【正文】 ed at the rib roots and advanced towards the area between the ribs. The indentations of the corrosion attack left on the specimen surface after removal of the oxide layer increase in dimensions and depth with increasing duration of the exposure. (84K) Fig. 2. Stereoscopic images (35) of (a) uncorroded specimen and (b) specimen exposed to salt spray corrosion for 10 days. The production of the oxide layer is associated to an appreciable loss of the specimen’s mass. The dependency of the obtained mass loss on the salt spray duration is displayed in Fig. 3. The derived dependency may be fitted by the Weibull function (2) The determined Weibull values C1 to C4 are given in Table 2. As it can be seen for salt spray duration of 90 days the mass loss of the corroded specimen is about 35% of the mass of the uncorroded specimen. It is worth noting that the involved salt spray test is an accelerated corrosion test which is performed at the laboratory. Although the salt spray test environment, to some extent, simulates qualitatively the natural corrosion in coastal environment, it is much more aggressive and causes a very severe corrosion attack in a short time. Currently, there is no direct correlation between the accelerated laboratory salt spray test and the natural corrosion of reinforcing steels such as to assess a realistic duration for the accelerated laboratory salt spray tests. Fig. 4 shows a photograph taken from a building constructed in 1978 at a coastal site in Greece. The corroded reinforcing bars indicated a severe mass loss. The mass loss of the corroded bars shown in Fig. 4 was as high as 18% which corresponds to an exposure of days according to the fitting curve in Fig. 3. The corrosion measured for the mentioned case appeared rather frequently during an extensive investigation on the integrity of older constructions at coastal sites in Greece. Even though the above results are by far not sufficient for establishing exact correlations between laboratory salt spray tests and natural corrosion, they clearly indicate that laboratory salt spray exposures for 40 days and longer are realistic for simulating the natural corrosion damage of steel bars which might accumulate during the service time of reinforced concrete structures at coastal sites. By assuming a uniform production of the oxide layer around the specimen and hence a uniform mass loss, the results of Fig. 3 can be exploited to calculate the reduction of the nominal specimen diameter with increasing duration of the salt spray test. The reduced diameter dr is calculated as (3) where a is the measured mass loss in percent and d is the nominal diameter of the uncorroded specimens (8 mm).The reduced values for the nominal specimen diameter are given in Table 3. The reduction specimen diameter with increasing salt spray exposure time is displayed in Fig. 5. The results in Fig. 5 were fitted using Eq. (2). The Weibull values C1 to C4 for Fig. 5 are given in Table 2. (18K) Fig. 3. Effect of the duration of corrosion exposure on mass loss. Table 2. Weibull values Mass loss Diameter reduction Yield stress reduction Ultimate stress reduction Energy density Elongation to failure C1 C2 C3 C4 (98K) Fig. 4. Photograph taken from building constructed in 1978. Table 3. Values of reduced specimen diameter Exposure to salt spray corrosion environment 0 10 20 30 40 60 90 Diameter (mm) 8 (18K) Fig. 5. Reduction of specimen’s diameter with increasing duration of corrosion exposure. It is essential to notice that the strength calculation of steel reinforced concrete structures according to the standards, . [24], occurs by using an engineering stress estimated by assuming the crosssectional area as (4) with d being the nominal diameter of the bars. For the bars of the present study, the nominal diameter was 8 mm. According to the valid standards, there is no special consideration for the reduction of the nominal diameter of the reinforcing steel, even when evaluating the strength of an older reinforced concrete structure indicating a severe corrosion damage of the reinforcing bars as shown in Fig. 4. Displayed in Fig. 6 and Fig. 7 are the apparent values of the engineering yield stress and ultimate stress over the duration of salt spray exposure by neglecting the reduction of the crosssection of the corroded specimens. In Fig. 6 and Fig. 7, the above values are referred to as Rpapp and Rmapp, respectively. The results have been fitted using the Weibull function of Eq. (2). The Weibull constants C1 to C4 for Fig. 6 and Fig. 7 are given in Table 2. As shown in the figures, the apparent values of Rm and Rp drop below the limits of Rm = 550 MPa for ultimate stress and Rp = 500 MPa for the yield stress, which are set by the standards [21] for involving reinforcing steels, after 32 and 27 days exposure to salt spray, respectively. Yet, the obtained degradation of the apparent strength values of the material reflects not only the effect of corrosion on the mechanical properties of the material but also a stress increase due to the reduction of the specimen’s crosssection. The effect of corrosion on the tensile strength properties of the reinforcing steel can be assessed when removing from the surface of the bars the corrosion products and also using in the calculation of the engineering stress the reduced nominal specimen diameters which are given in Table 3. The measured strength values are given in Table 4 and in the following will be referred to as Rpeff and Rmeff, respectively. For the uncorroded material, the effective strength values have been calculated by using the true crosssectional area: [22] (5) where G is the weight and l is the length of the specimen, whereas for the a