Contribution to modelling the behaviour of respiratory protective device cartridges: exposure to complex atmospheres of organic vapours and effect of utilisation cycles

The main aim of this study is to improve technical and scientific knowledge about the life spans of the absorbent beds that equip, in particular, respiratory protective devices (RPDs). Currently, the only data that is available for characterising the life span often results from Standard EN 14387 on anti-gas filters of respiratory protective devices. That standard is based on tests testing efficiency relative to test gases (e.g. cyclohexane). Users therefore only have patchy information: they know the utilisation time of any given RPD cartridge for an effluent loaded with a test pollutant over a range of concentrations, the conditions for the test being laid down by standard. The parameters related to the geometry of the system (mass, pass speed, etc.) and the composition of the test effluent (tested molecule, concentration, relative humidity, etc.) are well defined. The life span under other conditions cannot be certified, or sometimes even estimated. During a preceding study, INRS developed the bases of a predictive computer tool PREMEDIA in order to work on using modelling to extrapolate the results obtained with a test gas towards real pollutants. The results obtained to date make it possible to predict the utilisation time before puncturing for an RPD equipped with cartridges made by a given manufacturer from among the six best-selling brands in France. The current scope of validity is limited to organic products corresponding to cartridges of type A of the RPD classification, for single-exposure, with continuous flow, and for working atmospheres having relative humidity less than 50%.
The objectives of the proposed new study are firstly to finalise modelling the effect of humidity, and secondly to investigate the effects of exposure to mixtures of solvents and the effects of operating cycles (human breathing, re-use, storage, etc.).
The results of this study will make it possible potentially to enrich the predictive computation tool by extending its scope of application.