Demands for using electrical resistivity techniques (surface and bulk resistivity) as an alternative to the rapid chloride penetrability test (RCPT) have been growing, for example by a number of transportation agencies in North America, to give an indication of the relative penetrability of concrete. While resistivity measurements may reflect the quality of pore structure in the cementitious matrix, their accuracy might be affected by a multitude of parameters including the concentration of ionic species in the pore solution, particularly when supplementary cementitious materials (SCMs) are incorporated in the binder. Hence, a systematic investigation on the resistivity of concrete and its corresponding physical penetrability is warranted. Zinc sacrificial anodes are considered an effective and economical method to prevent the electrochemical corrosion of steel bars by providing cathodic current to bars, which can provide corrosion protection at low galvanic current densities in the range of 0.2 to 2 mA/m2. Sacrificial anodes are commonly used in RC structures particularly in bridge decks to mitigate a critical phenomenon that occurs in the original concrete beside the repaired patches, which is known as the ‘halo effect’. One of the key factors affecting the efficacy of zinc anodes is the resistivity of concrete or cementitious repair material in which these anodes are embedded. There is a general notion that the higher the electrical resistivity of concrete or repair material, the less likely that zinc anodes produce the target galvanic current for optimum protection of steel bars. However, no systematic data are available on the maximum allowable electrical resistivity of repair materials/concretes beyond which zinc anodes cannot properly function to prevent corrosion. In the first part of this thesis, a tripartite relationship (nomogram) to correlate surface resistivity with penetrability (migration coefficient) and porosity of concrete using a wide range ii of concrete mixtures, taking into account the effect of key mixture design parameters (water-to-binder ratio, air-entrainment, SCMs and type of cement) was established. Relationships between surface and bulk resistivity as well as migration coefficient and porosity of concrete were also introduced. In addition, a penetrability classification of concrete based on the corresponding ranges of surface resistivity, migration coefficient and porosity has been proposed. The nomogram and penetrability classification provided reasonable assessment for the condition of field cores extracted from newly constructed and aging concrete pavement. In the second part of this thesis, the functionality of zinc anodes at mitigating patch accelerated corrosion (halo effect) in repaired concrete was explored. Concrete slabs were cast to simulate the patch repair configuration in the field, and the main parameters in this study were changing the resistivity of the repair section in the slabs (5,000, 15,000, 25,000, 50,000 and 100,000 Ω-cm), and anode spacing (25, 100, and 250 mm) inside the repair patch. Analysis of current and potential data shows a high level of effectiveness of the anodes at controlling corrosion in this slab configuration up to 52 weeks under a wetting-drying exposure. / February 2017

Concrete construction methods and practices influence the final in-place quality of concrete. A low permeability concrete mixture does not alone ensure quality in-place concrete. If the concrete mixture is not transported, placed and cured properly, it may not exhibit the desired durability and mechanical properties. This study investigates the in-place permeation properties of low permeability concrete bridge decks mixtures used in the Commonwealth of Virginia. Permeation properties were assessed in both the laboratory and in the field using 4-point Wenner array electrical resistivity, surface air flow (SAF), and chloride ion penetrability (ASTM C 1202-97). Laboratory test specimens consisted of two concrete slabs having dimensions of 280 x 280 x 102-mm (11 x 11 x 4-in) and twelve 102 x 204-mm (4 x 8-in) cylinders per concrete mixture. Specimens were tested at 7, 28 and 91-days. Thirteen cylinder specimens per concrete mixture underwent standard curing in a saturated limewater bath. The simulated field-curing regimes used wet burlap and plastic sheeting for 3 (3B) and 7 days (7B) respectively and was applied to both slabs and cylinder specimens. Slab specimen were tested on finished surface using the SAF at 28 and 91 days, and 4-point electrical resistivity measurements at 1, 3, 7, 14, 28 and 91 days. Compressive strength (CS) tests were conducted at 7 and 28 days. Chloride ion penetrability tests were performed at 7, 28, and 91 days. Statistical analyses were performed to assess the significance of the relationships for the following: Total charge passed and initial current (ASTM C 1202-97); 3B resistivity and 7B resistivity; Slab and cylinder resistivity; Slab resistivity and ASTM C-1202-97 (Total Charge and Initial current); and Surface Air Flow and ASTM C-1202-97. Field cast specimens, test slabs and cylinders, were cast on-site during concrete bridge deck construction. The slab dimensions were 30.5 x 40.6 x 10.2-cm (12 x 16 x 4 in.), and the cylinders were 10.2 x 20.4-cm (4 x 8-in). In-situ SAF and resistivity measurements were taken on the bridge deck at 14, 42 and 91 days. In-place SAF and resistivity measurements on laboratory field cast slabs were taken at 7, 14 and 28-days. ASTM C 1202-97 specimens were prepared from field cast cylinders and tested at 7 and 28 and 42-days. The relationship between in-place permeation measures from field specimens was compared to laboratory data. Results indicated no difference in chloride ion penetrability (Figures 7.4 and 7.5) and 28-day compressive strength (Figure 7.2) with regard to differing simulated field curing regimes, for same age testing. There was no significant difference at the 95 % confidence level between 3B resistivity and 7B resistivity specimens tested at the same age (Figures 7.9 and 7.10). A well defined relationship was observed between total charge passed and initial current (Figure 7-6). An inverse power function was found to describe the relationship between charge passed/initial current and electrical resistivity for all laboratory mixtures used in this study (Figure 7.17 – 7.22). Field data was used to validate laboratory established models for charge passed/initial current and electrical resistivity. Laboratory established models were able to predict 30 to 50% of the field data (Figures 7.31 – 7.34). Results indicate that the SAF lacked the sensitivity to classify the range of concretes used in this study (Figure 7.24). / Ph. D.