Search

Executive Summary Cold Mix Asphalt (CMA) serves as a promising alternative for a wide range of paving applications from preventative maintenance and repair, to full-scale pavement construction. Compared to traditional Hot Mix Asphalt (HMA) and Warm Mix Asphalt (WMA), CMA provides a number of benefits that make it a promising alternative to conventional paving technologies, including reduction in heating energy and longer working time for transportation and placing. However, the reported higher air void content and higher moisture susceptibility of CMA relative to traditional HMA present legitimate concerns that has limited more widespread use of CMA to date. To mitigate the potential for moisture damage in CMA, it is necessary to include a simple and effective method for evaluating moisture susceptibility in the material selection phase of the mix design process. The most direct way to prevent moisture sensitivity is to require a minimum threshold of aggregate coating as part of CMA mix design in regards to both extent and quality of coating. This study refines a recently developed method to quantify aggregate coating extent in CMA using Digital Image Analysis Methods and applies the procedure to several aggregate-emulsion systems to isolate factors most directly affecting aggregate coating. After identifying the most significant factors, a regression analysis of the experimental results is used to develop quantitative models to predict coating of aggregates in CMA as a function of mix design parameters. Results show that not only Image Analysis Method serves as an effective and reliable tool in identifying aggregate coating, but also that the regression models developed to predict coating are sensitive to emulsion type, aggregate gradation and residual asphalt content. Including evaluation of coating quality is a generally accepted practice in mix design for conventional HMA, WMA, and CMA. This study introduces the concept that it is necessary to assess quality of coating as a means to ensure adequate bonding between the asphalt and aggregate to truly assess potential for moisture damage. To evaluate quality of coating, a Modified Boiling Test procedure on cured CMA loose mixture that involves application of the Image Analysis Method to quantify the coating before and after boiling was developed. Results reveal that after curing is achieved as indicated by negligible moisture loss in the mixture, the bonding between emulsion residue and aggregate continues to increases with curing time. Modified Boiling Test results on fully cured CMA loose mixture indicate that CSS-1hl performs much better than CSS-1h and CSS-2 emulsion, showing the best coating quality. The results were also verified by Binder Bond Strength (BBS) test on emulsion residues and asphalt emulsions. The relationship between coating quality and mixture performance was verified through comparison of boiling test results to mixture moisture damage as measured by the Tensile Strength Ratio (TSR). Prior to conducting this comparison a CMA mix design method was proposed to ensure a uniform process for selection of design emulsion content and preparation of mixture samples. The proposed CMA mix design method determines the design emulsion content (DEC) based on volumetric analysis and mixture indirect tensile strength. The Tensile Strength Ratio (TSR) test based on this mix design shows consistency with Modified Boiling test result and correlation between TSR test and Modified Boiling test is also established.

