**4. Discussion**

Both the qualitative results obtained through the measurement of corrosion potential and the quantitative results deduced from the corrosion rates (Figures 3 and 4) show that for chloride ion percentages of 0.4%, the bars are in a passive state but with higher corrosion rates in test specimens with LFS than in the standard test specimens. For test specimens with chloride ion percentages above 0.4%, the bars demonstrate very high states of corrosion both in standard test specimens and in those with LFS, being slightly larger in the former.

The symptoms produced by corrosion show as growing rust stains appearing over time and, therefore, with the increase in current in the standard test specimens, appearing first in the test specimens with 2% chloride ions. The stains began to coincide with the bar through which the external current passed, only to extend gradually along the edge and the lower face of the test specimens, but without cracking 608 days after the start of the experiment and having undergone 283 days of accelerated corrosion.

In the test specimens with LFS, the rust stains began to form later than in the standard test specimens, on the upper face of the test specimen, coinciding with its central bar. In the test specimens with 1.2% chloride ions, 174 days after the commencement of accelerated corrosion, cracks appeared on the lower face and on the edge of the test specimens; these increased in width and length over time until 282 days, when the upper face cracked. In the test specimens with 2% chloride ions by weight of cement, the cracks began 132 days after the commencement of the accelerated corrosion on both the upper and lower faces and on the edge of the test specimens, coinciding with the central bar of the test specimen. The passage of time and the increase in current increased the cracks, with a mesh of cracks appearing in the lower part of the test specimen after 250 days.

Breaking of the test specimens enabled observation of the penetration of corrosive products on the mortar pores in the test specimens with 2% chloride ions; this showed corrosion in the form of pitting produced by the presence of chlorides, beginning with the loss of corrugation in the bar and continuing with the loss of sections in a localized manner (Figure 8).

**Figure 8.** Corrosion products in the test specimens without (**a**) and with LFS (**b**), with 2% chloride ions by weight of cement.

Figure 9 shows the corrosion rates in standard test specimens and those with LFS for different chloride ion percentages within 546 days after the experiment onset. The results show that in test specimens with chloride ion percentages of 0.4%, the corrosion rates were slightly higher in the slag test specimens than in the standard ones. By contrast, for chloride ion percentages higher than the EHE Instruction Limit, the corrosion rates were similar and even higher in standard test specimens.

**Figure 9.** icorr values of test specimens attacked by chlorides, in standard test specimens (MCC) and test specimens with LFS (MCE) after 546 days.
