Page 56 - MODELING OF ASPHALT CONCRETE

34 Cha pte r T w o

∗
observed; the G value is reduced by 40 to 50 percent while the d value is increased by
only few degrees. Similar trends in changes were observed for the other types of polymers
that were used in the study. Considering the fact that energy dissipation and rate of
relaxation of binders are functions of sind or tand, it appears that effects of these
commonly used polymeric additives on binders at small strains or stresses are mainly
caused by change in rigidity while only secondary effects are caused by changes in elasticity.
It is, however, recognized that type of polymer and concentration can have different effects
(Bouldin et al. 1991; Brule et al. 1986 and 1988; King et al. 1992; Anderson and Lewandowski
1993; Collins and Bouldin 1991; Lesueur et al. 1997; Masson and Lauzier 1993).
Figure 2-7(b) depicts master rheological curves for an asphalt before and after
modification with a crumb rubber modifier (CRM), at 15 percent weight concentration
(Bahia and Davies 1994). The figure is in terms of loading frequency rather than
temperature. As discussed earlier, frequency and temperature are interchangeable; the
effect of high temperature corresponds to that of low frequencies, and vice versa. Changes
in mastercurves are similar to the changes observed for polymer modification shown in
∗
Fig. 2-7(a). G values increase at low frequencies (high temperatures) while they decrease
at intermediate and high frequencies (intermediate and low temperatures). The d values
are lower at low frequencies but higher at high frequencies. The relative changes of either
parameter are of the same order of magnitude as for the polymer modification. The effects
of CRM can, therefore, be also described as mainly changes in rigidity of the asphalt.
The mechanism by which CRM changes properties is, however, different; while for
most polymer modifiers the polymer is completely dispersed in the asphalt and causes
changes in the molecular structure of the asphalt, the CRM is observed to keep its physical
identity and behave as a flexible particulate filler in the asphalt. The overall effect of CRM
∗
on rheological mastercurve is reduction of dependency of G and d on frequency. This
effect is similar in nature to the effect of polymer modification despite the difference in the
nature of material. Polymer modification usually results in a more homogeneous binder
which is more favorable than the nonhomogeneous CRM modification. The trade-off,
however, is the relatively higher cost of the polymer modifiers compared to the CRM.
Various researchers have conducted extensive evaluation of CRM modifications (Oliver
1982; Chehoveits et al. 1982; Bahia and Davies 1994).

Effects of Modification on Failure Properties
Using the direct tension test developed by SHRP, the binders modified with the different
additives were tested at temperatures ranging between −30 and 0°C. The tests were
conducted at a deformation rate of 1.0 mm/min in three replicates and the stress and
strain at failure were calculated. To evaluate the effect of the modifiers, the stress and
strain values at failure of the base and the modified binders are compared. Figure 2-8
depicts the results for the same additives discussed in the pervious section.
Figure 2-8(a) shows strain at failure and stress at failure plots as a function of
temperature for an asphalt before and after modification with 3 and 6 percent of the SB-
based polymer. The strain curves show that the polymer increases the strain at failure
within the brittle and the brittle-ductile zones but converge to the same values as the flow
zone is approached. The effect can be considered as shifting the strain at failure curve
horizontally to lower temperatures without significant changes in shape of curve. The
effect of polymer addition is favorable as it tends to increase the strain at failure within the
critical region. The results shown also indicate that the effect is more favorable with higher
concentration of the polymer. The stresses at failure are similar for all binders which
indicate that the polymer does not result in significant changes in strength of binders.