Long-Term Field Performance of Warm Mix Asphalt Technologies (2017)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

47 Based on field and laboratory results, the findings of the long-term field performance of WMA pavements as com- pared with the corresponding HMA pavements for transverse cracking, wheel-path longitudinal cracking, and rutting and moisture susceptibility are summarized as follows. Overall, the HMA and WMA pavements perform similarly for pavements ranging between 4 and 10 years old. Transverse Cracking The findings for transverse cracking are as follows: • Transverse cracks were found to initiate from the top sur- face of the pavement, but often overlapped with trans- verse cracks in the existing asphalt layer, indicating that transverse cracking could be a combination of thermal and reflective cracking. • Field transverse cracking is mostly seen in pavements 4 years old or older. Younger pavements show less transverse cracking. • The majority of the HMA and WMA pavements showed comparable transverse cracking in the field. • The short-term transverse cracking resistance of the three major WMA technologies (chemical additive, organic additive, and foaming) is similar. However, in the longer term, the effect of aging on organic WMA technology is more prominent. The chemical and foaming WMA pavements appear to have better resistance to transverse cracking than the organic WMA pavements in the long term. • The mixture’s fracture work density property obtained from IDT testing at the low temperature (14°F) is pro- posed as a significant determinant for transverse cracking. A higher fracture work density value correlates with less transverse cracking. • Other potential material property indicators (significant determinants) for transverse cracking are the mix E* (14°F), mix IDT strength (68°F), mix creep compliance (32°F), and mix vertical failure deformation (68°F) values. • Statistically based crack initiation and propagation models were developed for transverse cracking and are applicable to both WMA and HMA pavements. These models indicate that, in addition to the identified material properties, the pavement age, climate, and traffic also have a critical impact on the initiation and development of transverse cracking. Wheel-Path Longitudinal Cracking The findings for wheel-path longitudinal cracking are as follows: • Wheel-path longitudinal cracks were found to initiate from the surface of the pavement and may be indicative of top-down fatigue cracking. • In general, the field wheel-path longitudinal cracks start to develop when the pavements are 4 years old; more longitu- dinal cracking is seen in HMA and WMA pavements that are 6 years old or older. • Based on the distress survey results, the majority of the HMA and WMA pavements exhibited comparable wheel- path longitudinal cracking. • The three WMA technologies (chemical additive, organic additive, and foaming) have similar resistance to wheel-path longitudinal cracking in relatively short service periods. However, in the longer term, the effect of aging on organic WMA technology is more prominent. The chemical and foaming WMA pavements appear to have better resistance to longitudinal cracking than the organic WMA pavements in the long term. • The mixture’s IDT strength at 68°F is proposed as the sig- nificant determinant for wheel-path longitudinal cracking. A higher IDT strength value correlates with more wheel- path longitudinal cracking, within the range of the IDT strength values measured during this project. C H A P T E R 6 Findings

