AHTD's Experience with Superpave Pavement Permeability

A presentation by Jerry R. Westerman , Assistant Materials Engineer, Arkansas State Highway and Transportation Department, for the Arkansas Superpave Symposium, January 21, 1998 in Little Rock, Arkansas.

Arkansas State Highway and Transportation Department
Materials Division

January 21, 1998

Background

Through 1997, the Arkansas State Highway and Transportation Department constructed or let to contract approximately 1,700,000 tons of Superpave 37.5 mm, 25 mm and 12.5 mm mix. All of the Superpave designs have been designed on the coarse side of the maximum density line and below the restricted zone. Initial reaction to the mixes dealt with the coarse texture and appearance; however, the pavements performed well and as anticipated. Limited test data on pavement rutting suggests that Arkansas’s Superpave mixes are less likely to rut than the typical Arkansas Marshall mixes.

After a wetter than normal August 1997, a Superpave project under construction exhibited signs of permeability. Several days after placement of the pavement, a Resident Engineer observed and reported water running from the pavement surface.

Permeability Analysis

A survey of 16 Superpave projects revealed that only two of the sixteen projects showed some signs of permeability. Samples were obtained from each of the 16 projects. Four inch cores were cut at 47 locations for permeability analysis. A falling head permeability apparatus was used to determine the permeability coefficient for each core.

The falling head permeability apparatus was developed by the University of Arkansas as part of the AHTD’s Transportation Research Project No. 82, "Asphalt Mix Permeability". TRC 82 evaluated ten Marshall mixes for permeability. The generated permeability data provides a basis for evaluating the permeability of the Superpave projects. The permeability measurements indicated that 60% of the Marshall mixes had permeability coefficient rates greater than 10^-4 cm/s. Over time, the Marshall mixes began to close, and the surface began to seal. Permeability dropped. The following permeability categories came from TRC No. 82.

For the Superpave permeability analysis, the Research Section and the Materials Division performed permeability and density testing. Correlation of permeability versus lift thickness and permeability versus density was performed. The correlation established the influence that lift thickness and/or density had in predicting pavement permeability. Results are plotted as Figures 1, 2, 3, and 4 respectively. Figure 1 shows the relationship between permeability coefficient and in place air voids. Figure 2 shows the relationship of permeability coefficient and lift thickness. Figure 3 shows the relationship of permeability coefficient versus inplace air voids and lift thickness. Figure 4 shows the lift thickness of the cores used in the analysis.

Figure 1:

Figure 2:

Figure 3:

Figure 4:

Conclusions

The Superpave permeability data and the TRC 82 data support the following conclusions:

0. The permeability coefficient 10^-4 cm/s is selected as the break criteria between high pavement permeability and poor pavement permeability;
1. Using a permeability coefficient of 10-4 cm/s as the base line for permeability, the Superpave mixes have been slightly more permeable (70% of the Superpave mixes had permeability coefficients > 10^-4 cm/s) than most of the Marshall mixes studied in TRC 82;
2. Pavements with density less than 94% had permeability coefficients greater than 10^-4 cm/s, as displayed in Figure 1;
3. For Superpave 12.5 mm mixes, a pavement with lift thickness less than 2 inches generally had permeability coefficients greater than 10^-4 cm/s;
4. The permeability to inplace pavement density correlation had an R value of 0.748;
5. The permeability coefficient (k) of Superpave pavements can be predicted using in place air voids and lift thickness in the following equation:
   

   k = (1.38 x 10^-7)(3.92%^AV)(0.61^{Lift Thickness});

6. Arkansas’s data is similar to the permeability data developed by Florida DOT.

   Recommendations

   Based upon the data from TRC No. 82 and the Superpave permeability analysis, the recommendations are:

   0. Minimum lift thicknesses should be greater than 2 inches or 4 times the maximum nominal aggregate size;
   1. Superpave mixes should be allowed to be free draining to allow water that might be in the pavement to drain;
   2. As new Superpave projects are constructed, permeability testing and loaded wheel testing should be performed as part of a regular testing program.
   3. Develop and use 9.5 mm mixes for projects with ESAL counts;
   4. Continue research efforts to establish acceptable permeability limits for Superpave mixes and to establish proper lift thickness to nominal maximum aggregate size for the various mix sizes.
   
   
   

   ---

   Return to Articles about Superpave
   Return to Superpave Home Page
   

   ---

   Project Manager: Dr. Yetkin Yildirim
   Website Administrator: Clair LaVaye
   © The Superpave Asphalt Research Program,
   The University of Texas at Austin.