109
International RILEM Conference on Materials, Systems and Structures in Civil Engineering
Conference segment on Service Life of Cement-Based Materials and Structures
22-24 August 2016, Technical University of Denmark, Lyngby, Denmark
every 5 cycles. The damage caused by PSA was assessed by means of mass loss. The
measurements were taken every 10 cycles.
Fundamental transverse, longitudinal, and torsional resonant frequencies of mortar specimens
was measured using forced resonance method according to ASTM C215 [17]. Ultrasonic
pulse velocity was measured in accordance with ASTM C597 [18] using transducers with the
resonant frequency of 54 kHz. Mass loss was measured using scales with the capacity of 3 kg
and precision of 10 mg.
3. Results and discussion
The deteriorated mortar samples after 100 thermal cycles are shown in Figure 1. As can be
recognized in Figure 1, the nature of damage caused by PSA is surface scaling. It can be seen
that tested mixtures exhibited degree of deterioration ranging from very high in mortar with
w/b ratio of 0.50 to minor in the mortars with slag. However, besides the size change, it is
very hard to rate the degree of deterioration by visual appearance. The surface of all samples
was affected by PSA. If significantly larger samples would be used in testing, it would be
hard to quantify visually the difference between the resistance to PSA of these mixtures. For
this reason, the rate of deterioration due to PSA was also assessed using mass loss, dynamic
modulus of elasticity by measured by means of UPV, fundamental transverse frequency and
fundamental longitudinal frequency, and modulus of rigidity determined by fundamental
torsional frequency. The comparison of the mass loss and the durability factors determined by
these methods for mortar with w/b ratio of 0.50 is shown in Figure 2.
Figure 1: Deteriorated samples after 100 thermal cycles
110
International RILEM Conference on Materials, Systems and Structures in Civil Engineering
Conference segment on Service Life of Cement-Based Materials and Structures
22-24 August 2016, Technical University of Denmark, Lyngby, Denmark
Figure 2: Comparison of different methods for evaluation of deterioration rate for mortar with
w/b ratio of 0.50.
Figure 3: Durability factor based on fundamental
transverse frequency
The mortar with w/b ratio of 0.50 was selected to show the differences between various
methods used to assess the rate of PSA, because it had the highest degree of damage, although
the same tendency was observed in all tested mortar mixtures. It can be seen in Figure 2 that
only mass loss reflects the full extent of degradation caused by PSA. It should be noted that
111
International RILEM Conference on Materials, Systems and Structures in Civil Engineering
Conference segment on Service Life of Cement-Based Materials and Structures
22-24 August 2016, Technical University of Denmark, Lyngby, Denmark
durability factor based on a fundamental frequency may be calculate by different ways. For
example, in ASTM C666 [19], durability factor is calculated based on the ratio of the square
of the fundamental frequency after and before the detrimental exposure. However, dynamic
properties depend not only on the fundamental frequency, but on the mass and dimensions of
the tested sample. In case of severe damage, mass loss and dimensional changes may
significantly affect the fundamental frequency. In Figure 3, the durability factor is calculated
based on the fundamental transverse frequency only, with corrections for mass and with full
correction for mass loss and dimensional changes are compared. As can be seen in Figure 3,
the reduction of dynamic transverse modulus of elasticity is about 20%, while durability
factor based solely on the fundamental frequency reduces by more than 50%. Because of high
mass loss, the durability factor with the correction for the mass shows degradation of almost
90%. However, as can be seen in Figures 2 and 3, due to the nature of damage the true elastic
properties of the mortar exposed to PSA are not impaired proportionally to the extent of
deterioration. Figure 3 shows the values calculated based on fundamental transverse
frequency as it is the most widely used method, but the same pattern appears in the other
dynamic properties. In addition, due to significant reduction of the cross-section, the
dimensional ratio of length to maximum transverse direction is shifted away from the
optimum. Thus, it may be concluded that mass loss is the most effective approach to
evaluation of the damage in PSA.
The comparison of mortar mass losses during PSA exposure is given in Figure 4. Here the
difference in PSA resistance between various mixtures can by quantified more easily than by
visual appearance. In addition, it can be recognized that the rate of deterioration during PSA
exposure was not constant. The rate of mass loss of mortars as function of thermal cycles
number is shown in Figure 5. It can be seen that all mixtures demonstrated a sort of induction
period in the beginning of exposure. Most mixtures increased the mass during the first 10
cycles, after which the mass loss started with an increasing rate. After this initial period the
rate of mass loss either stabilized or decreased. There may be two reasons for the decrease of
the rate of mass loss. First, the size of deteriorating specimens decreased, and the surface
available for PSA is reduced as well. Since the nature of the damage causing mass loss is
surface scaling, the rate of mass loss depends on the surface area exposed to PSA. Second, the
specimens during PSA exposure are submerged in sodium sulfate solution. This may promote
hydration, particularly in the mixtures containing SCMs. Most probably, this is the reason for
the reduction of the rate of mass loss in the mixtures with fly ash as the later stages of testing.
Thus, FA mixtures could benefit from a longer period of moist curing. Further investigation is
needed to confirm this hypothesis.
As expected, the reduction of w/b ratio increases the resistance of mortars to PSSA. The
difference in PSSA resistance between w/b ratios of 0.50, 0.45 and 0.40 are significant.
However, there is not much change in PSA resistance between w/b ratio of 0.40 and 0.35. The
possible reasons for this behavior maybe insufficient compaction, since the workability of
mortars decreased with w/b ratio since no plasticizing admixture was used. Comparing the
resistance of mortars with and without SCMs with the same w/b ratio, it can be seen that
GGBFS significantly improved the resistance to PSA, while it was reduced with FA. It is
interesting to note that the content of SCMs didn’t have significant impact on the behavior of
mortars in PSA.