48 • Other potential material property indicators (significant determinants) for wheel-path longitudinal cracking are (a) the mix vertical failure deformation (68°F) and hori- zontal failure strain (68°F) from the IDT test and (b) the mix creep compliance (14°F) values. • Statistically based crack initiation and propagation models were developed for wheel-path longitudinal cracking and are applicable to both WMA and HMA pavements. These models indicate that, in addition to the identified material properties, the pavement age, climate, and traffic also have a critical impact on the initiation and development of wheel- path longitudinal cracking. Rutting and Moisture Susceptibility The findings for rutting and moisture susceptibility are as follows: • Based on the rut depth survey results, the majority of the HMA and WMA pavements shows comparable rutting. • There is no statistical difference among WMA technolo- gies (i.e., chemical additive, organic additive, and foaming) with respect to rutting. • Field rut depth starts to build up as early as 3 years in some pavements. There is a general trend that, for older pavements, the rut depth is higher. For pavements 6 years old or older, the accumulated rut depths are more differentiable (more than 0.1 inches), for both HMA and WMA pavements. • The RRI value is proposed as the significant determinant for rutting. Relatively high RRI values correlate with low rut depth values in the field. • Other potential material property indicators (significant determinants) for rut depth are mix E* (86°F), mix creep compliance (86°F), binder PG, and binder non-recoverable creep compliance values. • No moisture-related distress was found during the distress surveys for either HMA or WMA pavements. However, most of the mixes without an anti-stripping agent exhib- ited SIPs in the HWT tests, suggesting that the use of an anti-stripping agent may be beneficial for both HMA and WMA mixtures. • A statistically based rut depth model was developed for both WMA and HMA pavements. In addition to the iden- tified material properties, the pavement age, climate, and traffic also have a critical impact on the accumulation of rut depth in both WMA and HMA pavements. Proposed Best Practices for WMA Technologies and Future Research Based on the investigation into long-term field perfor- mance, the overall proposed best practices of WMA tech- nologies from this project are similar to the ones reported in previous NCHRP studies, such as NCHRP Projects 09-43, 09-47A and 09-53, as well as in the National Asphalt Pave- ment Association’s Warm Mix Asphalt Best Practice report (Prowell et al. 2012). The performance of WMA mixtures can be treated as similar to HMA mixtures, except as otherwise specified in those reports and herein. Additional proposed best practices and future research are as follows: Mix Design and Evaluation. The identified significant determinants for transverse cracking, wheel-path longitu- dinal cracking, and rutting can be used to characterize the long-term performance of WMA pavements. They should be validated for additional local climate, traffic, and pavement conditions. Thresholds of the specific properties can be estab- lished for material screening and potentially be integrated into a performance-based mix design procedure for both HMA and WMA mixtures. The developed predictive models for field per- formance allow evaluation of the material properties (signifi- cant determinants) in conjunction with the effect of pavement structure, traffic, and climate. If the proposed significant deter- minants are difficult to implement by local agencies, as an alter- native, other potential material property indicators suggested in this report may be considered and further evaluated. It should also be noted that the significant determinants identified in this project are mainly based on field cores that have experienced short-term and some long-term field aging. To implement those significant determinants in the mixture design and mixture evaluation, appropriate aging procedures (short term and long term) should be used on laboratory pre- pared specimens so that the mixture properties (significant determinants) can be representative of field conditions. Pavement Performance. The long-term performance of the WMA pavements is found to be comparable with their companion HMA pavements. They may be used as an alterna- tive to HMA pavements for paving projects given the appro- priate WMA application practice is followed. Based on limited field data, the short-term performance (transverse cracking, wheel-path longitudinal cracking, and rutting) of the three major WMA technologies (chemical additive, organic addi- tive, and foaming) is similar. However, in the longer term, the effect of aging on organic WMA technology is more promi- nent. The chemical and foaming WMA pavements appear to have better resistance to transverse cracking and wheel-path longitudinal cracking than the organic WMA pavements in the long term. This finding and the mechanisms need further investigation through additional field projects. Agencies that have particular concerns with incidences of transverse and wheel-path longitudinal cracking may take this finding into consideration when selecting a WMA technology. No moisture damage was found in either HMA or WMA pavements that were up to 10 years old. However, laboratory HWT test results show that the mixtures without an anti- stripping agent, either HMA or WMA, can have some stripping

49 concerns while most mixtures with an anti-stripping agent show no sign of SIP. In general, the use of an anti-stripping agent seems to be beneficial for both HMA and WMA mixtures. Pavement Design. The current Pavement ME Design program is limited in its capability to predict the long-term performance of WMA and HMA pavements, especially with regard to transverse and wheel-path longitudinal cracking. The statistical based models developed in this project may be useful in this regard. If the Pavement ME Design program is used for the per- formance prediction or pavement design of WMA pave- ments, it is noted that the predictive quality of the program for WMA is similar to HMA pavements, regardless of predic- tion results. The WMA technology does not fundamentally alter the behavior of asphalt materials in pavement applica- tion. The research team suggests treating WMA as similar to HMA when using the Pavement ME Design program for performance prediction. Note that the corresponding WMA material properties should be used as input values.