National Institute for Occupational Safety and Health recommends the use of particulate respirators for protection against nanoparticles (<100 nm size). Protection afforded by a filtering facepiece particulate respirator is a function of the filter efficiency and the leakage through the face-to-facepiece seal. The combination of particle penetration through filter media and particle leakage through face seal and any component interfaces is considered as total inward leakage (TIL). Although the mechanisms and extent of nanoparticle penetration through filter media have been well documented, information concerning nanoparticle leakage through face seal is lacking. A previous study in our laboratory measured filter penetration and TIL for specific size particles. The results showed higher filter penetration and TIL for 50 nm size particles, i.e. the most penetrating particle size (MPPS) than for 8 and 400 nm size particles. To better understand the significance of particle penetration through filter media and through face seal leakage, this study was expanded to measure filter penetration at sealed condition and TIL with artificially introduced leaks for 20-800 nm particles at 8-40 l minute volumes for four N95 models of filtering facepiece respirators (FFRs) using a breathing manikin. Results showed that the MPPS was ~45 nm for all four respirator models. Filter penetration for 45 nm size particles was significantly (P < 0.05) higher than the values for 400 nm size particles. A consistent increase in filter penetrations for 45 and 400 nm size particles was obtained with increasing breathing minute volumes. Artificial leakage of test aerosols (mode size ~75 nm) through increasing size holes near the sealing area of FFRs showed higher TIL values for 45 nm size particles at different minute volumes, indicating that the induced leakage allows the test aerosols, regardless of particle size, inside the FFR, while filter penetration determines the TIL for different size particles. TIL values obtained for 45 nm size particles were significantly (P < 0.05) higher than the values obtained for 400 nm size particles for all four models. Models with relatively small filter penetration values showed lower TIL values than the models with higher filter penetrations at smaller leak sizes indicating the dependence of TIL values on filter penetration. When the electrostatic charge was removed, the FFRs showed a shift in the MPPS to ~150 nm with the same test aerosols (mode size ~75 nm) at different hole sizes and breathing minute volumes, confirming the interaction between filter penetration and face seal leakage processes. The shift in the MPPS from 45 to 150 nm for the charge removed filters indicates that mechanical filters may perform better against nanoparticles than electrostatic filters rated for the same filter efficiency. The results suggest that among the different size particles that enter inside the N95 respirators, relatively high concentration of the MPPS particles in the breathing zone of respirators can be expected in workplaces with high concentration of nanoparticles. Overall, the data obtained in the study suggest that good fitting respirators with lower filter penetration values would provide better protection against nanoparticles.