# CONCERNS WITH THE PG GRADING SYSTEM FOR ASHALTS

## Some Structural Problems

The PG grading system has been in place for many years and is used nearly universally throughout the Unites States. There are many positive things that can be derived from this system, however there are certain areas that are unfortunate and reflect the strong effect of the academic in setting the parameters. Following are some items of concern.

Going Against Mathematical Structures in Establishing the Grades.

Functions are set up such that y = f(x) where x is the independent variable and y is the dependent variable. Instead, the PG grades are set up with x = f(y) where x equals temperature and y equals a (phony) modulus. In all previous grading systems the temperature is held constant while the specification limits of a particular grading property is varied. In all previous grading systems only one test was needed to determine the grade. In the PG grading system, several tests are needed to determine the grade of an unknown such as recovered asphalt from RAP. This is fine for academic studies; however for forensic work it increases the testing load thus the cost.

The stiffness modulus from the dynamic shear rheometer at a shear rate of 10 rad/sec is G* = G’ + G” where G* is the complex modulus in kPa. G’, the elastic component is equal to G* x cosδ while G” the viscous component is equal to G* x sinδ. At 10 rad/sec G” is numerically equal to the viscosity, µ, in poises. G’ has two sources. In non-modified asphalt the energy is stored an enthalpy while in elastomer modified asphalts some of the energy is stored as temperature x entropy.

The specification calls out G*/ sinδ, which is a phony stiffness parameter. This means non- polymer modified asphalts, which  have a higher sinδ than modified asphalt at specification temperatures, may be found to be out of specification compliance at a viscosity at which a modified asphalt will be acceptable. The specification is therefore biased toward polymer modified asphalts as a lower viscosity at grading temperatures benefits the supplier in meeting low temperature properties.

Not having 60° C as one of the Grading Temperature.

The PG grading system range is from 46° to 82° C in 6° increments. In certain previous grading systems, 60° C was the temperature used. There is a large amount of data at that temperature in the literature that can be very useful in forensics and research. It appears that the academics purposefully set the temperatures such that past data would be difficult to use. If the mathematics had been used properly, 60° C could have been used for all grades with each grade having its own specification limit.

Using Initial Asphalt Properties for Grading.

When the PG grading system was introduced a point was made that the temperature ranges were set to relate to the temperatures of the area in which it would be used. However the asphalt viscosity in a pavement is not the same as that in the tank prior to mixing with the aggregate. It was because of that that the West Coast had introduced the AR grading system, which was based upon the expected viscosity of the asphalt on the ground. The PG grading system therefore does not represent that which it was claimed to do.

Low Temperature Testing.

On the other hand the long term aging (PAV) and use of flexural creep test (BBR) add very good tools to controlling non-load associated cracking. I think it would be wise to add those tests to chip seal emulsified asphalt specification in cold climates to control chip loss in the winter.

Robert L. Dunning chemistdunning@gmail.com, www.petroleumsciences.com