Page 62 - MODELING OF ASPHALT CONCRETE

40 Cha pte r T w o

due to mechanical working (repeated loading). In the field of asphalt emulsions,
technology is available to produce such thixotropic materials, such as high-float
emulsions. It is reasonable to assume that this same technology will find its way in the
production of paving grade asphalts.
The dependence of rheological response on loading rates and temperature is known
in many fields to be material specific. In the asphalt field, behavior of modified and
unmodified asphalts varies significantly in responding to change of traffic speed in the
field, or testing rate in the lab. Because of this variation, using common time-temperature
equivalency factors could result in significant errors in estimating the effect of traffic
speed on response of various asphalts. Figure 2-12 depicts that the relationship between
∗
effect of changing the loading rate and effect of changing temperature on G for typical
2
paving grade asphalts. The R values are 0.17 for the shift from 5 to 10 rad/s and 0.30
for the shift from 1 to 10 rad/s. This clearly indicates that there is hardly a correlation
∗
between the effects of loading time and temperature changes on changes in G values.
In other words, the practice of changing the temperature grade to compensate for traffic
speed cannot be justified. The polymer-modified asphalts appear to be significantly less
affected by a change in traffic speed (frequency) than the unmodified asphalts. Also, the
∗
effect on G of changing temperature by 6°C is highly variable, even for the unmodified
asphalts. For the asphalts examined, which include 32 sources tested during the SHRP,
the change ranged from 1.7 to 2.6 folds. This result is expected, since temperature
susceptibility is highly asphalt source specific. The simplification used in the current
specification is, therefore, not acceptable.
Regarding the assumption of homogeneity, the Superpave binder testing is done
using selected geometries for different temperatures and response types. These
geometries were selected to give stress fields that can be easily estimated and thus used
in calculating the material response. The concept is based on the assumption that
binders are homogeneous materials that exhibit isotropic behavior. Therefore, testing in
one mode of loading, using a single geometry, can give a comprehensive evaluation of
the material behavior under different geometric and loading conditions. For some
modified asphalts, however, the additive used results in an anisotropic binder and thus

FIGURE 2-12 Relationship between 6°C temperature change and loading rate.