J Richard Willis

Effect of rubber characteristics on asphalt binder properties

The use of scrap tyres in ground tyre rubber (GTR)-modified binders has continued to evolve since its introduction in the early 1960s. Currently, many states have developed recipe specifications requiring contractors to blend GTR with asphalt binders using specified rubber sizes, percentages, and grinding methods. While these specifications were developed based on early research, the GTR industry has developed new methods and techniques which might improve the quality of GTR in asphalt binders. The objective of this research was to assess how rubber properties affect the properties of an asphalt binder. This objective was completed by blending 12 unique GTR samples with a singular asphalt binder at a loading of 10% rubber. Two of the selected rubbers were additionally tested at 15% loading. These 14 GTR-modified asphalt binders were then tested using the performance grade (PG), multiple stress creep recovery, cigar tube separation test, and softening point methodologies. Statistical analyses were conducted to determine how particle size, grinding temperature, rubber chemistry, and surface area affected the four test results. GTR particle size was the most influential parameter on the majority of the test results. The smaller particle sizes improved the high- and low-temperature PG and particle separation as tested by cigar separation tubes and the softening point.

Full-Scale Structural Evaluation of Fatigue Characteristics in High Reclaimed Asphalt Pavement and Warm-Mix Asphalt

Warm-mix asphalt (WMA) and the inclusion of higher percentages of reclaimed asphalt pavement (RAP) are two strategies for developing environmentally friendly pavement structures that can also reduce pavement cost. Although demonstration projects have evaluated the constructability and short-term performance of these technologies in the field, questions remain about the structural characterization and longer-term field performance. As part of the 2009 National Center for Asphalt Technology test track, four sections were constructed; they included both WMA and high-RAP contents (50% RAP). A control section did not include either technology. Every section featured embedded instrumentation that facilitated strain versus temperature characterization under truck loading. Statistical comparisons of measured strain response were conducted at three reference temperatures. At the lowest temperature (50°F), the measured strain in the control section was not statistically different from the strains of the experimental sections. At the intermediate temperature (68°F), strains in the RAP with WMA section were statistically lower than the control, but the other sections were not distinguished from the control. At the highest temperature (110°F), the strain in the control section was statistically higher than all other sections. The WMA sections were next highest, and the RAP sections were the lowest. The base mixtures of each section underwent beam fatigue testing from which fatigue transfer functions were developed. Strains entered into their section-specific transfer functions showed that the RAP–WMA section would perform the best. Continued testing and monitoring of the sections are necessary to validate this finding.

Laboratory evaluation of high polymer plant-produced mixtures

Two high polymer mixtures (HPM) were placed at the 2009 National Center for Asphalt Technology (NCAT) Test Track as a field trial. These mixtures featured 7.5% polymer in contrast to the more typical 2–3% polymer contents. During construction, each mixture (one surface and one base mixture) was sampled for evaluation using laboratory performance tests. The results of the dynamic modulus test, asphalt pavement analyzer, flow number, bending beam fatigue, indirect tension creep compliance and strength test, energy ratio, and moisture susceptibility tests were compared with the results from comparable control mixtures placed in the same round of testing. The laboratory test results suggest HPM mixtures can be placed to develop more efficient (i.e., thinner) pavement cross-sections due to their enhanced fatigue and rutting resistance.

Performance of a Highly Polymer-Modified Asphalt Binder Test Section at the National Center for Asphalt Technology Pavement Test Track

Polymer modification of asphalt binders is commonly used to improve either the rutting or the cracking performance of asphalt mixtures. In most cases, 2% to 3% polymer is added to the binder; however, highly polymer-modified (HPM) mixtures can use 7% to 8% polymer to create a more integrated polymer chain network, which improves rutting and fatigue performance. In 2009, an HPM mixture test section was built at the National Center for Asphalt Technology Pavement Test Track; the test section was approximately 5.75 in. thick. A control test section was simultaneously built at 7 in. thick with the use of conventional paving materials. The objective of this study was to assess the long-term field performance of the HPM mixture and control sections placed at the test track. These sections were evaluated in regard to field performance and structural capacity. Falling weight deflectometer testing was conducted to determine in-place modulus over time, and weekly strain and pressure responses were collected to determine how the test sections behaved mechanistically under live trafficking. The performance of the test sections was monitored with the use of inertial profilers to measure rutting and ride quality, and visual inspection was used to quantify cracking. After 20 million equivalent single-axle loads of trafficking, the HPM mixture test section had less cracking and rutting than the control section and the ride quality had not diminished over time. These findings support the use of HPM mixes in agency paving programs where appropriate.

