Influence of sodium in calculating the strength of reactor unit elements

Abstract

Basis of the Calculation Method. The effect of the action of liquid sodium on constructional materials may be caused by a variety of elementary processes [i]. However, in calculating the strength of heat-exchanger reactor equipment parts it is normally limited to a study of the influence on the material of only certain of them such as thermal mass transport, the influence of oxygen in the sodium on corrosion, the reduction in stress-rupture strength of metals, the increase in creep rate of metals, decarburization of pearlitic steels, and carbon pick-up of austenitic sneels. In liquid sodium the rate of thermal mass transport is low. However~since heat-exchange equipment is designed for a long service life (10-20 years) a substantial decrease in part thickness as a result of the occurrence of even such a slow process is not impossible. Because of the linear character of thermal mass transport kinetics [i, 2] its influence on the strength of parts may be calculated by calculating the decrease in their thickness by multiplying the corrosion rate by the proposed service time of the part. At present it is impossible to give the corrosion rate of each type of steel and alloy individually both as a result of insufficient study of the thermal mass transport processes of constructional materials and for the reason of a significant spread in corrosion data. Therefore, in Strength calculations it is rational to group them by three types of materials characterized by close values of the thermal mass transport rate -- pearlitic steels, chromiumnickel austenitic steels with a nickel content up to 15%, and nickel base alloys -- and to represent them in the form of a log v vs I/T curve similar to that shown in Fig. 1 for austenitic chromiunr-nickel steels and drawn on the basis of the maximum thermal mass transport rates [3-7 et al.]. The mass transport rate of austenitic chromiunr-nickel steels and alloys containing from 15 to 50% nickel (an element capable of mass transport to a known degree) may be determined from Fig. 1 by multiplying by the coefficient ~ = v,/v (in accordance with Fig. 2), which is the ratio of the corrosion rate of steels and alloys with an increased nickel content to the corrosion rate of 12KhI8NIOT steel. The corrosion data presented refers to pure sodium, i.e., that containing up to i0-3~ 02. The higher values of corrosion rates in sodium containing oxygen may be calculated by using the coefficient K = vo/v (Fig. 3), i.e., the ratio of the corrosion rate of the material in sodium contaminated with oxygen to the rate in pure sodium [3]. If the part is at all times under the action of static stresses and at the same time experiences thermal mass transport the reduction in its life is more correctly determined from the results of laboratory investigations of the stress-rupture strength and creep of samples in liquid sodium. In this case the influence of the active medium is far more than that related to a simple decrease in cross section [i, 8]. However, such tests are complex, and the amount of the corresponding information is small. Therefore, until now the primary method of alloying for the reduction in stress-rupture strength and the acceleration of creep of materials is the introduction of safety factors to the values of stress-rupture strength and creep determined in air. Summarizing of experimental data makes it possible to propose a number of safety factors for the stress-rupture strengths (105 h) and creep (i0-5%/h) (Table I). As experience in the use of sodium loops has shown, one of the most intense corrosion processes is decarburization of pearlitic steels, which is especially intense if austenitic steels are in contact with the sodium at the same time. This process leads to the formation oN the surfade of the parts of a layer with lower short-term and stress-rupture strengths than the original. In view of the fact that in decarburization the carbon content decreases gradually in the direction of the steel surface, in calculating the depth of the decurburized layer it is necessary to select a conditional maximum relative value of the carbon content in it. This may be selected as 0.7, i.e. , the layer in which the carbon content drops