*3.1. Conventional Properties*

The effects of APAO/PPA on conventional characteristics, such as the penetration, softening point, ductility and Fraass breaking point of the modified binder are shown in Figure 2a–d. It is evident from Figure 2a that the penetration of asphalt binder declines with increases in APAO content when the PPA concentration is constant. When APAO content increases from 0 wt.% to 6 wt.%, the reduction in penetration is 25.6%, 14.8%, 13.6% and 11.1% when the PPA content is 0 wt.%, 1.0 wt.%, 1.5 wt.% and 2.0 wt.%, respectively. This indicates that when APAO concentration increases by 6 wt.%, the reduction in penetration decreases with the increase in PPA content. Moreover, the reduced level of penetration through the incorporation of APAO is lower than that from PPA. The compound modification of APAO and PPA leads to a minimum penetration of 32 (0.1 mm), which is a reduction of 58 (0.1 mm) compared with the base binder. Figure 2b displays an ascending trend of softening point with increases in APAO or PPA concentration, indicating that both APAO

and PPA could reinforce the high-temperature performances of asphalt binder. When the PPA dosage is 0 wt.%, 1.0 wt.%, 1.5 wt.% and 2.0 wt.%, a 6 wt.% increase in APAO leads to an 11.4%, 14.7%, 16.1% and 22.5% increase in softening point, respectively. This shows that the increment in the softening point increases with increases in PPA content when the APAO increases by 6 wt.%. Similarly, when the APAO dosage is 0 wt.%, 2 wt.%, 4 wt.% and 6 wt.%, the softening point increases by 39.7%, 40.8%, 49.5% and 53.7%, when the PPA content increases by 2.0 wt.%. It is noteworthy that the increase in softening point caused by PPA is greater than that by APAO.

Figure 2c,d present the variations of ductility at 10 ◦C and Fraass breaking point. Figure 2c shows that APAO significantly decreases the ductility of asphalt binder. When APAO increases by 6 wt.%, the reduction in ductility is 89.3%, 38.8%, 17.6% and 17.4%, when the PPA content is 0 wt.%, 1.0 wt.%, 1.5 wt.% and 2.0 wt.%, respectively. This indicates that when the APAO concentration increases by 6 wt.%, the reduction in ductility decreases with an increase in PPA content. After the incorporation of PPA, the ductility is further reduced on the basis of APAO. The compound modification of APAO and PPA results in a marked reduction in ductility, indicating that the low-temperature performance decreases. The literature [43,44] reveals that the incorporation of APAO is detrimental to the low-temperature properties of a mixture. The Fraass breaking point denotes the turning point of asphalt binder, where it transforms from a viscoelastic state to a brittle state. Figure 2d shows that APAO is harmful to the Fraass breaking point, while PPA is beneficial to the Fraass breaking point. When the PPA dosage is 0 wt.%, 1.0 wt.% and 2.0 wt.%, a 6 wt.% increase in APAO leads to a 15.4%, 20.0% and 22.2% increase in Fraass breaking point, respectively. Meanwhile, a 2.0 wt.% increase in PPA content contributes to a decrease by 38.5%, 30.8% and 27.3% in the Fraass breaking point, when the APAO content is 0 wt.%, 2 wt.% and 6 wt.%. The results show that PPA has a greater effect on the Fraass breaking point. The combined effect of APAO and PPA leads to a decrease in the Fraass breaking point, indicating that APAO/PPA could reduce the turning point of the binder's brittle state. In summary, the compound modification of APAO and PPA could enhance the high-temperature properties but decrease the low-temperature flexibility.

