Inaccurate Use of Boron Concentration for Reactivity Feedback

A deficiency has been identified with the use of boron concentration for formulating reactivity feedback while the power of a PWR in response to boron injection in the aftermath of a scram failure was simulated. The US NRC Consolidated Code (TRAC-M) was used to simulate this transient, in which the boron injection was expected to shutdown the reactor. The results of this study, which employed the point-kinetics model of TRAC-M, reveal that the use of core-average boron concentration for formulating the reactivity feedback is inadequate for transients when there is substantial coolant void inside the core. It is recommended that the macroscopic boron density be used for more accurate predictions of the boron feedback as the use of solute concentration is only adequate when no voiding occurs in the core. (authors)