Page 6 - Open-Access-Dec-2019
P. 6
TECHNICAL PAPER
Displacement-controlled uniaxial tensile tests were conducted 3. RESULTS AND DISCUSSIoN
on dog-bone specimens with an Instron 5569 universal testing
machine. The loading rate of the tensile test was 0.2 mm/min. 3.1 Mechanical Properties of Cover Materials
Load-controlled compressive tests were used to determine the Figure 1 shows the uniaxial tensile stress versus strain (for ECC)
compressive strength of cube specimens, where the loading rate or displacement (for FRCs after matrix cracking) curves of the
was 100 kN/min. four mixes. As can be seen, FRCs show post-crack tension
softening behavior with only one single crack while ECC exhibits
Since the purpose of this study was to evaluate and compare tensile strain hardening behavior with the formation of multiple
corrosion of steel reinforcement in different cover material micro-cracks. The strain capacity of ECC prepared in this study
against corrosion-induced expansion, accelerated corrosion is about 1.5%. Moreover, the crack width of FRC continues
was used as a means to induce corrosion expansion in a to widen with displacement while the crack width of ECC is
shorter period. A similar approach has been reported in other self-controlled and maintains at a stable level of about 100 μm
studies [21-27] . The reinforced prism specimens were partially during the strain hardening region. Nevertheless, the four mixes
immersed in a 3.5% sodium chloride solution. A constant have similar first cracking strength. Among the three FRCs, the
potential of 30 volts was imposed on the steel reinforcement toughness increases with increasing fibre dosage. It is expected
which served as the working electrode and connected to the that ECC provides the highest resistance followed by FRC 3,
positive output terminal of the direct-current power supply, FRC 2, and FRC 1 against corrosion-induced expansion. The
whereas the steel mesh near the specimen served as the 28-d compressive strengths of the four mixes are summarized in
auxiliary electrode and attached to the negative output terminal. Table 3. As can be seen, all four mixes have similar compressive
Accelerated corrosion up to 4 days, which induced obvious strength due to comparable mix design of the matrix. This
corrosion damage, was adopted to evaluate the corrosion ensures that the performance difference in corrosion (if any)
behavior of steel reinforcement. of the four mixes is not attributed to the difference in their
compressive strength, rather it is a result of their tensile
After the accelerated corrosion, four-point bending test was properties.
carried out to determine the residual flexural load capacity and
failure mode of reinforced prism. For each accelerated corrosion
duration, two specimens were tested to determine the average F F
residual flexural load capacity. The specimen was orientated so F F
that steel reinforcement was at the bottom surface with a cover F Óx F
F
thickness of 25 mm. The support span length was 240 mm and F
the center span length was 80 mm. The specimens were loaded /iÃi ÃÌÀiÃÃ *>® Óä
under displacement control at a rate of 0.5 mm/min until failure.
The load-deflection curves were recorded, and the ultimate >Ãà Ãà ¯® >Ãà Ãà ¯® £x
flexural load capacity was determined.
(4% £ä
(4%
After the bending test, the steel reinforcement was taken out (4%
from the prism specimen to determine the mass loss of steel '%%
reinforcement. For each accelerated corrosion duration, two
specimens were tested to determine the average mass loss /iÃi ÃÌÀ> ¯® À VÀ>V «i} ® , £ , Ó , Î LiÀ `Ã>}i ¯®
of steel reinforcement. The corroded steel reinforcement Figure 1: Tensile stress-strain (for ECC) or crack opening (for FRCs)
was cleaned with 100 mol/m hydrochloric acid solution to curves.
3
remove rust. The weights of steel bar before casting (m i) and
after corrosion (m f) were recorded to calculate the mass loss 3.2 Crack Pattern of Reinforced Prism
(4%
(Xcorr) and mass loss rate (Ycorr) according to Eqns. (1) and (2), (4% F (4%
respectively. Specimens after Accelerated Corrosion F ® (4%
®
Þ
(4%
(4%
>
Figure 2 shows the typical crack patterns of reinforced prism ® Þ > F
`
À i ` F '%%
'%%
specimens after 4-d accelerated corrosion. As can be seen, À i Þ Ì
«
Xcorr = m i-m f x 100% (1) longitudinal cracks parallel to the reinforcement form due « V >
¯
m i i ¯ « ä°äÈxÝ
Ì
to the expansion of corroded steel reinforcement. For FRC i > V Þ r ÓȰÎni
>
À Ã Ì > À ` > ,Ô r ä°ÎÎx
specimens, crack width reduces with increasing fibre content.
Ã
Ycorr = Xcorr (2) The main crack width in FRC 1 and FRC 2 specimens is larger à à >
t
> Ã Ã Ã À Õ
than 2 mm, while that in the FRC 3 specimen is reduced to be à > Ý i
where t is accelerated corrosion duration in d. less than 2 mm with few micro-cracks on the cover surface. While
12 The IndIan ConCreTe Journal | deCember 2019
ÕÀ>Ì v >VViiÀ>Ìi` VÀÀà `>Þî , £ , Ó , Î >Ãà Ãà v ÃÌii ÀivÀViiÌ ¯®
(4% (4% (4%
(4% (4% (4%
(4% (4% (4%
'%%
'%% '%%
>` ® >` ® >` ®
iviVÌ ® iviVÌ ® iviVÌ ®

