Particle Loading Time and Humidity Effects on the Efficiency of an N95 Filtering Facepiece Respirator Model under Constant and Inhalation Cyclic Flows

摘要

It is necessary to investigate the efficiencies of filtering facepiece respirators (FFRs) exposed to ultrafine particles (UFPs) for long periods of time, since the particle loading time may potentially affect the efficiency of FFRs. This article aims to investigate the filtration efficiency for a model of electrostatic N95 FFRs with constant and 'inhalation-only' cyclic flows, in terms of particle loading time effect, using different humidity conditions. Filters were exposed to generated polydisperse NaCl particles. Experiments were performed mimicking an 'inhalation-only' scenario with a cyclic flow of 85 l min(-1) as the minute volume [or 170 l min(-1) as mean inhalation flow (MIF)] and for two constant flows of 85 and 170 l min(-1), under three relative humidity (RH) levels of 10, 50, and 80%. Each test was performed for loading time periods of 6 h and the particle penetration (10-205.4 nm in electrical mobility diameter) was measured once every 2 h. For a 10% RH, the penetration of smaller size particles (< 80 nm), including the most penetrating particle size (MPPS), decreased over time for both constant and cyclic flows. For 50 and 80% RH levels, the changes in penetration were typically observed in an opposite direction with less magnitude. The penetrations at MPPS increased with respect to loading time under constant flow conditions (85 and 170 l min(-1)): it did not substantially increase under cyclic flows. The comparison of the cyclic flow (85 l min(-1) as minute volume) and constant flow equal to the cyclic flow minute volume indicated that, for all conditions the penetration was significantly less for the constant flow than that of cyclic flow. The comparison between the cyclic (170 l min(-1) as MIF) and constant flow equal to cyclic flow MIF indicated that, for the initial stage of loading, the penetrations were almost equal, but they were different for the final stages of the loading time. For a 10% RH, the penetration of a wide range of sizes was observed to be higher with the cyclic flow (170 as MIF) than with the equivalent constant flow (170 l min(-1)). For 50 and 80% RH levels, the penetrations were usually greater with a constant flow (170 l min(-1)) than with a cyclic flow (170 l min(-1) as MIF). It is concluded that, for the tested electrostatic N95 filters, the change in penetration as a function of the loading time does not necessarily take place with the same rate under constant (MIF) and cyclic flow. Moreover, for all tested flow rates, the penetration is not only affected by the loading time but also by the RH level. Lower RH levels (10%) have decreasing penetration rates in terms of loading time, while higher RH levels (50 and 80%) have increasing penetration rates. Also, the loading of the filter is normally accompanied with a shift of MPPS towards larger sizes.