Several different measures of permeation capacity involve the ability of concrete specimens to transmit direct current. Of these methods, the so-called `chloride permeability’ measurement is by far the most widely used, and is generally familiar to concrete researchers and technologists in North America and elsewhere. Standardized as ASTM Method C1202-97,56 the measurement quantifies the total passage of current (in coulombs) during a six-hour period across a 50 mm-thick slice of saturated concrete in contact with NaCl and NaOH electrode solutions; the voltage is maintained throughout the test at 60 volts DC. The range of values for archetypical concretes (as measured at the 28 days specified in the standard) is between ca. 6000 coulombs and ca. 1500 coulombs, depending mostly on w:c ratio. Like water permeability, the measured coulomb values are smaller for concretes of lower w:c ratios, and smaller yet for concretes incorporating supplementary cementing components. Concretes tested periodically as hydration proceeds show progressive reductions in the coulomb values measured. The method is rapid and fairly reproducible, but the high voltage used produces heating effects in the more permeable concretes that lead to com- plications. The procedure was originally developed to provide a means to rank the effectiveness of different treatments designed to prolong the service life of bridge deck concretes against steel corrosion; it was not specifically designed to measure either electrical conductivity or chloride diffusivity per se. More fundamental electrical measurements of permeation capacity in con- cretes have been developed over the last 25 years. Early work by Whittington et al.57 provided a basis for understanding the principles of electrical conduction in concrete. It was shown that electrical conductivity was almost entirely dependent on the conductivity of the cement paste binder and, as expected, was found to be higher for higher w:c ratio concretes; conductivity decreased with time as the binder became more impermeable with continued hydration. Christensen et al.58 later summarized the principles involved in the conduc- tion of current through the cement paste binder phase. `Bulk’ paste conductivity was generally found to be several orders of magnitude smaller than the conductivity of the alkali hydroxide pore solution within it. The governing relationships were expressed as the joint product of three factors: the con- ductivity of the pore solution; the effective volume fraction of current-carrying `capillary pores’ in the paste; and a `ß factor’, which constitutes an inverse measure of the tortuosity of these current-carrying channels. While the conductivity of the expressed pore solution may increase slightly as hydration proceeds, this is outweighed by the reduction in the effective volume of pores through which ions carrying the current can pass, and especially by the increase in the tortuosity of the flow path (reducing `ßThe net result is that electrical conductivity progressively reduces with hydration, paralleling the reduction shown in other methods of measuring permeation capacity. DC conductivity thus can serve as a convenient index of permeation capacity of a concrete, especially if its pore solution conductivity can be established or estimated. These ideas have been recently applied by Nokken and Hooton59 to formulate a procedure for the routine measurement of permeation capacity by measuring DC conductivity, with the goal of providing a basis for concrete durability specifications. Their procedure used the widely available ASTM C1202-97 apparatus and specimen configuration, but details of the measurement differed significantly from those of the ASTM procedure. To avoid heating effects, the voltage was reduced to 15 volts and the measurement time to 15 minutes; and 0.3M NaOH was used to provide electrical contact at both concrete faces. Bulkelectrical conductivity values measured this way for archetypical concretes ranged from roughly 800 mS/cm for permeable concretes to ca.100 mS/cm for impermeable ones. Values for silica fume and fly ash-bearing concretes were lower. Concrete pore solutions were expressed and analyzed by Nokken and Hooton,59 and solution conductivities were calculated for them by the method of Snyder et al.9 It was found that, while the pore solution conductivities were several orders higher than the bulk conductivities, normalizing the latter values (i.e. them dividing by the solution conductivity) provided only slightly better correlation with water permeability and other measures of permeation capacity than did the raw bulk conductivity values.
