Relation of the coefficient of linear expansion of massive concrete to negative temperatures

Abstract

Hydraulic construction in regions of Eastern Siberia and the North :has brought up new problems regarding the change of the physical and mechanical properties of concrete during its freezing. This problem is extremely timely, since as early as during the construction period the concrete of a structure is subjected to negative temperatures, which have an effect on its thermal-stress state. During the operating period the temperature determines the magnitude of deflections and opening of horizontal joints on the downstream face and thereby influences the behavior of the structure. The standards presently in effect (SNIP II-A 10-62) do not take into account the change of the physical and mechanical properties of concrete at negative temperatures. In particular, the recommended coefficient of linear expansion 1  10 "s dee -I, cannot be extended to concrete exposed to negative-temperature effects. All presently known investigations were carried out on specimens made of mortar of a simulated composition of concrete, whereby the rate of freezing exceeded by 1-2 orders the real rate of temperature changes in hydraulic structures. As investigations [1, 2] showed, nonobservance of one of the conditions characteristic for the temperature and moisture regime of massive concrete (absence of moisture exchange with the external environment, presence of large aggregate fractions, low rate of freezing) makes the resvlts of laboratory and full-scale investigations incomparable. During recent years the B. E. Vedeneev All-Union Scientific-t~Jesearch Institute of Hydraulic Engineering has been analyzing the data of full-scale investigations of temperature deformations in the dams of the Bratsk and Ust'-llimsk hydroelectric stations with simultaneous laboratory tests, the technique of which is sufficiently close to the conditions existing in massive concrete. The results of the fu11-scale and laboratory investigations proved to be completely identical [3, 4]. The average minimum temperature at a distance of 0.5 m from the downstream face of the Bratsk dam (the mixing water is retained at this depth in the concrete) was from -15 to -18"C, having dropped during the severe winter of 1967-68 to -80"C. The minimum temperature of the specimens ranged from --36 to --40"C. The composition of the experimental blocks and laboratory specimens covered all grades of concrete placed in the construction of the Bratsk and Ust'-Ilimsk stations, cement manufactured by the Krasnoyarsk, Irkutsk, and Aehinsk plants (slag-Portland, sand, and pozzolana cements), and various aggregates. Therefore, the results obtained can be extended to any concrete used in hydraulic construction. The only restriction is the petrography of the aggregate, since the concretes of the Bratsk, Ust'-llimsk, and I<rasnoyarsk dams were made on aggregates of extrusive rocks. In all, 122 prisms of 11 concrete compositions were tested, and the data of full-scale observations at 230 measurement points in 65 blocks of the Bratsk and Ust'-Ilimsk dams were analyzed. The relation of the deformations to negative temperatures (Fig. 1) was determined over the course of 5-8 years in the structure and during a period of prolonged freezing and thawing lasting 60-90 days under laboratory conditions. The relation ~ = f(T*) was determined for 100 points, on the average. Generalizing the results obtained from the many years of data from full-scale observations and from specially conducted tests, we can determine the effect of negative temperature on the coefficient of linear expansion of the concrete in the following manner. 1. The coefficient of linear expansion( c0 of massive concrete at positive temperatures ranges from 0.8 to 1.0 x 10 --~. deg -1. on the average 0.9 x 10-s deg-I which is somewhat lower than that adopted in GOST.