Having read the article by V. Ya. Bredikhin and A. A. Zmitrodan "Monitoring interloop seal-tightness of propulsion nuclear power systems with pressurized water coolant" (Atomnaya Énergiya, Vol. 98, No. 3, 170-175 (2005)) I want to call attention to the following.Specialists became interested in tritium as a potential reference radionuclide for monitoring the seal-tightness of equipment in nuclear power plants with VVÉR reactors 25 year ago [1]. Experience in operating nuclear power systems shows that using tritium as a reference is justified only in certain situations. The authors are undoubtedly aware of the method for monitoring the seal-tightness of steam generators on the basis of nonvolatile γ-ray emitting radionuclides by measuring the exposure dose rate in the charge in ionite filters used for cleaning steam condensate [2]. The method makes it possible to monitor the seal-tightness of a steam generator even in a stopped reactor with much more information obtained, much more quickly, and much less expensively [3].The article contains assertions which are not adequately argued. For example, data on the content of tritium in the first-loop coolant are presented as proof of the tritium transparency of the ionite filters in the second loop (see Table 1). Formulas are presented for calculating the total leakage which are similar to formulas which the authors present in another publication. However, the data in Table 2 are different from the data published in a previous work [4]. In a note to Table 2, the authors explain the discrepancy (~10%) between the injection computed according to the technical documentation and the value measured from the tritium activity by the "inaccuracy in determining the flow of first-loop coolant through the metering unit." The most acceptable explanation of this is that the dilution of tritium as a result of natural irreplaceable losses of the medium in the second loop and subsequent makeups was neglected. During the tests, the coefficient of exchange n of the water in the second loop, defined as n = M leak /M 2 , varied in the range 0.03-1. Taking this into account gives better agreement between total leakage according to the technical documentation on the dispenser and the value computed according to the tritium content in the steam condensate:where G is the total leakage of the first-loop coolant into the second loop, kg; A 2 is the specific activity of tritium in the steam condensate, Bq/kg; M 2 is the amount of water in the second loop, kg; n is the coefficient of exchange of the water in the second loop; A 1 is the specific activity of tritium in the first-loop coolant, Bq/kg.The authors do not consider the probability of environmental tritium entering the second loop, since tritium oxide will pass through all water-preparation barriers and enter the second loop with makeup water or during initial filling.It should be noted that the activity of tritium in liquid samples of the medium from the first and second loops of a propulsion nuclear power system is measured by the method...
