Page 37 - MODELING OF ASPHALT CONCRETE

Modeling of Asphalt Binder Rheology and Its Application to Modified Binders 15

independent of rate of loading or stress level. Asphalts exhibit Newtonian behavior
only at very high temperatures, (above softening point), or at very low shear rates,
which are seldom seen by asphalts in their applications in road pavements. At low
temperatures, or short loading times asphalts are not Newtonian and cannot be
described by an absolute value of coefficient of viscosity. To solve this problem and
make it more relevant to the domain of application conditions, the “apparent viscosity,”
a measure dependent on shear rate, was introduced. Questions then arise regarding
where, in the time, stress and temperature senses, that the measurement should be
taken. The selection of appropriate ranges varied amongst researchers and became a
matter of experimental convenience. Traxler and his coworkers (Traxler and Schweyer
1936; Romberg and Traxler 1947; Traxler 1947) selected a temperature level of 77°F and
a constant power input (constant value of the product of stress times strain rate) was
proposed. For these experimental studies a power input of 1000 ergs was used because
aged and unaged asphalts could be measured at that value of power input, with the
available viscometer without any extrapolation (Romberg and Traxler 1947). Several
researchers used this approach in asphalt aging studies, good examples are given in
Moavenzadeh and Stander (1967), Majidzadeh (1969), and Page and coworkers (1985).
With the introduction of the Shell sliding plate viscometer (Griffin et al. 1955),
−1
apparent viscosity at 77°F and a constant shear rate of 0.05 s was introduced for
experimental convenience and suitability of the device for this measurement. A great
number of research groups followed this approach; apparent viscosity at 77°F
−1
and 0.05s became the most common measure for evaluating asphalt rheology of aged
and unaged asphalts. Early as well as recent studies have used this measure (Heithus and
Johnson 1958; Gallaway 1957; Kemp and Predoehl 1981; Button and Epps 1985).
Despite the wide acceptance of the constant strain rate viscosity measure Mack (1965)
indicated that stress level is equally important as the shear rate, and that apparent viscosity
should not only be compared at a constant temperature and strain rate but also at a
constant stress level. Chipperfield and Welch (1967) suggested that using a constant stress
level is more accurate than constant strain rate. Based on an extensive field study, the
authors indicated that a constant stress apparent viscosity, although not the ultimate
choice, is a much better indicator of asphalt hardening due to aging than the constant
strain rate apparent viscosity. This other approach also had its followers (Schmidt 1972).
The apparent viscosity was one of the measures adopted into the American Society
for Testing and Materials (ASTM ) standards. It is, however, subject to many questions.
The common method of determining apparent viscosity is by incremental creep tests,
where a series of loads are added in sequence and the strain is measured with time. At
each load level the strain is monitored until it shows a constant rate which is selected to
calculate viscosity from the corresponding stress application. Then, the next load is
added and the procedure is repeated to calculate the viscosity at the new shear rate.
With several measured viscosities at different strain rates, the viscosity is calculated by
interpolation (ASTM D 3205). For a non-Newtonian material, such as asphalts at
temperatures below 140°F, the strain rate is a strong function of loading time and
becomes more so as the temperature is reduced. It may take several hours or even days
of loading time so that the strain rate reaches a constant and an asphalt starts behaving
like a truly viscous material. Moreover, the proximity to the steady-state viscosity is
highly asphalt specific.
The other fundamental problem with apparent viscosity is the possibility of reaching
the nonlinear region because of the geometry of the specimen or the stress level being