweaving of calcium carbonate crystals and C-S-H during the carbonation produces greater adhesive forces to steel, consistent with the adhesive forces between steel and the control calcium carbonate specimen. CH turned into calcium carbonate by reacting with carbon dioxide in air. An increase in adhesive forces was found between steel and calcium carbonate in the old cement than those between steel and CH in the new cement. Particle probe SFM is able to measure the adhesion in bimaterials. For bituminous materials, this methodology provides opportunities to evaluate the effects of different processing methods and to generate quantitative information for the development of multi- scale asphalt mixture cracking models. For cementitious materials, these studies opened new avenues to study the interactions between steel and cement at microscale under a variety of environmental conditions and can be formulated as crack initiation and propagation criteria incorporated in multiscale models for reinforced concrete structures.

This doctoral thesis presents results of literature review on classical and contemporary aspects of stripping, as well as experimental investigations on moisture damage as influenced by bituminous materials. Previous research in the area of moisture damage was reviewed and synthesized into a state-of-the-art. Important parameters linked to moisture sensitivity, like bituminous material characteristics, dynamic loads from heavy vehicles, environmental factors, construction practice and nature of anti-stripping additives, are presented. The state-of-the-art in current test methods is summarized and given. The experimental work involved investigations of the influence of bitumen and aggregate composition on water susceptibility. The influence of aggregate mineralogy and chemistry was evaluated using eleven aggregates and one bitumen, followed by studying the interactive effect of four bitumens and four aggregates. Moisture sensitivity was evaluated in accordance with (EN 12697-12:2003) for conditioning, ASTM D 4123 for resilient modulus determination, and (EN 12697-23:2003) for indirect tensile strength testing. Furthermore, thermal stability of two liquid amine anti-stripping additives mixed with two bitumens of varied acidity was investigated using potentiometric titration and Fourier Transform Infrared (FTIR) Spectroscopy. Lastly, a technique based on Fourier Transform Infrared Spectroscopy-Attenuated Total Reflectance (FTIR-ATR) was developed and used for studying transport of water across thin bitumen films, as well as stripping at bitumen/substrate interfaces. Bitumens from different sources and three substrates (silicon, germanium and zinc selenide crystals) were used. Based on the results obtained, it was concluded that aggregates containing sodium and potassium in alkali feldspars generally showed high moisture sensitivity. In contrast, presence of calcium, magnesium and iron was associated with aggregates with low moisture sensitivity. Contrary to several previous findings, one aggregate with practically 100% quartz exhibited low moisture sensitivity. No linear relationship between moisture sensitivity and the contents of SiO2 and Al2O3 in the aggregates studied was established. Results of the interactive influence of bitumen and aggregate composition showed that high acid and low penetration bitumens exhibited high dry strength for all the aggregates studied. On the other hand, for a given bitumen, the wet strengths were found to be aggregate specific. The results of tests on thermal stability of amine additives showed that usefulness of these additives reduces considerably, when the more alkaline additive was mixed with the high acid bitumen, followed by storing the blends under pronounced conditions of time and temperature (24 hours and 140ºC, or more, in this study). Much less interaction occurred when the less alkaline additive was blended with the low acid bitumen. Even if a correlation was found between the results of potentiometric titration and Fourier Transform Infrared spectroscopy, the latter was not considered good enough at detecting amine additives, especially at low dosages. The technique based on FTIR-ATR developed in this study distinguished between good and bad bitumens with regard to stripping. The effectiveness of amine-based additives in reducing stripping was also shown by the method. Three likely processes occurred during the test, namely water diffusion, film break, and displacement (stripping) of bitumen from the substrate surface. The results also indicated that the diffusion process of water into the bitumen/substrate interface does not obey Fick’s law. / QC 20100831