The main purpose of asphalt mixture production is the coating of mineral aggregates with asphalt binder within a reasonable time frame. Coating is required as it affects mixtures workability (e.g., lubricant effect of binder film), as well as durability (e.g., preventing mixtures from moisture damage). However, mixing time needs to be controlled since it greatly influences both plant productivity (i.e., amount of mixture produced per hour) and binder aging, which is caused by the loss in volatiles due to temperature and air flow during mixing. In light of this, mixing temperature is selected to properly reduce binder?s viscosity below a limit to ensure proper coating in a time frame of 90 seconds. Current viscosity limits used for selecting mixing and compaction temperatures are valid only for unmodified binders, which generally exhibit a linear Newtonian behavior; furthermore, the literature indicates that these limits were introduced solely based on practical considerations without a fundamental understanding of the physical and chemical processes driving the production of this material. A guideline for the selection of compaction temperatures for modified binders, which are nowadays very popular, is not available to practitioners. Mixing temperatures calculated based on current recommendations for viscosity limits are often higher than 150�C and they become even more excessive when modified binders are used. The usage of these temperatures raises concerns in terms of energy cost, sustainability and safety. Thus, transportation agencies and practitioners are currently challenged with the need of reducing the mixing temperatures during production without negatively affecting the level of coating or performance of mixtures. This can only be accomplished by providing a clear understanding of the main factors and variables that affect coating, and by showing the influence of coating on the susceptibility of asphalt mixtures to moisture damage. Reducing mixing temperatures will result in increasing viscosity. The effect of increased viscosity on both coating and the adhesive bond between the binder and aggregates has not been properly investigated in the past. In this thesis, effects of coating are considered based on two different points of view: (a) extent of coating, which is defined as the amount of dry aggregates coated by binder after mixing for a given time before compaction; and (b) the quality of coating which is defined as a qualitative description of the adhesive bond between binder and aggregate. Initially a procedure based on aggregates water absorption (ASTM C127 ? 07) was developed for determining the extent of coating. Results from this procedure showed that the extent of coating is affected by low shear viscosity and asphalt content, but highlighted that the extent of coating is not sufficient for assessing moisture susceptibility of mixtures. It was determined that moisture damage of mixtures is more dependent on the binder-aggregate bond strength rather than the extent of coating. For this reason, a modified boiling test procedure based on the ASTM D3625 ? 1996 was introduced and developed for evaluating the quality of coating. Results from the modified boiling test have shown that the quality of coating is highly affected by low shear viscosity, binder modification, and aggregate type. Furthermore, a time-dependency effect of the quality of coating was observed for certain combinations of asphalt binders and aggregates. The analysis of the boiling test results , which is based on visual inspection of aggregates surface post boiling, was validated using the acid ? base titration procedure recommended in the UNI EN 12697-11 2012 (E). In addition, the Bitumen Bond Strength (BBS) test (AASHTO TP 91-12), was included in the experimental study to evaluate if the application/production temperature affects the moisture susceptibility of the inderaggregates systems. Results showed that the modified boiling test and the BBS test are correlated. Furthermore, the boiling procedure was validated using the Indirect Tensile Strength (ITS) test, (AASHTO T 283). A preliminary correlation between the boiling test and the ITS showed that the proposed procedure is promising for the evaluation of the quality of coating and can be used as a surrogate for estimating the susceptibility of asphalt mixtures to moisture damage. The ITS results in combination with the analysis from the boiling test were used to provide preliminary design limits. Results from this thesis show that quantifying the extent of coating is not sufficient for predicting moisture susceptibility. Indeed, the quality of coating needs to be considered when assessing moisture damage of mixtures. Findings from the modified boiling test have pointed out that the quality of coating is affected not only by viscosity, but also by binder modification and aggregate type. In summary, the developed modified boiling test can be used as a simple procedure for selecting mixing temperature that will ensure sufficient extent and quality of coating, both of which have significant effect on asphalt-aggregate mixture moisture susceptibility.

Part of asphalt pavement production is the compaction process of loose mix. The quality of compaction process has a profound impact on the quality of the final product. With the advent of Warm Mix Asphalt (WMA) additive use, the compaction process became more efficient. In 2010 estimated 4 million tons of WMA asphalt were produced using the WMA additive technology (Hansen and Newcomb 2011). The additives allowed for improved compaction (requiring less compactive effort) and compaction at lower-than-typical temperatures. This thesis is focused on quantifying the improved mixture compaction through novel binder testing. The science of tribology has been used to understand and quantify the way certain WMA additives improve compaction process. It has been found that chemical WMA additives are capable of creating boundary lubrication films protecting aggregate particles from direct solid-solid contact, hence reducing friction between the particles. The new Asphalt Boundary Lubrication test was developed to quantify the boundary lubrication of binders in terms of boundary coefficient of friction. The test quantified the influence of aggregate mineralogy on the lubricating properties of certain WMA additives. It was also discovered that WMA modified binders improve boundary lubrication as compared with control binders. The binder asphalt boundary lubrication test results and mixture volumetric density data show good ranking. Additionally, at high testing temperature (145�C), boundary coefficient of friction provided explanation for changes in mastic viscosity, which can be well correlated with mixture volumetric data.