The delayed hydrogen cracking behavior of cold-worked Zr-2.5Nb at temperatures above about 423 K depends upon the direction of approach to test temperature. Cooling to the test temperatures results in an increase in crack growth rate, da/dt, with increase in temperature, given by the following Arrhenius relationship da/dt = 6.86 × 10−1 exp(−71500/RT) Heating from room temperature to the test temperature results in the same increase in da/dt with temperature, but only up to a certain temperature, TDAT. The temperature, TDAT, increases with the amount of hydride precipitated during cooling to room temperature, prior to heating, and with cooling rate. The results obtained can be explained in terms of the Simpson and Puls model of delayed hydrogen cracking, if the hydride precipitated at the crack tip is initially fully constrained and the matrix hydride loses constraint during heating.

No abstract

Zirconium alloys are susceptible to a mechanism for crack initiation and propagation called delayed hydride cracking. From a review of component failures and experimental results, we have developed the requirements for preventing this cracking. The important parameters for cracking are hydrogen concentration, flaws, and stress; each should be minimized. At the design and construction stages hydrogen pickup has to be controlled, quality assurance needs to be at a high enough level to ensure the absence of flaws, and residual stresses must be eliminated by careful fabrication and heat treatment.

No abstract

Smooth and notched cantilever beams and round-notched bars were machined from pressure tubes of cold-worked Zr-2.5Nb and Zircaloy-2. They were loaded in the temperature range 290 to 520 K. After two thermal cycles and at high stress, cracks were initiated in smooth beams of cold-worked Zr-2.5Nb. Under the same test conditions, cold-worked Zircaloy-2 plastically deformed with no cracking. When notches were present, cracks propagated at the same rate in both materials by delayed hydrogen cracking. In cold-worked Zr-2.5Nb, the crack velocity followed an Arrhenius plot with an apparent activation energy of 42 kJ/mol. Below 420 K, the threshold stress intensity factor for delayed hydrogen cracking was about 5 MPa m. Therefore, cracking can be prevented by keeping tensile stresses very low.