The evolution of friction coefficient with time at different
normal loads and rotational speeds was analysed for each sample.
A comparison of transient values of the friction coefficient is
In general, a stationary value of the friction coefficient was
achieved after approximately 5–6 min when using the
biopolymer-based greases as lubricants, whereas commercial
samples needed longer times (8–10 min). The bar diagram shown
in Fig. 6 portrays the steady-state friction coefficient values
obtained by applying normal loads of 10, 20 and 40 N at constant
angular velocity of 10 rpm and temperatures of 25 and 125 °C, for
each grease used as lubricant in the tribo-contact. Samples such as
LI and CH display the same tendency at 25 and 125 °C, i.e. the
friction coefficient decreases with the normal load. This is not the
expected normal force influence in the boundary and mixed
lubrication regimes [54] but it is consistent with the higher
rheological resistance offered by the grease to the rotational
motion at lower normal loads [28]. Moreover, for both samples,
the friction coefficient values are even lower at higher temperature,
in accordance with the lower rheological resistance offered as
a result of the thermally-induced softening of greases. At the same
time, microstructure is better preserved at 125 °C in these two
samples, showing the higher values of the consistency index
(Table 2). As a consequence, it is presumed that at high temperature
the rheological resistance to the motion is reduced but the
entrainment of the whole lubricant into the contact takes place.
On the other hand, greases CA, CP and MC follow the former trend
at 25 °C, but completely the opposite at 125 °C. This effect may be
attributed to the fact that these microstructures are more strongly
affected by shear and temperature, as the lower values of the
consistency index suggest (Table 2), resulting in a significant
reduction of the effective viscosity and, therefore, in the lubricant
film thickness, which favours wear, especially at higher normal
loads, as discussed in the next section. In fact, this temperature is
rather close to the dropping point in sample CA (see Table 1) and
therefore a more significant softening and oil bleeding is expected.
This hypothesis is supported by the fact that a single Stribeck
master curve was obtained for CA and MC samples when using the
base oil viscosity to estimate the Stribeck parameter. Alternatively,
MC and, in lower extent, CP microstructures may release much
more oil at high temperature and normal loads, as observed after
performing the frictional tests, thus locally producing, in some
parts of the tribo-contact, an increase in the effective concentration
of the cellulosic thickening agent, which can interact with the
metallic contact surfaces, causing higher friction and wear.
Fig. 7 shows the values of the stationary friction coefficient by
applying 20 N normal force and rotational speeds of 0.15, 10 and
400 rpm, at both 25 and 125 °C. The friction coefficient obtained
when using the commercial lubricating greases, LI and CA, and
formulation CP as lubricants, at 25 °C, decreases with the rotational
speed, whereas the opposite tendency is observed in the
case of CH at 25 °C. The sample MC at 25 °C displays a particular
behaviour that first exhibits a reduction of the friction coefficient
from 0.15 to 10 rpm, but a notably higher value at 400 rpm. At
125 °C, the tendency followed by almost all the samples, excepting
for CP, reveals an increase in the friction coefficient from 0.15 to
10 rpm and a diminution from 10 to 400 rpm. In the case of the
formulation based on the cellulosic pulp, at 125 °C, a remarkably
high value at 0.15 rpm can be observed; however, from 10 to
400 rpm, the friction coefficient raises with the rotational speed.
These tendencies followed by the stationary friction coefficient in
every case are absolutely consistent with the dynamic frictional
measurements previously discussed, i.e. the sliding velocity
curves, and reflect the different lubrication regimes.