Validation of leak test models for pharmaceutical isolators

Abstract

In this study, we performed three leak tests on a side-recirculated air isolator: a particle-count test, performed at a variety of positive and negative pressures; the SF₆ trace gas method, performed at a variety of airflow velocities; and a pressure decay method, performed at a variety of initial pressures. From the particle-count test, we measured particle leak ratios of 30–34% for pressures of −18 to −25 Pa. In contrast, using the SF₆ tracer gas method. we measured leak ratios of 7.8%, 5.9%, and 3.8% for airflow velocities of 0.61, 0.51, and 0.41 m/s, respectively. The particle-count test detected leaks quickly and easily, but its leak ratio was overly high because it included particles from outside of the chamber as well as from the filter. Although the SF₆ tracer gas method took more time, it was more sensitive and accurate. The leak quantity would be affected by air velocity and the pressure. Leak concentration acted as sterilization phase of isolator that would result in concentration distributed nonhomogeneous phenomena; concentration distributed from outside of the chamber can be acted as a leakage diffusion situation of contaminants leak. Using the pressure decay method to compare downward airflow through a raised-floor and side-wall return air isolators, the fall time, leak quantity, and leak area of the downward airflow through a raised-floor isolator was five times greater, four times les, and 4.5 times less than that of the side-wall return-air isolator, respectively. According to idea gas law, keeping the temperature and pressure of the chamber constant will effectively reduce the leak quantity.