Issue 49

Y. Liu et alii, Frattura ed Integrità Strutturale, 49 (2019) 714-724; DOI: 10.3221/IGF-ESIS.49.64 717 which was set to 12mm). The carbonization experiments were carried out on the reinforced concrete test blocks with four mix proportions under long-term bending load. The carbonization test was carried out in the standard fast carbonation box in the laboratory. The environmental parameters in the carbonization box include CO 2 concentration of (20 + 3) %, relative humidity of (70 + 5) %, and temperature of (20 + 2) ºC. The concrete blocks were carbonized for three different periods, namely, 7, 14 and 28 days. During the test, the concrete blocks were put into the carbonization box and carbonized to the corresponding period. After that, they were taken out for the measurement of carbonation depth. Specifically, each test block was split along the mid-span direction, and phenolphthalein, 1% w/v in alcohol, was dripped evenly on the block surface with a rubber head dropper. The carbonized part of the block did not change color, while the non-carbonized part became purple red. According to the change of cross-section color, the carbonation depths at 9 points were measured within 60mm in the middle of tension zone and the compression zone, and the average value was taken as the final carbonization depth of the area. To prevent the stress loss of the long-term load device, two preventive measures were taken in this experiment: (1) The pre-tension was carried out before the bending load was applied. First, a torque which was 1.05-1.1 times the corresponding stress was applied to the test block for 2-5min, and then the load was removed and the corresponding torque was applied. (2) During the carbonization process, the applied torque was measured every 7 days and any reduction of torque would be compensated. R ESULTS AND DISCUSSION Effect of the pouring surface on the carbonization resistance of concrete n the experiment, the carbonization depths of each block were measured from four longitudinal sides, and it was found that the carbonization depths were different from different sides, showing the different carbonization resistances of the concrete. Through analysis, it was found that the carbonization resistances of the pouring surface (top surface) and the non-pouring surface (bottom surface) were different. Some literatures [26-30] have found that the segregation of concrete during vibration lead to the unevenness of the block from the top to the bottom, and that the pouring surface of the concrete in actual construction has a certain effect on the carbonization resistance of concrete. To compare the difference in the carbonization performance between the pouring surface and the non-pouring surface, the carbonization influence coefficient of the pouring surface, denoted as K , is defined as follows: p b X K X  (1) where X p is the carbonization depth of the pouring surface (top surface) while X b is the carbonization depth of the non- pouring surface (bottom surface). With the data in Tab. 2, the variation curve of the carbonization influence coefficient K of the pouring surface with different fly ash contents can be determined (as shown in Fig. 3). As shown in Fig. 3, K decreased as the carbonization period increased. In the first 14 days, the decreasing rate was relatively fast, while in the subsequent 14 days, the declining trend became relatively gentle. In addition, the value of K was always greater than 1. This shows that the carbonization speed of the pouring surface was greater than that of the bottom surface under the same mix proportion. As both the top and the bottom were reinforced in the same way, the difference in the carbonization depth might be caused by the different compactness of the top and the bottom surfaces of the concrete. Therefore, the cross sections of the concrete test blocks were selected and the tops and the bottoms of the cross sections were observed, as shown in Fig. 4. As indicated in Fig. 4, the top of the cross section showed less coarse aggregate and denser pores, compared to the bottom of the test block. In addition, there was a thin layer of mortar at the top of the block. Due to the gravity and vibration when the block was poured and vibrated, the cohesion between the cement paste and the aggregate was not enough to resist against the external vertical downward force acting on the aggregate, so the heavy aggregate would sink and the light cement slurry would float. As a result, there was an increase in the coarse aggregate and a decrease in the cement paste from top to bottom along the cross section of the block. The water-to-cement ratio at the top became larger and that at the bottom smaller. There were more capillary pores and larger cracks at the top in comparison with those at I