## *3.2. Temperature Sensitivity*

The effects of APAO and PPA on changes in viscosity are shown in Figure 3. Notably, the viscosity increases as the content of APAO and PPA increases. The viscosity increases by 1.8 times when the APAO proportion increases from 0 wt.% to 6 wt.%, and the viscosity improves by 4.4 times as the PPA content increases by 2.0 wt.%. The results demonstrate that PPA has a greater effect on the increase in viscosity than APAO. Viscosity is a vital indicator to reveal the rheological behavior of asphalt binder, whereby SHRP requires that the 135 ◦C viscosity should be less than 3.0 Pa·s to guarantee construction workability. From Figure 3a,b, when the PPA proportion remains at 1.0 wt.% and 1.5 wt.%, and the APAO content reaches 6 wt.%, the viscosity meets the SHRP requirement. However, the viscosity exceeds 3.0 Pa·s when the PPA content increases to 2.0 wt.% and APAO content reaches 4 wt.%. Thus, to ensure the workability of asphalt binder, the APAO content could range from 0 wt.% to 6 wt.% when the PPA content is less than 2.0 wt.%, and the APAO proportion should not exceed 4 wt.% when the PPA content reaches 2.0 wt.%. As a result, two types of modified binders are chosen: one is 1.5 wt.% PPA combined with different contents of APAO, and the other is 2 wt.% APAO mixed with different concentrations of PPA.

**Figure 2.** *Cont.*

**Figure 2.** Conventional properties of all specimens: (**a**) penetration; (**b**) softening point; (**c**) ductility; (**d**) Fraass breaking point.

**Figure 3.** Variations in viscosity with temperature for all specimens: (**a**) composite modified binders with 0 wt.% and 1.0 wt.% PPA combined with 0-6 wt.% APAO; (**b**) composite modified binders with 1.5 wt.% and 2.0 wt.% PPA combined with 0-6 wt.% APAO.

The temperature sensitivity is an important rheological index for asphalt binder, which is usually determined by the viscosity–temperature curve. The asphalt binder's temperature sensitivity determines the characteristics of an asphalt pavement. Low temperature sensitivity of asphalt binder indicates better asphalt pavement performance. Figure 4a,b display the variations in viscosity of composite modified binder at different temperatures. Curve fitting is performed to show the correlation between viscosity and temperature and, from which, the slopes and coefficient of determinations (R2) are obtained, as listed in Table 5. The R2 value is greater than 0.99, which indicates that lg(lg(viscosity)) has a good linear correlation with lg(T). A lower slope value indicates that the asphalt binder is less susceptible to temperature changes. In comparison to neat binder, the modified asphalt binders A0P1.5 and A2P0 have lower slope values. Notably, the slope value of A2P0 is higher than that of A0P1.5, which indicates that the enhancement against temperature sensitivity of PPA is greater than that of APAO. It can be seen that the slope value of the

fitted curve becomes smaller with increases in APAO or PPA content, indicating that APAO and PPA contribute to the reduction in temperature sensitivity. Furthermore, the mixing and compaction temperatures of all samples are determined according to AASHTO T312 viscosity requirements of (0.17 ± 0.02) Pa·s and (0.28 ± 0.02) Pa·s [45]. Table 6 shows the mixing and compaction temperatures for all tested samples. It is clear that the mixing temperature and compaction temperature of composite modified binder increase with increases in APAO and PPA content. The difference between the mixing temperature and the compaction temperature of neat binder is 11.21 ◦C, and the temperature difference increases with increasing content of APAO and PPA. The temperature difference is 13.40 ◦C for A6P1.5 and 13.17 ◦C for A2P2.0. This indicates that a higher temperature is needed to reach the same viscosity, which demonstrates that APAO and PPA reduce the temperature sensitivity of asphalt binder. Therefore, the incorporation of APAO and PPA has the potential to strengthen the asphalt pavement performance.

**Figure 4.** Viscosity−temperature relationship for composite modified binders: (**a**) composite modified binders with various contents of APAO; (**b**) composite modified binders with various contents of PPA.


**Table 5.** The slope and intercept values of fitting curves.


**Table 6.** The mixing and compaction temperatures for all samples.
