2.1. Chemical Leaching of Si4+ and Ca2+
Different leaching trends were observed for the various media (Figure 1
). The leaching of Si4+
appears to increase with exposure duration (R
= 0.98 and R
= 0.89, respectively) for H2
at pH 3.0 (M0-1) over 270 days; in other media, less Si was leached out over time. This was also naturally true for Ca2+
and is in accordance with the literature [29
]. Regarding the media-leaching efficiency, the highest quantities of dissolved Si4+
were measured in H2
with a pH of 3.0, whereas the lowest concentrations were observed in fresh water throughout the experiment.
leaching trends of concrete samples with silica fume are illustrated in Figure 2
for various sulfate environments as well as fresh water. Quantities of dissolved Si4+
correspond to 1 g of concrete samples.
The most intensive leaching of Si4+
) during the 270 days of exposure was observed for concrete sample M1-1 exposed to an aggressive environment of H2
with a pH of 3 as observed in Figure 2
a. The concentration of dissolved Si4+
in leachates was the lowest for sample M1-4 immersed in a solution of MgSO4
with a concentration of SO42−
a). Senhadji et al.
] attributed the effect of silica fume on sulfate resistance more to chemical effects than reduced permeability while investigating the resistance of concrete to decomposition in MgSO4
solutions. Zelic et al.
mentioned that a silica fume replacement enhances the durability of mortar exposed to magnesium sulfate attack by lowering the lime content, thereby increasing the initial compressive strength; this occurs due to the pozzolanic reaction [30
The maximum amount of dissolved Ca2+
) was observed in the leachate of sample M1-1 after 270-day experiments (Figure 2
b). The concrete sample M1-5 exposed to fresh water was found to have the lowest values of leached-out Ca2+
during the experiment. Similar to concrete samples without silica fume, different leaching courses have been identified in concrete samples with silica fume. H2
(pH 3) was confirmed to be the most aggressive towards concrete, exhibiting a linear correlation between the leached Si4+
concentrations with R
= 0.83 and 0.92 exposure times, respectively.
The leaching courses of the other media exhibited an increasing trend until 150 or 180 days of the exposure, after which leaching decreased. The lower concentrations of the Si4+
in the leachates at the end of the experiment compared with the maximum at days 150 and 180, respectively, could be explained by the precipitation of newly formed compounds containing Ca2+
on the concrete surfaces as observed using X-ray powder diffraction (XRD). Traces of gypsum and quartz on the surface of concrete samples were also confirmed (Figure 3
The similarity between leaching courses of Si4+ and Ca2+ for both silica fume- and non-silica fume-based concrete samples was confirmed for the media used in the chemical corrosion simulation.
2.3. Comparison of Chemical and Biological Corrosion
shows a comparison of quantities of leached Si4+
due to both chemical and biological corrosion from samples made of two different mixtures after 270-day experiments corresponding to a 1-g concrete sample.
By comparing two different mixtures of cement composites made of ordinary Portland cement without silica fume (M0) and with silica fume (M1), the concrete mixture with silica fume was found to be more durable, in terms of Si4+ leaching, when exposed to aggressive environments of H2SO4 with a pH of 3, both MgSO4 solutions, and a diluted bacterial medium. However, lower durability, after the evaluation of Si4+ leaching, was detected in the H2SO4 with a pH of 4.0, concentrated bacterial medium, and fresh water.
When comparing the chemical and biological corrosion (Figure 5
leaching was more significant when subjected to bacterial exposure, with the exception of H2
with a pH of 3 (M0-1 and M1-1 samples).
As for Ca2+
leaching, the concrete mixture with silica fume was found to be more durable in the aggressive environment of H2
with a pH of 3, concentrated bacterial medium, and fresh water than in the other aggressive environments, as can be observed in Figure 5
b. Based on the leached-out masses of Ca2+
after the experiment, bacterial exposure was found to be the most significant compared with the chemical exposure with the exception of H2
with a pH of 3. However, the Si4+
concentrations at the end of the 270-day experiment likely do not represent the total amounts of dissolved ions. Therefore, the Si4+
leaching rates were calculated by considering the maximum measured amount of Ca2+
) in the leachates. The leaching rate Vd
was calculated by dividing the measured mass of Si4+
in a particular aggressive environment according to the corresponding time of exposure, as shown in Equation (1), based on the work of Ikeda et al.
Vd: Si4+ (or Ca2+) leaching rate per unit area (μg·h−1·cm−2);
Xd: maximum amount of Si4+ (or Ca2+) leached out during the experiment (μg);
T: period of test [=24 × days of leaching (hours)]; and
S: area of exposure surface (cm2).
leaching rates are reported in Table 2
Lower leaching rates have been identified for concrete samples with the addition of silica fume (M1) as opposed to samples without silica fume (M0) in corresponding media (Table 2
). As reported by Lee et al.
