【正文】 of the Marshall paction,for characterizing the PFC mixes[21–24]. and scope The main objective of this investigation was to characterize the PFC mixes for three levels of the Marshall paction levels investigated include 35,50,and 75 blows applied on each end of the effect of each paction level was studied on eight different PFC mixes,which corresponded to four different aggregate gradations(G),and two binder contents (BC)of and %by mass of total mix,using the neat bitumen of 85–100 peration selection of gradation,binder type,and binder content were based on the previous studies performed on similar mixes[25,26]. shows the details of the four aggregate gradations 1 provides details of the coding method adopted for designating the mixes for various paction levels and mix positions. 2. Specimen preparation and test plan Straightrun bitumen and crushed stone aggregates are the major constituents of PFC straightrun paving grade bitumen used in the present investigation was supplied by Mangalore Refinery and Petrochemicals Limited(MRPL),Manga granite stone aggregates obtained from local stonecrushing plants were used in this 2 shows some of the physical properties of bitumen and aggregates tested in accordance with the requirements of the ASTM D 946[27]and ASTM D7064[15], Portland Cement(OPC)was used as a part of the mineral filler,constituting 2%by mass of total aggregates weighing 1000 g. The procedure adopted for preparation of the PFC specimens was quite the same as that adopted for dense graded asphalt,as suggested in Asphalt Institute Manual Series2[28].The properties of pacted mixes investigated include bulk specific gravity (Gmb),air voids content(Va),stoneonstone contact condition,per meability(K),moisture susceptibility,and unaged abrasion properties were evaluated in accordance with the guidelines of the ASTM D 7064[15].Table 3 provides details on the tests performed for various experimental observations were made on each experimental mix. 3. Characterization of PFC mixes specific gravity of pacted mix The tests for bulk specific gravity of pacted mixes(Gmb) were performed on 24 mixes as described in Table 3,with three replicates for each Gmb of each pacted mix was determined using the geometric measurements of diameter,height,and the mass of the specimen in air,in accordance with ASTM D 7064[15].The individual and mean Gmb values were found to be in the range of –,and –,respectively,as shown in is evident that an increase in the paction effort will lead to densification of the mix,resulting in higher ,it is evident from the 95%confidence intervals that there is no significant difference between the mean Gmb values among the mixes with binder contents of %and %when pared,for a particular gradation and paction mixes with gradations G1 and G4 exhibited higher and lower Gmb values,respectively,especially since G4 prised of coarser mean Gmb values for the mixes with gradations G2 and G3 are approximately the same,as the minor differences in gradations have not affected the Gmb values significantly. voids The air voids(Va)content in a pacted mix is related to the Gmb and the theoretical maximum density(Gmm)of the unpacted mix,determined in accordance with the ASTM D 2041[30].The increase in the paction effort results in higher Gmb, resulting in a decrease in Va. shows the individual plot of Va for each also indicates the mean Va of the individual mixes tested,and the 95%confidence interval for the line connecting the mean Va shows the trend in variations of the Va with the paction individual and mean Va values were found to be in the range of 10–22%,and 10–20%, per the standard requirements of ASTM D 7064[15],PFC mixes should have a minimum Va of 18%.The mean Va corresponding to the mixes M19(),M20(),and M21()satisfied this requirement,while,the mixes M7(),M8(),M9 (),M13(),and M15()seem to satisfy this requirement at a confidence level of 95%.However,in the mixes tested for 75 blows,the individual,mean,and upper limits of 95%confidence interval of mean Va failed to satisfy the minimum Va mixes corresponding to the gradation G4,.,M19–M24 were found to have mean Va of more than15%.The reduction in the mean Va due to change in the paction level from 50 blows to 75 blows was found to be in the range of –%.Similarly,change in paction level from 50 to 35 blows,resulted in an increase in mean Va in the range of –%. of stoneonstone contact condition The presence of stoneonstone contact condition in the coarse aggregate skeleton is considered to be must,for the best performance of PFC[15].In the pacted PFC mix,stoneonstone contact condition was verified based on the percent voids in coarse aggregate of the pacted mixture(VCAm)and the percent voids in coarse aggregate of the coarse aggregate alone(VCAd)in dryrodded test,conducted according to the ASTM C29/C29 M[31].The stoneonstone contact conditions were confirmed,when the ratios of VCAm/VCAd were below VCAd and VCAm values were puted using the following equations: VCAd=(Gcaγw－ γs)／ (Gcaγw) VCAm=100－ (GmbPcA/GcA) where GCA=bulk specific gravity of the coarse aggregate。 γs =bulk density of the coarse aggregate fraction in the dryrodded condition。and PCA=percent coarse aggregate in the total mixture. Generally,PFC mixes characterized with higher Gmb,higher PCA,or a bination of these,result in lower VCAm,ensuring better stoneonstone contact ratio of VCAm/VCAd for each replicate mix is shown in can be noticed that each replicate of the mixes M7–M24(corresponding to gradations G2,G3,and G4),satisfied the stoneonstone contact condition,even under lowest paction level of 35 ,the mixes corresponding to the gradation,between the bands of G2 and G4,are expected to offer best mixes corresponding to gradation G1 failed to satisfy the might