Fatigue Performance of Highly Modified Asphalt Mixtures in Laboratory and Field Environment

High levels of SBS polymer modification lead to a bituminous binder with improved resistance to both rutting and fatigue cracking. Beam fatigue and modeling predict that, using this binder, pavement thickness can be reduced and still achieve equal or superior long term performance. To test this, a section at the National Center for Asphalt Technology (NCAT) was paved using a binder with a nominal grade of PG 88-22 for all three lifts. This pavement was constructed in August 2009 at a thickness of 145 mm compared to a standard thickess of 180 mm for the control and other pooled-fund study sections. In August 2010 an unrelated section that experienced failure was rehabilitated with a similar structure. This paper reports beam fatigue analysis and modelling and compares the results with observed field performance over both structurally sound and weak subgrades. The beam fatigue data predicts a very high endurance limit for the mixtures. Although conclusions are premature, to date neither structure shows surface distress.

Effects of Changing Virgin Binder Grade and Content on High Reclaimed Asphalt Pavement Mixture Properties

Most highway agencies have decades of experience with hot-mix asphalt whose percentage of reclaimed asphalt pavement (RAP) has remained low to moderate because of the general perception that RAP mixtures may be more susceptible to various modes of cracking. As the RAP proportion increases, so does the potential for an increase in mixture stiffness and a decrease in resistance to cracking. Two proposed ways to increase the durability of RAP mixtures are to (a) increase the amount of virgin binder in the asphalt mixture and (b) decrease the performance grade of the virgin binder. To assess these options, 0%, 25%, and 50% RAP mixtures at optimum asphalt content were designed with a standard PG 67-22 virgin asphalt binder. These mixtures were tested to evaluate surface cracking, reflection cracking, and rutting with the use of an energy ratio, overlay tester, and asphalt pavement analyzer, respectively. These tests also were conducted on RAP mixtures with 0.25% and 0.50% higher asphalt contents and at the optimum asphalt content with the use of a softer virgin binder. In addition, the linear amplitude sweep methodology was used to assess the fatigue properties of the blended binders. The results showed that, to improve resistance to cracking, the amount of virgin asphalt should be increased by 0.1% for every 10% of RAP binder in the mixture for up to 30% RAP binder. Once the RAP binder exceeds 30%, a softer grade of asphalt should be used to increase the mixtures resistance to cracking. All mixtures should be assessed for rutting susceptibility.

Long term performance of a highly modified asphalt pavement and application to perpetual pavement design

This paper reports on the final surface characterization and full forensic analysis of a highly modified asphalt pavement versus conventional Hot Mix Asphalt (HMA) at the National Center for Asphalt Technology (NCAT). The highly modified asphalt pavement is 20% thinner than a series of companion sections. The sections were subjected to 20 million Equivalent Single Axle Loads (ESALs) over a period of 5 years. At the end of two full track cycles, the thinner highly modified section has outperformed the others in permanent deformation and bottom up fatigue cracking. Modeling results give quite reasonable agreement with observed performance. The performance results validate pavement modeling reported at the International Conference on Perpetual Pavements (ICPP) in October and demonstrate that substantial thickness reduction is possible while retaining and even improving long term performance. Equally important, material properties of highly modified mixes may be used to adjust the damage model calibration factors in the AASHTOWare/r/ Pavement ME Design software so that appropriate pavement thickness can be determined through rational design. This methodology has been put into practice and so far the performance results on commercial projects have validated the design predictions.

