Page 38 - MODELING OF ASPHALT CONCRETE

16 Cha pte r T w o

used. Keeping the load for long times so as to reach an approximately constant strain
rate, may easily put the material in the material nonlinearity state (high stains), or in the
geometrically nonlinear region which complicates the interpretation of the measurement.
Increasing the load level to very high values, in order to reach a constant strain rate,
may also result in stress nonlinearity.
These fundamental problems with apparent viscosity are not new subjects; the same
researchers who promoted the measure have warned continuously of these problems.
Wood and Miller (1960) indicated that testing with the sliding plate microviscometer
may require several hours before a constant strain rate may be reached. Labout and van
Ort (1956), Griffin and coworkers (1955), Fink and coworkers (1961), and Evans and
Griffin (1963), all suggested using certain geometries for the sliding plates and used
different film thicknesses to limit the strain rates and to show that the measure may be
significantly affected by the geometry or the total amount of deflections are exceeded.
Others have also reported on the effect of stress level and more specifically the stress
history on the apparent viscosity; Puzinauskas (1967, 1979), and Majidzadeh and
Schweyer (1965) are among those who thoroughly evaluated the effect of stress history
on apparent viscosity and reported sufficient evidence of the stress dependence.
Shortfalls of the traditional tests have been recognized for some time, and various
attempts were made to use these tests to estimate fundamental rheological properties.
These attempts started more than 50 years ago and have demonstrated that it is best
that asphalt cements be characterized as linear viscoelastic materials that are thermor-
heologically simple. However, the difficulty of conducting fundamental rheological
testing has led many to simplify the rheology of asphalts, and depend on what is
available in the traditional labs, despite the empiricism involved and the insufficient
characterization that followed. Van der Poel introduced his widely used nomograph in
1954 and indicated that asphalts can be successfully characterized using simple
rheological indices, derived from empirical measures (Van der Poel 1954). The concepts
of using the so-called “temperature susceptibility parameters, and shear susceptibility
parameters” was introduced to model behavior of asphalts.

Asphalt Susceptibility Parameters
To advance modeling of asphalt, rheological derivatives of the traditional tests were
introduced. The motivation was to derive a better relationship with performance that
the traditional tests fail to predict. These estimated, pseudorheological properties, can
be divided into two sets: the temperature susceptibility parameters and the shear
susceptibility parameters.

The Temperature Susceptibility Parameters
The change in consistency with temperature is the general definition of temperature
susceptibility. Several types of such parameters have been proposed and used. They
vary basically in two aspects: the type of consistency measurement used and the range
of temperature covered.
The early works used penetration as the consistency measure. The penetration at
different temperatures was measured and the ratio of penetration, the difference in
penetration, the temperature required to increase the penetration by a certain number, or
the slope of log penetration versus temperature have been used to characterize
temperature susceptibility (Pfieffer and van Doormaal 1936; Van der Poel 1954; Neppe
1952; Barth 1962). More fundamental approaches used viscosity as the consistency