Page 65 - MODELING OF ASPHALT CONCRETE

Modeling of Asphalt Binder Rheology and Its Application to Modified Binders 43

stability and composition characteristics that are not typical of unmodified asphalts and
thus require special testing.
In this section a new approach to test performance related properties in the linear and
nonlinear domains is described. The approach is based on the concept of characterizing
resistance to damage as a better alternative to link lab testing results to pavement
performance.
To understand the causes of pavement distresses and study their effects on the
pavement performance has been the goal of asphalt researchers attentions for several
decades (Kim et al. 1997; Majidzadeh et al. 1972; Dijk 1975; Dijk and Visser 1977;
Moavenzadeh et al. 1974; Monismith and Deacon 1969). Among these distresses, rutting
and fatigue, which are recognized as associated with the increasing traffic volume, have
led to an apparent reduction in the long-term performance of flexible pavement. To
predict these damages efficiently, many research projects have been designed to evaluate
the effects of different factors on the performance of the asphalt pavement. Although it
is recognized that these distresses are mainly caused by the deformation and/or damage
within the asphalt binders, very few studies have used binder testing to evaluate
damage behaviors of binders under simulated testing conditions (Bahia et al. 2001).
While it is recognized that mixture factors and pavement structure factors can have
important effects, efforts to understand damage behaviors are very limited.
∗
In the existing Superpave specifications, the parameter G sind is used to rate the
∗
binder contribution to fatigue damage resistance, while G /sind is used to evaluate the
rutting damage resistance (Bahia and Anderson 1995). Both parameters were selected
based on the dissipated energy concept as applied to linear viscoelastic range. However,
there is a significant lack of information about the role of binder composition or
rheological properties of binders in damage progression under cyclic loading.
As part of the NCHRP 9-10 project (Bahia et al. 2001) two sources (gravel and crushed
limestone) and two gradations (12.5 mm coarse and 12.5 mm fine) of aggregates were
used to study the relationship between mixture damage behavior and the Superpave
binder parameters. One asphalt binder content was used throughout the testing. Nine
modified asphalts were included in the study. All of these asphalts were modified from
one base asphalt.
Five asphalts were modified with elastomers: ethylene-propylene diene terpolymer
(Ethylene terpoly), styrene-butadiene-styrene (SBS) radial, styrene-butadiene diblock
(SB), styrene-butadiene-styrene (SBS) linear, and styrene-butadiene rubber (SBR). One
asphalt was modified using stabilized polyethylene (PE) and three oxidized asphalts
were also included in this study. The oxidized asphalts were produced by steam
distilled, oxidized by back blending (BB) and oxidized by straight run (SR). All of the
asphalts were aged using an RTFO prior to conducting the binder tests.
To study the effect of modified asphalts on the rutting behavior of asphalt mixtures
the Repeated Shear Constant Height (RSST-CH) test (Chap. 10) was used. In order to
evaluate the effect of binder modification on the results of the RSCH test results, certain
mixture behavior indicators are needed. These indicators are commonly derived from
the typical power law model recommended for representing rutting by the SHRP.
The model, defined in Eq. (2-7), includes an initial strain factor (e ) and a slope (S)
p(1)
factor.
log ε = log ε + S log N (2-7)
p p() 1
where e is the total accumulated permanent strain and N is the number of cycles.
p(1)