Adaptation and validation of stochastic limiting strain distribution and fatigue ratio concepts for perpetual pavement design

Traditional perpetual pavement thickness design is based, in part, on controlling strain levels at the bottom of the asphalt concrete layer below an endurance limit to prevent bottom-up fatigue cracking (FC). A field-based limiting strain threshold was developed from cumulative distributions of field-measured tensile strains in the 2003 and 2006 research cycles at the National Center for Asphalt Technology Pavement Test Track to understand the limiting strain necessary to control FC. Additionally, the fatigue ratio, the ratio of the nth percentile strain to the fatigue endurance limit, was developed. Both the tensile strain distributions and fatigue ratios showed a clear difference between sections that experienced bottom-up FC and those that did not. However, it is necessary to adapt these thresholds to strains predicted by perpetual pavement design tools. PerRoad, a stochastic perpetual pavement design programme, was used to predict strains for the same 2006 sections. Previously developed strain distrib...Traditional perpetual pavement thickness design is based, in part, on controlling strain levels at the bottom of the asphalt concrete layer below an endurance limit to prevent bottom-up fatigue cracking (FC). A field-based limiting strain threshold was developed from cumulative distributions of field-measured tensile strains in the 2003 and 2006 research cycles at the National Center for Asphalt Technology Pavement Test Track to understand the limiting strain necessary to control FC. Additionally, the fatigue ratio, the ratio of the nth percentile strain to the fatigue endurance limit, was developed. Both the tensile strain distributions and fatigue ratios showed a clear difference between sections that experienced bottom-up FC and those that did not. However, it is necessary to adapt these thresholds to strains predicted by perpetual pavement design tools. PerRoad, a stochastic perpetual pavement design programme, was used to predict strains for the same 2006 sections. Previously developed strain distributions and fatigue ratios were adjusted to reflect observed differences in predicted and measured strains. Cumulative distributions and fatigue ratios based on predicted strains for the 2009 research cycle validated the updated limiting strain distribution and maximum fatigue ratios for designing perpetual pavements to resist bottom-up FC.

Laboratory and Field Evaluation of Florida Mixtures at the 2012 National Center for Asphalt Technology Pavement Test Track

The Florida Department of Transportation recently conducted a study at the National Center for Asphalt Technology Pavement test track to study cracking potential in asphalt mixtures. While cracking was the primary objective, the Florida Department of Transportation also wanted to ensure that no rutting would occur. Four mixtures were placed in test sections 100 ft in length. The four mixtures included two virgin mixtures: one used a polymer-modified binder and the other used a hybrid ground tire rubber and polymer-modified binder. The two other mixtures used combinations of polymer-modified binder with reclaimed asphalt pavement (RAP) or recycled asphalt shingles (RAS) or both. Conventional laboratory testing for rutting and cracking was performed on each mixture. In addition to the laboratory testing, field performance was assessed for rutting, cracking, and ride quality. Laboratory correlations were developed with field performance; however, because of the rapid crack propagation in the RAP–RAS test section, the correlations between 5 and 10 million equivalent single-axle loads (ESALs) changed significantly. The best correlations were developed between the resilient modulus of the mixture and cracking at 5 million ESALs and also the dissipated creep strain energy of the mixtures and cracking at 10 million ESALs. Overall, none of the mixtures experienced severe cracking in 10 million ESALs of trafficking.

Evaluation of Nanotechnology Additive on Tack Coat Moisture Resistance and Bond Strength

This study evaluated the laboratory bond strength and moisture susceptibility of a nanotechnology (derived from organosilane) modified emulsion against a control CSS emulsion. This additive is designed to convert the surface of the treated material (untreated aggregate, soil, or HMA) from a water loving (hydrophilic) surface to a water repelling (hydrophobic) surface while providing equivalent bond strength compared to an unmodified emulsion with a higher residual AC content. Two-layer slabs were produced in the laboratory with the following variables: emulsion type, surface type (new HMA or milled), and application rate (three different rates). Cores were obtained from each slab to evaluate the effect of the critical variables using a monotonic shear bond strength test. The results of the bond strength evaluation showed the modified emulsion provided equivalent bond strength to a control emulsion with 3 times the residual AC content. Additionally, an experimental evaluation was conducted to assess the moisture susceptibility of the tack coat materials. Additional slabs were fabricated at the optimum application rate for each surface type and emulsion application. The AASHTO T283 procedure was utilized to condition these specimens with one or two freeze-thaw cycles prior to determining their bond strength. The results of this evaluation were mixed. The data suggested moisture damage was occurring in the new HMA surface specimens, but not in the specimens with a milled surface interface. For the new HMA specimens, the control and nanotechnology modified emulsions had equivalent resistance to moisture damage as freeze-thaw cycles were applied to the specimens.