This thesis presents a continuum model for asphalt concrete incorporating non- linear viscoelasticity, viscoplasticity, mechanically-induced damage and moisture- induced damage. The Schapery single-integral viscoelastic model describes the nonlinear viscoelastic response. The viscoplastic model of Perzyna models the time- dependent permanent deformations, using a Drucker-Prager yield surface which is modified to depend on the third deviatoric stress invariant to include more complex dependence on state of stress. Mechanically-induced damage is modeled using continuum damage mechanics, using the same modified Drucker-Prager law to determine damage onset and growth. A novel moisture damage model is proposed, modeling moisture-induced damage using continuum damage mechanics; adhesive moisture- induced damage to the asphalt mastic-aggregate bond and moisture-induced cohesive damage to the asphalt mastic itself are treated separately. The analytical model is implemented numerically for three-dimensional and plane strain finite element analyses, and a series of simulations is presented to show the performance of the model and its implementation. Sensitivity studies are conducted for all model parameters and results due to various simulations corresponding to laboratory tests are presented. In addition to the continuum model, results are presented for a micromechanical model using the nonlinear-viscoelastic-viscoplastic-damage model for asphalt mastic and a linear elastic model for aggregates. Initial results are encouraging, showing the strength and stiffness of the mix as well as the failure mode varying with moisture loading. These initial results are provided as a an example of the model's robustness and suitability for modeling asphalt concrete at the mix scale.

Polymer modified bitumen is a binder obtained by the incorporation of various types of polymers in bitumen using mechanical mixing or chemical reactions. There are several factors affecting the properties of polymer modified bituminous composites such as / chemical composition of bitumen, kind of polymer and filler, compatibility of bitumen and polymer, amount of bitumen, polymer and filler, particle size of filler and process conditions. The main objective of this study is to determine the effects of polymer type and concentration on mechanical, thermal, properties and morphologies of bitumen based composites. It was also aimed to determine the effect of process temperature on mechanical and thermal properties of bituminous composites. Bituminous composites were prepared by using Brabender Plastic Coder, PLV 151. Mixing was made at two different temperatures (150 &ordm / C and 180 &ordm / C) at 60 rpm for 20 minutes. Three different kinds of polymer and four different polymer concentrations were used to understand the effect of polymer type and polymer concentration on bituminous composites properties. Low density polyethylene (LDPE), styrene-butadiene-styrene copolymer (SBS) and ethylene-vinyl-acetate (EVA) were chosen as polymer. The compositions were adjusted as the polymer volumes were equal to 5%, 10%, 20% and 50% of bitumen volume. According to the test results, addition of polymer increases the mechanical properties, reduces the melt flow index and thermal conductivity values of bituminous composites. Morphological analysis results show that, fibrillation occurs at tensile fractured surfaces of composites which contain LDPE and EVA when the polymer concentration reaches 20% of bitumen volume.

There are many different types of fillers used for bitumen modification such as / silica, limestone, basalt, mica, oyster shells. Filler gives rigidity, stiffness or hardness, regulates thermal expansion and shrinkage, improves heat resistance, and modifies rheological properties of bituminous composites. The main objective of this study was to determine the effect of filler type and ratio on mechanical, thermal properties and morphologies of bitumen based composites. It was also aimed to improve the heat resistivity of the bituminous composite to obtain a material with good mechanical and heat isolation properties. Bituminous composites were prepared by using Brabender Plasti-Corder, PLV 151. Mixing was made at 180 &ordm / C with 60 rpm for 15 minutes. Two grades of bitumen as 20/30 and 50/70 penetrations were used. CaCO3, CaO, mica, baryte, kieselguhr and silaned kieselguhr were used as fillers in this study. Ethylene vinyl acetate copolymer, styrene&amp / #8211 / butadiene rubber, and styrene&amp / #8211 / butadiene&amp / #8211 / styrene block copolymer were used as polymers. According to the test results, using mica at low percentages had the effect of decreasing the viscosity of the bitumen due to its flow alignment property. Baryte gave high heat capacity and low heat conductivity to bituminous material. EVA containing samples showed the best combination on mechanical properties. The silanation process decreased the pore sizes as observed in mercury porosimetry experiments. A decreased amount of bitumen impregnation was obtained by the silanation process, clearly observed in SEM micrographs.