The Effects of Cement Alkalinity upon the Pore Water Alkalinity and the Chloride Threshold Level of Reinforcing Steel in Concrete

Abstract

Cement of three alkalinities (equivalent alkalinities of 0.36, 0.52 and 0.97) was employed in fabricating a set of classical G109 type specimens. To-date, these have been subjected to a one week wet-one week dry cyclic ponding using 15 w/o NaCl solution. At the end of the dry period, potential and macro-cell current were measured to indicate whether the top reinforcing steel was in the passive or active state. Once this bar became active, the specimen was autopsied and the extent of corrosion was documented. Subsequent to visual inspection, concrete powder samples were collected from the upper region of the top rebar trace; and at a certain times concrete cores were taken from non-reinforced specimens. Using these, determinations were made of (1) critical chloride concentration for corrosion initiation (Cl-th), (2) effective chloride diffusion coefficient (De), and (3) pore water alkalinity ([OH-]).The pore water alkalinity was strongly related to the alkali content of cement that was used in the mix. The chloride concentration, [Cl-], was greater at active than at passive sites, presumably as a consequence of electro migration and accumulation of these species at active site subsequent to corrosion initiation. Accordingly, [Cl-] at passive sites was considered indicative of the threshold concentration for corrosion initiation. The Cl-th was increased with increasing Time-to-corrosion (Ti). Consequently, the HA(High Alkalinity) specimens exhibited the highest Cl-th and the NA(Normal Alkalinity) was the least. This range exceeds what has previously been reported in North America. In addition, the effective diffusion coefficient, De, was about 40 percent lower for concrete prepared with the HA cement compared to the NA and LA (Low Alkalinity) ones.