, the incorporation of 10% silica fume in ordinary Portland cement matrix showed that the total reduction in strength was greater for mortar specimens without silica fume compared with those with silica fume [26
]. Similarly, Ganjian and Pouya discovered that the performance of pastes and concrete specimens with silica fume exposed to simulation ponds and a site tidal zone were inferior to those without the silica fume replacement [34
]. However, Hekal et al.
reported that a partial replacement of Portland cement by silica fume (10%–15%) did not show a significant improvement in sulfate resistance of hardened cement pastes [35
Higher rates of Si4+ than Ca2+ leaching were detected for all samples subjected to bacterial attack. A comparison of the Si4+ and Ca2+ leaching rates due to H2SO4 (pH 4) attack and a bacterial medium with the same pH of 4 revealed that the bacterial attack was more aggressive.
To confirm the superior performance of silica fume-based concrete in an aggressive sulfate environment, the mass percentage of the dissolved ions (Table 3
) was also calculated. The percentage of dissolved ions was calculated by dividing the maximum quantity of dissolved ions by the total quantity of ions in concrete samples prior to the experiment.
The superior performance of concrete samples based on silica fume in terms of both Si4+ and Ca2+ leachability was confirmed for all concrete samples with the exception of samples immersed in fresh water. The most significant difference was noticed for samples subjected to bacterial attack. The calculated leachable fraction of Si4+ was 5.2- and 3.3-fold higher for samples without silica fume compared with the samples with silica fume after bacterial inoculation. Significantly lower leachable fractions of both Si4+ and Ca2+ ions were also observed for silica fume-based samples exposed to H2SO4 with a pH of 3 (2.9-fold for Si4+ and 2.3-fold for Ca2+).
The results of the mass changes of the analyzed concrete samples prior to and after the experiments are given in Table 4
A decrease in mass was noticed for all concrete specimens after the chemical corrosion experiments, whereas an increase in mass was detected for all samples after the biological corrosion experiments, as can be observed in Table 4
. The increase in mass for all samples under bacterial exposure is likely a result of the formation of massive precipitants on the surfaces of the concrete.
A histogram of mass changes of the concrete specimens with and without the addition of silica fume prior to leaching and after 270 days of leaching is shown in Figure 6
The percentage of mass changes for concrete samples varied from 0.12% (sample M0-3) to 2.42% (sample M1-1). The highest decrease in concrete mass was detected for samples exposed to the most aggressive environment, represented by H2SO4 with a pH of 3, which corresponds with the findings regarding the leaching of Si4+ and Ca2+. Surprisingly, higher mass changes were found for all concrete samples with the addition of silica fume than samples without silica fume. As is known, the deterioration of concrete can be caused by both mechanisms: (i) a dissolution of the cement paste constituents and its subsequent removal from the paste matrix due to its inherently high solubility and (ii) chemical reactions within the paste, e.g., salt crystallization, resulting in concrete volume expansion. The decrease in mass is linked with the leaching process, whereas an increase in mass can be linked with the penetration of sulfate solutions either by simple diffusion or capillary suction, which causes some salts to undergo cycles of dissolution and crystallization. The mass changes after chemical exposure indicate that the leaching process dominates in silica fume-based samples, whereas the crystallization process dominates in concrete samples without silica fume.
The formation of easily visible corrosion-induced cracks on the surface of concrete samples exposed to different aggressive media has been observed (Figure 7
and Figure 8
). No significant changes were identified in concrete samples M0-5 or M1-5 immersed in fresh water with a pH of 7.2.
The surfaces of the concrete samples under chemical exposure contained only traces of precipitates, whereas the surface of the concrete samples under bacterial exposure was nearly completely covered by white crystalline compounds. The concrete samples with and without silica fume exposed to H2
with a pH of 4 (M0-2 and M1-2 samples) and concentrated bacterial medium with a pH of 4 (M0-6 and M1-6 samples) were also analyzed using SEM and EDX, similar to our previous work [36
]. The presence of the new surface products containing SO42−
(gypsum and thaumasite) was observed via SEM (Figure 9
) and EDX (Figure 10
The surface products observed were analyzed by EDX to confirm the presence of Ca2+
compounds (Figure 10
As can be observed in Figure 10
c,d, the presence of Ca, Si, O, and S in the surface compounds was confirmed. Based on the EDX and XRD analyses presented above, the presence of thaumasite (Ca3
O) and gypsum (CaSO4
O) can be assumed to be acting on the concrete surfaces. As reported by Schmidt et al
], despite the fact that thaumasite is thermodynamically favorable and more stable at lower temperatures, it can also be detected at 20 °C at low concentrations after sulfate interaction. Secondary gypsum forms parallel to thaumasite at high concentrations of SO42−